159 Approaches for Characterizing the Cellular Transcriptional Response to Adenoviral Transduction Molecular Therapy �������� ��� ���� ���������������� �������� ���� ������© ����������� �!����� ����"�[.]
TARGETING AD VECTORS AND HOST RESPONSE 156 Silencing of Coxsackievirus and Adenovirus Receptor (CAR) with Small Hairpin RNA (ShRNA) 157 Prevalence of Anti-Adenovirus and 35 Neutralizing Antibodies in the Gambia, South Africa and United States Adult Population Shaonin Ji,1 Suzanne Oparil,3 Zeng-Bian Zhu,1,2 Yosuke Kawakami,1 David T Curiel.1,2 Division of Human Gene Therapy, Departments of Medicine, Pathology, and Surgery, University of Alabama at Birmingham, Birmingham, AL; 2The Gene Therapy Center, University of Alabama at Birmingham, Birmingham, AL; 3Hypertension Program, Division of Cardiovascular Disease, Department of Medicine, University of Alabama at Birmingham, Birmingham, AL Edward C Nwanegbo,1 Andrea Gambotto,1 Paul D Robbins,2 Wentao Gao,1 Hilton Whittle,3 Eftyhia Vardas,4 Hujie Sun.1 Surgery, University of Pittsburgh, Pittsburgh, PA; 2Molecular Genetics and Biochemistry, University of Pittsburgh, Pittsburgh, PA; 3Virology, Medical Research Council(MRC) Laboratories, Banjul, KMC, Gambia; 4HIV/AIDS Vaccine Division, University of the Witwatersrand, Soweto, Soweto, South Africa While mortality of cancer patients increases dramatically with advanced metastasis, attempts to deliver adenoviral (Ad) vectors via systemic circulation to such lesions are severely hindered by liver sequestration of Ads This is likely due to high levels of CAR expression in liver, as well as the discontinuous endothelial lining in the sinusoidal liver capillaries, which provide a gateway for Ad In contrast, tight endothelial lining in most other tissues is not permissive to viral entry The use of CAR-binding ablated vectors in retargeting may be limited by the inability to bind soluble-CAR containing retargeting conjugates and to disrupt tissue barriers This is because subgroup C adenoviruses (Ad2 and A5) exploit CAR for two important but distinct steps in their life cycle: binding to host cells and escape across the epithelial barrier CAR is localized to tight junctions of both epithelium and endothelium Viral infection in polarized epithelium requires disruption of tight junctions, and disruption of homophilic interaction between CAR molecules in tight junctions by the calcium chelator EDTA renders endothelial cells much more susceptible to Ad infection Our approach to interfere with tight junction formation is to use the small interfering RNA (SiRNA), a short 21 bp RNA duplex with 2-nucleotide 3’ overhang, whose structural is recognized by the RNA-induced silencing complex (RICS) SiRNA guides the related mRNA to RICS for degradation One may introduce SiRNA by transfecting mammalian cells with a plasmid expressing the small hairpin RNA (ShRNA) ShRNA is a single strand RNA with two self-complementing sequences in a head-to-head arrangement separated by a specific sequence constituting a loop in the secondary structure Cleavage of the hairpin loop by the enzyme Dicer produces SiRNA Here we hypothesize that blocking CAR protein expression in liver will lead to liver untargeting of subsequently injected Ads, and will disrupt tight junctions between endothelial cells and facilitate viral entry into other tissues First, we designed DNA templates of several ShRNAs targeting mouse and human CARs (CARShRNA) We cloned the CAR-ShRNA templates into a plasmid vector pSUPER containing the human H1 RNAse promoter, which is appropriate for expressing ShRNA We transfected plasmids expressing CARShRNA (pSUPER-CAR-ShRNA) by electroporations with more than 50% efficiency, as suggested by green fluorescent protein (GFP) expression in control cells FACS analysis of high-CAR C33A cells and medium-CAR Hela cells showed that CAR levels were reduced following transfection with pSUPER-CAR-ShRNA Also, Ad infection was reduced in both low-CAR rhabdomyosarcoma RD-1 cells and C33A cells following transfection Therefore, silencing of CAR expression can be achieved by expressing specific ShRNA from a plasmid vector This approach may be useful for liver untargeting by ablating CAR expression in liver and disruption of the endothelial barrier One of the major limitations to the use of adenoviruses as gene therapy vectors is the existence of preformed immunity in various populations Recent studies have linked morbidity, mortality and failure of adenovirus gene therapy trials to the presence of antiadenoviral neutralizing antibodies (Nab) Understanding the distribution and specificity of the antibodies may assist in the design and success of recombinant adenovirus gene based therapies Here, we have studied the ability of serum samples from three populations to prevent in vitro the infection of A549 cell line by the adenovirus serotypes and 35 Serum samples from adult immuno-competent individuals in The Gambia, South Africa and United States were serially diluted and incubated with adenovirus or 35 encoding the enhanced green or yellow fluorescent proteins respectively and analyzed by flowcytometry Using this technique, we have found, as expected, high prevalence of neutralizing antibodies against the adenovirus serotype in The Gambia, South Africa and American subjects at both low and high titers Conversely, all three populations show low neutralizing antibody prevalence to the adenovirus serotype 35, which, when present, were seen only at low titers 158 Adenoviral Transduction of Mouse Hematopoietic Stem Cells Steven B Bradfute,1,3 Margaret A Goodell.2,3 Immunology; 2Pediatrics; 3Center for Cell and Gene Therapy, Baylor College of Medicine, Houston, TX Hematopoietic stem cells (HSC) are quiescent, self-renewing cells that can give rise to all blood cell lineages Although human HSC can be transduced with high MOI of adenovirus (Ad), Ad transduction of mouse HSC has not been thoroughly studied Here we show that a population of mouse bone marrow highly enriched for HSC (side population, or SP cells) can be transduced with adenovirus serotype (Ad5) at low MOI and short culture times After 6-16 hours of coincubation of bone marrow and 10 MOI Ad5, approximately 32% of mouse SP cells can be transduced with Ad5 Transduced SP cells have normal in vitro myeloid differentiation capacity compared to mock-transduced SP cells Transduced SP cells retain substantial but reduced in vivo long-term repopulating activity and contribute to all blood cell lineages Ad5 transduction of SP cells is dependent upon coxsackie and adenovirus receptor (CAR), as a blocking antiCAR antibody greatly decreases transduction Therefore, we have described a transient gene expression system in mouse HSC utilizing Ad5 159 Approaches for Characterizing the Cellular Transcriptional Response to Adenoviral Transduction Viraj P Mane,1 Gabriele Toietta,1 Brendan Lee.1,2 Molecular and Human Genetics, Baylor College of Medicine, Houston, TX; 2Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX Toxicity associated with systemic administration of adenoviral vectors occurs in three phases: acute, intermediate and chronic Chronic toxicity has largely been abolished with the application of S62 Molecular Therapy Vol 7, No 5, May 2003, Part of Parts Copyright © The American Society of Gene Therapy NAKED DNA: METHODS helper-dependent vectors (HDV) However, the acute toxicity induced by both early generation and helper-dependent adenoviral vectors continues to be a major obstacle for clinical application in humans We are utilizing microarray technology to determine alterations in gene expression patterns that are brought about in vitro and in vivo by infection with both first generation and helperdependent adenoviral vectors Signaling pathways that are activated by Ad infection may contribute to the host innate immune response These same pathways would be targets for pharmacologic manipulations aimed at decreasing acute toxicity and improving the therapeutic index of HDVs We have performed preliminary in vitro analyses using RNA collected 24 hours after infection of Human Umbilical Vein Endothelial Cells (HUVEC) with PBS, First Generation or HelperDependent adenoviral vectors This RNA was reverse transcribed and hybridized to either Affymetrix oligo arrays or NFκB pathway arrays HUVECs represent a relevant target cell type based on substantial endothelial cell transduction following intravenous adenoviral vector administration observed in nonhuman primates, while the NFκB pathway has been implicated in host innate immune response to adenoviral vector by our lab and several others The Affymetrix chips offer a genome-wide analysis of expression pattern alterations, while the NFκB targeted array has been specifically designed to include 112 genes associated with the NFκB signaling network Data from the NFκB pathway array suggests that of the 112 spotted genes, approximately 40 genes undergo at least a two-fold induction in helper-dependent vector-infected HUVECs as compared to mock-infected HUVECs These genes include several TLRs, TRAFs and interferons alpha and beta On first pass, these data support modulation of NFκB signaling in target cells as a potential approach for addressing acute toxicity of Ad vectors Current in vitro experiments are addressing additional cell types including hepatocytes and dendritic cells from both mouse and human NAKED DNA: METHODS 160 Enhanced Systemic Transgene Expression after Non-Viral Salivary Gland Transfection Using a Novel Endonuclease Inhibitor/DNA Formulation Beth M McMahon,1 Edmund J Niedzinski,1 Yen-Ju Chen,1 Judy Dove,1 William Frazier,1 Roland Scollay,1 David C Olson,1 Michael J Bennet.1 Genteric, Inc., Alameda, CA Gene transfer to the major salivary glands is an attractive method for the systemic delivery of therapeutic proteins To date, non-viral gene transfer to these glands has resulted in inadequate systemic protein concentrations We believe that identification of the barriers responsible for this inefficient transfection will enable the development of enhanced non-viral gene transfer in salivary glands and other tissues One potential barrier is the degradation of plasmid DNA by endonucleases To test this hypothesis, we co-administered two endonuclease inhibitors [(zinc and aurintricarboxylic acid (ATA)], with plasmid DNA, containing the secreted alkaline phosphatase gene (SEAP), to the submandibular glands of rats The effect of zinc and ATA on SEAP expression, tissue accumulation of plasmid DNA, and plasmid DNA stability was then characterized We observed that mixtures containing zinc/DNA, ATA/DNA, and zinc/ATA/DNA significantly enhanced both systemic transgene expression and the amount of plasmid DNA associated with treated tissues The relative endonuclease inhibitory activity of zinc, ATA, and zinc/ATA correlated with observed effects on transfection efficacy The use of zinc/ATA enhanced the efficacy of salivary Molecular Therapy Vol 7, No 5, May 2003, Part of Parts Copyright © The American Society of Gene Therapy gland transfection by 1000-fold versus DNA alone Importantly, this improved performance resulted in systemic secretion of protein at concentrations (>0.4 ng/ml) considered therapeutically relevant for several diseases 161 Analysis of Target Site Selection for the Sleeping Beauty Transposon in Mouse Liver Stephen R Yant,1 Brian Garrison,1 Bernie Daigle,1 Mark A Kay.1 Pediatrics and Genetics, Stanford University School of Medicine, Stanford, CA, United States The Sleeping Beauty transposase and retroviral integrases are among a large family of evolutionarily related recombinase proteins that share a common catalytic D,D(35)E domain This highly conserved core motif mediates all of the polynucleotidyl transfer reactions involved in genomic insertion of transposable elements and retroviral DNAs Recently, a large study of retroviral insertion events in cultured human cells presented clear evidence that the HIV-1 integrase protein preferentially targets retroviral cDNA insertion into active genes (Schroder et al, 2002) Based on the structural and functional similarities between these two recombinases, we have initiated a large-scale examination of SB-mediated transposon insertion in adult mice in order to better understand the factors which influence target site selection during DNA transposition To this, we delivered plasmids encoding a kanamycin-resistant transposon and a Sleeping Beauty transposase expression cassette to the livers of C57Bl/6 mice by hydrodynamic-based transfection We then used a bacterial growth-dependent plasmid recovery strategy to isolate a large number of unselected transposon insertion events from mouse liver tissue and then sequenced the DNA flanking these elements in order to characterize each insertion site at the molecular level In all, we have identified 120 independent transposon insertion events that could be mapped to a specific locus in the mouse genome following homology-based searches of the NCBI mouse BLAST database Although SB transposition from host cell chromosomes strongly favors re-insertion into closely linked loci, we find here that the use of an extrachromosomal plasmid-based system exhibits a significantly more random distribution of target site acquisition Indeed, although certain chromosomes appeared to be hit more frequently than others, our data indicate that every chromosome can serve as a target for SB-mediated transposon insertion in vivo Among these 120 events, we observed a strong preference for integration into a palindromic TA consensus repeat, ATATATATA, in which the central TA is the canonical target site Furthermore, based on the observation that 69 of these insertion events (58%) mapped within noncoding regions of predicted transcriptional units, our data also suggest that transposase-mediated DNA insertion exhibits significant bias toward integration into transcriptional units (p