The following abstracts were late abstract submissions that are being presented at the American Society of Gene & Cell Therapy's 15th Annual Meeting in Philadelphia, Pennsylvania. These abstracts are scheduled in one Oral Abstract Session and three Poster Sessions as noted below.
Late Abstracts Poster Session I: Thursday, May 17, 2012
766. Real-Time MRI Tracking of Iron Oxide labeled Human Neural Stem Cells: Pre-Clinical Studies toward First in Human Clinical Trial
Alexander J. Annala1, Margarita Gutova1, Joseph A. Frank2, Kelsey Herrmann1, Michael Barish1, Rex A. Moats3 and Karen S. Aboody1,4
1Department of Neurosciences, 4Division of Neurosurgery, Beckman Research Institute of the City of Hope, Duarte, California, USA; 2National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda, Maryland, USA, 3Division of Radiology, CHLA/Keck School of Medicine, University of Southern California, Los Angeles, California, USA
Human neural stem cells (NSCs) possess intrinsic tumor-tropic properties, making them promising vehicles for targeted delivery of therapeutic genes to many types of cancer. Monitoring the time course, migration and distribution of NSCs following administration to patients would provide critical information for optimization of gene therapy regimens. No clinically compatible cell-tracking system has gained widespread acceptance. A highly promising technique involves pre-loading NSCs with ultrasmall superparamagnetic iron oxide nanoparticles (USPIOs) to enable repeated noninvasive MRI imaging of NSC migration and distribution in vivo. A combination of heparin (H), protamine sulfate (P) and ferumoxytol (F), termed HPF, was used to label HB1.F3.CD NSCs expressing the cytosine deaminase gene. HB1.F3.CD NSCs home to cancer cells and convert the oral prodrug 5-FC to the potent chemotherapeutic drug 5-FU at tumor sites. HPF labeling did not affect cell viability, growth kinetics or tumor tropism in vitro. MRI imaging of NSC tropism in orthotopic glioma mouse models revealed dynamic in vivo NSC tumor tropism at multiple time-points following intracerebral or intravenous injection that correlated with histological analysis. We also evaluated the pre-clinical safety of intracerebrally administered HPF-labeled NSCs in support of potential clinical use. After intracerebral administration of iron nanoparticles delivered by HPF-labeled NSCs in mice, no significant clinical or behavioral changes were observed and no neuronal or systemic toxicities were detected. These studies support the safety of intracerebrally delivered HPF-labeled NSCs in a clinical setting for the purpose of MRI visualization and tracking. Our request for FDA approval to amend our ongoing first-in-human NSC-mediated enzyme/prodrug recurrent glioma clinical trial (IND # NCT01172964) to include USPIO labeling of NSCs is currently pending. If approved, this would be the first-in-human use of MRI imaging of Ferumoxytol for cell tracking in the brain.
767. Targeted Gene Addition to a Predetermined Site in the Human Genome Using a ZFN-Based Nicking Enzyme
Jianbin Wang, Geoffrey Friedman, Yannick Doyon, Nathaniel S Wang, Carrie Jiaxin Li, Jeffrey C Miller, Kevin L Hua, Jenny Jiacheng Yan, Joshua E Babiarz, Philip D Gregory, Michael C Holmes
Sangamo Biosciences Inc., Richmond, CA
Zinc-finger nucleases (ZFNs) drive highly efficient genome editing by generating a site-specific DNA double-strand break (DSB) at a predetermined site in the genome. Subsequent repair of this break via the non-homologous end-joining (NHEJ) or homology-directed repair (HDR) pathways results in targeted gene disruption or gene addition, respectively. Here we report that ZFNs can be engineered to induce a site-specific DNA single-strand break (SSB) or nick. Using the CCR5-specific ZFNs as a model system we show that introduction of a nick at this target site stimulates gene addition using a homologous donor template, but fails to induce significant levels of the small insertions and deletions (indels) characteristic of repair via NHEJ. Gene addition by these CCR5-targeted zinc finger nickases (ZFNickases) occurs in both transformed and primary human cells at efficiencies of up to ~1-8%. Interestingly, ZFNickases targeting the AAVS1 “safe harbor” locus reveal similar in vitro nicking activity, a marked reduction of indels characteristic of NHEJ but stimulated far lower levels of gene addition – suggesting that other, yet to be identified, mediators of nick induced gene targeting exist. Introduction of a site-specific nicks at distinct endogenous loci provide an important tool for the study of DNA repair. Moreover, the potential for a SSB to direct repair pathway choice (i.e. HDR but not NHEJ) may prove advantageous for certain therapeutic applications such as the targeted correction of human disease-causing mutations.
768. Enrichment of melanoma-reactive cells from the fresh tumor digest through selection of CD8 T cells expressing PD-1, LAG-3, Tim-3 and 41BB
Alena Gros, Simon Turcotte, Eric Tran, Takashi Inozume, Ken-ichi Hanada, Qiong Wang, Mark E. Dudley, John R. Wunderlich, James C. Yang, Steven A. Rosenberg
Surgery Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD
Due to the unknown antigen specificities of melanoma-derived TIL, enrichment of tumor-reactive cells from the fresh tumor digest remains a challenge. The identification of a specific phenotype within the fresh tumor digest that would include the tumor-reactive cells and separate them from the non-reactive ones would enable us to enrich the tumor-reactive cells without the need for screening for tumor-reactivity. Several negative co-stimulatory molecules have been proposed to be expressed on tumor-reactive T cells in tumors as a result of chronic antigen stimulation, including PD-1, Tim-3, and LAG-3. 41BB, on the other hand, is a positive co-stimulatory molecule which is up-regulated on CD8 T cells upon specific antigen recognition. The main objective of our work was to characterize the phenotype of TIL in fresh melanoma tumor digests and to assess the usefulness of the markers studied to enrich for tumor-specific cells. In order to do this, we studied the expression of PD-1, Tim-3, LAG-3 and 41BB on CD8 T cells in the fresh melanoma digest, as well as their differentiation stage (CD62L, CCR7, CD45RO, CD27, CD28 and CD57). We found that CD8+ cells in melanoma tumors were enriched in effector memory-like cells (CD62L- CD45RO+) compared to peripheral blood. In addition, CD8 TIL displayed higher frequencies of Tim-3 (15%), PD-1 (13%),LAG-3 (8%) and 41BB (2%) expression compared to peripheral blood. Tim-3, PD-1 and LAG-3 were co-expressed on a small subset of TIL. Furthermore, Mart-1 Tetramer+ cells infiltrating melanoma lesions displayed an effector memory-like phenotype (CD62L-, CCR7-, CD45RO+) and also expressed higher levels of PD-1, LAG-3 and Tim-3 compared to the Tetramer- population, suggesting some of these markers might be useful for enriching tumor reactive cells. Most importantly, we performed functional experiments in which we isolated distinct T cell populations present in the fresh tumor digest, expanded them in vitro and tested the re activity of these populations against their autologous tumor cell lines. Tumor-reactivity was found preferentially in effector cells derived from the cells expressing PD-1+, LAG-3+, Tim-3+ and 41BB+ in the fresh tumor digest but not in the cells derived from PD-1-, LAG-3-, Tim-3- and 41BB-. Positive selection of cells expressing these markers resulted in a considerable enrichment of tumor reactive cells compared to the bulk CD8 T cells expanded from the fresh tumor digest. Our results suggest that tumor-reactive T cells in the fresh melanoma digest express PD-1, LAG-3, Tim-3 and 41BB and thus, these markers can be used to enrich for melanoma-reactive cells.
769. First in Human Clinical Trial of Neural Stem Cell-Mediated Gene Therapy for High-Grade Glioma: Proof of Concept
KS Aboody1,2, TW Synold3, SF Lacey4, B Badie2, X Wu3, J Najbauer1, M Gutova1, M Metz1, M D'Apuzzo5, M Chen2, J Portnow6
1Department of Neurosciences, 2Division of Neurosurgery, 3Clinical and Molecular Pharmacology, 4Clinical Immunobiology Correlative Studies Laboratory, Department of 5Pathology and 6Medical Oncology and Experimental Therapeutics, City of Hope National Medical Center and Beckman Research Institute, Duarte, California, USA
Human NSCs are inherently tumor-tropic, making them an attractive drug delivery vehicle. This first-in-human phase I study is assessing the safety and feasibility of using genetically-modified NSCs for tumor selective enzyme/prodrug therapy (clinicaltrials.gov ID # NCT01172964). An immortalized, clonal NSC line (HB1.F3) was retrovirally-transduced to stably express cytosine deaminase (CD), which converts the prodrug 5-FC to 5-FU, producing chemotherapy locally at sites of tumor in the brain while minimizing systemic toxicities. In this ongoing dose-escalation study, patients with recurrent high-grade gliomas undergo intracranial administration of NSCs during resection or biopsy of tumor. Four days later, 5-FC is administered orally every 6 hours for 7 days. Patients receive only one course of NSCs and 5-FC treatment. Correlative studies include analysis of intracerebral levels of 5-FC and 5-FU (via microdialysis cathether placed at the time of surgery), and correspondin g blood levels. Immunogenicity studies are also being performed on peripheral blood mononuclear cells collected on days 4, 10, 32, and 60. Analysis of serum antibodies using flow cytometry-based NSC-binding assays and CD4/CD8 degranulation assays are conducted to assess for NSC immunogenicity. qPCR is being performed on PBMC DNA to detect vMyc sequences as a marker for NSCs in peripheral circulation, and retroviral envelope sequences as a marker for replication-competent retrovirus (RCR). Accrual to dose levels 1, 2 and 3 has been completed. Thus far, there have been no dose limiting toxicities. Microdialysis data demonstrate the presence of 5-FC and 5-FU in the brain throughout the entire 5-FC dosing interval. Analysis of plasma samples to-date demonstrate average 5-FC levels that are 2.5-5 fold higher than brain levels. Plasma levels of 5-FU were undetectable in all but one subject, and in that patient, plasma 5-FU levels were well below levels in the brain. Anti-NSC antibody and T-cell responses have not been detected. No NSC or RCR have been detected in PBMC. Conclusions: Although accrual is ongoing, dialysate and plasma data from the first 2 patient cohorts demonstrate the proof-of-concept that the NSCs are converting 5-FC to 5-FU locally in the brain. For data analyzed from cohorts 1 and 2, levels of 5-FU in the brain increase in a 5-FC dose-dependent manner. No immune responses to these allogeneic NSCs have occurred after first exposure. (Supported by NCI 1R21 CA137639-01A1).
770. AAV9-Sulfamidase Vector Delivery to the Cerebrospinal Fluid Corrects Brain and Somatic Pathology in MPSIIIA Mice and Results in Detectable Enzyme Levels in Dogs
Virginia Haurigot1,2, Sara Marcó1,2, Albert Ribera1,2, Miquel García1,2, Albert Ruzo1,2, Pilar Villacampa1,2, Eduard Ayuso1,2, Sonia Anyor3, Anna Andaluz3, Mercedes Pineda5, Gemma García Fructoso5, Maria Molas1,2, Lucca Maggioni1,2, Sandra Motas1,2, Jesús Ruberte 1,4, Federico Mingozzi6, Martí Pumarola1,3 and Fatima Bosch1,2
1Center of Animal Biotechnology and Gene Therapy, Departments of 2Biochemistry and Molecular Biology, 3Medicine and Animal Surgery and 4Animal Health and Anatomy, School of Veterinary Medicine, Universitat Autònoma de Barcelona, Bellaterra and 5Hospital Sant Joan de Deu, Barcelona, Spain and 6Children's Hospital of Philadelphia, Philadelphia, PA, USA.
For most lysosomal storage diseases (LSD), particularly those affecting the central nervous system (CNS), there is currently no cure. The blood-brain barrier (BBB) that limits bioavailability of drugs administered systemically, combined with the short half-life of certain lysosomal enzymes, hamper the development of effective therapies for LSDs. Direct injection of gene transfer vectors in the brain parenchyma can potentially overcome these obstacles, but the limited diffusion from the point of injection of both vector and transgene product requires invasive procedures. Mucopolysaccharidosis Type IIIA (MPSIIIA) or Sanfilippo A syndrome is an autosomic recessive LSD presenting with CNS and peripheral disease caused by deficiency in sulfamidase, a sulfatase involved in the stepwise degradation of the glycosaminoglycan (GAG) heparan sulfate. Here we demonstrate that intra-CSF administration of a serotype of AAV vector with broad biodistribution profile, AAV9, encoding for the sulfamidase transgene mediates widespread correction of both CNS and peripheral pathology in a mouse model of MPSIIIA disease. Four months after intracisternal administration of AAV9-Sulfamidase, increased enzyme activity was detected throughout the brain of treated MPSIIIA mice, resulting in correction of GAG storage, lysosomal distention, cellular ultrastructure, and neuroinflammation. In the periphery, high sulfamidase activity was detectable in the liver and in the circulation with whole-body normalization of GAG accumulation and lysosomal pathology. Importantly, treated MPSIIIA mice had normal locomotor activity, showed correction of behavioral deficits, and had extended lifespan with a mean survival similar to that of healthy littermates. Scale up to healthy Beagle dogs was performed by both intracisternal and intracerebroventricular delivery, resulting in transgene expression throughout the CNS and liver, and increased sulfamidase activity detectable in the CSF. The results presented herein support the clinical translation of this approach for the treatment of MPSIIIA and other LSD with CNS involvement.
771. Synergistic Gene Therapy for Leukemia Using AAV/siRNA Hybrid Vectors
Soo Kyung Cho, Young Jik Kwon
Department of Chemical Engineering & Materials Science, Department of Pharmaceutical Sciences, Department of Biomedical Engineering, and Department of Molecular Biology & Biochemistry, University of California, Irvine, Irvine, CA
Despite their superior gene delivery efficiency to nonviral (synthetic) counterparts, use of viral vectors has been hampered by high immunogenicity and limited flexibility for modifications. In contrast, it is facile to molecularly tune nonviral vectors for versatile therapeutic and diagnostic applications. Incorporating synthetic polymers and peptides to viral vectors have frequently resulted in significantly compromised gene delivery efficiency with minimally reduced or newly developed immunogenicity. Longevity and proliferation of cancer cells are often attributed to not only over-expression of a pro-survival gene but also suppressed pro-apoptotic pathways. Therefore, synergistic cancer gene therapy requires simultaneous expression of the pro-apoptotic gene and silencing of the pro-survival gene. In this study, adeno-associated virus (AAV) was engineered with an stimuli-degradable polymeric shell that 1) shields the AAV core from immune response, 2) facilitates the release of the AAV core from the endosome, and 3) co-delivers siRNA along with the AAV core, in order to achieve safe, efficient, and synergistic gene therapy via simultaneous gene expression of a transgene and silencing of abnormal gene. The resulting AAV/siRNA hybrid vectors significantly enhanced transduction of human lymphoma B cells (gene expression), compared with unmodified free AAVs, and simultaneously silenced another target gene. Importantly, AAV-neutralizing antibodies did not affect the gene delivery by the AAV/siRNA hybrid vectors and conjugation of sialic acids to AAV/siRNA hybrid vectors enabled targeted gene delivery to CD22+ B cells. The facilitated intracellular trafficking of the AAV core and efficient cytosolic release of siRNA were confirmed by confocal microscopy. Synergistic gene the rapy for a human BCR-ABL+ leukemia model using AAV/siRNA hybrid vectors with the pro-apoptotic Bim-encoding AAV core and the pro-survival MCL-1 siRNA will also be presented in this talk to demonstrate the high potential of this novel AAV-based gene carrier for clinical translation.
772. Efficient Treatment of Rhabdomyosarcoma-initiating Cells by Survivin-responsive Conditionally Replicating Adenovirus: Promising m-CRA Strategy for Treating Cancer Stem Cells
Kiyonori Tanoue1,2, Yuqing Wang1, Minako Ikeda1, Kaoru Mitsui1, Takao Setoguchi3, Setsuro Komiya3, Shoji Natsugoe2, Ken-ichiro Kosai1
1Department of Gene Therapy and Regenerative Medicine, 2Department of Digestive Surgery, Breast and Thyroid Surgery, 3Department of Orthopaedic Surgery, Kagoshima University Graduate School of Medical and Dental Sciences, Japan
The accumulated data in the recent studies have suggested that a small population of cancer stem cells, also called tumor-initiating cells (TICs), which self-renews and is highly tumorigeic and metasitatic, is the main cause of poor prognosis in cancer patients. A matter of great importance is that TICs are generally resistant to the conventional chemotherapy and radiotherapy, suggesting that the development of a novel therapeutic strategy against TICs is an urgent need. Conditionally replicating adenoviruses (CRAs), selectively replicating in and killing cancer cells, are attractive anticancer agents. We previously developed a novel method to efficiently construct diverse CRAs that have multiple cancer-specific factors (m-CRAs) (Gene Ther 2005), and demonstrated that m-CRA strategy can furthermore increase the cancer-specificity without reduced anticancer effects (Cancer Gene Ther 2011). We also generated survivin-responsive m-CRA (Surv.m-CRA), in which expression of the wil d-type of E1A gene is regulated by the promoter of survivin, a new member of the inhibitor of apoptosis gene family (Cancer Res 2005). Surv.m-CRA efficiently treated most types of cancer and was superior to telomerase-responsive m-CRA both in terms of cancer specificity and efficiency. In this presentation, we demonstrate that m-CRA would be an ideal therapeutic strategy for efficiently treating not only the bulk of heterogenous tumor cells but also TICs. We first identified that fibroblast growth factor receptor 3 (FGFR3) is the marker of rhabdomyosarcoma-initiating cells (RICs) (Br J Cancer 2009); for instance, single FGFR3-positive KYM-1 rhabdomyosarcoma cells formed tumor in vivo whereas FGFR3-negative ones did not. The population of FGFR3-positive KYM-1 cells was increased from 0.7% to 7.4% after changing from the serum-plus to the serum-free medium. Both the expression level of survivin mRNA and the activity of the survivin promoter were significantly higher on the latt er (RICs-enriched) condition than the former one. This tendency was furthermore prominent when they were compared between FGFR3-positive RICs and FGFR3-negative cells that were purified using the cell sorter. Surv.m-CRA efficiently replicated and potently induced cell death to all the populations of KYM-1 cells, and the in vitro cytotoxic effects were more prominent on the serum-free (RICs-enriched) condition or the purified FGFR3-positive RICs than those of each counterpart. An injection of Surv.m-CRA into a tumor nodule, which was established by an inoculation of the purified FGFR3-positive RICs, induced significant tumor death and inhibition of tumor growth. All these results, together with our previous findings, importantly suggest that Surv.m-CRA has therapeutic advantages of not only treating most types of tumor cells in the tumor-specific manner but also rather increasing the potential therapeutic effect against more malignant TICs, which the conventional treatments fa il to treat. We also present other supportive data.
773. Novel method for the identification and visualization of CD8+ T Cell Mediated IFN-γ Signaling in Target Cells during an Antiviral Immune Response in the Brain: Implications for Neurological Gene Therapy
M Puntel1,2, Robert Barrett1, Nicholas S. R. Sanderson1,3, Kurt M. Kroeger1,†, Niyati Bondale1, Sean Kennedy1, Chunyan Liu1, MG Castro1,3 and PR Lowenstein1,3
1Gene Therapeutics Research Institute, Cedars-Sinai Medical Center, and Departments of Medicine, and Molecular and Medical Pharmacology, UCLA, Los Angeles, CA, USA; 2CONICET-Fundacion Instituto Leloir, Av. Patricias Argentinas 435, C1405BWE, CABA, Argentina; 3University of Michigan School of Medicine, Department of Neurosurgery, Department of Cell and Developmental Biology, Ann Arbor, MI, USA. †Diseased
The immune response to viral vectors is an important challenge to gene therapy, as killing of transduced cells could hamper therapeutic benefits. Long term gene transfer in humans and pre-clinical animal models can be limited by preexisting anti-adenoviral (Ad) immunity, as the adaptive immune response reduces transgene expression. In the brain, however, good transduction, even in the presence of anti-adenoviral immunity has been detected reliably. We have previously shown that during an anti-Ad immune response, CD8+ T cells infiltrate the brain, recognize cells expressing adenoviral antigens, become activated, and secrete IFNγ through the formation of specific immunological synapses. As part of this process IFN-γ becomes polarized at intercellular junctions between T cells and adenovirus infected target astrocytes. Currently, there are no methods available to recognize individual target cells that have been exposed to IFNγ secreted by antigen specific T cells, in vivo. Thus, we designed a model to identify those brain cells that become targeted by interferon-gamma during a systemic anti-Ad immune response. This method can identify individual cells that respond to IFNγ secreted by specific anti-adenoviral cytotoxic T cells. To this end, we constructed adenoviral vectors encoding a transcriptional response element that is selectively activated by IFNγ signaling, i.e., the gamma-activated site (GAS) promoter element. In our final vector the GAS element drives expression of a transgene, Cre recombinase (Ad-GAS-Cre). Upon binding of IFNγ to its receptor, the intracellular signaling cascade activates the GAS promoter, which drives expression of the transgene Cre recombinase. In ROSA26 mice bearing a STOP sequence flanked by LoxP sites which is located upstream of a β-galactosidase (β-gal) encoding sequence within the ROSA locus, Cre recombinase activity induces β -gal expression. We demonstrate that upon activation of a systemic immune response against adenovirus, CD8+ T cells infiltrate the brain, interact with target cells, and cause an increase in the number of cells expressing Cre recombinase. This method can be used to identify, study, and eventually determine the long term fate of infected brain cells that are specifically targeted by IFNγ secreted from cytotoxic T cells during an antiadenoviral immune response. This system will allow the characterization of the fate of Ad infected brain cells that are targets for anti-viral CD8+ T cells which infiltrate the brain; and will facilitate novel insights into the cellular and molecular mechanisms underlying brain immune responses and their impact on long term neurological gene therapy using Ad vectors.
774. AAV shRNA Mediated Knock-down of a Liver Target, PCSK9, in Non-human Primates
Heather Zhou1, Ser Mien Chia2, Wen Kang1, Lars Friis Mikkelsen2, Shuo Quan1, Weihua Ni1, Jennifer Shapiro1, Donald Chu1, Li Chun Huang2, Shannie Wong2, Amit Kulkarni1, Daphne Szeto1, Gail Forrest1, Thomas Roddy1, Nicole Boo2, Shian-Jiun Shih2, Andrew Nichols1, Guangping Gao3, Brian Henry2 and Thomas F. Vogt1
1Merck Research Laboratory, USA; 2Merck Research Laboratory, Singapore; 3University of Massachusetts Medical School, Gene Therapy Center, Worchester, MA
Efficient target modulation in a translational animal model is necessary to evaluate candidate target drug targets and to inform decision. Recombinant adeno-associated viral vector serotype 8 (AAV8) is a very attractive means for delivery of cDNAs to the liver in preclinical species including non-human primates (NHP) and most recently in human clinical trial of hemophilia disorders. Our group has achieved significant success in using AAV8 delivery of short hairpin RNAs (shRNA) to knockdown genes of interest in mouse liver. To investigate the applicability of this approach in a large translational animal model we sought to use AAV delivery of shRNA in the nonhuman primate. To assess the feasibility and safety of AAV-RNAi in NHP liver, we selected proprotein convertase subtilisin/kexin type 9 (PCSK9), a well validated LDL cholesterol lowering target, for the proof-of-concept study. PCSK9, a secreted protein mainly from liver, binds and directs low density lipoprotein receptor (LDLR) to lysosomes for degradation, and thus controlling plasma LDL cholesterol level. A number of shRNAs were designed against cynomolgus monkey Pcsk9 and cloned into AAV cis plasmid. Three Pcsk9 shRNA leads were identified through initial screening in cells stably expressing cynomolgus monkey Pcsk9 and selected for AAV8 vector production. The lead shRNA was selected by subsequent ranking in primary cynomolgus hepatocytes and ultimately testing in an in vivo mouse co-transduction system. After rigorous in vitro and in vivo quality-control, self complementary AAV8 expressing Pcsk9shRNA driven by H1 promoter (AAV8-shPcsk9) was administrated into six 2-4 year old male cynomolgus monkeys that were selected for the lack of detectable neutralizing antibody to AAV8 capsid. AAV8- shPcsk9 treatment was in general well tolerated in all monkeys. Transduction of AAV8- shPcsk9 resulted in up to 88% reduction of plasma PCSK9 and up to 62% reduction of LDL-c relative to baseline, starting from day 3, while HDL cholesterol level of animals treated with AAV control vector remains unchanged. The data from the ongoing study revealed that plasma PCSK9 inhibition is maintained at 57% at 8 weeks following single AAV-shRNA dose injection. Our study demonstrates the safety and efficacy of AAV8-shRNA mediated inhibition of plasma PCSK9 and concurrent LDL cholesterol lowering. These results indicate that in addition to successful AAV-cDNA overexpression, AAV-RNAi is a promising approach to achieve long-term silencing of liver target genes in primates.
775. AAV8-mediated Over-expression of Cynomolgus LCAT Raises HDL Cholesterol Levels in Cynomolgus Monkeys
Heather Zhou1, Ser Mien Chia2, Donald Chu1, Wen Kang1, Sarah Tiu2, Ti Ling Chang2, Shuo Quan1, Sabry Hamza2, Lars Friis Mikkelsen2, Hannes Hentze2,Weihua Ni1, Jennifer Shapiro1, Lichun Huang2, Shannie Wong2, Pamela Pereira2, Adeline Tee2, Andrea Peier1, Mihajlo Krsmanovic1, Priscilla Chan2, King Shung Khoo2, Hui Ling Peng2, Jose Castro-Perez, Gail Forrest1, Thomas Roddy, John Thompson1, Nicole Boo2, Shian-Jiun Shih2, Guangping Gao3, Brian Henry2 and Thomas F. Vogt1
1Merck Research Laboratory, USA; 2Merck Research Laboratory, Singapore; 3University of Massachusetts Medical School, Gene Therapy Center, Worchester, MA
Lecithin:cholesterol acyltransferase (LCAT) is a key liver-produced circulating enzyme responsible for high density lipoprotein (HDL) cholesterol esterification, HDL maturation, and potentially reverse cholesterol transport. We have previously demonstrated that AAV8-mediated sustained high level expression of human LCAT in hCETP; Ldlr+/- mice resulted in marked increase in HDL cholesterol level and HDL particle size during a 32-week long-term study. While hCETP; Ldlr+/- mice express CETP, the lipoprotein metabolism of non-human primates more closely resembles that of human. To determine the feasibility of over-expressing cynomolgus LCAT above endogenous physiological levels in normal cynomolgus monkeys and to assess the effect of LCAT over-expression on modulating HDL cholesterol level, we generated an AAV8 vector expressing cynomolgus LCAT (AAV8-cyno LCAT) driven by a liver specific promoter. Prior to transducing cynomolgus monkeys, a proof-of-concept study in mice was conducted and the effectiveness of AAV8-cyno LCAT vector in raising HDL cholesterol was demonstrated. AAV8-cyno LCAT was administrated into 2-3 year old male cynomolgus monkeys (n=5) that were selected for the lack of detectable neutralizing antibody to AAV8 capsid by both in vitro neutralizing antibody assay in Huh7 cells and passive adaptive transfer assay in mice. AAV8-cyno LCAT was well tolerated in all animals, with normal liver enzyme levels and hematology parameters throughout 256 days of the study duration. All animals treated with AAV8-cyno LCAT exhibited an increase of LCAT enzyme activity as determined by cholesterol ester production and LCAT protein expression assayed by western blot analysis. There was a concurrent and significant increase of HDL cholesterol in AAV8-LCAT treated animals. The dosed animals maintained ~30% higher HDL cholesterol relative to its own baseline level and to untreated control animals at approximately 8.5 months after single dose of AAV8-cyno LCAT treatment. The bio-distribution evaluation conducted at the end of the study revealed that the AAV8 vector genome was persistent in liver 8.5 months post single dose treatment. In summary, the data of AAV8-mediated cynomolgus LCAT overexpression and concomitant HDL cholesterol-raising in normal cynomolgus monkeys suggested the safety and efficacy of expressing a target gene above endogenous level to modulate physiological phenotype in non-human primates. These results in the non-human primate provide support for translation of LCAT raising modalities in humans.
776. Epigenetic Control of Mouse Chromosome 12qF1 miRNA Expression
Biswajoy Roy-Chaudhuri, Paul N. Valdmanis, Mark A. Kay
Departments of Pediatrics and Genetics, Stanford University, Stanford, CA
Many imprinted genes show differentially methylated regions (DMRs), which carry methylation on only one parental allele. The Dlk1-Dio3 imprinting cluster on chromosome 12 in mice (12qF1) is extensively methylated in an allelic fashion and contains the intergenic germline-derived DMR (IG-DMR) and the Gtl2-DMR. Imprinting of the Dlk1–Dio3 cluster is also regulated by histone acetylation. Maternal-specific histone acetylation has been reported at the Gtl2 DMR. The Dlk1-Dio3 region also harbors a ~800 kb region expressing ~ 50 miRNAs that are specifically expressed from the maternal chromosome. Interestingly, the maternal allele also encodes a transcript, Rtl1as, expressing six miRNAs that cleave the Rtl1 transcript encoded by the paternal allele in an Argonaute 2 –dependent fashion. Work from our lab has shown that the cluster of ~50 miRNAs and the Rtl1as cluster of miRNAs are specifically upregulated in a developmental fashion in mouse. Here, we report the effects of DNA methy lation and histone acetylation on the expression of these miRNAs using known inhibitors of DNA methyltransferases and histone deacetylases. Since these miRNAs have downstream targets involved in oncogenesis and development, our work might provide insightful knowledge on the use of currently available inhibitors of DNA methyltransferases and histone deacetylases as therapeutic agents for cancer.
777. The 4.5Å cryo-EM Structure of AAV-DJ: An Engineered Gene Therapy Vector
Thomas F. Lerch1, Qing Xie1, Jason K. O'Donnell2, Nancy L. Meyer1, Kenneth A. Taylor2,3, Scott M. Stagg2,4 and Michael S. Chapman1
1Department of Biochemistry & Molecular Biology, Oregon Health & Science University, Portland, OR, 2Institute of Molecular Biophysics, 3Department of Biological Science, and 4Department of Chemistry & Biochemistry, Florida State University, Tallahassee, FL
Adeno-associated virus (AAV) is a leading candidate vector for gene therapy. Therapeutic gene delivery using AAV vectors is hindered by broad tissue tropism, low transduction efficiency, and a prevalent immunity to the natural virus. Recently, a capsid library created by random shuffling of several AAV serotypes was put under selective pressure in hepatocytes using pooled human polyclonal sera. A promising chimeric vector, AAV-DJ, emerged from this screen and has greatly enhanced liver specificity, while avoiding recognition by neutralizing antibodies. Our previous high-resolution crystal structures of several AAV serotypes have provided insights into the functions of AAV capsids, and serve as templates for the rational design of vectors with altered specificity. Crystallographic approaches, however, require large quantities of AAV and significant efforts to optimize crystal quality for high-resolution studies. In the current study, we overcome these limitations by usin g cryo-electron microscopy to determine the structure of the AAV-DJ capsid at 4.5Å resolution. The structure reveals insights into the novel properties of the vector, provides a foundation for further characterization and development of next-generation vectors, and provides mechanistic insights into the process of directed evolution.
778. Design of Hybrid Retroviral/Synthetic Gene Delivery Vectors
Rahul K. Keswani, Daniel W. Pack
Department of Chemical and Biomolecular Engineering; University of Illinois, Urbana-Champaign, Urbana, IL, USA
Gene therapy has the potential to revolutionize healthcare for millions of people. However, it has yet to become a common treatment for the variety of diseases that could benefit from the delivery of therapeutic genes. Limited progress is primarily due to the lack of a safe and efficient means of delivering genetic material. Viral vectors, for instance, are extremely efficient but potentially pathogenic, immunogenic and less amenable towards modified tropism. Non-viral vectors are non-pathogenic and non-immunogenic, yet lack high efficiency and are typically toxic at clinically useful concentrations. It is clear that the current implementation of gene therapy must be preceded by the development of vectors with improved characteristics. Murine Leukemia Viruses (MLVs) were one of the earliest class of retroviral vectors that were investigated for human gene therapy. They are RNA viruses capable of stable integration of the foreign gene into the target cell genome. We have developed a new class of hybrid vectors by developing synthetic polymer-based - chitosan (χ) and lipid-based (φ) envelopes for Moloney Murine Leukemia Virus-like particles (M-VLPs). M-VLPs are essentially intact viruses lacking the envelope protein most necessary for transfection, thus making them inactive. Both chitosan and liposomes composed of DOTAP, DOPE and cholesterol electrostatically associated with M-VLPs forming hybrid vectors (χ/M-VLPs and φ/M-VLPs). The transfection efficiency of the hybrid vectors was not only better than our earlier hybrid vectors (PEI/M-VLPs and PLL/M-VLPs) but also of the same order of magnitude as amphotropic MLVs (MLV-A). The size and transfection efficiency of χ /M-VLPs was dependent on certain design parameters such as the chitosan pH, pKa of acidic solvent, chitosan stock concentration and M-VLP particle density. High transfections and low sizes could be achieved via using chitosan dissolved at 0.5 mg/ml at 0.6% glutamic acid, pH 3 for conjugating with M-VLPs (>10 x 10^9 M-VLPs/ml). The toxicity of chitosan was negligible compared to using polymers such as PEI and PLL. Chitosan in low pKa acids such as HCl were unable to mediate any transfections with M-VLPs. Transfection efficiency of φ/M-VLPs was dependent on the lipid composition used for the synthetic lipid envelope. Liposomes with low DOTAP, low DOPE and high cholesterol content were able to mediate better transfections with M-VLPs. Cellular uptake of φ/M-VLPs was dependent on DOTAP content with high DOTAP mediating higher cell entry although successful transfections were not dependent on total uptake levels. The lipid toxicity was also dependent on the composition of the lipid envelope with higher cholesterol content leading to high toxicity. Significant size reduction of the φ/M-VLPs was achieved via liposomal extrusion prior to association with M-VLPs (~ 300-400 nm). φ/M-VLPs were able to provide stable transgene expression over a period of three weeks but exhibited poor serum stability. This simple hybrid vector design allows for easier re-targeting for cell-specific gene delivery such as folate-targeting for cancer cells along with addition of components into the synthetic envelope that could provide non-native functions into retroviral vectors.
779. Protein Mediated Transport of Biologically Active RNA to the Extracellular Space Provides Inhibition of Cell Signaling Events
Kevin J. Polach1, Leslie Wilkinson1, Elaine Brunhoeber1, Amy Pettigrew1, Kirby Wallace1, Jeff Sparks1, Majed Matar1, Jennifer Rice1, Khursheed Anwer1 and Jason G. Fewell1
1EGEN, Inc., Huntsville, AL
The functional versatility of small RNAs provides great potential for modulating the behavior of biological systems with much effort being directed at the development of RNA based therapeutics. The direct application of synthetically derived RNA reagents to target cells, either alone or in complex with various carriers, has been effective at modifying biological processes in vitro and in vivo. In vivo, RNAi applications are normally transient and typically involve a bolus administration of reagent that is followed by a period of biological activity associated with the RNA effector molecule. In addition the activity of RNAi is limited to the cells that are transfected. Traditional delivery approaches suffer from inefficient delivery and require high doses of reagent with associated toxicities. To overcome these limitations we have developed a system for secreting biologically active RNA from mammalian cells. This approach utilizes a specialized fusion protein that joins together proteins secreted through non-classical secretion pathways with a high affinity RNA binding domain derived from the bacteriophage lambda Protein N (PN). The RNA molecule intended for secretion (secRNA) is constructed by linking a biologically active aptamer to the Box B recognition stem loop (RSL), which is bound by the PN RNA binding domain. These expression cassettes have been cloned independently into a constitutively active plasmid vector under the control of a Simian Virus 40 (SV40) promoter. In vitro, we show secretion of RNA from HeLa cells that is dependent on the fusion protein and is not associated with cell lysis. In model systems, secreted aptamers were shown to act either in autocrine or paracrine type assays; the use of a secreted HER 3 aptamer resulted in ~40% inhibition of MCF7 cell growth compared to inactive aptamer control vectors. To our knowledge this is the first example of an engineered RNA secretion system for mammalian cells, a system that may offer novel approaches to overcoming RNA delivery challenges unattainable with the current delivery systems.
780. Genotoxicity and Integration Pattern of Foamy Virus Vectors with Strong Internal Viral Enhancers in In Vitro Immortalized Mutants from Primary Murine Hematopoietic Progenitor Cells
P Arumugam1, D Lynn1, A Loberg1, H Van der Loo1, P Dexheimer3, M Keddache2, TR Bauer4, D Hickstein4, DW Russell5, P Malik1
1 Exp. Hematology and Cancer Biology, 2Human Genetics, 3Biomedical Informatics, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, 4Exp. Transplantation and Immunology, NCI, NIH, Bethesda, MD, 5Div. of Hematology, University of Washington, Seattle, WA.
Clinical trials for X-linked chronic granulomatous disease using the spleen focus-forming virus (SFFV) based gammaretroviral vector (RV) showed therapeutic benefits, but resulted in insertional mutagenesis via MDS-Evi1 gene upregulation. The same gene enhancement results in immortalization of one in 50 primary murine hematopoietic progenitor cells in a well-established in vitro immortalization (IVIM) assay. The SFFV RV and LV are capable of immortalizing murine hematopoietic stem/ progenitor cells (HS/PCs) by insertional mutagenesis with >80% of insertions in the proto-oncogene Evi1. We and others have shown a 10-20 fold lower immortalization potential of the SFFV enhancer in lentivirus vectors (LV) compared to analogous RV. Foamy virus vectors (FV) have several advantages: it is non-pathogenic and carries a higher vector payload. Herein, we compared the genotoxicity of FV vectors carrying SFFV enhancer/promoter to analogous RV and LV using IVIM assay of primary HS/PCs. Insertional mutagenesis mediated immortalization of clones was detected at 2 weeks, and their replating potential and clonal fitness was reassessed at 5 weeks. SFFV FV and SFFV LV showed 200-fold and 12-fold lower frequency in generating IVIM clones respectively, compared to SFFV RV. MSCV RV showed 3 fold lower immortalization potential compared to SFFV RV. Comparison of the MSCV enhancer/promoter in LV and FV showed 5 fold and ~150 fold lower replating frequency respectively, compared to MSCV RV. Insertion sites in the SFFV and MSCV FV immortalized clones were determined using ligation amplification mediated polymerase chain reaction followed by next generation sequencing and compared them to reported insertional sites using the SFFV RV and LV. Of the FV integrations, most were in repeat sequences (69%); 7 out of 10 clones showed insertions in a non-genic region downstream of Meis1 locus suggesting, a single clone with insertion near Meis1 could have predominated the in vitro culture during expansion, as has been seen with SFFV RV where MDS-Evi1 insertions predominate. In MSCV FV vectors, the predominant insertions were in repeat regions; 5/10 IVIM clones showed insertion in the same gene locus which codes for Sialyltransferase 1 gene (Siat1) partial sequence, shown to be upregulated in squamous cell carcinomas. In contrast, MSCV LV showed >60% of insertions within genes. Using IVIM assay, we did not observe any insertions in or near MDS-Evi1 locus with SFFV FV, unlike that reported with SFFV RV and LV. No MDS Evi1 insertions were seen with MSCV FV vectors. Our data suggests that the genotoxic potential of FV vectors is nearly 150-200-fold lower compared to the RV or LV vectors even with strong enhancer elements, due to the propensity to integrate into repeat sequences. Vector insertions in the vicinity of proto-oncogenes do occur, are near transcriptional start sites as seen with RV, and lead to immortalization of primary progenitors. These integration patterns provide critical insights into the genotoxicity of FV with strong viral enhancers and suggest that FV backbone may be safer for gene transfer applications.
781. In utero interference RNA (RNAi)-based silencing of endogenous thrombopoietin receptor (cMpl) results in hematopoietic deficiencies
Emily Partridge, Jesse Vrecenak, Alan W Flake
Children's Hospital of Philadelphia, The Center for Fetal Research, Philadelphia, PA
Introduction: Thrombopoietin (TPO) is a key regulator of hematopoiesis, and its effects are mediated through the binding and intracellular signaling of the thrombopoietin receptor (cMpl), a member of the cytokine receptor superfamily. Studies of knockout mice lacking expression of TPO or cMpl revealed impaired megakaryocyte and platelet production as well as significant deficiencies in hematopoietic stem cells (HSC), with reduced colony forming units in vitro and a reduction of the repopulating capacity of bone marrow cells in vivo. In utero hematopoietic stem cell transplantation is a promising approach for the treatment of a number of congenital hematological disorders. One obstacle to achieving robust levels of engraftment following transplantation is the high level of occupancy of the microenvironmental niche by large numbers of rapidly cycling HSCs. We hypothesize that transient depletion of cMpl by the application of non-virally packaged RNAi may provide a myeloablative approach to permit enhanced levels of donor HSC engraftment following in utero transplantation. Methods: Two Ambion™ in vivo interference RNA sequences targeting the cMpl transcript were prepared in Invivofectamine 2.0™ reagent (Invitrogen Life Sciences). Mononuclear cells isolated from E16 fetal livers were used for in vitro studies. Time-dated pregnant mice were subjected to laparotomy with exposure of the uterine horns, and RNAi was injected at a final dosage of 7ug/fetus. Animals were sacrificed at 96 hours post-injection, and fetal livers were harvested with the mononuclear cell fraction isolated by Ficoll-Paque centrifugation. 2x105 cells were plated in MethoCult media, with colonies counted after 10 days of incubation. The remaining fractions of cells were lysed for protein extraction and immunoblotting. Results: Positive control RNAi targeting Factor VII resulted in a reduction of Factor VII protein levels compared to controls in both in vitro and in vivo studies. RNAi sequences targeting cMpl resulted in reduced cMpl protein expression and colony forming units in vitro [116.7+/-22.0 – control; 66.3 +/-16.5 – RNAi sequence 1; 84.5+/-12.6 – RNAi sequence 2]. Following vitelline vein injection of RNAi sequences, fetal livers were resected en bloc and were found to be significantly smaller in treated animals compared to controls. Western blotting revealed reduced cMpl expression levels, and colony formation assays revealed reduced total number of colony-forming units [431.7+/-71.9 – control; 95.3 +/-15.0 – RNAi sequence 1; 168.5+/-27.6 – RNAi sequence 2], with proportionate reductions in granulocyte, macrophage and erythroid colonies and significantly greater reductions in megakaryocyte-positive colonies. Conclusions: Delivery of RNAi targeting cMpl in the murine fetus is effective in reducing expression levels both in vitro and in vivo. The observed deficiencies in colony forming units are consistent with those reported in the literature using the cMpl knockout mouse, and suggest significant impairment of hematopoiesis in treated fetuses. These results support our hypothesis that transient non-integrating RNAi targeting cMpl may provide a safe and well-tolerated myeloablative strategy in the fetus.
782. Tumor-Targeted Delivery of IL-2 by NKG2D Leads to Accumulation of Antigen-Specific CD8+ T Cells in the Tumor Loci and Enhanced Anti-Tumor Effects
Chien-Fu Hung, TC Wu
Departments of Pathology, Johns Hopkins University, Baltimore, MD, USA
Interleukin-2 (IL-2) has been shown to promote tumor-specific T-cell proliferation and differentiation but systemic administration of IL-2 results in significant toxicity. Therefore, a strategy that can specifically deliver IL-2 to the tumor location may alleviate concerns of toxicity. Because NKG2D ligands have been shown to be highly expressed in many cancer cells but not in healthy cells, we reason that a chimeric protein consisting of NKG2D linked to IL-2 will lead to the specific targeting of IL-2 to the tumor location. Therefore, we created chimeric proteins consisting of NKG2D linked to Gaussia luciferase (GLuc; a marker protein) or IL-2 to form NKG2D-Fc-GLuc and NKG2D-Fc-IL2, respectively. We demonstrated that NKG2D linked to GLuc was able to deliver GLuc to the tumor location in vivo. Furthermore, we showed that TC-1 tumor-bearing mice intramuscularly injected with DNA encoding NKG2D-Fc-IL2, followed by electroporation, exhibited an increased number of luciferase-exp ressing E7-specific CD8+ T cells at the tumor location. More importantly, treatment with the DNA construct encoding NKG2D-Fc-IL2 significantly enhanced the therapeutic anti-tumor effects generated by intradermal vaccination with therapeutic HPV DNA in tumor-bearing mice. Therefore, by linking NKG2D to IL2, we are able to specifically deliver IL-2 to the tumor location, enhancing antigen-specific T-cell immune response and controlling tumor growth. Our approach represents a platform technology to specifically deliver proteins of interest to tumor loci.
783. AAV2 Infection Requires Endocytosis through the CLIC/GEEC Pathway
Mathieu Nonnenmacher, Thomas Weber
Cardiovascular Research Center, Mount Sinai School of Medicine, New York, NY, USA
Adeno-Associated Viruses (AAVs) are non-enveloped, single-stranded DNA viruses that are non-pathogenic, and AAV-based vectors are promising delivery vehicles for gene therapy applications. Our knowledge of the endocytic and intracellular trafficking pathways of the AAVs is incomplete. Here, we report that infectious endocytosis of the prototypical AAV, AAV2, is independent of clathrin, caveolin and dynamin. Furthermore, AAV2 infection is sensitive to EIPA, a fluid-phase uptake inhibitor but is unaffected by Rac1 mutants or other macropinocytosis inhibitors. In contrast, AAV2 infection requires actin cytoskeleton-remodeling and membrane cholesterol and is sensitive to inhibition of Cdc42, Arf1 and GRAF1, factors known to be involved in the formation of clathrin-independent carriers (CLIC). AAV2 virions are internalized in the detergent-resistant GPI-anchored-protein-enriched endosomal compartment (GEEC) and translocated to the Golgi apparatus, similarly to the CLIC/GEEC marker cholera toxin B. Our results indicate that the pleiomorphic CLIC/GEEC pathway constitutes the major endocytic route for AAV2 infection.
784. TALE proteins may be used to induce specific gene expression to treat some diseases
Jacques P. Tremblay, Pierre Chapdelaine, Zoé Coulombe, Joel Rousseau
Department of Molecular Medicine, Faculty of Medicine, Laval University, Quebec, Canada and Centre de Recherche du Centre Hospitalier Universitaire de Québec.
Genes coding for Tal effector (TALE) proteins may be engineered to target specific DNA sequences. TALEs are fused with a transcription activator can be used to specifically induce the expression of a gene. This could lead to completely new therapies for several diseases. We have applied this potential therapeutic approach to Friedreich ataxia (FRDA) as an example. FRDA is due to a reduced expression of frataxin following an elongation of a trinucleotide (GAA) repeat in intron 1. Aim: To develop a potential treatment for FRDA by increasing the expression of the frataxin gene. Results: We have engineered 12 TALE genes (TALEFrat) coding for TALEFrat proteins each specifically targeting different 14 bp DNA sequences within the proximal region of the human frataxin promoter. When the genes of these TALEFrat were fused with a transcription activator, i.e., four VP16 peptides (i.e., VP64), the resulting TALEFrat-VP64 proteins induced the expression of a mCherry reporter gene fused t o a mini-CMV promoter able to be activated by the insertion of the frataxin proximal promoter upstream to the mini promoter. These TALEFrat-VP64 also increased by 2 to 3 folds the frataxin gene expression (detected by qRT-PCR) in the cells. Conclusion: TALEFrat proteins targeting the frataxin promoter are thus a method to increase the expression of frataxin mRNA and potentially could alleviate the symptoms of Friedreich ataxia. This new methodology of TALE effector opens a new field, which could be used to develop TALE proteins to treat other diseases by inducing the expression of specific genes.
785. Aptamer-Mediated Delivery of Splice-Switching Oligonucleotides to the Nuclei of Cancer Cells
Jonathan W. Kotula1, Elizabeth D. Pratico1, Xin Ming2, Osamu Nakagawa2, Rudolph L. Juliano2 and Bruce A. Sullenger1
1Departments of Surgery and Molecular Genetics and Microbiology, Duke University Medical Center, Durham, NC; 2Division of Molecular Pharmaceutics, UNC Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC
To reduce the adverse effects of cancer therapies and increase their efficacy, new delivery agents that specifically target cancer cells are needed. We and others have shown that aptamers can selectively deliver therapeutic oligonucleotides to the endosome and cytoplasm of cancer cells that express a particular cell surface receptor. Identifying a single aptamer that can internalize into many different cancer cell-types would increase the utility of aptamer-mediated delivery of therapeutic agents. We investigated the ability of the nucleolin aptamer (AS1411) to internalize into multiple cancer cell-types and observed that it internalizes into a wide variety of cancer cells and migrates to the nucleus. To determine if the aptamer could be utilized to deliver therapeutic oligonucleotides to modulate events in the nucleus, we evaluated the ability of the aptamer to deliver splice-switching oligonucleotides. We observed that aptamer-splice-switching oligonucleotide chimeras can alter splicing in the nuclei of treated cells and are effective at lower doses than the splice switching oligonucleotides alone. Our results suggest that aptamers can be utilized to deliver oligonucleotides to the nucleus of a wide variety of cancer cells to modulate nuclear events such as RNA splicing.
786. Cell- and Site-Specific RNAi Interference in Tumor-Associated Macrophages via ‘Click', Mannosylated Polymeric Nanoparticles
Shann S. Yu1, Cheryl M. Lau1, Whitney Barham2, Halina M. Onishko2, Christopher E. Nelson1, Fiona E. Yull2, Craig L. Duvall1, Todd D. Giorgio1,2
1Dept. of Biomedical Engineering, Vanderbilt University, Nashville, TN; 2Dept. of Cancer Biology, Vanderbilt-Ingram Cancer Center, Nashville, TN
Tumor-associated macrophages (TAMs) represent a promising therapeutic target in cancer because they have been shown to facilitate tumor growth, invasiveness, and metastasis. However, macrophage-specific drug delivery within tumor sites is a significant challenge, as systemic interference with macrophage behavior may lead to autoimmune manifestations. In this work, we designed and characterized mannosylated polymeric nanoparticles (ManNPs) in order to achieve CD206 (macrophage mannose receptor)-targeted siRNA delivery. CD206 is almost exclusively expressed on macrophages and dendritic cells, and upregulated in tumor-associated macrophages. The ManNPs are composed of tri-block co-polymers, including the following blocks: (1) an azide-displaying block for the attachment of alkyne-functionalized mannose via ‘click' chemistry, (2) a cationic block for the condensation of polyanions such as siRNA, and (3) a pH-responsive terpolymer block that facilitates endosomal disruption. This terpolymer is hydrophobic at pH 7.4, allowing these polymers to self-assemble into 25 nm micelles under physiologic conditions, as shown by transmission electron microscopy. However, they become protonated at lower pH ranges representative of endosomal compartments (5.8–6.2), leading to disassembly of the nanoparticles, and increased ability to disrupt endosomal membranes. Therefore, this pH-dependent behavior facilitates improved cytoplasmic delivery of siRNA, access to the intracellular silencing machinery, and consequently, knockdown of target gene expression. The glycoconjugates improved siRNA delivery into primary murine bone marrow-derived macrophages by 4.5-fold, relative to a non-mannosylated version of the same carrier. Internalization of these constructs can be blocked by co-incubation with mannose or suppressed by downregulation of CD206 via LPS pre-treatment, showcasing the specificity of the construct for CD206. This is particularly important for cancer applications because CD206 is upregulated in tumor-suppressed and non-activated macrophages, enabling more specific targeting of TAMs versus healthy macrophages in other tissues. Finally, the delivered siRNA retains its activity following delivery, resulting in 40±10% knockdown of a model gene within four hours of delivery, relative to non-transfected macrophages. We have also examined the behavior of the ManNPs in vivo by delivering them retro-orbitally into mouse models of breast cancer, caused by the expression of the polyoma middle T oncoprotein in the mammary epithelium. Within 24 h, the ManNPs facilitated improved delivery of siRNA into CD206-expressing cells in tumors, as shown by immunostaining of tumor frozen sections. Flow cytometry analysis also shows significant co-localization of the delivered siRNA with tumor-associated F4/80+ cells. The ManNPs described here present new opportunities to target TAMs in various cancers, providing an enabling technology for the modification of the immunosuppressive tumor environment by targeting TAM activity.
787. Transient Ischemia is Necessary for Efficient Adenovector Gene Transfer in the Heart
Weiwei Shi1, L. Susan Schmarkey1, Rong Jiang1, C. Collin Bone1, Marah E. Condit1, Dirck L. Dillehay2, Robert L. Engler3,4, Gabor M. Rubanyi4, Jakob Vinten-Johansen1
1Cardiothoracic Research Laboratory, Carlyle Fraser Heart Center, Emory University, Atlanta, GA; 2Department of Pathology and Laboratory Medicine and Division of Animal Resources, Emory University, Atlanta, GA; 3School of Medicine, University of California, San Diego, CA; 4Cardium Therapeutics, Inc., San Diego, CA
Background: Conflicting results in intracoronary (IC) adenoviral vector (Ad5) transfection studies might be due to the permeability of the vascular endothelium barrier. Endothelial permeability may be increased by sodium nitroprusside [SNP] nitroglycerin [NTG] or transient ischemia. Hypothesis: We tested whether transfection efficiency (TE) of IC Ad infusion is increased by SNP, NTG or various durations of ischemia. Methods: An angioplasty balloon catheter was fluoroscopically guided into the left anterior descending (LAD) and the left circumflex (LCx) coronary arteries in anesthetized Yorkshire pigs, through which Ad5 encoding the luciferase gene (Ad5Luc,1 x 1011vp) was selectively infused. Three or 14 days after infusion Ad5Luc was quantified by conversion of luciferin to fluorescent oxyluciferin (pg/g tissue). Protocol 1 (n=8): Ad5Luc was delivered IC to non-ischemic myocardium with or without IC co-administration of SNP (50 µg/min). Protocol 2 (n=8): IC Ad5Luc was delive red after one, or through the distal lumen during the second of 2 episodes (5 minutes apart) of 3 minute coronary artery balloon occlusion without or with concomitant NTG infusion. Protocol 3 (n=24): Ischemia was induced by 75 min of balloon inflation; Ad5Luc was infused 15 min after start of reperfusion. Results: SNP did not significantly increase the minimally detectable Ad over saline control (0.2 ± 0.16 vs ~0 ± 0, p=ns) in non-ischemic myocardium. Infusion during the second occlusion, but not after one short episode of coronary artery occlusion (Protocol 2) tended to increase TE compared to non-ischemic myocardium (35.88 ± 27.47 vs ~0 ± 0, p=0.06), but TE was significantly greater with IC co-administration of NTG (50ug/min) (82.27 ± 39.55 vs ~0 ± 0, p<0.001). In protocol 3, IC delivery of Ad5Luc into infarcted/reperfused myocardium caused no local inflammation or hemodynamic instability. Infarct size [% of area at risk (AAR)] between saline control and Ad5Luc groups were comparable at 3 (53±6 vs 58±3) and 14 (67±6 vs 52±7) days of reperfusion. TE in Ad5Luc group was significantly (p<0.05) greater in the AAR compared to border zone and non-ischemic zone at both 3 (90.4±26.5, 9.6±6.1 and 0.2±0.6 pg Luc/g tissue,) and 14 days (18.4±10.1, 0.90.5 and 0.5±0.5 pg Luc/g tissue) of reperfusion. The finding of several orders of magnitude greater TE during transient or after sustained ischemia may explain conflicting results in previous preclinical and clinical studies. Conclusion: In normal myocardium there was minimal Ad5 gene transfection. Two 3 minute coronary occlusions, especially with NTG infusion, markedly enhanced TE. The selective uptake of IC Ad5 to ischemia/reperfused myocardium without adenovector associated inflammation validates use of this delivery protocol in future human gene therapy trials in patients during transient or following sustained myocardial ischemia.
788. Prevention of Type I Diabetes in NOD Mice after liver-directed Gene Therapy Correlates with Increased Foxp3 Expressing Regulatory T cells
Mahzad Akbarpour1,2, Kevin Goudy1, Andrea Annoni1, Lucia Sergi Sergi1, Luigi Naldini1,2, Maria Grazia Roncarolo1,2
1San Raffaele Telethon Institute for gene Therapy (HSR-TIGET), Division of Regenerative Medicine, Stem Cells and Gene Therapy, San Raffaele Scientific Institute, Milan, Italy; 2Vita-Salute San Raffaele University School of Medicine and Surgery, Milan, Italy
The goal of this study was to determine whether a microRNA142 (142T)-regulated lenti-viral vector (LV) encoding the insulin β 9-23 region (InsB) can prevent diabetes onset in the human model for type I diabetes, the non-obese diabetic (NOD) mouse. To test this, a series of LV and LV.142T vectors encoding the InsB fragment or an irrelevant antigen ovalbumin (OVA) were cloned under the control of liver-specific enhanced transthyretin promoter (ET) or the ubiquitously expressed phosphoglycerokinase promoter (PGK). Virus was injected into prediabetic 8-10wk old female NOD mice and followed for diabetes. Mice treated with LV.PGK.InsB and LV.PGK.OVA.142T developed diabetes comparable to untreated controls while, ET.OVA.142T treated mice had delayed incidence of diabetes with 40% protection and LV.ET.InsB.142T injected NOD mice were 90% protected from diabetes at 40 weeks of age. Upon adoptive transfer of splenocytes from protected LV.ET.InsB.142T and diabetic mice, 62% of NOD/Scid recipient mice were protected from diabetes 50 days after transfer, while splenocytes of ET.OVA.142T treated mice did not alter disease development. Higher frequency of foxp3+ regulatory T cells (Treg) in liver and PLN after LV.ET.InsB.142T together with reduction in effector T cells population suggests LV.InsB.miR142T treatment promotes tolerogenic responses which could protect mice form disease onset. These findings demonstrates that liver-specific, miRNA regulated vectors encoding for an autoantigen is a successful approach to circumvent established autoimmune responses, and presents a promising therapeutic advance in designing clinical gene delivery systems to treat type I diabetes and other immune-related diseases.
789. Oncolytic HSV Containing Gamma-ICP34.5 Gene under Nestin Promoter Efficiently Induces Killing in Ovarian Cancer
Chelsea Bolyard1,2, Ji Young Yoo1, Selvendiran Karuppaiyah3, Doug McCarty2, Balveen Kaur1
1The Ohio State University - Comprehensive Cancer Center, Dardinger Laboratory of Neurosciences, Columbus OH; 2The Research Institute at Nationwide Children's Hospital, Department of Gene Therapy, Columbus OH; 3The Ohio State University - Obstetrics & Gynecology, Columbus OH.
Oncolytic herpes simplex virus I (oHSV) has been tested as an agent to treat a number of tumor types, including ovarian cancer. Most studies published to date use an attenuated “oncolytic” virus. To generate this virus, the viral ribonucleotide reductase ICP4 is mutated by inserting a GFP cassette. This makes the virus require host replication machinery, and thus limits its ability to infect non-dividing cells. Furthermore, these attenuated viruses contain deletions for the gamma-ICP34.5 gene. Viral ICP34.5 is a major neurovirulence factor, but it also prevents host cell anti-viral immune response by de-phosphorylating eIF2-alpha. Without ICP34.5, upon HSV infection, Protein Kinase R (PKR) activates a variety of cellular phophotases that modulate an anti-viral immune response. These phosphatases phosphorylate eukaryotic initiation factor 2-alpha (eIF2-alpha), a critical transcription initiating factor, resulting in the shutdown of cellular protein synthesis. oHSV vectors that lack of ICP34.5 are unable to mount a robust replication cycle in absence of cellular protein synthesis. In the study presented herein, we examine the use of an oHSV that expresses ICP34.5 under the nestin promoter. This oHSV vector also expresses vasculostatin-120 (VStat-120) under a HSV immediate early promoter. We call this vector 34.5 ENVE, viral ICP34.5 Expressed by Nestin promoter and Vstat120 Expressing. 34.5ENVE has been previously shown to elicit rapid and potent regression of glioblastoma cancer in murine models. Nestin is a class VI intermediate filament protein, first identified to be upregulated in central nervous system tumors. Since the initial publication, nestin has been shown to be upregulated in a number of cancers, with limited data published on increased nestin expression in cancer “stem cells” found in ovarian cancer. VStat-120 is the extracellular fragment from Brain Angiogenesis Inhibitor-1 (BAI1), and is able to induce antiangiogenic effects in vitro and in vivo. We have found that 34.5ENVE is cytotoxic to ovarian cancer cell lines PA-1, OV-4, SKOV-3, OVCAR-3 and cisplatin-sensitive A2780. We are assessing the relative nestin gene expression in these cells. PA-1, which displayed the highest level of nestin expression is markedly more sensitive to 34.5ENVE infection, relative to 34.5-deleted HSV-Q virus, with the ENVE yielding more potent cell killing. 34.5ENVE infection of these cell lines leads to robust VStat120 expression within 24 hours. Anti-angiogenic and anti-tumorigenic effects of 34.5ENVE are currently being assessed in model of late stage ovarian cancer.
790. Combined Delivery Carrier of Anticancer Drug and siRNA for the Treatment of Glioblastoma
Na Yi, Hyun Ah Kim, Minhyung Lee
Department of Bioengineering, College of Engineering, Hanyang University, Seoul, Korea
The R3V6 peptides, which are composed of 3-arginines and 6-valines, formed self-assembled micelles in aqueous solution. In addition, these peptides were able to deliver siRNA and hydrophobic anti-cancer drugs into cells. The aim of this study is to investigate the tumor suppressive effect of the complex of bis-chloroethylnitrosourea (BCNU) loaded R3V6 (R3V6-BCNU) and VEGF small interfering RNA (VEGF-siRNA) in the rat glioblastoma model. The cytotoxicity of VEGF-siRNA/R3V6-BCNU complex against rat glioblastoma cells in vitro suggested that VEGF-siRNA/R3V6-BCNU complexes had higher anti-cancer effect than BCNU alone. In vitro transfection assay in rat glioblastoma cells, VEGF-siRNA formed with R3V6-BCNU had the highest transfection efficiency at a 1:20 weight ratio (VEGF-siRNA:R3V6-BCNU). In a gel retardation assay, the BCNU/R3V6 completely retarded siRNA at a 1:20 weight ratio (VEGF-siRNA:R3V6-BCNU). The heparin competition assay showed that siRNA-VEGF with BCNU/R3V6 peptide a t 1:5 ratio (VEGF-siRNA:R3V6-BCNU). In an in vitro silencing assay in rat glioblastoma cells, VEGF-siRNA/R3V6-BCNU complexes had a higher transfection efficiency than lipofectamine or naked siRNA. To evaluate the anti-tumor effects of VEGF-siRNA/R3V6-BCNU complexes in vivo, VEGF-siRNA/R3V6-BCNU complexes were prepared in a 20 to 1 ratio and injected twice intracranially into the rat glioblastoma model. The tumor volumes were measured a week after first treatment. In the treatment with VEGF-siRNA/R3V6-BCNU complexes reduced tumor volume compared with the untreated tumor control group. Therefore, VEGF-siRNA/R3V6-BCNU complexes may have suppressive effect on tumor volume in of glioblastoma rat model.
791. Carrier-cells with Conditionally Latent Oncolytic Adenovirus
Hiroshi Nakashima, J. Bradley Elder, Imran Shalkh, E. Antonio Chiocca
The Ohio State University Medical Center, Columbus, OH
Carrier-cell systems have been utilized for delivery of oncolytic viruses (OVs) to tumors, such as brain tumors. In these approaches, the carrier cell is infected with the OV in ex vivo and then administrated to animals. However, the viral lytic cycle is likely to destroy several carrier cells before they reach the intended target cancer, releasing progeny OVs to other unintended loci and/or reducing the beneficial effect of utilizing a carrier-cell based delivery mode. We thus reasoned that conditional regulation of the OV lytic cycle could prolong carrier-cell survival and enhance anticancer efficacy. The overall strategy involved engineering a lytic viral genome that could be inserted into carrier cells where it would exist in a latent fashion until the carrier cell was exposed to tumor microenvironments that would signal the viral genome to re-enter into the lytic cycle. We utilized human mesenchymal stem cells (hMSCs) as carrier cells and hypoxia as the distinctive tumor factor. We thus engineered the oncolytic, Ad5F35δ24, adenovirus with a E1Aδ24 gene that was silenced by sequences located between the CMV promoter and the E1Aδ24 gene which could be excised by an hypoxia-responsive promoter-driven site-specific Flp recombinase. This new adenovirus (proAd5F35δ24) was significantly less toxic to hMSCs and glioma cells compared to replication-competent Ad5F35δ24 and remained silent/latent in the infected hMSCs. To activate E1A gene expression and start the viral lytic cycle, we utilized a replication-defective HSV amplicon that expressed flp recombinase under control of the Hypoxia Responsive Element (HRE) promoter. Introduction of this HRE-flp vector into proAd5F35δ24-infected hMSCs led to the generation of progeny oncolytic Ad5F35δ24 in hypoxic, but not normoxic conditions. In an intracranial glioma xenograft model, the combination of proAd5F35δ24-infected carrier cells with HRE-flp led to a significant anticancer effect by injecting at distant point (1mm) from tumor implantation when compared to Ad5F35δ24-carrier hMSCs without HRE-flp or parental oncolytic Ad5F35δ24 intracranially. This provides a platform for generation of carrier cells that express latent oncolytic viral genomes that can be reactivated upon presentation of a tumor-specific signal.
792. Prognostic and predictive factors in therapy with oncolytic adenoviruses
Anniina Koski1, Raita Heiskanen2, Kristian Taipale1, Kaarina Partanen3, Timo Joensuu3, Akseli Hemminki1–3
1Cancer Gene Therapy Group, University of Helsinki, Finland; 2Oncos Therapeutics Ltd, Helsinki, Finland; 3Docrates Hospital, Helsinki, Finland
Disclaimer: A.H. is shareholder and consultant for Oncos Therapeutics Ltd.
286 patients with advanced chemotherapy refractory solid tumors were treated in a personalized manner in the Advanced Therapy Access Program. Presented here is an analysis of clinical, immunological and radiological prognostic and predictive factors. Statistically significant prognostic factors for overall survival included WHO performance score, treatment dose, concomitant low-dose cyclophosphamide, absence or severe adverse reactions, smaller magnitude of increase in neutralizing antibodies and lower degree of virus replication as measured with qPCR from blood. Factors predictive of disease control in post-treatment imaging included intracavitary injection, smaller magnitude of increase in neutralizing antibodies and lower degree of virus replication as measured with qPCR from blood. Interestingly tumor load also seemed to influence treatment outcome as patients with smaller tumor masses were more likely to have a positive result in post-treatment imaging. Moreover, the location of metastases seemed to play a role; liver metastases were associated with unfavorable outcome while peritoneal disease resulted in a higher likelihood of disease control albeit at borderline statistical significance. These findings may influence design of clinical trials with oncolytic viruses.
793. Lentiviral Expression of SGSH Under the Human CD11b Promoter in transplanted Thaematopoietic Stem Cells Fully Corrects Behaviour and Neuropathology of the Mouse Model of Mucopolysaccharidosis IIIA
A Sergijenko1, A Langford-Smith1, FL Wilkinson1, KJ Langford-Smith1, A Lau1, SA Jones2, JE Wraith2, RF Wynn3, BW Bigger1
1Stem Cell & Neurotherapies, School of Biomedicine, Faculty of Medical and Human Sciences, University of Manchester, Manchester, UK; 2Genetic Medicine, St Mary's Hospital, Manchester, UK; 3Blood and Marrow Transplant Unit, Royal Manchester Children's Hospital, Manchester, UK
Mucopolysaccharidosis IIIA (MPSIIIA) is a lysosomal storage disorder caused by mutations in the sulfamidase (SGSH) gene, resulting in accumulation of heparan sulphate and progressive neurodegeneration in patients. We previously demonstrated improved neuropathology and behavioural correction of MPSIIIA mice using wild type haematopoetic stem cells transduced with lentiviral vector expressing SGSH under a viral SFFV promoter. In this context, we could not achieve behavioural correction using lentiviral transduced MPSIIIA cells, and this combined with the use of a viral promoter, limited clinical application of this approach. To eliminate the viral promoter we compared eGFP expressing lentiviral vectors under the ubiquitous PGK or myeloid-specific CD11b or CD18 human promoters. Lineage depleted bone marrow was transduced with these vectors and transplanted into WT mice. Six weeks after transplant we achieved significantly improved eGFP expression in B cells and monocytes from peripheral blood using the CD11b promoter over either the PGK or CD18 promoter. The CD18 promoter was more monocyte specific, but weak and was not analysed further. To improve gene expression, we codon optimised the hSGSH gene, significantly improving activity, and transduced lineage depleted MPS IIIA bone marrow with the PGK-coSGSH (LV-PGK) or CD11b-coSGSH (LV-CD11b) vectors. These were transplanted into busulfan conditioned MPSIIIA mice. At 6 months, the abnormal behaviour exhibited by MPSIIIA mice in the open field was fully corrected by LV-CD11b but not by LV-PGK. LV-PGK improved SGSH enzyme activity in the liver and spleen to 66% and 245% respectively whilst LV-CD11b improved this to 60%, and 160% of WT levels respectively. In the brain, LV-PGK increased enzyme activity to 7% of WT levels whilst LV-CD11b was significantly better at 11%. Abnormally elevated β-hexaminidase levels in the liver and brain of MPSIIIA mice were normalised by LV-CD11b and reduced by LV-PGK. LV-CD11b fully normalised brain glycosaminoglycans whilst LV-PGK had some residual glycosaminoglycan, but there was no significant difference between treatments. Neuroinflammation as measured by microglial infiltration into the cerebral cortex was completely normalised by LV-CD11b, whilst this was reduced but not normalised by LV-PGK. These data demonstrate that LV-CD11b mediated haematopoietic stem cell gene therapy in autologous MPSIIIA cells fully corrects neuropathology and behaviour of MPSIIIA mice whilst LV-PGK improves neuropathology but did not affect behaviour. This is the first demonstration of efficacy of the autologous stem cell gene therapy approach in MPSIIIA mice using a clinically relevant lentiviral vector and opens the door to clinical application in MPSIIIA patients. The human CD11b promoter is more efficient than the human PGK promoter in this context.
794. Characterization of Surface Markers, Clonogenicity and Gene Expression during Neutrophil Differentiation from Human Induced Pluripotent Stem Cells
Hongmei Wang, Colin Sweeney, Harry L. Malech
Laboratory of Host Defenses, NIAID, NIH, Bethesda, MD
We derived neutrophils from human induced pluripotent stem cells (iPSCs) through discontinuous culture over 32 days (18 days of embryoid body (EB) formation in hematopoietic stem cell (HSC) cytokines; EB dissociation and 7 days co-culture with OP9 stromal cells in HSC cytokines; transfer of non-adherent cells for final 7 days OP9 co-culture in G-CSF). CD45 hematopoietic surface marker appeared gradually during the EB stage (4% of cells weakly CD45+ at day 18), but rose rapidly upon OP9 co-culture (100% strongly CD45+ by day 22). CD34 surface marker exhibited biphasic expression, with ~11% CD34+ but CD45- at day 10, likely representing hemangioblast precursors of both hematopoietic and endothelial cells, correlating with increased microRNA-126 expression. At day 20-21 CD34 again peaked with 30% of cells strongly CD34+ and also weakly CD45+, indicating early HSCs. This correlated with emergence of cell capacity to generate hematopoietic methylcellulose colony forming units (CFU) at day 18, peaking at day 20 and dropping thereafter. Expression of HSC-associated microRNA-126 and microRNA-155 also peaked on day 20. During day 18-20 OP9 co-culture, adherent CD34+ cells expressed Flk-1 and VE-Cadherin endothelial markers, whereas non-adherent CD34+ cells rapidly lost endothelial markers. Transcription factor expression was analyzed by real-time PCR in cells FACS-sorted for surface antigens. GATA-2 and AML1/Runx1 were up-regulated in day 22 CD34+CD45+ presumptive HSCs compared with undifferentiated iPSCs, but were rapidly down-regulated with the later appearance of CD15. The myeloid-associated transcription factor PU.1 was highly expressed at early stages of neutrophil differentiation and retained in CD33+ cells and CD15+ neutrophils. Late-acting transcription factor C/EBPε was only detected in CD15+CD33+ cells, consistent with previous reports that C/EBPε is exclusive to neutrophils. Our study provides the first detailed characterization of surface markers correlated with functional assay and transcription factor expression during neutrophil differentiation from human iPSCs. This begins to define complete myeloid differentiation from human iPSCs, and will help to identify stages for further manipulating this process.
795. Targeted-PLGA nanoparticles encapsulating histone deacetylase inhibitors effectively control melanoma
Jeonga Shin1, Sunmi Cho1, Priti Kumar2, Sang-Kyung Lee1
1Hanyang University, Seoul, Korea; 2Yale University, CT, USA
Histone acetylation that is regulated by histone deacetylases (HDACs) and histone acetylases (HATs) has an important role in cell cycle, apoptosis and differentiation. HDAC inhibitors (HDACi) are promising anti-cancer agent as they can reactivate aberrantly silenced tumor suppressor genes and block tumor cell growth. However the targeted delivery of HDACi to cancer cell is essential to minimize the adverse activation of bystander cells. In this study, we used the FDA-approved, biodegradable polymer, PLGA to deliver HDACi specifically to cancer cells by conjugation to a peptide that binds acetylcholine receptors expressed abundantly on cancer cells. The 200 nm sized modified cancer targeting PLGA/HDACi nanoparticles inhibited B16F10 melanoma cell proliferation by more than 90% in in vitro. Further it also inhibited tumor growth in mice subcutaneously implanted with melanoma cancer cell in upon iv injection. This approach opens avenues for new safe cancer therapeutics with HDACi.
796. Degradation of plasmid DNA by DNase I and EndoG in blood plasma and relations between the endonucleases
Daesong Jang, Xiaoying Wang, Todd Fite, Tariq Fahmi, Dmitry Zhdanov, Alexei G. Basnakian
University of Arkansas for Medical Sciences & Central Arkansas Veterans Healthcare System, Little Rock, Arkansas, USA
DNA endonucleases play important role in host cell and body defense by degrading intruding foreign DNA. This protective mechanism presents a fundamental problem for gene therapy causing poor gene transfer efficiency and expression, in particular, after intravenous injection of DNA. However, the endonucleases responsible degradation of foreign DNA in blood plasma has not been identified. In this study, we explored whether two most abundant cytotoxic/apoptotic endonucleases, deoxyribonuclease I (DNase I) and endonuclease G (EndoG) are responsible for the degradation of foreign DNA in blood plasma during gene delivery. For this, supercoiled covalently-closed pECFP plasmid DNA was exposed at 37°C for up to 12 hours with blood serum extracted from DNase I knockout (KO), EndoG KO, and their counterpart wild-type (WT) mice; and DNA degradation to open-circular, linear and fragmented DNA was assessed by agarose gel electrophoresis. The fastest DNA degradation was produced by serum from WT mice, which left no supercoiled DNA in 120 min. In contrast to WT, degradation of DNA in DNase I KO serum was decreased by ~90%, and degradation in EndoG KO serum was decreased by ~25%. We subsequently tested whether DNase I or EndoG is responsible for decreased transfection efficiency of the plasmid in human breast cancer HCC1954 cell line. Plasmid exposed with blood sera as described above was purified and the transfection efficiency was measured by cyan fluorescence of the cells caused by CFP expression. In this approach, exposure with serum from DNase I KO mice showed significant increase in transfection efficiency compared to WT serum, even at 120 min after the exposure. Serum from EndoG KO mice also showed some improvement compared to WT, but less than DNase I KO. This result was completely consistent with the observations of the electrophoresis data. It suggests the transfection efficiency of pECFP plasmid in HCC1954 cells depends on the level of degradation of the plasmid by serum DNase I and, to the lesser extent, by EndoG. The only unexplained phenomenon was that in total, degradation of the plasmid by the endonucleases was above 100%. To test whether the two endonucleases may activate each other, rat kidney tubular epithelial NRK-52E cells expressing both endonucleases were transfected with DNase I or EndoG gene, and expression of the two endonucleases was measured by real-time RT-PCR. This experiment showed that DNase I overexpression caused induction of EndoG, and the overexpression of EndoG caused induction of DNase I. Taken together, these observations present strong evidence that both DNase I and EndoG, while inducing each other, are required for the degradation of DNA in blood plasma, and may cause low transfection efficiency during gene delivery.
797. PABPN1 expression level regulates muscle function in aging and in OPMD
Vered Raz, Seyed Yahya Anvar, Yotam Raz, Andrea Venema, Peter AC't Hoen, Silvere M van der Maarel
Leiden University Medical Center, Leiden, The Netherlands
Introduction: Aging is influenced by multiple genetic and environmental factors resulting in progressive tissue degeneration. Key molecular regulators of this complex process are not fully elucidated. Aging-associated genetic disorders with accelerated tissue degeneration can provide clues to the normal process. Here we used the late-onset muscle weakness disorder oculopharyngeal muscular dystrophy (OPMD) in order to identify critical regulators of skeletal muscle aging. OPMD is an autosomal dominant caused by an expansion mutation in PABPN1. Methods: Genome-wide expression profiles were generated from quadriceps of OPMD patients and healthy age-matching control group in order to identify molecular pathways that are associated with the diseases. In addition, to identify key regulators of disease onset expression trends were compared between OPMD carriers at pre-symptomatic and symptomatic stages. Novel bioinformatics analyses were applied to compare changes in expression trends in all datasets. Validation studies were carried out by functional genomic tools in muscle cells. Results: Significant similarities were found between OPMD and normal muscle aging transcriptome datasets, with a major shift in expression profiles during the fifth decade. The transcriptional changes in OPMD were more pronounced compared with physiological aging, suggesting acceleration of muscle degeneration in OPMD. We identified significant deregulation of genes encoding for aggregated proteins, among which PABPN1 was identified. The expression of PABPN1 significantly declined only from the fifth decade in both muscle aging and OPMD, whilst a more pronounced decline was identified in OPMD. PABPN1 down-regulation in human myotube cultures progressively induced cellular senescence and muscle cell fusion. Moreover PABPN1 down-regulation levels gradually affected a decrease in muscle contraction genes expression. We found that changes in PABPN1 expression were significantly more pronounced in skeletal muscles compared with other aging tissues. Conclusions: Our data suggest that PABPN1 is a key regulator of muscle cells and its expression level can mark a decline in muscle performance during aging.
798. Busulfan-Conditioning Improves Engraftment of Hematopoietic Donor-Derived Cells in the Brain by Establishing Pro-Migratory Cues, Whilst Irradiation Promotes an IL-1 Driven Neuroinflammatory Environment
FL Wilkinson1, KJ Langford-Smith1, M Malinowska2, RF Wynn3, Bigger1
1Stem Cell & Neurotherapies, Faculty of Medical and Human Sciences, University of Manchester, UK; 2Faculty of Biology, University of Gdansk, Poland; 3Blood and Marrow Transplant Unit, Royal Manchester Children's Hospital, UK.
Hematopoietic stem cell gene therapy for neurodegenerative lysosomal disorders relies on transmigration of donor-derived monocytes to the brain, where they can engraft as microglia and deliver therapeutic proteins. Many mouse studies use whole-body irradiation to investigate brain transmigration pathways, but chemotherapy is generally used clinically. We compared hematopoietic donor-cell brain engraftment after bone marrow transplants in busulfan or irradiation-conditioned mice. After 6-9 months, significantly more donor-derived microglial cells engrafted in busulfan-conditioned brain compared to irradiated, whilst total microglial content was similar between groups. MCP-1, a key regulator of monocyte transmigration, showed long-term elevation in busulfan-conditioned brain, whilst irradiated brains showed long-term elevation of the pro-inflammatory chemokine IL-1α, with increased in situ proliferation of resident microglia, and significant increases in the relative numbe r of amoeboid activated microglia in the brain. This mechanistic insight to the role of busulfan and irradiation in trafficking and engraftment of monocyte derived microglia in the brain has significant implications for the choice of conditioning regimen to promote hematopoietic cell brain engraftment and emphasizes the limitations of using irradiation in mouse models of transplantation.
799. Rapid manufacture of custom TAL effectors for genomic editing and genetic circuits
Michael Poderycki, Matthias Arnold, Jian-Ping Yang, Veronica Blackston, Antje Bodemann, Matthias Arenskötter, Michaela Deinert, Vinita Kapur, Frank Notka, Ralf Wagner, Todd Peterson, Jonathan Chesnut
Synthetic Biology Research and Development Carlsbad, CA USA and Regensburg, Germany; Life Technologies Corporation, Carlsbad, CA, USA
The ability to target a functional protein to a user-defined DNA sequence is a long sought goal with the potential to enable genome engineering and genetic regulation of diverse cell types. The customizable DNA-binding specificity of TALE microbial proteins is largely governed by a simple code and represents a means to achieve this goal. We have developed a large scale manufacturing capability to rapidly assemble custom codon optimized TALEs as Gateway entry clones. We have designed flexibility into the process such that TALE length is not strictly pre-supposed, allowing for more precise targeting to genomic sites. Furthermore, TALE functionality can be user-determined. Examples of such functionalities could include DNA cleaving enzymes for genome editing and transcription factors for building genetic circuits. It is expected that the applications of custom TALEs will include the generation of cell-based models and therapies, and testing for causality between DNA polymorphisms and human disease.
Late Oral Abstract Session: Friday, May 18, 2012
800. The First Report on a rAAV.sFlt-1 Phase I/II Trial for Wet Age-Related Macular Degeneration (AMD)
EP Rakoczy1, CM Lai1, C Pierce2, A Magno2, D Dismuke3, J Grieger3, RJ Samulski3, TW Chalberg4, SD Schwartz5, MS Blumenkranz6, IJ Constable1
1Centre for Ophthalmology and Visual Sciences, The University of Western Australia, Crawley, Australia; 2Lions Eye Institute, Nedlands, Australia; 3Department of Pharmacology and Gene Therapy Centre, University of North Carolina, Chapel Hill, NC, USA; 4Avalanche Biotechnologies, San Francisco, CA, USA; 5UCLA Jules Stein Eye Institute, Los Angeles, CA, USA; 6Stanford University, School of Medicine, Stanford, CA, USA
The standard of care in treating wet AMD involves frequent intraocular injection of recombinant anti-VEGF proteins every 4-8 weeks. We developed a rAAV construct for a potent (Kd ~10 pM), naturally occurring anti-VEGF protein, soluble Fms-related tyrosine kinase-1 (sFlt-1), for the treatment of wet AMD. rAAV.sFlt-1 was produced in accordance with FDA and ICH guidelines at the UNC Vector Core Human Application Laboratory. Here we report on three patients who received rAAV.sFlt-1 in January 2012. The three enrolled subjects (mean age 77 years) all had active subfoveal choroidal neovascularization, with visual acuity of 20/80 to 20/400, and had previously received multiple intravitreal injections of ranibizumab. A final injection of ranibizumab was given at baseline, and on day 7, 1 x 10E10 vector genomes of rAAV.sFlt-1 in 100 ul volume was administered via subretinal injection. In all three cases, the bleb of sub-retinal fluid resolved within 4 hours. After 24 hours, most of the air in the vitreous had absorbed and only the retinal injection site remained visible. One patient developed a minor hemorrhage associated with the procedure that did not affect vision. As expected following vitrectomy, there was a transient increase in neutrophil counts that returned to normal by 14 days post injection. Vector sequence was found in the tears of one subject at one day post injection that cleared by day 30. Other than this single occurrence, AAV2 was not detected in any of the subjects' blood, saliva or urine samples either by qPCR or ELISA to date. Background levels of the naturally occurring sFlt-1 protein showed a high baseline variation in the urine, serum, and saliva with no increase following treatment. sFlt-1 levels in the vitreous also varied among subjects (975-2085pg/ml). Blood biochemistry, complete blood count, and T-cell response, remained without any significant change compared to baseline. Subretinal injection of rAAV.sFlt-1 showed no clinically significant retinal toxicity as assessed by serial ophthalmic examinations over a two month period. No superficial, anterior segment or vitreous inflammatory signs were present in any of the subjects. At day 60 none of the patients required rescue treatment. There was no evidence of visual acuity loss, IOP elevation, retinal detachment, or any intraocular or systemic immune response in any of the patients. These results extend the promise of gene therapy for the large cohort of elderly patients (>75) suffering from wet AMD. Further efficacy results with the low dose (1 x 10E10 vg) cohort and safety on the high-dose (1 x 10E11 vg) cohort will be discussed.
801. Long-term airway epithelial transduction and induction of tolerance to transgene via intra-tracheal administration of airway-targeting lentivirus to mid-gestation fetal sheep
Marcus G. Davey1, Philip W. Zoltick1, Maria P. Limberis2, Carlyn A. Todorow1, Jason Sulkowski1, Jacqueline Tsai1, Emily A. Partridge1, Erik G. Pearson1, Jesse Vrecenak1, Alan W. Flake1
1The Center for Fetal Research, Children's Hospital of Philadelphia, Philadelphia, PA, USA; 2Gene Therapy Program, Department of Pathology and Laboratory Medicine at the University of Pennsylvania, Philadelphia, PA, USA.
Prenatal gene transfer to the respiratory epithelium may provide opportunities to overcome physical and immunological barriers limiting efficient gene transfer to the adult lung. Unlike the adult, airway stem and progenitor cells are; (1) present at higher frequencies during fetal life; and (2) more accessible to gene transfer as the airways are devoid of mucous and pulmonary macrophages. In addition, the immature immune system of the fetus provides the possibility of inducing immunological tolerance to the exogenous vector and transgene that might allow repeated postnatal delivery of the same vector / transgene. We recently demonstrated that a lentivirus (HIV-1-based) pseudotyped with Jaagsiekte sheep retrovirus envelop (JSRV-LV) expressing GFP administered into the fetal lung lumen at ~70days of gestation (full term ~145days) transduced ~30% of the fetal airways, between 100-300µm diameter, when examined 10 weeks post-transduction (1). We examined the durability of the respiratory cell transduction 6 months after birth and analyzed immunological tolerance to GFP following mid-gestation fetal lung vector administration. Fetal sheep (n=8) received an intra-tracheal injection (0.5mL) of JSRV-LV (~10^9 viral particles) expressing GFP under the control of the CMV promoter. Postnatal immunological tolerance to GFP was evaluated by measuring serum antibodies (IgG1/IgG2) against GFP following an intramuscular injection of adeno-associated virus (AAV) vector expressing GFP (AAV2/1.GFP). Pulmonary epithelial GFP transduction was assessed at 3 and 6 months after birth by immunohistochemical detection of GFP on sections of formalin fixed paraffin embedded tissue. At 3 months after birth, low levels of pulmonary GFP expression were observed in the lungs of 3 of 8 lambs and were restricted to distal epithelium of respiratory bronchioles; transduction was non-heterogeneous and located predominantly in upper and middle lung lobes. At 6 months of age, GFP expression persisted in 2 (twins) of the 3 lambs that were positive at 3 months. In lambs without persistent lung GFP expression (n=6), intramuscular injections of AAV2/1.GFP induced a marked humoral immune response as evidenced by increased serum levels of IgG1 antibodies against GFP. Significant influx of neutrophils and plasma cells surrounded GFP-positive myocytes. The 2 lambs with durable lung GFP expression failed to develop humoral or local immune responses to AAV2/1.GFP intramuscular injection while intense GFP immunostaining was observed in myocytes. Prenatal lung gene delivery can achieve durable postnatal transgene expression, the durability of which correlates with humoral and cellular tolerance to the transgene. These initial experiments in a large animal model demonstrate the therapeutic potential of in utero gene therapy for treatment of prenatally-diagnosed lung diseases. [1Davey et al., Gene Ther. 2012 Feb;19(2):201-9]
802. Large animal models of rare diseases: generation of a cohort of genetically modified swine by lentiviral transgenesis
Nicolas Grandchamp1, Corinne Kostic2, Simon Lillico3, Jacques Mallet4, Chamsy Sarkis1,4, Bruce Whitelaw3, Yvan Arsenijevic2
1NewVectys, France; 2UNIL, Switzerland; 3The Roslin Institute, UEDIN, UK; 4CRICM, France
Studies on large animal, although crucial for the understanding of physiopathological conditions and therapy development are mostly restricted to wild type animals, while studies in diseased background are usually limited to rodents. Thanks to the recent developments of engineering tools and reproductive methods, it is now feasible to generate genetically defined animal models in large species, opening the way to higher quality models for biomedical research and development. Large animal models are of particular interest in fields where rodents do not satisfy expectations, either in terms of size, life expectancy, physiology or biochemistry. One of these fields is ophthalmology, in particular diseases affecting cone photoreceptors, which are rare in rodent retina. To this regard, swine is particularly suitable to study cone diseases: the cone/rod ratio is closer to human than the rodent ratio and most importantly swine is the only non-primate specie containing a cone-rich reg ion resembling the human macula. We thus developed a swine cone dystrophy transgenic model. We first constructed the lentiviral vector using a dominant negative allele of the human Gucy2D gene, involved in cone dystrophies (GUCY2DE837D/R838S) under the transcriptional control of the swine Arrestine 3 promoter, specific of cones. Arr3-GUCY2DE837D/R838S lentiviral vector was injected in the zona pellucida of swine zygotes, which were further reimplanted in 3 pseudopregnant females. 30 animals were born from this experiment, including 19 transgenic animals with 1 to 6 copies of integrated vector (63% of transgenic). Transgene expression was further confirmed by RT-PCR. These animals were submitted to various physiological, behavioral and histological examinations during two years: electroretinography (ERG), behavioral tests, optical coherence tomography (OCT) and histological analyses. Results of ERG and behavioral tests reveal that transgenic animals have impaired visual function. Anatomic and histologic analyses by OCT measurement and on cryosections show no severe loss of photoreceptors, consistent with what is described in patients. However, significantly increased numbers of ectopic nuclei, localized in the photoreceptor segment layers, are detected in transgenic animal retinas. Most interestingly, the cohort of transgenic animals displays phenotypic variation as related to severity of the alteration of visual function, with almost unaffected to severely affected individuals, reflecting the heterogeneity observed in the human pathology. Lentiviral-directed transgenesis is a rapid and straightforward method to engineer transgenic pigs with dominant negative mutation, allowing modeling a physiopathologic condition with a wide range of phenotypes, reproducing the heterogeneity observed in patients. This approach could be used to generate animal models to study diseases which are poorly modeled in rodent, such as diseases affecting the central nervous system, the immune system or the cardiovascular system as well as cancers. Swine transgenic models will certainly contribute to improve the quality of preclinical studies, particularly in the cell and gene therapy fields where the size of the animal model is a limiting factor of success.
803. Liver gene therapy for PKU using minicircle-based naked-DNA vectors
HM Viecelli1, RP Harbottle2, M Chuah3, T VandenDriessche3, CO Harding3 and B Thöny1
1Department of Paediatrics, University of Zurich, Zurich, Switzerland; 2National Heart and Lung Institute, Imperial College London, London, United Kingdom; 3Free University of Brussels, Division of Gene Therapy & Regenerative Medicine, Brussels, Belgium; 3Departments of Molecular and Medical Genetics and Paediatrics, Oregon Health & Science University, Portland, Oregon, USA.
We have previously reported long-term correction of hyperphenylalaninemia and hypopigmentation of the PKU mouse model, C57Bl/6-Pahenu2, after liver-directed gene transfer with recombinant adeno-associated viral (AAV) vectors. However, questions of expression stability, treatment toxicity, potential for insertional mutagenesis, and safety required for targeting newborn and paediatric patients for potential life-long treatment remain a risk for viral vector-dependent approaches. To overcome these drawbacks, we are developing non-viral gene transfer methods for liver targeting. Here we report the successful use of minicircle (MC) technology to treat murine PKU. Our MC-DNA vectors contain a de novo designed liver-specific hybrid promoter-enhancer fragment, the mouse phenylalanine hydroxylase (mini-) gene based on the endogenous cDNA (mPah), or a codon-optimzed and intron-containing version (mcoPah), plus the bovine growth hormone polyA signal. Delivery of MC vectors was mediated by hydrodynamic tail vein (HTV) injection as a liver-targeted approach. MC-Vector titration studies showed that the blood phenylalanine levels were normalized in a dose dependent manner concomitant with reversion of hypopigmentation at optimal MC-DNA concentrations for up to several months without reapplication (ongoing experiment). Upon scarifying treated PKU mice, sustained transgene expression was confirmed by the presence of hepatic PAH enzyme activity. In conclusion, MC technology offers a better safety profile and has the potential for gene-therapeutic treatment of liver diseases.
804. Intra-patient variations in type 1 diabetes-specific iPS cell differentiation into insulin-producing cells
Tayaramma Thatava1, Yogish C. Kudva2, Ramakrishna Edukulla1, Karen Squillace1, Andre Terzic3, Yasuhiro Ikeda1
1Department of Molecular Medicine, 2Endocrinology, 3Division of Cardiovascular Diseases, Mayo Clinic, Rochester, MN, USA
Background: Nuclear reprogramming of adult somatic tissue enables embryo-independent generation of autologous, patient-specific induced pluripotent stem (iPS) cells. Exploiting this emergent regenerative platform for individualized medicine applications requires the establishment of bioequivalence criteria across derived pluripotent lines and lineage-specified derivatives. Our aims are to evaluate the utility of type 1 diabetes (T1D)-specific iPS cells for translational applications. Methods: From individual patients with type 1 diabetes (T1D) multiple human iPS clones were produced and prospectively screened using a battery of developmental markers to assess respective differentiation propensity and proficiency in yielding functional insulin-producing progeny. Results: Pluripotency expression patterns and the capacity to differentiate into SOX17- and FOXA2-positive definitive endoderm-like cells were comparable among individual iPS clones. Notable intra-patient variation was evident upon further guided differentiation into HNF4 alpha- and HNF1 beta- expressing primitive gut tube, and INS- and GCG-expressing islet-like cells. Differential dynamics of pluripotency-associated genes and pancreatic lineage-specifying genes underlined clonal variance. Successful generation of glucose-responsive insulin-producing cells required silencing of stemness programs and induction of stage-specific pancreatic transcription factors. Conclusions: Comprehensive fingerprinting of individual clones is mandatory to secure homogenous pools amenable for diagnostic and therapeutic applications of iPS cells from patients with T1D.
805. Directed Evolution of AAV in Pig Airway Epithelia In Vivo
David D. Dickey1, Boulos S. Nassar1, Katherine J.D.A. Excoffon2, James T. Koerber3, Jamie M. Bergen3, David V. Schaffer3, Joseph Zabner1
1University of Iowa, Iowa City, IA; 2Wright State University, Dayton, OH; 3University of California, Berkeley, CA
Recently, a cystic fibrosis (CF) pig model was developed that demonstrates many of the hallmarks of the human disease, including lung disease. We therefore hypothesized that directed evolution of adeno-associated viruses (AAV) could be used to generate a novel AAV gene therapy vector for the airways of the CF pig. We used DNA shuffling and error-prone PCR of the cap genes of AAV1, AAV2, AAV4, AAV5, AAV6, AAV8, and AAV9, along with an AAV2 binding loop insertion library to create a library of approximately 10^7 unique AAV capsid variants. This virus library was pre-selected in vitro in pig airway epithelial cultures for three rounds, followed by mutagenesis for further diversification, and then subjected to three rounds of selection in vivo in the lungs of wild-type pigs. We recovered a single variant that was entirely composed of the AAV2 cap gene with five point mutations, which we called AAV2H22. We found that AAV2H22 transduced pig airway epithelial cultures in vitro approximately 240-fold better than AAV2. In contrast to AAV2, AAV2H22 bound the apical surface of pig airway epithelia with specificity, suggesting altered receptor binding, although, in cells lines, heparan sulfate proteoglycan is still be utilized as a receptor for the virus. AAV2H22 transduced pig alveolar epithelia much less efficiently than AAV2, and did not demonstrate increased transduction of human airway epithelia compared to AAV2, suggesting that AAV2H22 is optimized specifically for the transduction of cells in the airway epithelia of pigs.
806. Long-Term Rescue of Skeletal and Cardiac Muscles in Dystrophic Mdx Mice by Peptide- Conjugated Morpholino
Bo Wu1, Peijuan Lu1, Caryn Cloer1, Mona Shaban1, Snimar Grewal1, Stephanie Milazi1, Sapana N Shah1, Hong Moulton2, Qi Long Lu1
1McColl-Lockwood Laboratory for Muscular Dystrophy Research, Neuromuscular/ALS Center, Department of Neurology, Carolinas Medical Center, Charlotte, NC, USA; 2Department of Biomedical Sciences, College of Veterinary Medicine, Oregon State University, Corvallis, OR, USA
Exon skipping has been demonstrated with capability to correct frame-shift and nonsense mutations of Duchenne Muscular Dystrophy (DMD). PhaseI/II clinical trials in UK and Netherlands have reported induction of dystrophin expression in muscles of DMD patients by local and systemic administrations of both phosphorodiamidate morpholino oligomers (PMO) and 2'O methyl phosphorothioate (2OMePS). Long-term rescue of dystrophin and muscle functions has also been achieved with PMO, but at considerable higher doses in animal models systemically than those reported in the clinic trials. Peptide-conjugated PMO (PPMO) offers significantly higher efficiency than PMO with the ability to induce near normal levels of dystrophin and restore functions in both skeletal and cardiac muscles. Here, we examined the long-term (one year) efficacy and toxicity of systemic administration of PPMO targeting exon 23 of dystrophin gene in mdx mice. Half life of the dystrophin expression was about 2 months in skeletal muscles but shorter in cardiac muscle. Biweekly injection of 1.5 mg/kg PPMO induced less than 5% dystrophin expression in skeletal muscles with limited muscle function improvements and no dystrophin in the cardiac muscle. Monthly injections of 30mg/kg PPMO restored dystrophin to more than 50% normal levels in all muscles including cardiac muscle. The levels of serum CK were reduced to near normal level with muscles showing significantly decreased CNF with no inflammatory cell accumulation and muscle fiber population became highly uniform. Significant improvement in skeletal muscle function was clearly observed by grip force measurement. Our result demonstrated for the first time that long-term repeated administration of PPMO could be safely applied to achieve significant therapeutic effect for long-term treatment of DMD with tolerable toxicity.
807. Long-term Transgene Expression with Rational Designed AAV Vector Demonstrates Similar Tropism across Multiple Species
R. Jude Samulski1, Alice F. Tarantal2
1Gene Therapy Center, University of North Carolina at Chapel Hill, Chapel Hill, NC; 2Center for Fetal Monkey Gene Transfer for Heart, Lung, and Blood Diseases, California National Primate Research Center, University of California, Davis, CA.
Adeno-associated virus (AAV) vectors are demonstrating therapeutic effect in a number of Phase I clinical trials. Although AAV has very broad tropism, most humans have neutralizing antibodies that restrict vector re-administration. To further enhance AAV vector tropism and identify capsid variants that escape neutralizing antibody, we generated synthetic AAV2 capsid by replacing a hexapeptide sequence in a previously identified heparan sulfate receptor footprint with corresponding residues from AAV8. The AAV2/AAV8 chimera designated AAV2i8 has displayed an altered antigenic profile, readily traversed the blood vasculature, and selectively transduced muscle tissues including the heart with high efficiency in animal models. More remarkable, this capsid variant reduced hepatic tropism. In a parallel study, we demonstrated that a single amino acid (aa) insertion of Threonine into AAV2 capsid would convey high muscle transduction and changed the capsid immune profile as well. To explore whether the single aa insertion tropism in AAV2 was distinct from AAV2i8 capsid we inserted aa at 265 (AAV2i8D), and injected AAV2i8D vector encoding the firefly luciferase transgene into mice via the tail vein. Compared to parental AAV2i8, liver tropism was rescued and muscle transduction was preserved. To address whether unique pharmacokinetic, pharmacodynamic, and antigenic features of these synthetic AAV variants studied in mice translated across species, we screened 33 gravid nonhuman primates for AAV antibodies and identified six for study. Pregnancies were monitored sonographically during gestation and newborns were delivered at term by cesarean-section. Three AAV/luc vectors (AAV9, AAV2i8, and AAV2i8D) were administered intravenously at birth (N=2 per construct). Infants were raised in a nursery and transgene expression was monitored monthly by bioluminescence imaging (BLI) immediately after the intravenous injection of D-luciferin. Blood samples were also collected from a peripheral vessel monthly to monitor complete blood counts (CBCs) and clinical chemistry profiles. Animals remained healthy during the study period, to date, and two years post-vector administration health, growth, CBCs, and clinical chemistry panels were within normal limits for the age group. Vector transgene expression determined by BLI has persisted without significant decline over time. Our results suggest that rational designed AAV variants can induce identical long-term transgene expression without safety concerns across various species (mice to nonhuman primates). Further studies comparing species-specific vector transduction tropism and immune profiles after systemic injection are in progress.
Late Abstracts Poster Session II: Friday, May 18, 2012
808. Stable and Selective Transgene Expression in the Mammary Epithelium via In Utero Gene Transfer
Jesse D. Vrecenak, Miroslaw Kozlowski, Carlyn A. Todorow, Antoneta Radu, Alan W. Flake
Dept of Surgery and Center for Fetal Research, Children's Hospital of Philadelphia, Philadelphia, PA, USA
Fetal gene transfer represents a promising approach to achieve sustained transgene expression within the hormonally responsive mammary epithelium through transduction of mammary stem cells. We have previously demonstrated the ability to transduce a regenerative mammary epithelial stem cell population through early gestational intra-amniotic injection of lentiviral vector. Because the genes responsible for inherited forms of breast cancer are cell cycle regulators, selective expression in the mammary epithelium is a prerequisite for clinical genetic correction. In mice, the mammary ridge forms around the eleventh gestational day (E10.5-11); thus, nearly all epithelial and neural crest progenitors are exposed to vector delivered into the amniotic sac. We hypothesized that selective mammary expression could be achieved by the early gestational intra-amniotic injection of lentiviral vector through the use of a mammary-specific promoter. To test this hypothesis, lentiviral vectors were constructed using: 1) the entire Mouse Mammary Tumor Virus (MMTV) LTR truncated by 25 bp at the 3' end (courtesy Philip Leder); 2) an 890 bp fragment of the Whey Acidic Protein (WAP) LTR or; 3) a 415 bp fragment of the WAP LTR with a 50 bp truncation at the 3' end replaced by a 65 bp fragment of the CMV promoter including the TATA box, all obtained by PCR using primers designed based upon GenBank sequence data. For purposes of detection, all vectors expressed GFP. An in vitro vector screen was completed using the NK-J breast cancer cell line, and constructs shown to express GFP in these cultures were selected for in vivo testing. Vectors were injected into the amniotic cavity (0.35 μL at 5.4 x 108 pfu/mL) of the E8 Balb/c fetus under ultrasound guidance. Surviving female pups were harvested at seven weeks and the mammary glands surveyed for GFP expression by fluorescence stereomicroscopy and immunohistochemistry. Though the WAP promoter alone had no activity in vitro or in vivo, the version modified with CMV elements expressed GFP at low levels in NK-J cells. The MMTV promoter drove moderate expression in NK-J cells and mammary glands from intra-amniotic injected mice showed strong GFP expression in both ducts and glands by fluorescence stereomicroscopy. This vector showed very low level and sporadic expression in some cortical elements of the brain, with no expression in any other tissue, including those of epithelial lineage. Immunohistochemistry confirmed transduction of both mammary ducts and glands using anti-GFP immunoperoxidase. Taken in conjunction with our previous results, the widespread and persistent expression of our MMTV and WAP-CMV vectors throughout the mammary epithelium supports the ability to achieve expression limited to mammary progenitors via early gestational intra-amniotic gene transfer. Because the regulatory activity of many proto-oncogenes complicates the production of transgenic animals, the ability to selectively modulate mammary gene expression in this manner might allow tissue-specific manipulation and study of genetic elements important in breast development and oncogenesis. Such selective expression also represents a critical first step toward the application of in utero gene transfer techniques to the correction of inherited breast cancers, which might eventually enable fetal gene therapy to correct an adult disease.
809. T cell receptor Vβ repertoire plays an important role in graft-versus-leukemia effect to acute lymphocytic leukemia with TCF3-PBX1
Haruko Tashiro, Yoko Oka, Ryosuke Shirasaki, Takuji Matsuo, Tadashi Yamamoto, Nobu Akiyama, Kazuo Kawasugi, Naoki Shirafuji
Department of Hematology/Oncology, Teikyo University School of Medicine, Tokyo, Japan
Aims: Graft-versus-leukemia (GVL) effect is a typical model of leukemia-immunotherapy. We recently experienced one acute lymphocytic leukemia (ALL) patient (case 1) with TCF3-PBX1 who relapsed after allogeneic hematopoietic stem cell transplantation (HSCT). This patient achieved complete remission (CR) with a discontinuation of an immunosuppressant, and the patient's graft-versus-host disease (GVHD) was controllable. This patient is still alive and well. We analyzed blood T cell receptors (TCRs) obtained from case 1 to characterize GVL effect. Materials and Methods: The institutional ethical committee approved this study, and he was given informed consent. His leukemic blasts, blood after achieving CR, and liver biopsy specimen when his GVHD worsened, were collected. Blood lymphocytes were co-cultured with the irradiated his ALL blasts (GVL model) or liver cells (GVHD model) for 14 days with interleukin-2. Then TCR Vα and Vβ usages were analyzed with RT-PCR. When the preferential usage of Vα and Vβ was demonstrated, another ALL patient with TCF3-PBX1 (case 2) who was transplanted and kept in CR, and one ALL patient with TCF3-PBX1 (case 3) during chemotherapy were also analyzed. Results: In GVHD model and GVL one, TCR Vα and Vβ were oligoclonally used. Vβ 6, 7, 10, 12, and 14 were preferentially used in GVL model but not in GVHD. In case 2 TCR Vβ 6, 10, and 14 were preferentially used. In case 3 TCR Vβ 6, 10, and 14 were not detected. Case 3's ALL relapsed molecularly soon after collecting sample. Discussion: In GVL model TCR Vβs were preferentially used, which were not used in GVHD model. And, TCR Vβ 6, 10, and 14 were also used in case 2. Currently we isolate single T cell clone that expresses these TCR Vβs, and determine the leukemic cell-specific antigens. We also observe their GVL effects using leukemic TCF3-PBX1 blast-burden non-obese diabetes/severe combined immunodeficiency mouse in vivo. Our identified TCR Vβs may play an important role in keeping CR of ALL with TCF3-PBX1.
810. Engineered nucleases for gene therapy
Fayza Daboussi1, Fabien Delacôte1, Christophe Perez1, Mikhail Zaslavskiy1, Marianne Duhamel3, Christelle Rochon3, Rachel Macmaster1, Laurent Poirot2, Roman Galetto2, Julianne Smith2, Aymeric Duclert1, Georges Silva1, Frédéric Pâques1, Philippe Duchateau1
1Cellectis and 2Cellectis Therapeutics, Paris, France; 3Ectycel, Genopole, Evry, France
Engineered site-specific endonucleases have emerged today as a promising approach to gene therapy. By targeting specific sequences within the genome, they can stimulate either Homologous Recombination (HR) or Non Homologous End Joining (NHEJ) at a predefined location making possible a wide range of modifications including the correction of mutated alleles, the insertion of a functional coding sequence at a precise location or the generation of functional gene knockouts. We have developed a combinatorial method to entirely redesign the DNA-binding interface of the I-CreI LAGLIDADG meganuclease to create site-specific endonucleases cleaving human genes. A comprehensive analysis of over 35 engineered meganucleases indicates that high levels of genome modification can be obtained with a strong correlation between the efficiency of targeted mutagenesis (up to 6%) and homologous gene targeting (up to 15%). In addition, for optimal activity the chromosomal context of the target sequence should be taken into account as both chromatin accessibility and epigenetic modifications of the DNA target can play an important role in determining the overall efficacy of engineered meganucleases. Finally, we have compared the activity of engineered meganucleases to that of Transcription Activator-Like Effector Nucleases (TALENs), a novel class of sequence-specific nucleases created by the fusion of transcription activator-like effectors (TALEs) to the catalytic domain of an endonuclease. TALENs display similar levels of homologous gene targeting while targeted mutagenesis frequencies are higher. However, including DNA end-processing enzymes such as terminal deoxynucleotidyltransferase (Tdt) or the Trex2 exonuclease, a non-processive 3' exonuclease, can enhance targeted mutagenesis for both meganucleases and TALENs, allowing frequencies of greater than 30% to be obtained. Data will be presented demonstrating the effective use of both engineered meganucleases and TALENs for genome engineering and will discuss their potential application for therapeutic applications.
811. Formulation Characterization and Intracellular Trafficking of Hybrid Retroviral/Synthetic Gene Delivery Vectors
Rahul K. Keswani, Mihael Lazebnik, Daniel W. Pack
Department of Chemical and Biomolecular Engineering, University of Illinois, Urbana-Champaign, IL, USA
Our research is focused on developing improved gene therapy vectors through development of synthetic polymer-based – chitosan (χ) and lipid-based (φ) envelopes for Moloney Murine Leukemia Virus (MLV)-like particles (M-VLPs). M-VLPs are essentially intact viruses lacking the envelope protein most necessary for transfection, thus making them inactive. Both chitosan and liposomes composed of DOTAP, DOPE and cholesterol electrostatically associated with M-VLPs forming hybrid vectors capable of highly efficient transfections to the same order of magnitude as amphotropic retrovirus (MLV-A). Both χ/M-VLPs and φ/M-VLPs provided stable transgene expression over three weeks. χ/M-VLP were also much more resistant to the presence of serum proteins during transfections. Both hybrid vectors also were able to maintain significantly better transfection efficiency 12 days post formation whereas MLV-As lost > 99% efficiency over the same period. The uptake of χ/M-VLPs was primarily via endocytic pathways but intracellular trafficking was dependent on the pH of the chitosan used for forming the hybrid vectors. On the other hand, uptake of φ/M-VLPs was via both endocytosis as well as passive fusogenicity with the plasma membrane. However, successful gene delivery required an active endocytic pathway. Intracellular trafficking of φ/M-VLPs was dependent on the lipid composition. φ/M-VLPs also had significantly faster trafficking kinetics as compared to χ/M-VLPs but slower than MLV-A which was confirmed by inhibition of reverse transcription and visualization via confocal microscopy. It can be concluded that the presence of the synthetic component of the hybrid vectors not only provided differing optimal trafficking mechanisms as compared to MLV-A, but also modulated the kinetics of release of M-VLPs within the cell.
812. Integration targeting to the ribosome RNA genes by a novel fusion protein consisting of HIV-1 integrase and the homing endonuclease I-PpoI
Diana Schenkwein1,4,a, Vesa Turkki1,4,a, Mervi Ahlroth5, Nirav Malani6, and Seppo Ylä-Herttuala1–3
1Department of Biotechnology and Molecular Medicine, A.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio, Finland; 2Department of Medicine, University of Eastern Finland, Kuopio, Finland; 3Gene Therapy Unit, Kuopio University Hospital, Kuopio, Finland; 4Ark Therapeutics Ltd, Kuopio, Finland; 5Finnish Rheumatism Association, Helsinki, Finland; 6Department of Microbiology, University of Pennsylvania School of Medicine, Philadelphia, PA, USA; aEqual Contribution to the work.
Stable addition of genetic material into human cells is the prerequisite for the efficient treatment of genetic disorders by gene therapy. Although several new methods capable of site-specific gene insertion and targeted DNA modification have been created, a generally accepted genomic safe harbor (GSH) that would allow transgenes to be stably and reliably expressed without adversely affecting endogenous gene structure and function, remains to be identified. We suggest the ribosomal RNA (rRNA) genes as a new GSH for therapeutic gene integration and present a modified lentivirus vector that directs gene insertion into the aimed site. Vectors targeting the rRNA gene were created by fusing the homing endonuclease I-PpoI into the C-terminus of the Human Immunodeficiency virus (HIV-1) integrase (IN). A point mutation (N119A) was introduced into I-PpoI to abolish unwanted DNA cleavage by the endonuclease. Vectors containing the IN-I-PpoIN119A fusion protein were found capable of spe cific integration targeting into the rRNA gene. The typical lentivirus vector integration pattern was also changed at certain aspects, resembling that seen in cells depleted of the cellular integration interactor LEDGF/p75. These results demonstrate the feasibility of targeting lentivirus vector integration with IN fusion proteins, and suggest the use of rRNA genes as available GSHs. In addition, the IN-fusion protein platform can be used as an efficient means to simultaneously deliver functional meganucleases and expressed transgenes to the nuclei of transduced cells.
813. Toxicology studies in support of the UK CF Gene Therapy Consortium's Multi-dose clinical trial
G McLachlan, U Griesenbach, D Collie, JA Innes, T Higgins, S Cheng, R Scheule, EWFW Alton, JC Davies, SC Hyde, DR Gill, D Porteous, AC Boyd
On behalf of the UK CF Gene Therapy Consortium, Imperial College London, University of Edinburgh, University of Oxford, Genzyme
In preparation for a Multi-dose clinical trial to assess whether 12 monthly doses of the CFTR gene can improve CF lung disease we have conducted two toxicology studies (TS). In addition to a GLP-mouse TS study (outsourced to a CRO) we also performed an in-house study in sheep. Mice received 12 doses of pGM169 (a plasmid carrying the CFTR gene) complexed to Genzyme lipid GL67A at fortnightly intervals by inhalation; exposure to the complex was for 0.5 (low= ~5x human dose), 2 (mid= ~ 20x human dose) or 6 hr (high dose=~60x human dose). Controls were exposed to air for 6 hr. Standard non-invasive assessments at regular intervals and post-mortem toxicology were undertaken 2 weeks and 3 months after the last dose. Approximately 600 mice were included in the study. Anesthetised sheep (n=8) received 9 doses (consistent with EMEA guidelines) of aerosolized pGM169/GL67A at monthly intervals. Similar assessments were made as for the mouse TS, in addition to lung function measurements and bronchoalveolar lavage (BAL) at intervals throughout the study. The amount (mg/kg) of pGM169/GL67A delivered to sheep approximately mimics the proposed human dose. Anesthetised controls (n=8) received air only. Results: All mice tolerated the treatment well. In the high dose group (~60x human dose) small increases in lung weight and circulating neutrophils were seen 2 days after the last dose, but not in the cohort sacrificed 14 days after the last dose. Histology showed scattered alveolar macrophages in the low and mid dose groups 2 weeks after administration of the last dose. In the high dose group multifocal alveolar foamy macrophage accumulations and occasional inflammation were noted. The presence of fat in macrophages was confirmed by Oil Red O staining. There was no evidence of structural remodelling. All other organs were unremarkable. 3 months after administration, findings in the low dose were not different from baseline, but were still observed in the mid and high dose. All sheep tolerated treatment well. In contrast to mice, haematology 1 or 15 days after dosing did not show any treatment related effects. Neutrophils were transiently increased in BAL fluid 1 day after administration, but had returned to baseline after 15 days. Observed changes in lung function or gas transfer measurements were mild and not test item related. There were no test item related histological findings in any organ. Conclusions: All animals tolerated the treatment well. The transient and dose-related systemic inflammatory responses and drop in lung function observed in the single-dose phase 1 trial were not replicated in non-CF sheep and non-CF mice only developed mild systemic inflammation at the highest dose (~50x human dose) possibly highlighting species-specific differences or an increased response to the lipid/DNA complexes in the inflamed human CF lung. Both TS support progression into a Multi-dose CF gene therapy trial.
814. Exploring the potential of antisense oligo-induced exon skipping for treatment of Alzheimer's disease
Reeteka Sud, Gerard D. Schellenberg
Dept. of Pathology & Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA
Most therapies are directed at fixing “what's wrong”, and for good reason; since numerous human diseases are caused due to errors – be it a mutation, a faulty protein, a missing gene, etc. In stark contrast, the strategy we have designed for Alzheimer's disease treatment is based on introducing mistakes. Our targets are the transcripts coding for tau protein in neurons. Known for supporting transport within nerve axons, tau protein aggregates as tangles in diseased brains. Limiting the synthesis of tau, we propose, would reduce free protein available for aggregation. To this end, we designed antisense oligos to interfere with splicing of constitutive exons in tau-encoding gene, MAPT. High-throughput screening of antisense oligos was performed in human neuroblastoma cell lines. Oligos were ranked in their efficiencies in excluding targeted exons. We also quantified whether tau levels were decreased in oligo-treated paradigms. We discovered that forced exclusion of exon 0, 1, or 5 significantly downregulated tau expression. Amplification of cDNA from antisense-treated cultures verified the exclusion or skipping of the exon targeted. Decreases were evident both at the level of transcripts as well as protein. To investigate the reasons for tau downregulation, we sequenced the ‘modified' transcripts, and found that MAPT sequence downstream of the skipped exons contained premature stop codons. These data indicate that the ‘modified' transcripts may be flagged for degradation by cells' quality-control mechanisms, consequently downregulating tau protein levels. These findings signify the potential of exon skipping as a promising therapy for Alzheimer's disease.
815. Highly Efficient Gene Delivery to Hepatocytes In Vivo
Yunxia Hu1, Younjee Chung1, Yang Yang1, Yuhua Wang1, Feng Liu1, Leaf Huang1
1Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, USA
A multifunctional core/membrane type of nanoparticle is developed for highly efficient delivery of plasmid DNA to the nucleus of hepatocytes in vivo. Here, amorphous calcium phosphate (CaP) nanoparticles (NPs) work as the solid core to encapsulate DNA and peptide together, and then is coated with an asymmetrical lipid bilayer membrane on its surface and densely PEGylated with a mixture of DSPE-PEG and DSPE-PEG-ligand. In vivo results show that most of the injected nanparticles are uptaken by hepatocytes, not the Kupffer cells, of the liver, due to the presence of dense PEG brush on nanoparticle surface allowing RES evasion of the NP and also galactose ligands engineered on the nanoparticle surface for targeting the asialoglycoprotein receptor of hepatocytes. More important, the decapeptide CRRRRRRRRC (CR8C) trapped together with the plasmid DNA played an important role of assisting plasmid DNA into the hepatocyte nucleus in vivo and thus improving gene transfection to observe strong red fluorescence protein expression when RFP plasmid was loaded in the nanoparticles and i.v. injected into mice. As far as we know, this is the first successful plasmid DNA delivery to post-mitotic nuclei in the liver by using such small NPs. Work supported by NIH grants CA129835, CA149363 and CA151652.
816. Creation of a Novel AAV2 Vector Showing AAV9-Like Transduction Properties by Displaying a Galactose Binding Motif on the Capsid
Kei Adachi1, Tatsuji Enoki2, Yasuhiro Kawano1,2, Hiroyuki Nakai1
1Department of Molecular and Medical Genetics, Oregon Health & Science University, Portland, OR; 2Center for Cell and Gene Therapy, TAKARA BIO Inc., OTSU, SHIGA, Japan
Recombinant adeno-associated virus (AAV) has many serotypes, and each serotype exhibits unique biological properties when made into gene delivery vectors. Although much is known about serotype-specific biological properties, a question remains as to which amino acids in the AAV capsid proteins are responsible for manifesting each property. To mechanistically address this question in a comprehensive manner, we have recently established a new approach, named “AAV Barcode-Seq” (Adachi K. et al., Abstract 397, ASGCT2011). The AAV Barcode-Seq integrates next generation sequencing technologies into high-throughput reverse genetics studies. With this new approach, it has become possible to investigate multiple phenotypic changes caused by a mutation in each of all the amino acids in the entire AAV capsid region with minimum time and efforts. Therefore, the AAV Barcode-Seq can significantly advance our understanding of the capsid amino acid sequence-phenotype relationships (Adachi K. et al., Abstract 10, ASGCT2012). Here we show that the AAV Barcode-Seq successfully identifies a motif in the AAV9 capsid that binds to its primary cell surface receptor (i.e., terminal galactose). Subsequently we demonstrate that AAV2 vectors exhibiting AAV9-like biological properties can be designed and created based on the knowledge obtained by the AAV Barcode-Seq. In the study, we produced a total of 191 double alanine-mutated AAV9 to scan the entire region of the carboxy-terminal half of the AAV9 capsid (i.e., amino acid positions 356 through 736). To identify AAV9 capsid amino acids required for binding to terminal galactose in the cell surface glycans, CHO Lec2 cells that express terminal galactose on the cell surface were incubated with AAV9 mutant libraries at 4 °C to allow viral particles to bind to cells. The viral genomes in the cell surface-bound virions were recovered and subjected to the AAV Barcode-Seq analysis. We found that more than ten mutants show ed significant impairment in binding to Lec2 cells. Based on the amino acid evolutionary conservation data and their topological locations, we narrowed down the amino acids to those that potentially constitute the galactose binding motif. Then we investigated whether the AAV9's biological properties can be transferred to AAV2R585E, a heparin binding mutant of AAV2, by substituting amino acids in such a way that AAV2R585E capsid acquires the galactose binding motif we identified. The resulting AAV2R585E mutant, AAV2R585E.9, showed a dramatic increase in its transduction efficiency in Lec2 cells compared to the parental AAV2R585E, while the transduction efficiencies of both AAV2R585E.9 and AAV2R585E were equally low in Pro5 cells devoid of expression of terminal galactose. These observations provide strong evidence that the amino acids we identified in the AAV9 capsid in fact function as the galactose binding motif. Importantly, the new AAV9-like AAV2 mutant, AAV2R585E.9, exemp lifies a more comprehensive knowledge-based approach to design a new AAV capsid. Thus, our study demonstrates that the AAV Barcode-Seq has enormous potential to elucidate the capsid amino acid sequence-phenotype relationships and builds an intellectual basis for rational design of new rAAV vectors with the most desirable biological properties.
817. Inhibition of thrombosis by nucleic acid scavengers without increasing the risk of bleeding
Shashank Jain, George A. Pitoc, Eda K Holl, Jaewoo Lee, Bruce A. Sullenger
Division of Surgical Sciences, Department of Surgery, Duke University Medical Center, Durham, NC, USA
Management of thrombosis during cardiac interventions such as CABG and PCI and vascular diseases such as DVT remain challenging despite the development of various anti-coagulant and anti-thrombotic agents. Recently a role for nucleic acids and other polyphosphates, such as linear chains of inorganic phosphates, have been implicated in thrombosis. Inorganic polyphosphates are 60-100 mer molecules which are stored in the dense bodies of platelets and activated platelets secrete polyphosphates which facilitates thrombosis. Here we explored if the nucleic acid scavenger polymer, PAMAM G-3, could inhibit nucleic acid/polyphosphate- mediated thrombosis in vivo and in vitro. Treatment of wild type C57BL/6J mice with nucleic acid binding polymer PAMAM G-3 resulted in the complete prevention of thrombus formation in carotid artery due to FeCl3 induced injury. The mean time to occlusion of carotid artery when treated with control normal saline was 4 min. 30 sec. while in 11 of 12 mice carotid artery didn't occlude for 40 min. after the FeCl3 patch removal when mice were treated with PAMAM G-3 (20mg/kg). In another model of thrombosis, collagen/epinephrine induced pulmonary thromboembolism, more than 85% of PAMAM G-3 (20mg/kg) treated mice survived for more than 30 min. after the administration of collagen/epinephrine as compared, 100% of vehicle treated mice died within 4 minutes after the administration of collagen/epinephrine. Moreover, we evaluated effect of PAMAM G-3 in standard tail bleeding assay. As would be expected if nucleic acids and other polyphosphates activate coagulation through the contact pathway, no significant difference was observed in blood loss due to distal tail transection between vehicle and PAMAM G-3 treated (20mg/kg) mice. Together these results suggest that nucleic acid scavengers such as PAMAM G-3 can prevent thrombosis without causing significant bleeding and present therapeutic potential to be used for the treatment of thrombotic conditions.
818. GUCY2C-targeted chimeric antigen receptor-expressing T cells mediate anti-tumor therapy in a mouse model of metastatic colorectal cancer
Michael S. Magee1, Adam E. Snook1, Adam R. Hersperger2, Glen Marszalowicz3, Scott A. Waldman1
1Department of Pharmacology and Experimental Therapeutics, Division of Clinical Pharmacology, Thomas Jefferson University, Philadelphia, PA, USA; 2Department of Microbiology and Immunology, Thomas Jefferson University, Philadelphia, PA, USA; 3School of Biomedical Engineering, Science & Health Systems, Drexel University, Philadelphia, PA, USA.
GUCY2C is a membrane-bound cyclase selectively expressed on the apical surfaces of intestinal epithelial cells facing the intestinal lumen, sequestered from the systemic compartment via tight junctions. GUCY2C expression is maintained throughout colorectal tumorigenesis and has been found on >95% of metastatic colorectal tumors, making it an ideal target for cancer immunotherapy. Adoptive T cell therapy (ACT) is an emerging therapeutic approach which has been successful in early stage clinical trials for the treatment of melanoma and B cell leukemias. However, the isolation of human colorectal cancer-specific T cells required for ACT has not been achieved. Here mouse monoclonal antibodies specific for GUCY2C have been used to create Chimeric Antigen Receptors (CARs) that possess an antigen recognition domain derived from antibody variable regions and T cell receptor signaling domains conferring recognition of native surface GUCY2C, rather than GUCY2C peptide-MHC complexes. We hypothesize that GUCY2C-specific CAR-expressing T cells can mediate regression of systemic colorectal cancer metastases, without damaging normal intestinal epithelial cells, reflecting normal mucosal compartmentalization of GUCY2C. Here we use retroviral transduction to genetically modify mouse T cells to express these CARs. Indeed, GUCY2C-specific CARs induce T cell cytokine production as well as direct anti-tumor activity in a GUCY2C-dependent manner in vitro. Moreover, GUCY2C-specific CAR-expressing T cells mediate anti-tumor therapy in a mouse model of metastatic colorectal cancer in vivo.
819. Pentostatin-based Host Conditioning and Syngeneic Th1/Tc1 Cell Transfer Each Contribute to the Regression of Established Murine Prostate Carcinoma Tumors
Miriam Mossoba1, Tania Felizardo1, Jason Foley1, Jeffrey A. Medin2, Daniel H. Fowler1
1Experimental Transplantation and Immunology Branch, National Cancer Institute (NCI), Bethesda, MD, United States; 2Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada.
Introduction: Host conditioning is often used to enhance adoptive T cell therapy, yet the independent (direct) effect of conditioning on anti-tumor responses remains an important therapeutic consideration. In murine models, we have determined that the combination of pentostatin [P] with cyclophosphamide [C] acted synergistically to eliminate host T cells for the prevention of fully MHC-disparate hematopoietic stem cell graft rejection (Mariotti et al. 2011). Pentostatin is a purine analog that inhibits adenosine deaminase (ADA), while cyclophosphamide is a DNA alkylating agent. Here, we have evaluated whether this PC regimen might also directly mediate anti-tumor responses in a murine model of established prostate carcinoma; of note, there are no reports in the literature to indicate that pentostatin mediates anti-tumor responses against solid tumors. In addition, we evaluated whether ex vivo manufactured syngeneic rapamycin-resistant Th1/Tc1 (T1.R) cells might mediate anti-tumor responses after PC conditioning. Previously, we showed that T1.R cells were apoptosis-resistant and had increased in vivo expansion potential. We hypothesized that (1) because solid tumors reportedly have elevated levels of ADA, anti-tumor effects may be attained using a PC conditioning regimen; and (2) T1.R cells may induce an anti-tumor effect that is enhanced by the PC regimen. Methods: Mice were inoculated with 2x10^5 murine prostate tumor RM-1 cells. When tumors were measurable by caliper (~50 mm3), subsets of mice (n=5) were injected with Pentostatin (1 mg/kg; once every 3 days) and/or Cytoxan (100 mg/kg/day for 6 days). For adoptive transfer studies, at least 5 mice per cohort were then transplanted with 5x10^6 T1.R cells 24-48 hrs post-conditioning. Tumors and lymphoid organs were assessed post-transplant for tumor growth, cell lineage depletion, and T cell function. Results: PC conditioning of host mice induced marked depletion of lymphoid immune cells in the spleen, lymph nodes, and bone marrow relative to non-conditioned hosts and hosts conditioned with each agent alone (P<0.05). P/C conditioning also consistently led to stabilization in tumor growth relative to no conditioning (P<0.05). Adoptive T1.R cell therapy of PC-treated mice led to rapid and sustained reductions in tumor volume (P<0.05). T1.R cell therapy also resulted in elevated absolute numbers of splenic IFN-γ-producing CD4+ and CD8+ T cells relative to recipients of no conditioning (P<0.05). Conclusions: We have made the novel observation that PC chemotherapy directly mediates anti-tumor effects against non-hematopoietic tumor cells. Host immune depletion using PC thus offers the additional benefit of controlling aggressive solid tumor growth prior to adoptive T cell therapy. Subsequent adoptive transfer of rapamycin-resistant Th1/Tc1 cells further induces significant tumor volume reduction. In conclusion, syngeneic T1.R cell therapy after P/C conditioning represents a new immune therapy approach.
820. Gene-Modified Placenta-Derived Adherent cells (PDACTM) as a Payload Delivery Platform for Cancer Therapy: in vitro Characterization of TRAIL-Modifications.
Wei Liu, Robert Hariri, Stewart Abbot, Bitao Liang
Celgene Cellular Therapeutics, Warren, NJ, USA
Introduction: Stem cell-mediated gene delivery is emerging as a strategy to improve the efficacy and minimize the toxicity of current gene therapy approaches. Human placenta-derived adherent cells (PDACTM) are readily obtained, are well characterized, can be expanded in culture to clinically-relevant scales, are minimally immunogenic and can distribute to multiple tissues in vivo. Hence they may represent an effective platform for gene-modification and potential delivery vehicle for gene-based anti-tumor treatments. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) has been documented to specifically induce apoptosis in malignant human cells while leaving normal cells unscathed. Expression of TRAIL in PDACTM could represent an effective component of future gene-modified cell-based cancer therapies. Methods and results: PDACTM cells were transduced with TRAIL-LentiORF (PDAC-TRAIL) and RFP-LentiORF (PDAC-RFP) vector. Phenotypic characterization of PDAC-TRAIL cells showed a strong induction of TRAIL mRNA (3x105 fold increase) detected via qRT-PCR, and soluble TRAIL (100pg/ml vs. undetectable) determined via ELISA compared to PDAC-RFP. Flow cytometry analysis indicated expression of membrane-bound TRAIL. Fluorescence activated cell sorting distinguished surface-TRAIL-High (PDAC-TRAILhi) and surface-TRAIL-low (PDAC-TRAILlo) profile. Conditioned media from both PDAC-TRAIL and PDAC-RFP showed no influence on viability of recombinant human TRAIL (rhTRAIL)-sensitive colorectal carcinoma line HCT116 in vitro. Tumoricidal activity in co-culturing conditions evaluated using Annexin V and PI staining resulted in comparable dose-dependent antitumor activity on HCT116 cells by both PDAC-TRAILhi and PDAC-TRAILlo populations. Pre-incubation with anti-rhTRAIL monoclonal antibody (mAb) abolished PDAC-TRAIL-induced apoptosis by both populations to a similar extent, implying that PDAC-TRAIL cells cause apoptosis of tumor cells in close contact even with low level of membrane-bound TRAIL. Interestingly, rhTRAIL-resistant human glioma line U-87MG was sensitive to PDAC-TRAIL, and co-incubation with anti-rhTRAIL mAb protected U-87MG from PDAC-TRAIL-induced cell death. This suggests that in close contact, PDAC-TRAIL could mediate much more potent tumor killing than soluble TRAIL protein and may involve different mechanisms underlying apoptosis induction by soluble and membrane-bound format of TRAIL. Since 50% of tested tumor cells so far are TRAIL-resistant due to aberrant regulation of apoptosis signaling pathway, combination use of TRAIL with other anti-cancer agents was evaluated. A protease inhibitor (Bortezomib) and histone deacetylase inhibitors (hydroxamic acid, trichostatin A and suberoylanilide hydroxamic acid) enhanced cell death in HCT116 and U-87MG lines, and sensitized TRAIL-resistant colon cancer HT-29 and prostate cancer LNCaP lines to PDAC-TRAIL-induced apoptosis. Conclusion: PDACTM can be modified to deliver anti-tumor transgenes, or potentially other pharmacologically-relevant payloads to induce apoptosis of tumor lines in vitro, suggesting their potential use as an efficient payload delivery platform for cancer therapy.
821. Improvement and Comparative Analysis of Methods for Vector Integration Site Mapping and Quantification
Aaron R. Cooper1, Matteo Pellegrini2,3, Donald B. Kohn2,4,5
1Molecular Biology Interdepartmental PhD Program, 2Molecular Biology Institute, and Departments of 3Molecular, Cell & Developmental Biology, 4Microbiology, Immunology & Molecular Genetics and 5Pediatrics, University of California, Los Angeles
Vector integrations permanently alter the genomic content of transduced cells and their progeny. Mapping and quantifying the unique junctions between vector and host genome DNA can determine the locations of genomic modification and the relative abundances of the progeny of transduced cells. This approach is of special interest in clinical gene therapy trials, in which vectors have served as novel inappropriate cis-regulatory sequences for integration-proximal proto-oncogenes. Additionally, viral integrations can be exploited as tags for powerful clonal tracking experiments in stem cell biology and development. Numerous molecular techniques for amplification of vector-genome junctions have been advanced in the past decade, but a comprehensive comparative study is lacking. We prepared and analyzed several reference samples probing hundreds of genomic sites to assess the bias and sensitivity of several popular techniques, including restrictive linear amplification-mediated polymerase chain reaction (LAM-PCR), non-restrictive LAM-PCR (nrLAM-PCR), and newer techniques based on pre-shearing input DNA. All methods were adapted to an Illumina paired-end sequencing procedure prior to comparison. Our results resoundingly confirm the superiority of non-restrictive approaches with respect to bias, and we further find that an optimized nrLAM-PCR protocol exhibits the highest sensitivity in addition to the previously reported reduced bias relative to LAM-PCR. The lowest quantification bias, however, is obtained with shearing methods, which eschew sequence read counts in favor of fragment length counts. We conclude that the choice of protocol depends on the relative importance of quantification bias, sensitivity, and ease of sample preparation in the experimental goals.
822. Potent Inhibition of CXCR4- and CCR5- tropic HIV-1 infection by lentivirally co-expressed shRNAs targeting CCR5 and HIV-1
Gene-Errol E. Ringpis1,2, Saki Shimizu2,3, Hubert Arokium1,2, Yiming Xie1,2, Joanna Camba-Colon4, Maria V. Carroll4, Bryan P. Burke4, Dong Sung An2,3, Masakazu Kamata2,5, Irvin S.Y. Chen1,2
1Department of Microbiology, Immunology, and Molecular Genetics, David Geffen School of Medicine, UCLA, Los Angeles, CA, USA; 2UCLA AIDS Institute, Los Angeles, CA, USA; 3UCLA School of Nursing, Los Angeles, CA, USA; 4Calimmune, Inc, Los Angeles, CA, USA; 5Department of Hematology and Oncology, David Geffen School of Medicine, UCLA, Los Angeles, CA, USA.
Downregulation of the HIV-1 coreceptor CCR5 holds significant potential for long-term protection against HIV-1 in patients. However, CCR5 knockdown will not be effective against existing CXCR4-tropic HIV-1 and emergence of resistant viral strains. As such, combination approaches targeting additional steps in the virus lifecycle are required. We screened a panel of previously published short hairpin RNAs (shRNAs) targeting highly conserved regions of HIV-1 and identified an shRNA targeting the R-region of the HIV-1 long terminal repeat. Here, we report that co-expression of two shRNAs targeting 1. CCR5 and 2. the R region of HIV-1 within a single lentiviral vector stably down-regulates CCR5 and inhibits CXCR4- and CCR5- tropic HIV-1 replication. Transduction with vectors bearing both shRNAs inhibited CXCR4- and CCR5- tropic viral infection in cell lines and peripheral blood lymphocytes. No obvious cytotoxicity or interferon response was observed. Transplantation of vector-tra nsduced hematopoietic stem/progenitor cells (HSPC) into humanized bone marrow/liver/thymus (hu-BLT) mice resulted in stable gene marking and CCR5 downregulation in human CD4+ T cells in peripheral blood and systemic lymphoid tissues, including gut-associated lymphoid tissue. CXCR4- and CCR5- tropic HIV-1 infection was effectively inhibited in hu-BLT-derived human splenocytes ex vivo. These results demonstrate that simultaneous delivery of shRNAs targeting both CCR5 and HIV-1 into HSPC may be a potential therapeutic reagent against HIV disease.
823. Oncogenesis following vector delivery in the fetal mouse appears to be independent of WPRE-presence in vector constructs and supports this animal model for assessment of vector-associated genotoxicity
*Ali Nowrouzi1 and *Wing Cheung2, Tingting Li3, Xuegong Zhang3, Anne Arens1, Anna Paruzynski1, Simon N. Waddington4, Emma Osejindu13, Safia Reja13, Christof von Kalle5, Yoahe Wang6, Faisal Al-Allaf2,7, Lisa Gregory2, Matthew Themis11, Maxine Holder8, Niraja Dighe2, Alaine Ruthe2, Suzanne Buckley4, Brian Bigger9, Eugenio Montini10, Adrian J. Thrasher11, Robert Andrews12, Charles Coutelle2, **Manfred Schmidt1 and **ƒMike Themis2,13
1National Centre for Tumorigenesis (NCT), Heidelberg, Germany; 2Gene Therapy Research Group, Section of Cell and Molecular Biology, Imperial College London, London, UK; 3Bioinformatics Division, TNLIST and Department of Automation, Tsinghua University, Beijing, China and Peking University Medical School, Peking, China; 4Institute for Women's Health, University College London, London, UK; 5Division of Experimental Hematology, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, USA; 6Cancer Research UK, Queen Mary's School of Medicine & Dentistry at Barts & The London John Vane Science Centre, London, UK; 7Department of Medical Genetics, Faculty of Medicine, Umm Al-Qura University, Makkah, Saudi Arabia; 8Apoptosis and Proliferation Control Laboratory, Cancer Research UK, London, UK; 9MPS Stem Cell Research Group, Faculty of Medical and Human Sciences, University of Manchester, Manchester, UK; 10San Raffaele Telethon Institute for Gene Therapy, HSR-TIGET, Safety of Gene Therapy and Insertional Mutagenesis Research Unit, Milan, Italy; 11Molecular Immunology Unit, Institute of Child Health, London, UK; 12Wellcome Trust Sanger Institute, Cambridge, UK; 13Brunel Institute for Cancer Genetics and Pharmacogenomics, Division of Biosciences, Brunel University, Uxbridge, UK (* Equally contributing first authors to this research; ** Equally contributing senior authors; ƒTo whom correspondence should be addressed: Tel: +44 1895 267252; Fax +44 1895 274318; Email: Michael.themis@brunel.ac.uk)
We previously reported high frequency hepatocellular carcinoma (HCC) development in mice treated with EIAV SMART2/3 based non-primate lentivirus vectors (npLVs) but not with a primate HR'SIN-cPPT-S-FIX-W LV (pLV) vector. Because a truncated form of the X gene carried on the WPRE in the npLV had the potential to be expressed in contrast to pLV where mutations were present to prevent X gene expression it was suggested that the X gene, which in full-length form is known to be involved in HCC, was the cause of oncogenesis. This follow-up report shows WPRE mutagenesis in EIAV npLV does not prevent oncogenesis by this vector and HCC development also occurs at similar frequency using an FIV based npLV. Here we discuss the insertion site profiles of the npLV and pLV vectors and show differences in insertion site selection that are due either to vector insertion preferences by npLV for gene dense regions and highly expressed genes during fetal development or clonal outgrowth as result of npLV mediated genotoxicity. Overall, we observe genes with npLV insertions reduced in expression and, using microarray and Ingenuity pathway analysis (IPA) related genes in HCC networks appear differentially expressed also. Our data indicate the fetal mouse model as useful to determine gene therapy associated genotoxicity and to reveal genes involved in liver oncogenesis.
824. Proliferation of genetically modified human cells on electrospunnanofiber scaffolds
Mandula Borjigin, Bryan Strouse, Pawel Bialk, Chris Eskridge, Rohina Amiri, Jingwei Xie and Eric B. Kmiec
Delaware State University, Dover, DE, US
Gene editing is a process by which single base mutations can be corrected, in the context of the chromosome, using single-stranded oligodeoxynucleotides (ssODNs). The survival and proliferation of the corrected cells bearing modified genes however are impeded by a phenomenon known as reduced proliferation phenotype (RPP); this is a barrier to practical implementation. To overcome the RPP problem, we utilized nanofiber scaffolds as templates on which modified cells were allowed to recover, grow and expand after gene editing. Here, we present evidence that HCT116-19, bearing an integrated, mutated eGFP gene and corrected by gene editing, proliferate on polylysine or fibronectin coated polycaprolactone (PCL) nanofiber scaffolds. In contrast, cells from the same reaction protocol plated on both regular plain surfaces and polylysine (or fibronectin) coated plate surfaces do not proliferate. Therefore, growing genetically modified (edited) cells on electrospunnanofiber scaffolds promotes the reversal of the Reduced Proliferation Phenotype and increases the potential o gene editing as an ex vivo gene therapy application.
825. Next generation tools for in vitro and in vivo miRNA functional analysis
A.V. Vlassov, N. Puri, C. Burnett, M. Li, A. Cheng, S. Magdaleno
Life Technologies, Austin, TX, USA
MicroRNAs (miRNAs) are endogenous 20-25 nt RNAs that play important gene-regulatory roles in animals and plants by pairing to the mRNAs of protein-coding genes to direct their posttranscriptional repression. There is an urgent need to develop additional tools for miRNA research - from isolation through discovery, profiling, quantitation, validation, and functional analysis. Here we report in vitro and in vivo performance of the next generation miRNA mimics and inhibitors that feature unique chemical modifications to enhance their potency and specificity. miRNA mimics are 18-25 bp double-stranded synthetic molecules with limited chemical modifications, mimicking the natural microRNAs. Evaluation in >10 in vitro reporter and endogenous assays (including miR1, miR21, miR23a, miR19a, let7c, let7a) indicated that novel mimics are highly efficient and their star strand is completely inactivated by chemical modifications. When miR1 mimic was administered in vivo (with liposomal delivery reagent, at 7 mg/kg), >2 fold down-regulation of PTK9 mRNA target was detected in the liver. miRNA inhibitors are chemically modified synthetic oligonucleotides capable of binding and inactivating the natural miRNAs. Evaluation in >10 in vitro reporter and endogenous assays indicated that novel inhibitors feature extremely high potency balanced with specificity. When miR122 inhibitor was administered in vivo (with liposomal delivery reagent, at 3 x 7 mg/kg), 2.5-4 fold up-regulation of AldoA, Hfe2, Slc35, Lass6 mRNA targets and corresponding proteins was observed in the liver. Neither oligonucleotides nor delivery reagent induced toxicity or immune response in animals as evidenced by blood chemistry panel (ALT, AST, ALP, BUN, Total bilirubin, glucose), hematology (WBC, RBC, platelets, hematocrit, hemoglobin, mean corpuscular volume) and cytokines (TNF-alpha, IL-6) at 6h through 2 weeks post-injection. This abstract for an oral presentation will present the use of this technology in cancer models. In summary, our chemically modified miRNA mimics and inhibitors, coupled with lipid based nanoparticle formulations, provide the complete solution for miRNA based functional validation in vitro and in vivo.
826. Ubiquitous high-level gene expression in hematopoietic lineages provides effective lentiviral gene therapy of murine Wiskott-Aldrich Syndrome
Alexander Astrakhan1, Blythe D. Sather2, Byoung Y. Ryu2, Socheath Khim2, Swati Singh2, Stephanie Humblet-Baron3, Hans D. Ochs4, Carol H. Miao4, and David J. Rawlings1
1Department of Immunology, University of Washington, Seattle, WA, USA; 2Center for Immunity and Immunotherapies, Seattle Children's Research Institute, Seattle, WA, USA; 3Center for Cellular and Molecular Therapy, GIGA-R, University of Liege, Liege, Belgium; 4Department of Pediatrics, University of Washington, Seattle, WA, USA
The immunodeficiency disorder, Wiskott-Aldrich Syndrome (WAS), leads to life-threatening hematopoietic cell dysfunction. We used WAS protein (WASp)-deficient mice to analyze the in vivo efficacy of lentiviral (LV) vectors utilizing either a viral-derived promoter, MND, or the human proximal WASp promoter (WS1.6) for human WASp expression. Transplantation of stem cells transduced with MND-huWASp LV resulted in sustained, endogenous levels of WASp in all hematopoietic lineages, progressive selection for WASp+ T, NKT and B cells, rescue of T cell proliferation and cytokine production, and substantial restoration of marginal zone (MZ) B cells. In contrast, WS1.6-huWASp LV recipients exhibited sub-endogenous WASp expression in all cell types with only partial selection of WASp+ T cells and limited correction in MZ B numbers. In parallel, WS1.6-huWASp LV recipients exhibited an altered B cell compartment, including higher numbers of λ-light-chain+ naïve B cells, development of self-reactive CD11c+FAS+ B cells and evidence for spontaneous germinal center responses. These observations correlated with B cell hyperactivity and increased titers of IgG2c autoantibodies, suggesting that partial gene correction may predispose towards autoimmunity. Our findings identify the advantages and disadvantages associated with each vector and suggest further clinical development of the MND-huWASp LV for a future clinical trial for WAS.
827. Enhancing therapeutic effects of oncolytic or gene therapy agents through external regulation of gene expression
Hannah Chen1, Juan J Rojas1, Padma Sampath1, Stephen H Thorne1
1Hillman Cancer Center, Department of Surgery, University of Pittsburgh, Pittsburgh PA
Modulation of gene expression from oncolytic or gene therapy vectors is often desirable for the purpose of controlling therapeutic gene delivery, control of immune response, or as a means of rendering viral replication conditional, as a safety mechanism. Vaccinia virus is a potent oncolytic agent currently demonstrating significant clinical potential; however, its cytoplasmic mode of replication means it is unfeasible to introduce transcriptional control elements to regulate transgene or endogenous viral gene expression. This limiation could be overcome by exploring post-transcriptional modification of proteins. A novel strategy to regulate protein stability involves using a small protein domain termed a destabilizing domain (DD) to confer instability to a protein of interest by fusing the DD to the protien's N or C terminus. Instability is reversed upon the exogenous administration of a small molecule that binds specifically to the DD. Using this strategy, we have demonstra ted conditional production/destruction of various transgene products expressed from vaccinia virus, including reporter proteins and therapeutic molecules such as immunomodulatory cytokines both in vitro and in vivo. Enhanced antitumor activity can be achieved in mouse tumor models following systemic delivery of an oncolytic vaccinia virus harboring a conditionally regulated cytotoxic protein, demonstrating the potential of this strategy to augment oncolytic virotherapy by offering reversible control of the timing and level of specific protein activity.
828. Apolipoprotein (apo) B mRNA Specific Hammerhead Ribozymes, Delivered via Recombinant Adeno Associated Virus AAV2/8 Vector is a Promising Strategy for Atherosclerosis Treatment
Hersharan Nischal, Hua Sun, Ba-Bie Teng
Research Center for Human Genetics, Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston, Houston, TX, USA
Atherosclerosis is a leading cause of coronary heart disease and ischemic strokes. Elevated levels of apolipoprotein B (apoB) are positively associated with coronary heart diseases. Our laboratory has previously shown that specifically designed hammerhead ribozymes, RB1 (GUC2326 ↓) and RB15 (GUA6679 ↓) cleaved apoB mRNA in HepG2 cells very efficiently and decreased its levels by 75% (RB1) and 63% (RB15), respectively. Moreover, RB15, delivered via adenoviral vector into mice, decreased apoB mRNA expression by 80%. In the current study, apoB mRNA specific ribozymes (RB1, RB15 and the combination of RB1/RB15) were delivered in vivo into hyperlipidemic LDb (Apobec-/- Ldlr-/-) mice using the recombinant AAV2/8 vector. At 4 months after treatment, apoB mRNA levels in LDb mice were significantly decreased by 37% (RB1), 49% (RB15) and 50% (RB1/RB15) as compared to the control vector RB15mutant treatment. Although there was no significant change in the total cholesterol levels in the plasma on treatment, free fatty acids and triglycerides levels reduced significantly in the three treatment groups at 4 months after treatment. This was in contrast to the control group (RB15mutant) that showed an increase in cholesterol, free fatty acids, and triglyceride levels from day 0 to day 120. Quantitative enface analysis of the mouse aorta, collected on day 120, showed a marked decrease in the atherosclerotic lesions (RB1 decreased 64%, RB15 decreased 50% and RB1/RB15 decreased 63%, compared to RB15 mutant treated animals). Our analysis on the aortic sinus also shows decreased atherosclerotic lesions in RB1 and RB1/RB15 treated group. Taken together, our study indicates that decreasing apoB mRNA using specific ribozymes results in decreased atherosclerosis development in mice and this warrants a therapeutic investigation in hyperlipidemia.
829. Critical Variables Affecting Clinical-Grade Production of the Self-Inactivating Gamma-Retroviral Vector for the Treatment of X-linked Severe Combined Immunodeficiency
Johannes C.M. van der Loo1, William P. Swaney2, Elke Grassman1, Ami Terwilliger1, Tomoyasu Higashimoto1, Axel Schambach3, Salima Hacein-Bey-Abina4, Diana L. Nordling1, Marina Cavazzana-Calvo4, Adrian J. Thrasher4, David A Williams8, Lilith Reeves9, Punam Malik1
1Division of Experimental Hematology and Cancer Biology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH; 2Siteman Cancer Center, Barnes Jewish Hospital, Washington University St. Louis School of Medicine, St. Louis, MO; 3Institute of Experimental Hematology, Hannover Medical School, Hannover, Germany; 4Centre Immunodeficiency, UCL Institute of Child Health, London, UK, and Department of Immunology, Great Ormond Street Hospital NHS Trust, London, UK; 5Dept. of Biotherapy, Hopital Necker Enfants-Malades, and Clinical Investigation Center in Biotherapy, Groupe Hospitalier Universitaire Ouest, AP-HP, INSERM, Paris, France; 6INSERM U-768, Hopital Necker Enfants-Malades, AP-HP, Paris, France; 7Université Paris Descartes, Paris, France; 8Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA; 9Indiana University School of Medicine, Indianapolis, IN.
In two earlier trials, patients with X-linked Severe Combined Immune Deficiency (SCID-X1) were successfully cured following gene therapy with a gamma-retroviral vector (gRV) expressing the common gamma chain of the interleukin-2 receptor (IL2RG). However, 5 of 20 patients developed leukemia from activation of cellular proto-oncogenes by viral enhancers in the long-terminal repeats (LTR) of the integrated vector. These events prompted the design of a gRV vector with self-inactivating (SIN) LTRs to enhance vector safety. We have previously demonstrated that GMP-grade SIN gRV can be produced at high titers using transient transfection in bioreactors (van der Loo et al, Gene Ther 2011). Herein we report on the large-scale production of a clinical-grade SIN IL2RG gRV pseudotyped with the Gibbon Ape Leukemia Virus envelope for a new gene therapy trial for SCID-X1, and highlight variables that were found to be critical for transfection-based large-scale SIN gRV production. We demons trate that ammonia build-up from serum free medium (X-Vivo 10) that contains pre-added glutamine is detrimental to vector titer and ultimately led to loss of cells in the bioreactor post-transfection. Successful clinical production required careful selection of culture medium without pre-added glutamine, reduced exposure of packaging cells to cell-dissociation enzyme, and presence of cations in the wash buffer used post-transfection prior to the media change. Adjustment of these variables led to the successful production of the IL2RG gRV clinical vector that was high titer and transduced 68-70% Colony forming units derived from human mobilized peripheral blood CD34+ cells. Transduced cells engrafted in NOD.CB17-Prkdcscid/J (NOD/SCID) and NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) mice; and resulted in the production of T cells in vitro from human SCID-X1 CD34+ cells. The vector was certified and released for treatment of SCID-X1 in a multi-center international phase I/II trial which is currently recruiting patients in Paris, London, Los Angeles, Boston and Cincinnati.
830. Herpes Simplex Virus (HSV) Vector Mediated Enkephalin Expression Alleviates Pain in a Varicella Zoster Virus (VZV) Rat Model of Post-Herpetic Neuralgia (PHN)
William F. Goins1, Mingdi Zhang1, Michael Yee2, Jean-Marc Guedon2, James R. Goss1, Joseph C. Glorioso1, and Paul R. Kinchington1,2
1Department of Microbiology & Molecular Genetics and 2Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
Post-Herpetic Neuralgia (PHN) is an extremely painful disorder that is the result of Varicella zoster virus (VZV) reactivation from latency within peripheral nervous system neurons. PHN occurs in 20 to 33% of the estimated 1,000,000 herpes zoster cases occurring yearly in the US. The severity, incidence and duration of the pain response are dependent on increased patient age and decreased patient cellular immune status. The most debilitating aspect of PHN is allodynia, which is pain invoked by light innocuous tactile stimulation. Most drug therapies for VZV-induced PHN pain have only been partly successful with 30-50% of patients showing no improvement with any current approaches such as amytryptiline and Gabapentin that have worked in treating other pain states. Thus, PHN-related pain remains a significant public health issue in need of improved therapeutic approaches. The mechanism(s) by which VZV induces PHN are poorly understood as it remains unclear which VZV gene products play a role in this process in humans and this issue has been complicated by the lack of a good animal model of herpes zoster. We have further developed a preclinical rat model in which VZV induces prolonged pain behavior to gain insight into this process and begin to explore the use of the novel therapeutics. We have shown that rats reliably demonstrate extended thermal hyperalgesia (TH) and mechanical allodynia (MA) for 6-8 weeks following injection of cell-associated VZV into the rat hindpaw and transport of the virus back to the dorsal root ganglia (DRG) that innervate the injection site. L4 and L5 DRG that innervate the hind-paw contained VZV DNA and expressed limited VZV proteins in neurons at early times post inoculation, notably the major regulatory protein IE62 in the absence of VZV replication within the DRG. Although a minor immune infiltrate can be detected, L4-L5 DRGs show upregulation of various markers of neuropathic pain. VZV-PHN rats demonstrating characteristic MA and TH pain responses were injected with herpes simplex virus (HSV) replication-defective vectors expressing either the natural opioid molecule enkephalin that we have previously used to treat various rodent models of chronic pain, or with an isogenic control virus expressing the GFP reporter gene. We found HSV vector-mediated enkephalin significantly reduced VZV-induced PHN pain out to 1-month post vector injection into the ipsilateral hind-paw, whereas control vector showed no effect in altering the PHN pain response. Vector dosing and re-administration studies are currently underway. Based on the promising results of another enkephalin-expressing HSV vector in a Phase-I human clinical trial to treat cancer pain, it may be possible that this novel therapeutic regimen may in the future be amenable for treatment of zoster patients suffering from chronic retractable PHN pain.
831. EGF Modulates Transcriptional Activity of Egr-1 Promoter on Human Primary Fibroblasts
Francisco Martíne-F1,2; Araceli Barrera-L1, Hugo Sandoval-Z3, Christian Lomelí-R1, Catalina Machuca-R3, Erika Rivera-D4; Adrián Vizcaíno-D1, Juan Antonio Madinaveitia1, Hilda Villegas-C3
1Molecular Biotherapeutic Program, Skin & Tissue Bank, National Institute of Rehabilitation., Ministry of Health, México; 2Department of Pharmacology, School of Medicine, National University of Mexico; 3Department of Molecular and Cellular Morphology, National Institute of Rehabilitation, Ministry of Health, México; 4Intensive Care Unit, National institute of Neurology and Neurosurgery, “Manuel Velasco Suárez,” Ministry of Health, México
Introduction: Epithelial Growth factor (EGF) is a key protein for life balancing, cell differentiation and migration of fibroblasts and epithelial cells. Signal transduction pathway of EGF stimulates migration and contractibility of fibroblasts and regulates metal proteases activity involved for scaring and epithelization. In this regard, mitogenic activity is a final biological event for EGF and Egr-1; however, transcriptional regulation on the egr-1 promoter has not been described. Hereby we evaluate the effect of EGF on Egr-1 promoter in human primary fibroblasts transduced with adenoviral vector Adegr1- Luc. Materials and Methods: Cells and adenoviral vectors: No replicative recombinant adenovirus (AdEgr1-Luc) were packaged at large scale in HEK-293 cells and purified according to the current protocol of the core facility of PBM-INR, based on cesium chloride gradient protocol for in vivo application. Viral stock was titled by plaque assay and OD. Human primary fibroblasts (HPF) were obtained from skin foils samples of the Skin and Tissue Bank of the INR, based on collagenase digestion protocol. Dermal derived cells were cultured in D-MEM Media supplemented with 10% HI-FBS and antibiotics under standard culture conditions for 1 week and stored for experimental procedure. 5x104 fibroblasts were seeded/well. After 12 hrs, cells were infected with AdEgr1-Luc at 50 MOÍs during 2 hs in a serum reduced media. After infection, cells were keeped in 1% of FBS for 24 hours into environment standard conditions for culture, until exposition to EGF (5ng/ml). Protein extraction was performed at 2, 4 and 6 hrs based on Cell Glo Lysis Buffer (Promega corp.) and luciferasa activity was quantified using a multidetector DTX-880. Results: Luciferase activity assay of not transduced cells not shows luciferasa activity (0.33-0.5 LC/s). The EGF exposition of human fibroblasts transduced with AdEgr1-Luc induces positive activity of Egr-1 promoter. Luminescent activity was observed at 2, 4 and 6 hs (1,490 LC/S; 2904.66 LC/s and 29511.33 LC/s, respectively). Transcriptional activity induced by EGF (29,511.33 LC/s) was over the positive control with 10% of FBS (23,915.33 LC/s). Conclusion: EGF activates transcriptional activity of egr-1 promoter in fibroblast cells. However, this activity could be a final event of no genomic pathway, and could be discerned by other studies using proteomic tools. Acknowledgments: This research project is granted by the National Council of Science and Technology of México. Grant: FOSIS/CONACYT-161624 and the project PIFUT08-169.
832. Gene Expression From Pseudouridine and 5-methylcytidine Modified Messenger RNAs
Jiehua Zhou1, Julie R. Escamilla-Powers2, Anton P. McCaffrey2, John J. Rossi2
1Department of Molecular and Cellular Biology, Beckman Research Institute of the City of Hope, Duarte, CA; 2Trilink Biotechnologies, San Diego CA
Recently, there has been significant interest in the use of messenger RNA (mRNA) based expression systems for gene therapy applications as well as for the generation and manipulation of stem cells. Several groups have shown that mRNAs are attractive vehicles for therapeutic gene expression in mammals (Kormann et al. Nat. Biotechnol (2011) 29, 154; Kariko et al. Molecular Therapy (2012) Epub ahead of print). Additionally, Warren et al. demonstrated highly efficient induced pluripotent stem cell (iPSCs) generation by transfection of mRNAs encoding reprogramming factors (Warren et al. Cell Stem Cell (2010) 7, 618). Because they are RNAs, mRNAs do not have any risk of insertional mutagenesis and subsequent oncogenesis. For this reason, the authors suggested that iPSCs generated in this manner should be safer than iPSCs derived by plasmid transfection or viral transduction. A key insight for the development of mRNA expression system was the recognition that mRNAs induce innate immune responses in transfected cells. Kariko et al. showed that substitution of uridine and cytidine residues with pseudouridine and 5-methlycytidine dramatically reduced innate immune recognition of mRNA (Kariko et al. Molecular Therapy (2008) 16, 1833). They also showed that pseudouridine modified RNA was translated more efficiently and had increased nuclease resistance. These studies highlight the importance of development of stable, non immunogenic mRNA to support these applications. In this current study, we synthesized capped pseudouridine and 5-methlycytidine modified eGFP mRNAs at milligram scales. Fluorescence activated cell sorting (FACS) demonstrated >95% transfection in HEK-293 cells. Expression in human CEM T-cells and CD34+ hematopoietic stem cells was also examined. Transfected mRNAs showed a surprising long duration of expression (up to 8 days post-transfection). Gene therapy applications of mRNA will require scalable purification methods that are able to produce mRNAs at gram scales. Recently, it was shown that purification of mRNA by high-performance liquid chromatography (HPLC) dramatically reduced innate immune responses relative to unpurified mRNA (Kariko et al. Nucleic Acids Research 39, e142). Here we compare mRNA purified by classical silica membrane chromatography to HPLC purified materials.
833. Targeting Lung Cancer using GD2-specific T cells
Zakaria Grada, Alexia Ghazi, Claudia Gerken, Vita S Brawley, Donald R Shaffer, Eric Yvon, Helen E Heslop, Stephen Gottschalk, Nabil Ahmed
Center for Gene and Cell Therapy, Baylor College of Medicine, Houston, TX, USA
Background: Approximately 160.000 patients die annually of lung cancer, the leading cause of cancer death in the United States. GD2 is expressed in majority of primary lung cancer tissues, making it an attractive target for adoptive immunotherapy. The safety of targeting GD2 using genetically-engineered T cells was established through human studies performed by our group. Methods: Using flowcytometry and immunohistochemistry, we characterized the surface expression of GD2 on a panel of human small cell- and non-small cell- lung cancer lines. We genetically engineered T cells to express a third generation GD2-specific chimeric antigen receptor (CAR). To evaluate the in vitro effector functions of GD2-specific T cells, we performed cytotoxicity assays and analyzed the expansion and cytokine release of these cells against GD2-positive, GD2-negative lung cancer cell lines and the appropriate controls. Results: We identified a number of small cell- and non-small cell- lung cancer cell lines that uniformly express GD2 on the cell surface. Using a retroviral system we successfully transduced primary human CD3/CD28 T cells to express the GD2-specific third generation CAR 14g2a.CD28.OX40.zeta. Transduction rates ranged from 35% to 75%. In 51Cr release cytotoxicity assays, GD2-specific T cells lysed all GD2 expressing lung cancer cell lines. Analysis of supernatants from cocultures of GD2-specific T cells and GD2-positive lung cancer cell lines using ELISA revealed a Th1 cytokine release pattern. Mock-transduced T cells failed to exhibit any measurable effector functions. Similarly, GD2-negative cells failed to induce T cell activation. Conclusion: GD2 represents a promising target for adoptive immunotherapy in lung cancer.
834. Serial treatment with oncolytic adenovirus results in redistribution of T-cell subsets in humans and mice.
Anna Kanerva1,2, Petri Nokisalmi1, Siri Tähtinen1, Anniina Koski1, Lotta Kangasniemi3, Iulia Diaconu1, Sari Pesonen1, Vincenzo Cerullo1, Timo Joensuu4, Akseli Hemminki1,3,4
1Cancer Gene Therapy Group, University of Helsinki, Finland; 2Department of OB/GYN, Helsinki University Central Hospital, Finland; 3Oncos Therapeutics Ltd., Helsinki, Finland; 4International Comprehensive Cancer Center Docrates, Helsinki, Finland
(Disclaimer: A.H. is shareholder and consultant for Oncos Therapeutics Ltd. A.Ka., A.Ko., I.D. and S.P. hold Oncos Therapeutics options.)
We studied T-cellular immunological effects against tumor and viral epitopes following single or multiple administration of Ad5/3-D24-GMCSF (CGTG-102), a 5/3 capsid modified adenovirus coding for granulocyte macrophage colony-stimulating factor (GMCSF). CGTG-102 was used to treat patients with advanced solid tumors refractory to standard therapies in the context of an Advanced Therapy Access Program regulated by FIMEA according to EY 1394/2007. T-cell kinetics were studied further in a murine B16-OVA melanoma model serially treated with CGTG-102. The phenotypic panel of circulating white blood cells was analyzed in a cohort of single treated patients. 15 out of 23 patients displayed an increase in the total count of CD8+ cytotoxic lymphocytes, while only 6/23 showed an increase in CD4+ T-cells. The total CD3+ T-cell population increased in 10/23 patients. The mean ratios of CD8+/CD4+ and CD8+/CD3+ increased 46 % (P=0.117) and 10 % (P=0.235). Based on the trend for increase in CD8+ T-cells, adenovirus- and tumor-specific T-cell responses were investigated with ELISPOT. Induction of tumor-specific immunity in patients was estimated by detecting anti-survivin, anti-CEA+Ny-ESO-1 (pooled), anti-c-Myc +SSX2 (pooled), anti-MAGE-3 and anti-WT-1 T-cells. Increase in the blood frequency of anti-tumor T-cells against one or several of these epitopes was seen in 7/15 (47 %) single treated patients while the frequency was 1/4 (25 %) in the serial group. Putative evidence of trafficking of T-cells to the tumor (ie. decrease in frequency in blood) was seen in 8/15 (53 %) and 3/4 (75 %) of single and serial treated patients. There was 18/19 (95 %) concordance in behavior of anti-tumor and anti-viral T-cells in blood. In mice, tumors showed virus-induced accumulation of tumor-antigen specific cytotoxic T-cells over time, concomitant with an initial reduction in lymph nodes, spleen and blood. Later, T-cells increased also in these tissues. The data in both humans and mice is compatible with an initial virus induced recruitment of anti-tumor T-cells to the tumor, followed by subsequent amplification of the response. Moreover, the data suggest that analysis of anti-tumor T-cells in just blood is unreliable due to overlapping processes of trafficking to tumor and response amplification. Our results suggest that serial treatment could increase potentially beneficial immunological reactions, whereas first-in-humans correlation between anti-viral and antitumor T-cell responses provided tantalizing preliminary evidence of epitope spreading in the context of oncolytic virus treatment of humans. These results have led us to utilize multiple rounds of injection in our ongoing clinical trials with CGTG-102.
835. Prospects for the Gene Therapy of Friereich's Ataxia using Neurotrophic Factors
Yurika Katsu, Frida Loria and Javier Diaz-Nido
Departamento de Biologia Molecular (Universidad Autonoma de Madrid); Centro de Biologia Molecular Severo Ochoa (UAM-CSIC); CIBER de Enfermedades Raras (CIBERER)
Friedreich's ataxia (FA) is the most common hereditary ataxia and there is currently no effective cure or treatment. FA is caused by autosomal recessive mutations that produce a deficiency in frataxin, a protein which is thought to play an important role in the regulation of mitochondrial function. FA is characterized by a severe neurological deterioration resulting from neurodegeneration which predominantly affects the dorsal root ganglia, spinal cord, brainstem and cerebellum. Neurotrophic factors have been shown to inhibit neurodegeneration in a variety of experimental models of neurodegenerative diseases. However, the possible impact of neurotrophic factors into the progression and therapy of FA has not been studied so far. Here we have used a herpesviral vector carrying the cDNA encoding for brain-derived neurotrophic factor (BDNF) to drive its overexpression in neuronal cells and test for its effect on frataxin-deficient neurons. Gene transfer of BDNF to primary cultures of mouse neurons prevents the apoptosis which is triggered by the knockdown of frataxin gene expression. The neuroprotective effect of BDNF gene transfer is specific since it can be abolished by the treatment of neurons with a neutralizing antibody to BDNF or a chemical inhibitor of BDNF receptor (TrkB) kinase activity. Interestingly this neuroprotective effect of BDNF is also observed in vivo. The stereotaxic injection into the mouse cerebellar cortex of a lentiviral vector carrying a minigene encoding for a frataxin-specific shRNA triggers a frataxin deficit which is accompanied by significant neuronal apoptosis and a marked atrophy of Purkinje cells in parallel to a loss of motor coordination of treated mice as assayed by the rotarod test. Co-injection of a herpesviral vector carrying the cDNA encoding for BDNF efficently prevents both neuronal apoptosis and Purkinje cell atrophy in addition to rescue the ataxic phenotype. In view of these data, we suggest that BDNF gene transfer might be of therapeutic use in FA since BDNF may protect frataxin-deficient neurons from degeneration. A gene therapy approach based on the transfer of neurotrophic factor genes has also the advantage that the number of “rescued” neurons may be much higher than the number of “transduced” neurons, since neurotrophic factors are secreted proteins which may act on neurons other than those directly expressing it.
836. Broad Analysis of Naturally-Occurring C2H2 Zinc Finger Proteins Reveals a Multitude of Novel Recognition Motifs
Metewo Selase Enuameh1, Yuna Asriyan1, Adam Richards1, Victoria Hall1, Jessie A. Brasefield1, Matthew D. Basciotta1, Ryan G. Christensen2, Gary D. Stormo2, Michael H. Brodsky1,3, Scot A. Wolfe1,4
1Program in Gene Function and Expression, University of Massachusetts Medical School, Worcester, MA, 2Department of Genetics, Washington University School of Medicine, St Louis, MO, 3Department of Molecular Medicine, University of Massachusetts Medical School, Worcester, MA, 4Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, Worcester, MA, USA
Cys2His2 zinc finger proteins (C2H2 ZFPs) constitute the most prominent group of transcription factors in higher metazoans. As such a complete characterization of these C2H2 ZFPs and their associated cis-acting sequences is pivotal to fully annotate their genomes. As a first step in this process, we have characterized the DNA-binding specificities of 133 C2H2 Zinc finger sets from Drosophila melanogaster by using our bacterial one-hybrid system. This data set contains the DNA-binding specificities of 73 unique C2H2 ZFP gene products, 22 additional alternate splice isoforms and 38 ZFP finger subsets of the former group, which have been crucial in defining the correct orientation and register of the zinc fingers on their defined binding sites. This represents the largest block of characterized C2H2 ZFPs from any organism described to date. Analysis of this data enhances our understanding of DNA-recognition by C2H2 ZFPs, allowing new recognition helix specificity determinants to be elucidated that aid in discerning the recognition preferences for C2H2 ZFPs. Using this information we employed subsets of these Drosophila fingers in combination with our in-house archive of zinc finger modules to create ZFPs with novel DNA-binding specificity. These finger combinations can be used to create functional Zinc finger nucleases with novel recognition sequences.
837. Immunological behavior of long-term thoracic duct lymph drainage (LT-TDLD): I. Dissociation of immunodeficiency in cell mediated immunity
Hiromitsu Kimura
National Research Institute for Child Health and Development, Tokyo, Japan
Employing enhanced green fluorescent protein transgenic rat as a donor impact of long-term thoracic duct lymph drainage (LT-TDLD) on general immune suppression was examined, in particular, skin and bone marrow cell (BMC) allografts. Background: Early classical studies by J.L. Gowans and his associates demonstrated that inasmuch as lymphocytes responsible for allograft rejection in general belong to re-circulating white blood cells (WBC), a week chronic drainage of thoracic duct lymph (TDL) resulted in profound immunodeficiency. Nevertheless this is so limited that small rodents including mouse and rats are not able to enjoy the operation more than ten days. Indeed when major histocompatibility antigen complex (MHC) differences prevail, application of the TDL drainage failed to induce allograft tolerance. Indeed experimental setting has a limited value, i.e., successful allograft tolerance was only induced in MHC compatible combinations judged by test skin acceptance. With this respect we have established a simple modification of long-term TDL drainage method, more than 20 days to 40 days and its impact on physiological value was re-evaluated. Materials and Methods: Inbred strains of DA (MHC:RT1avl), LEW (MHC:RT1l) and their enhanced green fluorescence transgenic counter parts EGFP-LEW and EGFP-DA were employed. Surgical procedures; standard TDL drainage, adult thymectomy, full thickness skin, BMC, and heterotopic cardiac transplantation were likewise employed. Cell and molecular analysis; flowcytemetric analysis and quantitative PCR were employed. Reagents: Cytokine; human recombinant G-CSF, Antobody; rabbit anti asialo GM1 Results and Discussion; 20 to 30 days long-term TDL drainage resulted in both test skin as well as cardiac allograft acceptance in MHC incompatible combinations. This LT-TDLD combined with adult thymectomy became so efficient and stable that 20 days TDL resulted in nearly 100 % allograft acceptance in both skin and cardiac allograft even in strong MHC antigen differences prevailed. However, when it applied to allogenic BMC transplantation, cellular chimerism was not observed under the experimental conditions, i.e., 100 x 10^6 to 300 x 10^6 cell infusion. BMC transplantation combined with G-CSF therapy was not able to improve the creation of donor cell chimerism in peripheral blood at ten days following BMC transplantation. Preliminary attempts to combined therapy of LT-TDLD with anti-asialo GM1 antibody was able to induce stable cellular chimerism. Taken together our studies strongly indicates that T cells responsible for allograft rejection and natural killer (NK) cells responsible for allogeneic BMC belong to distinct population. Unlike naïve lymphocytes NK does not re-circulate.
838. Abstract Withdrawn
839. Down regulating HIV latency by transcription activation
Suresh K. Arya and Agnes Holczbauer
National Cancer Institute, National Institutes of Health, Bethesda, MD, USA
One difficulty in curing HIV/AIDS is the tendency of the virus to enter into latency, where minimal or no viral gene expression occurs. The sequestered virus is thus immune to chemotherapy. One way to subject the virus to therapy again would be to activate viral gene expression. Latency could be due to blockade both of transcriptional initiation and transcript elongation. We have tested this idea in a cell culture model of latency where lymphocytic Jurkat cells harbor latent env- and GFP+ provirus. We subjected these cells to treatment, singly and in combination, with inducers of initiation - NFkB activation with prostratin (5 uM), chromatin remodeling with hydroxamic acid (SAHA) (5uM), and promoter demethylation with Aza-deoxycytidine (AzaCdR) (1 uM)- and also deblocking of elongation by transactivation with Tat-1 by way of lentiviral vector (50 ul) transduction. The results (see Table) support the idea that latent virus expression can be activated and in some case it is advantageous to activate both transcriptional initiation and elongation.
The opinion expressed in this abstract are those of the authors and do not necessarily represent views of the National Cancer Institute.
840. Noval functions of Foxp3 isoforms in human CD4+T cells
Sunil Martin, James L Riley.
Abramson Family Cancer Research Institute, Dept. of Microbiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia PA, USA
Regulatory T (T-reg) cells play a major role in immune. The transcription factor Foxp3 is the master regulator of T-reg phenotype and function. Interestingly, murine T cells express a single isoform of Foxp3, whereas human T cells express two co-dominant versions of Foxp3 of which the smaller form is missing exon 2 (δ2). We have previously published that Foxp3δ2 fails to up regulate the Pim 2 kinase and others have reported that members of the ROR family bind to this region of Foxp3. Nevertheless, whether these isoforms differentially affect T-reg function is unclear. Here, we show that introduction of Foxp3 and Foxp3δ2 into primary human CD4+CD45RA+CD25- naïve T cells induced suppressive activity that was indistinguishable from natural T-regs. However, when the same vectors were introduced into CD4+CD45RA-CD25- memory T cells, we observed that the full-length Foxp3 construct was unable to introduce a suppressive phenotype whereas Foxp3δ2 retained its ability to promote suppressive activity. Furthermore memory cells transduced with full length Foxp3 abrogated the suppressive functions of naïve cells transduced with full length Foxp3 but not Foxp3δ2.Cytokine profiling indicated 3 fold increase in the IL-2+IFN-γ+ cells in the memory cells transduced with Foxp3 and neutralization of IFN-γ and IL-2 restored the suppressive phenotype of these cells. Altogether these data demonstrate an isoform dependent alternate function of Foxp3 in the CD4+ memory T cell compartment. These studies support the use of naïve T cells for strategies that transduce effector T cells with Foxp3 expressing vectors as a means to rapidity generate T-regs for use in adoptive T cell therapy. Moreover, these studies suggest that the Foxp3 isoforms expressed in humans do play distinct roles in modulating suppressive activity in T-reg subsets and alterations of the ratio of one Foxp3 isoform to another may have implications for tumor and autoimmune disease.
841. Clinical application of automation of phage display and SELEX technology, the basis for the development of targeted therapeutics
Mekhled Alzaza
Al-Ghad International Colleges for Applied Science, Saudi Arabia
Define a ligand/receptor-based molecular map of human vasculature, which have been useful in functional genomics and proteomics, to translate the automation of phage display and SELEX technology into clinical applications and turn out to be the basis for the development of targeted medicine. The capillary thermodynamic DSMA-4 will be used to illustrate the mechanism of binding HIV-1 Rev in HeLa cells to CMR1, importin 9 which are the major nuclear import receptors for HIV-1 Rev in HeLa cells. Then, by means of capillary electrophoresis and automation in combination with high throughput screening methods, the smart DNA aptamers with a wide range of predefined values of Kd, high selectivity and response will be generated. New therapeutics such as novel zinc-finger DNA binding proteins, zinc finger enzymes that modify the genome and antibodies which target angiogenesis of receptors, as well as, deliver radioisotopes to receptors will be created.This will allow to inhibit receptors and block the Nup 358(also known as RanBP2), the gateway and access DNA of HIV into the nucleus. Also, will destroy cancer cells by increasing of antibody-antigen interactions.
842. An Improved qPCR Method for Titering Recombinant Adeno-associated Virus
Feng Wang, Ruian Xu
Engineering Research Center of Molecular Medicine, Ministry of Education & Institutes of Molecular Medicine, Huaqiao University, Quanzhou, Fujian, China
Recombinant adeno-associated virus (rAAV) vector has been proved to be a promising gene delivery vector by a number of clinical trails. It is time now to establish a common and comparable titration method in order to elucidate the results of different rAAV clinical trails. qPCR has been the mostly applied method for quantifying rAAV genome. However, our evidence revealed that the traditional qPCR method underestimated rAAV titers, this was consistent with a previous report. Since inverted terminal repeat (ITR), the only one remaining element from wild AAV genome left in all rAAV vectors, could form special configurations that would impair qPCR titration. Especially for self-complementary rAAV (scAAV), it contains a mutant ITR in the middle of genome and the complementary sequence would induce head-to-head or tail-to-tail dimer structure. This led to a great variation in titration of scAAV genome. To avoid such a problem, an endonuclease Sma I whose recognizing site exists in ITR was used to digest the ITR before qPCR, the titer of scAAV was found to be increased and more stable. This improved strategy would become a more reliable qPCR method for titration of all rAAV genome.
Late Abstracts Poster Session III: Saturday, May 19, 2012
843. Lentiviral protein transduction with genome-modifying integrase-I-PpoI fusion protein
Vesa Turkki1,2,a, Diana Schenkwein1,2,a, Mervi Ahlroth3, Seppo Ylä-Herttuala1
1Department of Molecular Medicine and Biotechnology, A. I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio, Finland; 2Ark Therapeutics group Plc, Kuopio, Finland and London, UK; 3Suomen Reumaliitto Ry, Helsinki, Finland; aEqual contribution to work.
In addition to gene transfer, lentiviral vectors can also be harnessed for protein transduction purposes. We have previously presented a cis-packaging method for foreign protein incorporation into HIV-1 derived lentiviral vectors while retaining the gene transfer activity. Here we demonstrate the feasibility of the cis-packaging method in genome editing using sequence-specific integrase(IN)-I-PpoI fusion protein with DNA-cleaving activity. I-PpoI is a small dimeric homing endonuclease which has a 15 bp recognition site residing in the 28S rRNA gene of eukaryotes. The cytotoxicity of meganuclease was studied in different cell lines through protein transduction. To test the effects in target cells, third generation self-inactivating lentiviral vectors were produced in 293T cells with standard cotransfection of four plasmids, one of which contained the I-PpoI cDNA fused C-terminal to IN in the gag-pol sequence. Correct packaging of the fusion protein and its ability to cut its cellular target sequence were verified. The fusion protein was able to enter the nucleus and cause specific double-strand breaks. While testing the effects and cytotoxicity associated with DNA-cleavage upon transduction with IN-I-PpoI containing lentivirus vectors, we found the tested cancer cell lines to recover inefficiently from the meganuclease-induced double-strand breaks, whereas non-cancerous cells suffer only a minor loss of viability. The effect was compared to available literature on cell line characteristics. Notable cytotoxicity of IN-I-PpoI LVV for cancer cells lines in vitro led to the testing of the fusion protein in vivo, where the effect was pre served and growth of subcutaneous tumours in nude mice was restricted with an efficiency comparable to the well-studied thymidine kinase transgene combined with ganciclovir injections. These results strengthen the applicability of the cis-packaging method in genome editing and lentiviral protein transduction.
844. The Therapeutic Administration of CRAd loaded NSCs Shows Survival Benefits in Diverse Models of Glioblastoma Multiforme
Bart Thaci, Alex L. Tobias, Atique U. Ahmed, Yu Han and Maciej S. Lesniak
The Brain Tumor Center, The University of Chicago, Chicago, IL, USA
Background: Glioblastoma (GBM) remains amongst the most deadly types of human cancers, with a median survival rate of only ~15 months. The short prognosis for patients with GBM is do in part to several factors including their variable characteristic, phenotypic, and invasive profiles as well as the ability of a small population of cancer stem cells (CSCs) to become resistant to conventional therapies. Our group has taken advantage of the intrinsic tumor homing property of neural stem cells (NSCs) to deliver conditionally replicative adenoviruses (CRAds) to the disseminated tumor burden, leading to tumor lysis. Methods: We tested the potential of FDA approved NSC (HB1.F3-CD) loaded with CRAd-S-pk7 (5x105 cells loaded with 50IU/cell) to prolong survival of GBM bearing nude mice in four different glioma models. These glioma cell lines differ from each other in terms of vascularity, invasive potential and cancer stem cell content. U87 cell line forms highly vascularized xenografts in the mice brain; GBM43FL and GBM12FL are primary glioma cell lines that are kept propagating in mice so that their gene profile remains constant. Moreover, we tested our loaded carrier cells in xenografts established by CSC enriched populations. Results: The highest increase in survival was noted in the U87 xenografts. Peritumoral (PT) injection of loaded NSCs increased median survival by 49 days (***p<0.001) as compared to PBS treated animals. At the same time, since GBM is highly invasive and cells migrate away from the main tumor bed we tested the capability of our loaded NSC to target distant foci by injecting them in the contralateral hemisphere of U87 xenografts bearing mice. Although only about 20-30% of the loaded NSC migrated to the glioma bearing hemisphere, they were enough to increase median survival by 44 days (*p<0.05). When injected in mice brains bearing GBM43FL xenografts, established by implanting low (5x103) or high (3.5x105) amount of cells, our loaded NSC injected PT increased median survival by approximately 7 days (**p=0.0013, and **p=0.0116, respectively). Also, recapitulating U87 findings, contralateral injection of therapy increased median survival similar to the peritumoral administration (6.5 days; *p=0.0342). In the case of the orthotopic xenograft established with 1.5x105 GBM12FL cells, a human derived glioma cell line expressing wild-type EGFR, the median survival was increased by 11 days (*p<0.05). Lastly, loaded NSCs prolonged survival in glioma bearing mice established with cells enriched for CSCs, specifically CD133+ GBM43FL cells. Sixty percent of the mice in the CRAd-S-pk7 loaded NSC group were still alive while all the PBS treated mice had succumbed to the disease (*p<0.05). Conclusion: Based on these different orthotopic xenograft glioma models and alternative injection protocols, representative of the wide variability amongst GBM in humans, NSCs loaded with CRAds are claiming their right to be tested in clinical trials. More importantly, the variability between pre-clinical models should be recognized and addressed properly before advancing to the clinic.
845. Combined preconditioning and in vivo chemoselection with 6-thioguanine: strategies for reduction of HPRT expression to allow selection of genetically modified hematopoietic stem cells in vivo
Katrin Hacke1, Janet A Treger1, Robert H Schiestl2, Alok V Joglekar3, Megan D Hoban3, Roger P Hollis3, Gay M Crooks2, Dong Sung An3, Scott G Kitchen3, Donald B Kohn3, Andreas Reik4, Pei-Qi Liu4, Philip D Gregory4, Michael C Holmes4, Noriyuki Kasahara1
1Department of Medicine, David Geffen School of Medicine, University of California Los Angeles Los Angeles, CA; 2Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA; 3Department of Microbiology, Immunology and Molecular Genetics, University of California Los Angeles Los Angeles, CA; 4Sangamo BioSciences, Richmond, CA
Purine analogs such as 6-thioguanine (6TG) cause myelotoxicity upon conversion into nucleotides by hypoxanthine-guanine phosphoribosyltransferase (HPRT). We have recently developed a novel and highly efficient strategy using 6TG as a single agent for both preconditioning and in vivo chemoselection of HPRT-deficient hematopoietic stem cells (HSC). At appropriate doses, 6TG induced selective myelotoxicity without any adverse effects on extra hematopoietic tissues in HPRT wild-type mice, while HSC from HPRT deficient mice were highly resistant to its cytotoxic effects. In a murine model of bone marrow (BM) transplantation, combined 6TG conditioning and post-transplantation chemoselection consistently achieved ~95% engraftment of HPRT-deficient donor BM, without significant toxicity and in the absence of any other cytotoxic conditioning regimen. Long-term reconstitution of immunophenotypically normal BM was achieved in both primary and secondary recipients, indicating that 6TG selection amplified the self-renewing, pluripotent HSC population from HPRT-deficient donor BM. We are now pursuing different strategies for reducing HPRT expression and thereby conferring protection against 6TG myelotoxicity to primary murine and human hematopoietic progenitor cells. Accordingly, we have constructed and validated a lentiviral vector expressing short-hairpin RNA (shRNA) that targets a consensus sequence present in both human and murine HPRT genes, and which can achieve highly effective and long-term knockdown of HPRT expression in cells from either species in vitro. Furthermore, we have developed an alternative approach using engineered zinc finger nucleases (ZFN) to target the mouse and human HPRT loci for gene disruption. Enhanced gene modification rates of up to >80% were observed in human CD34+ cells after HPRT disruption and 6TG selection in vitro. In on-going studies, both strategies are currently being tested in conjunction with ex vivo genetic correction for application to HPRT-wt donor-derived hematopoietic progenitor cells transplanted and amplified with combined 6TG preconditioning and in vivo chemoselection.
846. Integrase stability and topography of nuclear integration regulate wild type HIV-1 and HIV-1-based vector integration
Bruna Marini, Hashim Ali, Marina Lusic, Mauro Giacca
Molecular Medicine Laboratory, International Centre for Genetic Engineering and Biotechnology (ICGEB), Trieste, Italy
A number of studies performed over the last several years have indicated that HIV-1 and HIV-1-based vectors preferentially integrate into actively transcribed genes and have indicated that some cellular factors assist the viral cDNA in the integration process. The reasons for this preference and the molecular mechanisms involved, however, have been elusive so far. Here we explore the role of nuclear topology in HIV-1 integration site selection. We show that the virus preferentially integrates into the active genes that are localized in the periphery of the nucleus and that the presence and integrity of viral integrase is essential in this process. By 3D Immuno-DNA FISH (Fluorescence In Situ Hybridization) with DNA probes corresponding to the human genomic regions that are mostly targeted by HIV-1 integration, we discovered that most of these regions are preferentially located in the outer shells of the nucleus. Here, HIV-1 avoids the Lamin Associated Domains (LADs), as revealed by bioinformatic analysis, while it specifically integrates into the DNA within the nuclear pore compartment. By chromatin immunoprecipitation using HIV-1 infected cells, we found that the RNA Pol II-bound, transcriptionally active HIV-1 provirus specifically associated with various nucleoporines of the inner basket of the Nuclear Pore Complex (NPC), some of which were already identified as important factors for integration. We have previously shown that stability of viral IN is essential for efficient viral integration and that this process depends on the specific phosphorylation of integrase Ser57 (Manganaro, L, et al. 2010. Nat Med 16, 329). Viruses carrying a mutation at integrase Ser57, or other mutations impairing integrase function, loose connection with the nuclear pore and are found as 2LTR circle episomes all throughout the nuclear space. An identical mislocalization occurs when the cells are infected with wild type HIV-1 and treated with the integrase inhibitor raltegravir. Collectively, these results indicate that the integrity of HIV-1 integrase is an essential determinant for HIV-1 positioning in close proximity to the nuclear pores and that this is mediated by the direct interaction of integrase with different nucleoporins. The significance of these findings with respect to the regulation of wild type HIV-1 and HIV-1 vector gene expression will be discussed.
847. Using Unlocked Nucleic Acid (UNA) Substitutions to Alter Strand Selection By the RNA Induced Silencing Complex. “Making a Bad siRNA Good.”
Nicholas M. Snead1, Julie R. Escamilla-Powers2, Sabrina Shore1, Natasha Paul2, John J. Rossi1, Anton P. McCaffrey2
1Department of Molecular and Cellular Biology, Beckman Research Institute of the City of Hope, Duarte, CA, USA; 2Trilink Biotechnologies, San Diego CA, USA.
Entry of the desired (antisense) strand into the RNA Induced Silencing Complex (RISC) is a critical determinant of small interfering RNA (siRNA) potency and specificity. For most synthetic siRNA duplexes, both strands enter RISC to some degree. Undesired RISC sense strand entry can cause off-target silencing of genes with sense strand homology. Off-target effects can result from strand cleavage or microRNA-like interactions between the seed region of the incorporated strand and seed matches in off-target mRNAs. Here, we utilized Unlocked Nucleic Acid (UNA) substitutions to alter strand selection to favor incorporation of the desired antisense strand. UNAs are non-nucleotide analogs in which the C2'-C3' bond of the ribose ring is absent. Previous studies have shown that strategic UNA substitutions at the 5' end of the passenger strand and at position seven of the guide strand can reduce off-target effects by >90% (NAR (2011) 39, 1823). Previously, we conducted a detailed examination of the IC50 for the antisense and sense strands of a series of siRNAs targeting hnRNPH (NAR (2010) 39, 1510). We found that shifting an siRNA by a single nucleotide dramatically altered the relative potency of the antisense and sense strand as measured by luciferase target reporters. Here, we tested whether UNA substitution could “make a bad siRNA good.” As a test case, we chose hnRNPH siRNA H5, for which strand selection is highly biased toward undesired sense strand incorporation (IC50 of 71.6 pM for the sense strand and 9.75 pM for the antisense strand). 5' UNA substitution of the sense strand greatly reduced sense strand silencing while increasing on-target silencing by the antisense strand; in effect, converting a non-functional siRNA to a functional one. Thus, UNA substitution can be used drive incorporation of the antisense strand into RISC for improved on-target silencing and presumably reduced off-target effects by the sense strand. When targeting highly conserved viral regions or single nucleotide polymorphisms, the potential sequence space for siRNA selection is highly constrained. In these cases, UNA modification may be a useful tool to expand the selection of potential siRNA sequences. The antisense strand can also mediate off-target effects through miRNA-like interactions between the seed region of the antisense strand and seed matches in off-target mRNAs. Several reports have suggested that UNA modification of antisense position seven could reduce seed based off-target effects (NAR (2010) 38, 5761). In one report, combination of a UNA substitution at antisense position seven with 5' UNA sense strand modification, dramatically reduced off-target effects (NAR (2011) 39, 1823). Using antisense and sense strand reporters, we confirmed that UNA modifications at these positions did not compromise the activity of an ApoB siRNA.
848. Therapeutic potential of in vivo administered pseudouridine-containing mRNA
Katalin Karikó1, Hiromi Muramatsu1, Jason M Keller2, Drew Weissman1
1Perelman School of Medicine, University of Pennsylvania, Philadelphia, USA; 2National Institute of Dental and Craniofacial Research, NIH, Bethesda, USA
In vitro-transcribed mRNA has great therapeutic potential to transiently express the encoded protein without the adverse effects associated with viral- and DNA-based deliveries or direct protein supplementation. We have demonstrated that HPLC-purified pseudouridine-containing mRNA is non-immunogenic and translates very efficiently in cultured primary cells. We tested Transit-mRNA-complexed erythropoietin (EPO)-encoding mRNA in mice as a model therapeutic. We observed that a single injection of 100 ng (0.005 mg/kg) of mRNA elevated serum EPO levels significantly by 6 h and levels were maintained for 4 days. In comparison, mRNA containing uridine produced 10 to 100-fold lower levels of EPO that returned to baseline after 1 day. EPO translated from pseudouridine-mRNA was functional and caused a significant increase of both reticulocyte counts and hematocrits. Weekly injection of 100 ng of EPO mRNA containing pseudouridine was sufficient to maintain significantly increased hema tocrits without inducing the production of inflammatory cytokines or anti-EPO antibodies. Extending this approach to macaques, we could detect significantly increased serum EPO levels following intraperitoneal or subcutaneous injection of pseudouridine-containing mRNA coding for rhesus EPO. These results demonstrate that HPLC-purified, pseudouridine-containing mRNAs encoding therapeutic proteins have great potential for clinical applications.
849. One Step Human iPS Cell Induction and Enhanced Gene Targeting Using Viral Vectors And TALE Nucleases
Ko Mitani1, Emi Aizawa1, Claudio Mussolino2, Manami Ohtaka3, Ken Nishimura3,4, Toni Cathomen2, Mahito Nakanishi3
1Gene Therapy Div., Research Center for Genomic Medicine, Saitama Medical Univ., Saitama, Japan; 2Institute of Experimental Hematology, Hannover Medical School, Hannover, Germany; 3Research Center for Stem Cell Engineering, National Institute of Advanced Industrial Science and Technology (AIST), Ibaraki, Japan; 4Labratory of Gene Regulation, Faculty of Medicine, University of Tsukuba, Japan
Prolonged culture during induction and manipulation of induced pluripotent stem cells (iPSCs) has been shown to increase point mutations and copy number variations in the iPSC genome. This raises safety concerns for therapeutic applications of iPSCs, especially in gene repair therapy. In order to minimize cell culture for iPSC induction and gene repair, thus reducing the risk of introducing new mutations, we examined a strategy of simultaneous iPSC induction and gene targeting without cell cloning between the two processes. For this purpose, we utilized Sendai virus (SeV) vector-mediated iPSC induction and helper-dependent adenoviral vector (HDAdV)-mediated gene targeting. Male TIG3 fibroblasts were subject to iPSC induction by using the Sendai virus vector, SeVdp(KOSM). Instead of isolating and characterizing individual iPSC colonies before gene targeting, iPSC-like colonies were grown as a cell population and transducted with a HDAdV targeting the HPRT locus, HDAdV-hHPRT-neo. This HDAdV was designed to introduce the PGK-neo cassette into the exon 6. Cells were subjected to G418/6-thioguanine (6TG) selection to isolate gene-targeted iPSC colonies and correct targeting in G418/6TG double-resistant colonies was verified by PCR analyses. The gene targeting frequencies were 1.8 - 3.1 x 10e-7 per cell and 3 - 5% of chromosomal integration (G418-resistance). These results were similar to those in authentic human iPSCs, which were 1.9 - 6.3 x 10e-7 per cell and 3 - 12% per integration. From the start of iPS cell induction (infection with SeV) to isolation of drug-resistant colonies, it took only 6 weeks. We are currently characterizing pluripo tency and undifferentiated states of these potential gene-repaired iPSCs. Recently, it was reported that expression of tailor-made sequence-specific nucleases, such as zinc-finger nucleases and TAL effector nucleases (TALENs), introduce DNA double-strand breaks (DSBs) at the target chromosomal sites, thus stimulating homologous recombination-mediated gene repair. In order to further enhance gene targeting efficiencies, we tested a strategy of a combination of TALEN expression and HDAdV-mediated delivery of donor DNA for gene targeting at the HPRT locus in human iPSCs. A pair of TALENs was designed to introduce a DSB at the exon 6 of HPRT locus. Human iPSCs were transfected with the TALENs, followed by infection with the HDAdV-hHPRT-neo. The cells were then selected for G418 (to detect chromosomal integration) and 6TG (to detect HPRT knockout). In the absence of TALENs, only 1 out of 7 G418-resistant colonies were G418/6TG double resistant, while 3 out of 12 colonies were G418/6TG resistant by expressing TALENs. Correct targeting in all of the G418/6TG double-resistant colonies was subsequently verified by PCR analyses. These results suggest that, by combining efficient methods for iPSC induction and gene targeting, gene repair therapy using patients' iPSCs becomes more realistic.
850. Improved Hematopoietic Stem Cell (HSC) Based Gene Therapy
Benjamin Goebel1, Christian Brendel2, Sabrina Kneissl3
1Med. Clinic II, KGU, Frankfurt, Germany; 2Georg-Speyer-Haus, Frankfurt, Germany; 3Paul-Ehrlich-Institut, Langen, Germany
The potential of gene therapy as a curative treatment for monogenetic disorders has been clearly demonstrated in a series of recent phase I/II clinical trials. Generally, CD34+ cells are used for genetic modification in gene therapy trials. However CD34+ cells consist of a heterogeneous cell population including a large amount of cells with limited long-term repopulating capabilities, resulting in low long-term engraftment levels in particular in those diseases, in which gene modified cells lack a proliferative advantage over non-modified cells. Therefore modifications in gene transfer vectors and gene transfer strategies are required to improve engraftment levels and long-term clinical benefit in gene therapy patients. One particular attractive approach to solve this problem is the improvement of HSC based gene transfer by specifically targeting cells with long-term engraftment capabilities. We constructed lentiviral gene transfer vectors (LV) targeting specifically CD133+ cells, a cell population with recognized long-term repopulating capabilities. Targeting is achieved by incorporating the measles virus (MV) glycoproteins hemagglutinin, responsible for receptor recognition, and fusion protein into LVs. The hemagglutinin protein is blinded for its native receptor and displays a single-chain antibody specific for CD133 (CD133-LV). Transduction of CD34+ cells with CD133-LV vectors resulted in an expansion of gene marked CD34+ cells in vitro, while the number of VSV-G-LV transduced CD34+ cells decline over time. Competitive repopulation experiments in NSG mice showed a significantly improved engraftment of CD133-LV transduced HSCs. At 8 weeks post-transplantation, gene marked hematopoiesis in transplanted animals was dominated by the progeny of CD133-LV transduced cells in 11 out of 14 animals. This new strategy may be promising to achieve long-term engraftment in patients treated by gene therapy.
851. Transgene-Free iPSCs Generated from Small Volume Peripheral Blood Non-mobilized CD34+ Cells
Randall K. Merling1, Colin L. Sweeney1, Uimook Choi1, Suk See DeRavin1, Timothy G. Myers2, Francisco Otaizo-Carrasquero2, Jason Pan1, Gilda Linton1, Lifeng Chen1, Sherry Koontz1, Narda L. Theobald1, Harry L. Malech1
1Laboratory of Host Defenses and 2Research Technologies Branch, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA
A variety of somatic cells can be reprogrammed to induced pluripotent stem cells (iPSCs), but the small number of CD34+ hematopoietic stem cells (HSCs) present in non-mobilized peripheral blood (PB) would be a convenient and desirable starting target. We report here a simple method for targeting derivation of iPSC from non-mobilized PB CD34+ HSCs using immunobead purification and 2-4 day culture to achieve enrichment of CD34+ HSCs to 80±9%, followed by reprogramming transduction with loxP-flanked polycistronic (Oct4, Klf4, Sox2, and c-Myc) STEMCCA-loxP lentivector at an MOI of 2. Our yield was 4.7±2.2 iPSC colonies (n=12) per 20 mL non-mobilized peripheral blood, where most colonies had single copy STEMCCA-loxP that was easily excised by transient transfection expression of Cre. Resultant iPSC clones expressed pluripotent cell markers, had genomic methylation pattern closely matching embryonic stem cells, and generated teratomas containing tissues of all three germ lineages i n immunodeficient mice. Furthermore, we conclude that these iPSC are derived from the non-mobilized CD34+ HSCs enriched from PB rather than from any lymphocyte or monocyte contaminants because they lacked somatic rearrangements typical of T or B lymphocytes, and because we demonstrated that purified CD14+ monocytes do not yield iPSC colonies under these reprogramming conditions.
852. AAV2/8 transduces skeletal muscle in non-human primates at efficiencies equal to or greater than that observed in mice
Jenny A. Greig, Rebecca L. Grant, Erin Bote, Hui Peng, C. Angelica Medina-Jaszek, Omua Ahonkhai, Peter Bell, Roberto Calcedo, Anna Tretiakova, James M. Wilson
Gene Therapy Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USA
Prior to the use of gene therapy in clinical trials, the efficiency of the vector needs to be tested in a variety of animal models across a number of species. The goal of this study was to evaluate the efficiency of skeletal muscle transduction in mice and non-human primates using AAV2/8 vectors. Two secreted transgene products were tested in immune deficient and immune competent mice as well as rhesus macaques and a squirrel monkey. Experiments were performed with rhesus alpha-fetoprotein (rhAFP) as a transgene in an effort to prevent or minimize host immune responses to the transgene product in rhesus macaques, which would hamper the maximal achievable gene expression in this species. Intramuscular (IM) administration of AAV2/8.CMV.rhAFP at a dose of 1x1011 GC per mouse (5x1012 GC/kg) into the right and left gastrocnemius muscles of RAG KO mice led to production of 1.4 ± 0.2 µg/ml rhAFP at day 28 post-injection. In comparison, rhesus macaques which were administered with 5x1012 GC/kg of the vector IM (injected as 1 ml of vector dilution per kilogram body weight into the left and right vastus lateralis muscles) produced 10-fold higher levels of rhAFP (13.3 ± 6.7 µg/ml) at day 28. Similar studies were repeated with a different AAV2/8 vector, AAV2/8.CMV.201Ig IA, which expresses an immunoadhesin construct of the SIV antibody 201Ig. At day 28 post-administration of 1x1011 GC per mouse (5x1012 GC/kg) of the vector by IM injection, RAG KO mice expressed 378 ± 52 µg/ml 201Ig IA. Four rhesus macaques were administered with a lower dose of the same vector (3x1012 GC/kg). At day 28 post-vector administration, the rhesus macaques produced similar levels of expression of the 201Ig IA with the average level of expression for the four animals being 458 ± 53 µg/ml 201Ig IA. One squirrel monkey was also administered the AAV2/8 vector expressing 201Ig IA at a dose of 3x1012 GC/kg IM (injections were performed in the same manner as that for the rhesus macaques). The squirrel monkey expressed higher levels of 201Ig IA at day 28 of 610 µg/ml, with a peak of expression reaching 1.5 mg/ml 201Ig IA. In conclusion, AAV2/8 vector transduced skeletal muscle in non-human primates at efficiencies equal to or greater than that which was observed in mice.
853. Anaphylaxis and Immune Toxicities Are Dose Dependent Following ERT with Recombinant Alpha Glucosidase and Affected by Type of Mutation in Pompe Disease
S Nayak1, S Porvasnik1, D Cloutier1, P Doerfler1, R Khanna3, K Valenzano3, G Walter1, RW Herzog2, BJ Byrne1
1Pediatrics, Powell Gene Therapy Center, University of Florida, Gainesville, FL; 2Pediatrics, Molecular and Cellular Therapy, University of Florida, Gainesville, FL; 3Amicus Therapeutics Inc. La Jolla, CA
Pompe disease is a lysosomal storage deficiency caused by mutations to the acid alpha glucosidase (GAA) gene. Early onset patients are characterized by severe cardiomegaly, respiratory distress and muscle weakness, resulting in mortality within a year of birth without treatment. Recombinant human acid alpha glucosidase (rhGAA) enzyme replacement therapy (ERT) is the current standard of care; however, these patients develop severe immune responses to the therapy that reduces the efficiency of treatment. Immune responses include infusion associated reactions (IAR) and high anti-rhGAA titer that further limits efficiency and can alter the course of treatment. In GAA knockout (129SVE GAA-/-) mice, we determined that the dose of rhGAA ERT plays a vital role in the severity of the immune response. Current human therapy in GAA-/- mice with 20 mg/kg rhGAA resulted in the most severe Th2 immune response (anti rhGAA IgG1 6x10e5 ng/ml seven weeks after initiation of rhGAA injections) and the response reduced 12-fold to 4x10e4 ng/ml in the 5 mg/kg rhGAA cohort and further reduced by 500-fold to 1x10e3 ng/ml in the 1 mg/kg group. Anti-rhGAA IgG2a response was low but followed the same pattern. Increased immune responses and hypersensitivity was associated with a significant drop in oxygen saturation and breath distention (effort required to breathe) in the 20 mg and 5 mg/kg groups while remaining normal in the 1 mg/kg and GAA-/- cohort. Hypercoagulation, possibly amounting to disseminated intravascular coagulation, was observed in association with hypersensitivity in GAA-/- mice. Bleeding time was significantly reduced within 5 min after the 5th rhGAA intravenous injection and activated partial thromboplastin time test (aPTT) indicated depletion of clotting factors in plasma after three 20 mg/kg rhGAA IV injections. A significant drop in platelet count and significant increase in platelet width distribution (PWD) were observed. H&E stains indicated extensive residual blood in the heart, kidney and liver of the 20 mg/kg mice that was absent in untreated mice. Anaphylaxis associated mortality was observed in the 20 mg/kg and 5 mg/kg rhGAA cohorts of 129SVE GAA-/- mice but not in the 1 mg/kg cohort. Additionally, mice carrying the human missense mutation (P545L) were found to mount lower immune responses to 20 mg/kg rhGAA (1100-fold lower than the 129SVE GAA-/- mice at 7-weeks). They did not develop reduced oxygen saturation or breath distention even at the high 20 mg/kg rhGAA dose and showed no anaphylactic responses. The amount of residual enzyme and type of mutation has a role in the immune responses to rhGAA. These data support the hypothesis that reducing the dose of ERT required for therapeutic intervention (possibly by increasing the efficiency of lysosomal uptake of rhGAA or the use of stabilizing chaperones) will reduce the rhGAA associated immune responses and toxicities in ERT for Pompe disease.
854. DNA Damage Response Pathway and Replication Fork Stress During Oligonucleotide Directed Gene Editing
Melissa Bonner, Bryan Strouse, Mandula Borjigin, Pawel Bialk, Rohina Niamat, Chris Eskridge, and Eric Kmiec
Delaware State University, Dover, DE
Single-stranded DNA oligonucleotides (ODNs) can be used to direct the exchange of nucleotides in the genome of mammalian cells in a process known as gene editing. Once refined, gene editing should become a viable option for gene therapy and molecular medicine. Gene editing is regulated by a number of DNA recombination and repair pathways whose natural activities often lead to single- and double-stranded DNA breaks. It has been previously shown that introduction of a phosphorotioated ODN, designed to direct a gene-editing event, into cells results in the activation of γH2AX, a well-recognized protein biomarker for double stranded DNA breakage. Using a single copy, integrated mutant enhanced green fluorescent protein (eGFP) gene as our target, we now demonstrate that several types of ODNs, capable of directing gene editing, also activate the DNA damage response and the post-translational modification of proliferating cell nuclear antigen (PCNA), a signature modification o f replication stress. We find that the gene editing reaction itself leads to transient DNA breakage, perhaps through replication fork collapse. Unmodified specific ODNs elicit a lesser degree of replication stress than their chemically modified counterparts, but are also less active in gene editing. Modified PTOs are detrimental irrespective of the DNA sequence. Such collateral damage may prove problematic for proliferation of human cells genetically modified by gene editing.
855. Synergistic Transcriptional Activation by Combinations of Engineered TALEs
Pablo Perez-Pinera1, Alicia M Farin1, Jonathan M Brunger1,2, Katherine A Glass1, David G Ousterout1, Farshid Guilak1,2, and Charles A Gersbach1
1Departments of Biomedical Engineering and 2Orthopaedic Surgery, Duke University, Durham, NC, USA
A principal goal of gene therapy and synthetic biology is to be able to fine tune levels of gene expression. However, this poses a significant challenge for many applications, particularly the activation of endogenous genes in their native chromosomal context. Typically, knockdown and strong overexpression of genes have been the most common strategies for regulating gene function. These two methods interrogate biological systems at only two discrete points that are determined by experimental feasibility and not by biological significance. Therefore, the dependency of phenotype on levels of gene expression may not be accessible due to the limitations intrinsic to available experimental tools. Recently, alternative approaches to modulate gene expression have emerged that mimic the combinatorial regulation of natural promoters by various transcription factors. These systems are based on observations that multiple transcription activators acting simultaneously results in greater activity than the sum of the effects of each working individually. For example, individual transgenes may be controlled by synthetic promoters that incorporate multiple transcription factor binding sites. Although this approach is successful for regulating engineered systems, it is not applicable to endogenous cellular genes controlled by their native promoters. To overcome these limitations we have developed an alternative approach that uses multiple Transcription Activator-Like Effector (TALE) transcription factors acting simultaneously in the 5' UTR of genes of interest to induce different levels of expression. The DNA binding domain of the TALE was cloned in frame with a VP64 transcriptional activation domain to rapidly generate multiple transcription factors targeting MyoD1, IL1RN, HER2, OCT4 and KLF4 that display various degrees of activity in luciferase reporter systems. When used in combination, these TALE transcription factors have a synergistic effect that is sufficient to activate transcription up to 500-fold. We demonstrate that the number and position of the target sequences in relation to the start codon permits facile manipulation of the levels of gene expression from luciferase reporter constructs or the endogenous locus as demonstrated by qRT-PCR, Western blot, and ELISA. The results of these experiments support current hypotheses of mechanisms for synergistic transcriptional regulation, postulate new guidelines to achieve activation of endogenous gene expression by artificial transcription factors, and provide a new experimental platform for gene therapy, functional genomics, and synthetic biology.
856. Mathematical model for radial expansion and conflation of intratumoral infectious centers during curative oncolytic virotherapy
Kent Bailey, Shruthi Naik, Michael Steele, Linh Pham, Elizabeth Hadac, Rebecca Nace, Suzanne Greiner, Kah Whye Peng, Stephen J Russell
Mayo Clinic Departments of Molecular Medicine and Health Sciences Research, Rochester, MN
Oncolytic viruses destroy tumors by diverse mechanisms and with different kinetics. Previous (dynamic, non-spatial) mathematical models developed to guide protocol design and predict outcomes have been theoretical and complex. Here, we developed a simple spatial mathematical model of oncolytic virotherapy induced from detailed immunohistochemical analyses of plasmacytomas regressing after a single intravenous injection of an oncolytic vesicular stomatitis virus (VSV). Intratumoral virus spread dramatically outstripped tumor growth during the first 48 hours after virus administration, but was then rapidly curtailed by host antiviral defenses (these were syngeneic tumors in fully immunocompetent mice). The new model calculates the probability of any tumor cell surviving after intravenous VSV administration based on three key parameters; the size of the tumor, the density of initially infected cells in the tumor, and the average maximum size attained by the infectious centers. Predicted effects of virus dose modification, antiviral immune response modulation and tumor size were investigated through additional in vivo experimentation and the results are in good agreement with the predictions of the model. This simple, convenient and well-validated new spatial model gives helpful insights into the process of oncolytic virotherapy will be useful to guide the further optimization of the approach.
857. Effective Gene Delivery System with Sonoporation for Cancer Gene Therapy
Ryo Suzuki1, Yusuke Oda1, Yoichi Negishi2, Kazuo Maruyama1
1Department of Drug Delivery System, School of Pharmaceutical Sciences, Teikyo University; 2Department of Drug and Gene delivery System, School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, Hachioji, Tokyo, Japan.
Objective: Sonoporation is an attractive technique to develop non-invasive and non-viral gene delivery system. However, simple sonoporation using only ultrasound (US) is not enough to establish effective cancer gene therapy because of low efficiency of gene delivery. Therefore, we improved this problem by the combination of US and novel US-sensitive liposome (Bubble liposome) which was a liposome containing US imaging gas (perfluoropropane). This was an effective gene delivery system with collapse (cavitation) that was induced by US exposure to Bubble liposome. In this study, we assessed the ability of this system in cancer gene therapy using IL-12 cording plasmid DNA. Material and Methods: B6C3F1 mice were intradermally inoculated with mouse ovarian cancer cells (OV-HM cells) into flank. After 7 days, Bubble liposome (2.5 ug) and mouse IL-12 cording plasmid DNA (10 ug) were injected into tumor, and US (1 MHz, 0.7 W/cm2, 60 sec.) was transdermally exposed toward tumor. In addition, gene delivery with conventional lipofection method using Lipofectamine 2000 was also examined as control of non-viral vector. IL-12 expression was assessed with RT-PCR for IL-12p40 mRNA. Anti-tumor effects were evaluated by measuring tumor volume. Results and Discussion: IL-12p40 mRNA expression level in gene delivery with Bubble liposome and US was higher than that with conventional lipofection method. Additionally, in IL-12 gene delivery with conventional lipofection method, the suppression of tumor growth was not observed. On the other hand, IL-12 gene delivery with Bubble liposome and US was dramatically suppressed tumor growth. Therefore, we concluded that the combination of Bubble liposome and US would be a good non-viral vector system in cancer gene therapy.
858. Developing and Validating New Design Rules for Transcription Activator-Like (TAL) Effector Nucleases
Yanni Lin, Eli Fine, Thomas J Cradick, Gang Bao
Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, US
TAL Effector Nucleases (TALENs) have become an important type of designer nucleases for genome editing. Although the code of nucleotide binding specificity has been deciphered, there are still questions on how to select optimal TALEN targeting sites for efficient cleavage. The design guidelines based on naturally occurring TAL effector target sites (Cermak et al., NAR 2011), have proved fruitful, but have not been systematically tested. We have designed and constructed more than 80 TALENs targeting the human sickle cell β-globin gene and other targets. These TALEN target sequences differ in their nucleotide composition, length, spacer length and composition at their 5' and 3' ends. Synthetic homodimeric and heterodimeric target plasmids allow testing TALENs individually and in pairs, using both in vitro and cell culture assays. Based on the results from individual TALEN designs, we refined the design rules for TALENs, and derived a scoring function that predicts the activities of TALENs based on their target sequences. The scoring formula was derived by comparison of the activity of the first 40 TALENs tested, and gives an excellent correlation between prediction and the experimental results. An additional 40 TALENs were designed and tested to refine the scoring function and determine its predictive ability. Our Scoring Algorithm for Predicted TALEN Activity (SAPTA), which was programmed into a user- friendly search tool to determine the best target sequences of the gene of interest, will greatly assist in the design of custom TALENs that can be widely used for gene targeting applications.
859. Direct selection of internalizing RNAs
Amy Yan, Maria de Lourdes Borba Magalhães, Linsley Kelly, Samantha E. Wilner, Keith Maier, Brian Wengerter, Matthew Levy
Department of Biochemistry, Albert Einstein College of Medicine, Bronx, NY, USA
Aptamers represent a promising alternative to antibodies and other protein-based targeting agents for targeted delivery of therapeutic and diagnostic cargoes to cells. We have developed a selection scheme to identify nucleic acid sequences that are internalized by mammalian cells. To illustrate the approach, we performed internalization selections against two separate cell lines, one mouse and one human, using two different aptamer libraries. Surprisingly, each selection yielded a single highly functional sequence that share a common core motif, and both are internalized by a variety of different cell types including primary cells. The internalization selection approach can also be used to facilitate the selection of aptamers that target specific cell surface receptors that are known to be endocytosed. As a demonstration of this, we have applied this method to identify aptamers which specifically recognize the human transferrin receptor (hTfR). Using minimized versions of these aptamers, we have generated aptamer targeted siRNA-laden liposomes. Aptamer targeting enhanced both siRNA uptake and target gene knockdown in cells grown in culture when compared to non-modified or non-targeted liposomes. Our selection approach and these targeting molecules should prove useful for the development of novel cell imaging and targeted drug delivery agents.
860. Intracellular Trafficking Kinetics of Nucleic Acid Delivery Vehicles in Three Dimensions
Nilesh P. Ingle
Department of Chemistry, University of Minnesota, Minneapolis, MN, USA
Recent scientific findings in trafficking of polymeric nucleic acid delivery vehicles in mammalian cells has led to an improved understanding of their affinity towards a specific endocytosis mechanism and also their subsequent trafficking pathways. This has also stimulated the development of novel polymeric structures to navigate these systems in a cell. However, much remains to be explored about the en route physical characteristics of these vehicles in the intricate cytoplasmic milieu. Therefore the goal of this study was to understand the intracellular evolution pertaining to unique physical characteristics of polyplexes with time. In the proposed method the objective was to visualize as well as to quantify this polyplex evolution at different time points in three spatial dimensions. The quantified parameters were the distributions of size and distance of these polyplexes with respect to the nucleus of cell. In this study four polymeric systems were studied. The polymers that were synthesized in our laboratory consisted on trehalose and pentaethylenetetramine subunits. In addition there were two commercially available polymers GlycofectTM and jetPEITM. The trehalose polymer consisted of two degrees of polymerization namely 55 (Tr4-55) and 77 (Tr4-77). Experimental Method: HeLa cells were plated on glass bottom dishes and transfected with polyplexes comprised of polymers and FITC-pDNA at N/P ratios (1) jetPEITM at 5, (2) GlycofectTM at 20, (3) Tr4-55 at 7 and (4) Tr4-77 at 7. The cells were then fixed at 4, 8, 12 and 24 hours post transfection. DAPI was used to label nuclei. The image z-stacks acquired on Zeiss LSM Meta 510 confocal microscope were deconvoluted and surface rendered using Huygens Essential software. Results and Discussion: The 3D image analysis quantified (1) volume distribution, (2) distance distribution from surface of nucleus, and (3) distance between two nearest polyplexes. Statistical analysis indicated a higher probability (0.3 to 0.6), for smaller polyplexes of < 0.05 µm3 (spherical diameter: 457 nm), to exist as compared to larger particles > 0.05 µm3 at any given time point. And there was a higher probability for polyplexes to exist at a closer distance of < 2 µm from the surface of the nucleus as compared to > 10 µm at any given time point. This may mean that polyplexes closer to nucleus at < 2 µm may be representative of a population of smaller polyplexes of size up < 0.05 µm3. Indeed we also observed that the polyplexes existing at larger inter-polyplex distance of 5 to 10 µm decreased by 24 hours. This may be indicative of dominance of smaller size polyplexes in the intracellular space at distance closer to the nucleus. Glycofect™ polyplexes appeared to be endocytosed at a higher number of polyplexes of volume < 0.01 µm3 (spherical diameter: 267 nm), as compared to Tr4-55 and Tr4-77 at 4 hours (P value < 0.05). Tr4-55 polyplexes appeared to gradually assemble in perinuclear region of < 2 µm from 4 hours to 24 hours. We thus propose a method to understand the kinetics of intracellular trafficking in 3D.
861. Expression of the Transcriptional Co-Activator PGC1α by a Lentiviral Vector in Human Neural Progenitor Cells Arrests Cell Proliferation, Induces β-tubulin III Expression and Causes a Microtubule Catastrophe – Balancing ATP Production with Microtubule Stability
Manfred Schubert1, Paul S. Fishman2
1National Institute of Neurological Disorders and Stroke, NIH, Bethesda MD; 2Veterans Administration Research Services and Department of Neurology, University of Maryland, Baltimore MD
The transcriptional co-activator PGC1α, in conjunction with several transcription factors, plays an important regulatory role in mitochondrial biosynthesis and metabolic homeostasis. It also provides neuroprotection under oxidative and metabolic stress. Lack of functional PGC1α can progress neurodegenerative diseases. Here we describe for the first time that the overexpression of PGC1α in neural cells impacts microtubule stability, which may contribute to neurodegeneration. Neurons depend on oxidative phosphorylation as their main source for ATP, and PGC1 α facilitates mitochondrial replication. To examine whether PGC1α plays a role at early stages during neural differentiation, proliferating human neural progenitor cells (hNPCs) were infected with a PGC1α -encoding vector or control vectors. Unexpectedly, PGC1 α expression arrested hNPC proliferation. Within 24 hrs, the cells became spherical in shape, but remained viable for at least several days. hNPCs infected with the EGFP control vector continued to divide and express the reporter gene. Under differentiation conditions, hNPC populations became heterogeneous with the development of neurons, multipolar and radial glial cells. In contrast to neurons, radial glia and multipolar cells were efficiently infected by lentiviral vectors. Infection with the PGC1α -encoding vector, however, led to a dramatic retraction of their extensive cellular processes. hNPCs are shaped by the internal microtubule network, which in the case of hNPCs contained α-tubulin I and β-tubulin II, as well as glial fibrillary acidic protein (GFAP) and actin. Neurons did not express GFAP, but expressed the neuron-specific isoform β -tubulin III, which was absent from radial glia and multipolar cells. Interestingly, the PGC1α-induced catastrophe of the microtubule network in radial glia and multipolar cells was accompanied by immunofluorescent staining for β-tubulin III and GFAP, which were co-expressed in these “hybrid” cells. The addition of the microtubule stabili zing agent paclitaxel to PGC1α -expressing hNPCs prevented the complete retraction of the cellular processes with the microtubule insertion of β -tubulin III in PGC1α-expressing hNPCs. In conclusion, the expression of PGC1α in hNPCs surprisingly revealed an exciting new ability of PGC1α to induce β-tubulin III expression in hNPCs. β-tubulin III differs from other β-tubulin isoforms with respect to its effect on microtubule dynamics. Counter-intuitively, β-tubulin III can destabilize microtubules, which may have contributed to the observed microtubule catastrophe. β-tubulin III is considered neuron-specific, except for cells of the testis and several cancers. In cancer cells, β-tubulin III expression correlates with high invasiveness and increased paclitaxel resistance. The induction of β-tubulin III by PGC1α will be discussed with respect to its potential effect on microtubule stability during neurodegeneration and the switch from oxidative phosphorylation to glycolysis in cancer cells.
862. Combined delivery of HMGB1 box A peptide and S1PLyase siRNA for acute lung injury
Binna Oh, Minhyung Lee,
Department of Bioengineering, College of Engineering, Hanyang University, Seoul, Korea
Acute lung injury (ALI) is an inflammatory disease, which is characterized by the increased lung vascular permeability and alveolar flooding. ALI leads to morbidity and mortality. High mobility group box-1 (HMGB-1) is a DNA binding non-histone nuclear protein and pro-inflammatory cytokine. However, it was previously reported that the recombinant box A peptide of HMGB-1 (rHMGB-1A) reduced the production of the pro-inflammatory cytokines by antagonizing wild-type HMGB-1 (wtHMGB1). Moreover, rHMGB-1A binds to and delivers siRNA into the cells, since rHMGB1-A is a cationic peptide. Therefore, rHMGB-1A may have dual functions as a therapeutic agent and a siRNA delivery carrier. S1PLyase modulates LPS-mediated p38 MAPK and NF-kB. The delivery of the S1PLyase siRNA can inhibit LPS-mediated p38 MAPK and NF-kB and reduce the LPS-induced interleukin-6 (IL-6) secretion. In this study, rHMGB-1A was proved as a S1PLyase siRNA carrier with anti-inflammatory effects for acute lung injury. rHMGB-1A was expressed and purified by nickel chelate affinity chromatography, cationic exchange chromatography and polymixin B chromatography. The purified rHMGB-1A was analyzed by SDS-PAGE. In vitro treatment assay with lipopolysaccharide (LPS) activated macrophages showed that rHMGB-1A had an anti-inflammatory effect, reducing IL-6 secretion in the cells. In gel retardation assay, rHMGB-1A formed a stable complex with siRNA at a 1:3 weight ratio (siRNA:rHMGB-1A). Furthermore, MTT assay showed that rHMGB-1A is non-toxic to the cells, suggesting that rHMGB-1A is a safer siRNA carrier than polyethylenimine (PEI) and lipofectamine. Therefore, rHMGB-1A may be useful as a carrier an S1PLyase siRNA delivery as well as anti-inflammatory agent for the treatment of acute lung injury
863. Microcell-mediated chromosome transfer of human artificial chromosome following cryopreservation for the ready-made use.
Narumi Uno, Katsuhiro Uno, Susi Zatti, Kana Ueda, Masaharu Hiratsuka, Motonobu Katoh, Mitsuo Oshimura
Department of Biomedical Science, Institute of Regenerative Medicine and Biofunction, Graduate School of Medical Science and Chromosome Engineering Research Center, Tottori University, Tottori, Japan
Microcell-mediated chromosome transfer (MMCT) is a technology which enables to transfer a single and intact mammalian chromosome or its fragment containing some Megabase-sized stretches from donor to recipient cells. Human artificial chromosomes (HACs) for genetic correction or modification have been transferred to various cell types e.g., iPSCs and MSCs by fusing microcells with the recipient cells. Polyethylene glycol (PEG) has conventionally been used. However, PEG is not suite to all type of cells as a fusogen, because it has cytotoxicity against some cell types. The colony efficiency of fusion between microcells and recipient cell is about 1 x 10-6 ~5 x 10-5. Measles virus fusogen envelop proteins that are hemagglutinin (H) and fusion (F) proteins were expressed on the surface of microcells. These proteins can mediate to fuse microcells and recipient cells. Hence, the cytotoxicity was reduced and improved the efficiency of MMCT to 1 x 10-4. The conventional MMCT method had been performed immediately after purification of microcells. Here, we established a cryopreserved method to store microcells at -80 degree. Furthermore, we compared the conventional and the cryopreserved method for the efficiency of MMCT and the stability of human artificial chromosome when transferring to human HT1080 cells. Drug-resistant cells appeared after selection in culture with the tagged selection marker gene on the HAC. The chromosome transfer efficiency was determined by counting the total number of stable clones expressing EGFP obtained in each experiment. The presence of the HAC in microcell hybrids was confirmed by FISH analyses. There is not a significant difference between the two methods for the chromosome transfer efficiency and retention rate of HAC. Thus, the cryopreserved method with the MV-H and F proteins as a fusogen is an improved simple MMCT protocol.
864. Procalcitonin Gene Expression is a True Biomarker of Aging Process
Mehmet Ali Soylemez
Marmara Medical School, Istanbul, Turkey
Objective: Inflammation is the single greatest precipitator of aging and age-related diseases such as, diabetes, heart disease, alzheimer, arthritis, certain forms of cancer, as well as wrinkled, sagging skin. Inflammation, which takes place on a cellular level, is triggered by a wide variety of factors such as the oxidative stress, ingestion of toxins, excess exposure to ultraviolet radiation, hormonal changes and eating a pro-inflammatory diet (Hyperglycemic diet). The Procalcitonin (PCT) gene, referred to as Calc-1, is located on chromosome 11p15.4 and was sequenced in 1989. The procalcitonin gene has sites for basal transcription factors but more interestingly, also has sites for NF-kappaβ (Nuclear Factor kappaβ) and AP-1(Activator protein-1), factors induced under inflammatory conditions. Hyperglycemia is associated with oxidative stress and elevation of advanced glycation end products (AGEs). AGEs are produced by a non-enzymatic, maillard reaction between reducing sugars and either proteins or lipids. AGE's interacts the receptor for advanced glycation end products (RAGE) and, RAGE activation is caused by elevation of transcriptional factors NF- kappaβ and AP-1. These factors induce procalcitonin gene expression. The aim of this study was to determine whether or not procalcitonin is a specific marker of aging process. Research Design and Methods: Fourteen type II Diabetes Mellitus (DM) patients within hyperglycemia were studied along with age and sex matched with normal non-hyperglycemic subjects (7 male and 7 female). Blood samples were taken for measurements of procalcitoninin. Serum procalcitonin levels were analized with kryptor analizator. Kryptor – PCT is a kit designed for kryptor automated immunofluorescent assay of procalcitonin in serum. Procalcitonin levels were elevated in diabetic patients within hyperglycemia when compared with normal non-diabetic normoglycemic subjects. There was a statistically significant difference in serum procalcitonin of diabetic patients within hyperglycemia versus normal- normoglycemic subjects(P<0.01). Conclusion: Our study revealed a raise in serum procalcitonin in diabetic patients within hyperglycemia. Hyperglycemia associated with increased circulating concentrations of procalcitonin may be a mechanism which explains many of the clinical and biochemical features of aging process. The PCT as aging biomarker can save drug companies huge amounts of money and time. Procalcitonin is a new aging biomarker which can help for discover new drugs to combat aging and DM, as well as new uses for already approved drugs.
865. AAV Mediated Neonatal Gene Delivery to the Mouse Retina Via the Palpebral Vein
D Chung, J Bennicelli, L Bryant, Z Wei, J Bennett
F.M. Kirby Center for Molecular Ophthalmology, Scheie Eye Institute, University of Pennsylvania, Philadelphia, PA, USA
Objective: To demonstrate AAV mediated gene transfer to the neonatal mouse retina using an intravenous palpebral vein injection. Methods: Cohorts of neonatal (P0-1) C57Bl6 wild type mice received intravenous injections of AAV2/9.eGFP/Luciferase into the right palpebral vein. The neonatal right palpebral vein was identified with direct lighting and transillumination and 15 ul of vector was injected. The left eye was used as a non-injected control. Animals were euthanized 14 days post injection. Eyes were enucleated, cryoprotected and sectioned. Additionally, other organs were harvested and GFP and luciferase expression was examined with fluorescence microscopy and bioluminescence imaging, respectively. Results: Examination of the right eye following injection with AAV2/9.eGFP/luciferase showed significant eGFP in the inner and outer retina. Significant eGFP expression was found in the retinal pigment epithelium, ganglion cells and other cells of the retinal cell types. Expression was also found in the heart, and trace expression in the lungs, and liver. No eGFP expression was found in the kidneys or spleen. Luciferase imaging also demonstrated similar expression. Conclusion: Neonatal uni-ocular retinal gene transfer can be accomplished by an intravenous injection into the corresponding palpebral vein. This method may provide an alternative non-intraocular route of gene delivery to the neonatal eye.
866. Novel Synthetic Human Telomerase Reverse Transcriptase DNA Vaccine Delivered by Electroporation Induces Strong Cytotoxic Cellular Immune Responses and Antitumor Immunity
Jian Yan1, Panyupa Pankhong2, Thomas Shin2, Matthew P. Morrow1, Amir S. Khan1, Niranjan Y. Sardesai1, David B. Weiner2
1Inovio Pharmaceuticals Inc, Blue Bell, PA, USA; 2Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USA
Disregulated telomerase is a ribonucleoprotein that enables cancer cells to maintain their telomeres, facilitating countless proliferation that is a hallmark of tumor growth. It contains two components: an RNA component (human Telomerase RNA, hTR) and the catalytic subunit (Human Telomerase Reverse Transcriptase, hTERT). High levels of hTERT have been detected in more than 85% of all human cancers, while normal somatic cells showed undetectable levels of telomerase expression. Immunological analysis indicates that hTERT is a widely applicable target recognized by T cells and can be potentially studied as a universal cancer vaccine. We engineered a novel hTERT DNA vaccine (phTERT) and introduced several genetic modifications to improve transgene expression and immune potency - notably a Kozak sequence was introduced to increase translation initiation and a highly efficient leader sequence was fused in frame upstream of the start codon to facilitate expression. In addition, several mutations were incorporated into the hTERT DNA in order to potentially break tolerance and obtain more potent antitumor immunity. Codon/RNA optimization was performed to further enhance the expression level of hTERT. Finally, the vaccine delivered using electroporation to enhance its immunogenicity. Here, we demonstrate that immunization with phTERT in mice induced strong hTERT-specific CD8-mediated cellular immune responses with enhanced magnitude of CD107a, IFN-γ and TNF-α secretion. Significantly, hTERT vaccination resulted in delayed tumor growth in a murine HPV16 associated tumor challenge model. Due to these encouraging results, we next studied phTERT in a non-human primate model to determine and characterize the immune responses elicited by enhanced DNA vaccination. We found that all immunized monkeys exhibited robust IFN-γ ELISpot responses. After four immunizations, the average number of antigen-specific IFN-γ secreting cells in response to stimulation with hTERT peptide pools was 2417 + 593/per million PBMCs. As Perforin is one of the main cytolytic proteins of cytolytic granules responsible for lysing target cells, we further performed a perforin ELISpot assay to determine actual lytic granule release from PBMCs taken from the phTERT-immunized monkeys. Results demonstrated that vaccination with phTERT induced robust antigen-specific perforin release. Whether the induction of hTERT-specific CTLs are capable of killing target cells in an antigen specific manner is under investigation by lytic granule loading assay and granzyme B cell-killing assay. Taken together, these data support the further development of phTERT as a cancer DNA vaccine therapy candidate.
867. Histone polyplexes utilize caveolar uptake and traffic through the Golgi and ER during gene transfer
Meghan J. Reilly1, John D. Larsen2, Millicent O. Sullivan1
1Department of Chemical Engineering, University of Delaware, Newark, DE, USA; 2ABS Global, Madison, WI, USA
Improved control over the subcellular trafficking of polyplexes has been widely identified as a key hurdle towards improving their activity in the in vivo setting, where gene transfer inefficiencies cannot be easily offset by increasing the concentration of the delivered vehicle. It has become increasingly apparent that the endocytic uptake route for polyplexes is a fundamental determinant of intracellular trafficking and activity. Our recent findings suggest that caveolar uptake may be involved in transfection by polyplexes comprised of pDNA, a histone H3 sequence involved in nuclear localization and chromatin activation, and branched poly(ethylenimine) (PEI). We have found that polyplexes containing ~90% (w/w) H3 tail peptides and ~10% PEI produced significantly more robust transfection and lower cytotoxicity than PEI-pDNA polyplexes. We hypothesized that the improved activities of the hybrid H3/PEI-pDNA polyplexes as compared with PEI-pDNA polyplexes were caused by their altered intracellular trafficking and enhanced nuclear delivery. H3/PEI-pDNA polyplexes exhibited slower uptake and a reduced dependence on endocytic pathways that trafficked to the lysosome in comparison with PEI-pDNA polyplexes. Furthermore, treatment with the macrolide antibiotic bafilomycin A1 significantly reduced H3/PEI-pDNA transfection, but had less substantial effects on PEI-pDNA transfection. The V-type ATPase binding partners of bafilomycin A1 colocalize with caveolin-1, and bafilomycin A1 alters retrograde trafficking from the Golgi complex. Thus, these results may suggest that trafficking through caveolae and the Golgi complex represents a productive pathway that enhances H3/PEI-pDNA polyplex delivery to the nucleus.
868. Pharmacological Modulation of Transcriptional Activity of Adegr1-Luc in Human Glioma Cells
Francisco Martínez-F1,2, Araceli Barrera-L1, María Simona Ustoa1, Hugo Sandoval-Z3, Hilda Villegas C3, Karina Guzmán-M1, Bruno Escobedo-G1, Kathleen Sakamoto4
1Molecular Biotherapeutic Program, Skin & Tissue Bank, National Institute of Rehabilitation, Ministry of Health, México; 2Department of Pharmacology, School of Medicine, National University of Mexico; 3Department of Molecular and Cellular Morphology, National Institute of Rehabilitation, Ministry of Health, México; 4Department of Hemato-Oncology, Childreńs Hospital of UCLA, USA
Introduction: Mechanism action of steroid drugs is pleiotropic on several signal pathways. In the case of steroid hormones, a specific receptor is involved for nuclear mechanisms and genetic regulation of several target genes. Egr-1 is a transcriptional factor with mitogen activity in response at physical, chemical and pharmacological response. Hereby we describe the effect of steroid drugs in glioma cells of different malignant degree. Materials and Methods: Recombinant adenovirus AdEgr1-Luc and AdCMV-GFP were packaged and purified at large scale according to the current protocol of core facility of the Molecular Biotherapeutic Program at INR. Multiplicity of infection were assayed for CH235-GM (Grade I); U373-GM (grade III) and D54-GM (Grade IV) human cell lines using the reporter gene of Ad-CMV-GFP and quantified by LSM meta 100 (Carl Zeiss). All cell lines were cultured in D-MEM/F12 (50:50) media supplemented with 10% of heat inactivated FBS and antibiotics in environment standard conditions for culture. 1x105 glioma cells were transduced at 75 MOÍs for 2 hs in reduced serum media in six well plates by triplicates. After adenoviral infection, cells were starved from serum for 24 hrs and exposed to β-estradiol (50 ng/ml), progesterone (100 ng/ml) and betametasona (25 ng/ml). Cells were cultured until protein extraction for luminometry analysis at 3, 6 and 12 hs according to manufacturer instructions (Cell Glo lysis Buffer, Promega Corp). Luciferasa activity was performed based on the Glo-luciferase system assay and quantified in a multimode detector DTX-880. Results: Progesterone and b-estradiol, shows a pattern of positive regulation a 12 hs in CH235 cell line (4,156.68 CL/Seg and 4,534.70 CL/Seg respectively at 12 hrs), and b-estradiol shows a negative activity on transcriptional activity (range: 105.81-283.78 LC/s). In U373 cell line, b-estradiol and progesterone shows positive regulation with maximum activity a 6 hs (2,923.92 LC/s) on the transcriptional control of Adegr1-Luc at 6 hrs of exposure but betametasone shows negative activity compared with both steroid hormones. In case of D54 cells activity not show significant difference compared to the control cell line. Conclusion: Steroid hormones show activation of egr-1 promoter inversely proportional to the progress of malignant degree. This finding could be explained by missing mechanism control of steroid response machinery or pathway signal transductions. This fact correlates with the phenomenon of resistance at drugs mediated by presence or absent receptor in other kind of cancer resistance. Acknowledgments: This research project is granted by the National Council of Science and Technology of México. Grant: FOSIS/CONACYT-161624 and the project PIFUT08-169.
869. Administration of Bone Marrow Derived Mesenchymal Stem Cells into the Liver: Potential to Rescue Pseudoxanthoma Elasticum in a Mouse Model (Abcc6–/–)
Qiujie Jiang, Shunsuke Takahagi, Dian Wang and Jouni Uitto
Department of Dermatology and Cutaneous Biology, Thomas Jefferson University, Philadelphia, PA, USA
Aberrant mineralization of peripheral connective tissues is a characteristic feature of pseudoxanthoma elasticum (PXE), a heritable metabolic disorder caused by mutations in the ABCC6 gene which is primarily expressed in the liver. There is currently no effective treatment for PXE. In this study, we characterized bone marrow derived mesenchymal stem cells (MSCs) and evaluated their ability to contribute to liver regeneration, with the aim to rescue PXE phenotype. The marrow derived stem cells were isolated from GFP-transgenic mice by magnetic cell sorting using a hematolymphoid lineages depletion antibody cocktail as well as stem cell antigen-1 (Sca-1) antibody for negative and positive selection, respectively. MSCs have high potential for hepatic differentiation, with expression of Abcc6, in culture containing hepatocyte growth factor (HGF), oncostatin M and fibroblast growth factor (FGF). MSCs (0.5 x 106) at passage 6 were transplanted into the livers of 4-week old immunodeficient Abcc6-/- mice by intrasplenic injection one day after partial hepatectomy, when peak expression of the stromal cell derived factor-1 (SDF-1) was observed. Fluorescent bio-imaging analysis indicated the presence of transplanted MSCs in the liver at day 1 and day 7 after transplantation. Significant numbers of GFP-positive cells were confirmed in the liver by immunofluorescence as well. Moreover, enhanced engraftment efficiency was observed with MSCs with high expression levels of the chemokine receptor CXCR4, a receptor for SDF-1. However, at week 8 after engraftment, GFP-positive cells were no more detectable in the recipient livers, and vibrissae mineralization, a characteristic feature of Abcc6-/- mice, was not altered in the MSCs transplanted mice when compared with the control mice injected with PBS alone. These data suggest that purified MSCs have the capability of differentiating into hepatic lineages and homing to the liver by the interactions between SDF-1 and CXCR4, but not able to reside in the target liver for a long term amelioration of the mineralization in PXE mice.
870. Effects of Trehalose Polycation End-group Functionalization on Plasmid DNA Uptake and Transfection
Kevin Anderson1, Antons Sizonvs2, Mallory Cortez1, Theresa Reineke1
1Department of Chemistry, University of Minnesota, Minneapolis, MN; 2Department of Chemistry, Virginia Tech, Blacksburg, VA
In this study, we have synthesized six analogs of a trehalose-pentaethylenehexamine glycopolymer (Tr4) that contain (1A) adamantane, (1B) carboxy, (1C) alkynyl-oligoethyleneamine, (1D) azido trehalose, (1E) octyl, or (1F) oligoethyleneamine end groups and evaluated the effects of polymer end group chemistry on the ability of these systems to bind, compact, and deliver pDNA in cultured HeLa cells. The polymers were synthesized in one-pot azide-alkyne cycloaddition reactions with an adaptation of Carothers equation for step-growth polymerization to produce a series of polymers with similar degrees of polymerization. An excess of end-capping monomer was added at the end of the polymerizations to maximize functionalization efficiency, which was evaluated with GPC, NMR and MALDI-TOF. The polymers were all found to bind and compact pDNA at low N/P ratios and form polyplexes with plasmid DNA. The effects of the different end group structures were most evident in polyplex uptake and delivery efficiency in transfections containing serum, as determined by flow cytometry and luciferase gene expression respectively. The Tr4 polymers end-capped with carboxyl groups (N/P = 7), octyne (N/P = 7), and oligoethyleneamine (N/P = 7), were taken into cells as polyplex and exhibited the highest levels of fluorescence, resulting from labeled reporter plasmid. Similarly, the polymers end-functionalized with the carboxyl groups (N/P = 7), octyl groups (N/P = 15) and the oligoethyleneamine (N/P = 15) yielded dramatically higher reporter gene expression. This study yields insight into how very subtle structural changes in the polymer chemistry such as end groups can yield very significant differences in the biological delivery efficiency and transgene expression of polymers used for pDNA delivery.
871. A Complete and Efficient Platform for the Production and Use of Adenoviral Vectors
Thomas P. Quinn, Steve Oh, Lily Lee, Mei Fong, Michael Haugwitz, Andrew Farmer
CLONTECH Laboratories, Inc., A Takara Bio Company, Mountain View, CA, USA
Adenoviral vectors are widely accepted for gene delivery because they can be cultured to high titers and may be used to efficiently transduce a wide variety of cell types in vitro and in vivo. Despite this versatility, many users experience difficulty in efficiently producing and monitoring high-titer adenoviral stocks. Here we present an efficient platform for the production of adenoviral vectors going from vector construction to useable virus in only 7 days. Previously, we developed a rapid method of adenoviral vector construction to seamlessly and directionally clone PCR fragments into a linearized adenoviral vector utilizing a 30 minute, ligation-free, sequence-independent reaction. This rapid and highly efficient approach generated an intact, transfectable adenoviral vector in only 3 days, compared to the 6 to 8 days required by other cloning methods that use a donor or shuttle vector. With this method, we observed >80% cloning efficiency and nearly 100% cloning success with little or no background, as measured by colony PCR screening and restriction analysis. Plasmids cloned with this method were transfected into low-passage HEK 293 cells to produce infectious virus with cytopathic effects typically observed within 4 days. To rapidly ascertain the status of adenovirus rescue post-transfection and to ensure a productive rescue of the adenoviral vector into infectious virus, we developed a hexon-specific lateral flow strip, whereby a small amount of crude supernatant (50 µl) is added to the strip to produce a clear adenovirus-specific signal after approximately 3 min. Using these strips, we confirmed that adenovirus particles were being produced post-transfection but even before cytopathic events were observed. These results confirmed that virus would be produced predictably in 2 to 3 days and could be used for downstream processes such as amplification, purification, and titration. To ensure maximum yields during the amplification process, we employed these same lateral flow strips to non-invasively determine the optimum time to harvest virus from infected cells. Supernatant was tested each day following infection of cells with master stock by placing 20 µl of culture supernatant on the lateral flow strip. Signal was observed by the second day and maximum signal and highest yield were seen by day 4. Amplified virus was then purified from the harvested pellet using a chromatography-based system that produced as much as 4.68 x 1010 particles and yielded >72% of our starting material in one hour. To quantitatively and rapidly determine the final yields of multiple stocks simultaneously, we employed a 4 hour adenovirus vector-specific qPCR assay. This assay facilitated the normalization of infectious units between stocks and permitted transductions to be performed using this information within the same day. Taken together, this platform can decrease the time it takes to obtain infectious virus, reduce repeat transfections and provide higher yields through simple and robust monitoring, purification and titration methods that ensure more consistent and reliable results.
872. HPV induces EMT that is TWIST dependent and can promote carcinogenesis and metastasis of cervical cancer
Wang Wei, Li Yan
Nanfang Hospital, People's Republic of China
Purpose: Human papillomaviruses (HPVs) are very common and now are recognized as the etiological agent of cervical cancer. A group of transcription factors, including Twist1, Twist2, Snail2, ZEB1, and ZEB2, have been shown to induce epithelial mesenchymal transition (EMT), thus promoting tumor progression and dissemination. Little is known about the role of Twist in EMT induced by HPV16 E6/E7 in cervical cancer. This study aims to investigate the mechanisms associated with HPV 16 E6/E7 and cervical cancer progression and metastasis. Experimental Design: We used microarray analysis to determine which cellular genes are regulated by the human papillomavirus type 16 (HPV-16) E6/E7 oncoprotein. Then we confirmed the role of Twist in EMT induced by HPV16 E6/E7 respectively in cervical cancer tissues and cells. Results: We first showed that Twist expression was correlated with HPV 16 E6/E7 in microarray analysis. We also showed for the first time that direct induction of Twist is essential for HPV16 E6/E7 induced EMT in vivo and vitro. Conclusion: This study shows that HPV induced EMT is Twist dependent and can promote carcinogenesis and metastasis of cervical cancer.
873. Reductive degradation of poly(ethylene oxide)-S-S-poly(ε-caprolactone) assemblies for drug and siRNA delivery
Núria Sancho Oltra, Karthikan Rajagopal, Dennis E. Discher
Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, USA
Worm-like micelles from amphiphilic diblock copolymers represent a very attractive structure for drug delivery due to both high drug loading per carrier and longer in vivo circulation times compared to spherical assemblies.[1,2] The performance of polymeric carriers can in principle be enhanced by the introduction of functionalities responsive to disease environments such as a reductive environment found in tumors. Here, we describe a diblock copolymer containing a disulfide group connecting both blocks of poly(ethylene oxide)-S-S-poly(ε-caprolactone), or PEO-S-S-PCL which self-assembles into worm-like micelles and undergoes degradation under reducing conditions. Besides the delivery of hydrophobic drugs and the attachment of Near-Infrared dyes useful for in vivo imaging, this system can also be used to deliver siRNA by incorporating an oligonucleotide-S-S-PCL block copolymer in the assembly. The applicability of such a drug-delivery vehicle is significantly broadened b y such responsive functionality.
[1]D. A. Christian, S. Cai, O. B. Garbuzenko, T. Harada, A. L. Zajac, T. Minko, D. E. Discher, Mol. Pharm. 6: 1343-1352 (2009).
[2] S.M. Loverde, M.L. Klein, D.E. Discher. Advanced Materials DOI: 10.1002/adma.201103192 (2011).
874. Development and Validation of a qPCR Analysis of Male Porcine Cell Biodistribution in Female Pigs
Maoxiang Li1, Emily Hill1, Jeffrey Grover1, Valerie Steenwinckel2, Mark Wolfe1, Mark Johnson1 and Haiyan Ma1
1Department of Cellular and Molecular Biology, MPI Research Inc., U.S.A; 2Cardio3 Biosciences S.A., Belgium
Stem cells, whether gene-modified or not, present a number of safety concerns after administration into human subjects. Major questions have been raised as to where stem cells migrate to after administration, whether the cells engraft and differentiate, and what their long-term fate is. Therefore, GLP-compliant preclinical biodistribution and persistence animal studies are recommended by the United States Food and Drug Administration (FDA) to evaluate the safety of potential stem cell-based products. Quantitative PCR is currently the most sensitive detection method for biodistribution of cell and gene therapy products. For this purpose, a quantitative real time polymerase chain reaction (qPCR) method for biodistribution analysis of male mesenchymal stem cells (MSC) in female pig tissues has been developed and validated. In this study, the specificity of the TaqMan probe/primers to the male porcine sex-determining region Y (SRY) gene sequence was determined by carrying out real-time PCR on genomic DNA isolated from male porcine MSC in the absence and presence of female pig matrix genomic DNA (gDNA). The condition of the qPCR analysis was optimized. The low limit of detection (LLOD), low limit of quantification (LLOQ), upper limit of quantification (ULOQ), and the potential effects of female porcine matrix gDNA were evaluated by conducting a series of qPCR experiments on total female pig DNA spiked with known amounts of male MSC DNA. The PCR efficiency was estimated through the linear regression of the dilution curve. The intra-assay precision, inter-assay reproducibility, and accuracy were evaluated over five separate assays. For each assay, two sets of quality control (QC) samples were prepared independently and each set included three concentrations of male pig MSC DNA (high, middle and low) in triplicate. The assay specificity, linear range, intra- and inter-assay precision, accuracy and reproducibility, and acceptance criteria are reported.
875. Cell culture models of non-viral transgene expression are not indicative of in vivo lung outcome
Cathy A. Oliveira, Stephanie G. Sumner-Jones, Ian A. Pringle, Stephen C. Hyde & Deborah R. Gill
Gene Medicine Research Group, University of Oxford, John Radcliffe Hospital, Oxford, UK
Gene therapy for chronic lung diseases will require vectors capable of persistent transgene expression. This has been achieved for several non-viral vectors by judicious promoter selection and removal of CG dinucleotides from the plasmid DNA. Plasmid pG4-hCEFI-soLux directs persistent, high-level transgene expression for at least 6 months in the mouse lung after aerosol delivery, utilising the hCEFI promoter, which is a hybrid of CpG-free versions of the human CMV enhancer (hC) and human Elongation Factor 1a (EF) promoter sequences [Hyde et al, 2008, Nat Biotech 26(5):549]. However, replacement of the 302bp human CMV enhancer with the equivalent murine sequence (mC) in plasmid pG4-mCEFI-soLux resulted in only transient expression lasting for a few days. To further study the cellular and molecular mechanisms involved, we investigated the potential use of two cell culture models to mimic the transgene silencing observed in vivo: the A549 lung cell line (ATCC) and human air-liqu id interface (ALI) primary respiratory epithelial ex vivo cultures (Epithelix, Sarl, Switzerland). A549 cells were transiently transfected with both plasmids (5E4 cells, 3µg plasmid DNA, 0.1M polyethylenimine (PEI)) and luciferase activity in cell lysate was measured at t=6h, 12h, 24h, 48h and 72h (n=3). Although luciferase activity differed at 12h (p<0.01), 24h (p<0.001) and 48h (p<0.001) (2-way ANOVA with Bonferroni posttest), neither plasmid directed persistent transgene expression. ALI cultures were transiently transfected with both plasmids (5E5 cells per insert; 10µg plasmid DNA; 25µl lipofectamine) for 6h, after which the transfection mix was removed. Addition of 100ul of D-Luciferin at 30mg/ml was added prior to measurement of luciferase activity using the IVIS100 (Caliper Life Sciences) bioluminescent imaging system. Imaging was repeated at 12h and 24h after transfection and daily thereafter for 7 days. Average radiance in initial studies was very low with both plasmids at all time-points. To increase transgene expression, ALIs were treated with EGTA to disrupt tight junctions transiently, via pre-treatment with 10mM EGTA for 30 minutes, or co-treatment with 10mM EGTA present in the transfection mix. EGTA pre-treatment did not lead to increased transgene expression compared with non-treated controls, but co-treatment with EGTA resulted in a 63-fold increase in luciferase activity at 12h (p<0.001, 2-way ANOVA with Bonferroni posttest) over pre-treated inserts. Using the EGTA co-treatment transfection conditions, a time-course experiment was performed to determine transgene expression in ALI cultures when transfected with pG4-hCEFI-soLux or pG4-mCEFI-soLux (n=7 ALIs per plasmid). Luciferase activity directed by the two plasmids was different at 24h (p<0.001), 48h (p<0.001) and 72h (p<0.05) (2-way ANOVA with Bonferroni posttest). However, in each case, peak expression rapidly declined within a few days. Together these data demonstrate that non-viral transgene expression observed in mouse lungs following aerosol delivery is not recapitulated in A549 lung cells or in human ALI ex vivo cultures. Currently the reasons for this are unknown, but may be due to species and cell type differences.