- Email: [email protected]
1198. Inhibition of Angiogenesis and Tumor Progression by Hydrodynamic Co-Transfection of Angiogenesis K1-3, Endostatin and Saxatilin Genes
NAKED DNA: APPLICATIONS 1196. Ectopic Expression of CFTR in Adult Airway Stem Cells Inhibits Capacity To Regenerate and Differentiate Xiaoming Liu,1 Yulong Zhang,1 Meihui Luo,1 Gary S. Mansfield,2 Lloyd G. Mitchell,2 John F. Engelhardt.1 1 Gene Therapy Center, The University of Iowa, Iowa City, IA, United States; 2Intronn Inc, 9700 Great Seneca Hwy, Rockville, MD, United States. Using a human bronchial xenograft model, we have assessed the extent to which airway stem cells can be reconstituted with either a full-length CFTR or LacZ transgene. Primary human airway cells infected in vitro with either RV.CBLacZ or RV.CBCFTR retroviral vectors demonstrated similar levels of transduction (10-20%). However, when these cells were seeded into bronchial xenografts, clonal expansion of CFTR expressing stem cells was more than 100fold less than that seen for LacZ, as detected using in situ hybridization with a vector-specific probe. In contrast to LacZ, the few CFTR expressing clones observed were predominantly composed of undifferentiated non-ciliated cells. Together, these results suggested that CFTR expression in human adult airway stem cells may confer a selective disadvantage in terms of proliferation and/or differentiation. To more clearly test this hypothesis, we attempted to more closely reconstitute endogenous patterns of CFTR expression using Spliceosome-Mediated RNA Trans-splicing (SMaRT), reasoning that regulated expression of CFTR might minimize the toxic effects seen following ectopic expression. To this end, we constructed three different VSV-G pseudotyped retroviral vectors (RV.CBLacZ, RV.CBCFTR, RV.CFTRPTM-24), infected human bronchial airway epithelial cells in vitro and seeded these cells into human bronchial xenografts. RV.CFTRPTM-24 encoded a pre-trans-splicing molecule (PTM) consisting of human CFTR exons 10-24 preceded by an intron 9 trans-splicing domain. Taq-man PCR was used to normalize transduction of primary cells between the various vector groups as well as to quantify vector genome persistence at six weeks following cellular reconstitution of tracheal xenografts. Results from this study demonstrated a substantially higher level of epithelial reconstitution and differentiation in xenografts infected with RV.CBLacZ and RV.CFTRPTM-24 vectors as compared to RV.CBCFTR. A striking difference in differentiation was seen in the histology between RV.CFTRPTM-24 and RV.CBCFTR infected xenografts. Ciliated cells were abundant in RV.CFTRPTM-24 infected xenografts but were completely absent in RV.CBCFTR infected xenografts, which were largely devoid of epithelial cells and contained patches of squamous epithelia. Although the abundance of ciliated cells was similar between RV.CBLacZ (control vector) and RV.CFTRPTM24 infected xenograft epithelium, the height of the epithelium was slightly reduced in RV.CFTRPTM-24 infected xenografts, which is consistent with a reduced but not absent toxicity from the PTM-24 construct. Differences in epithelial reconstitution between the various vectors were also reflected in the abundance of integrated viral genomes in xenograft epithelia as determined by Taq-man PCR. These data demonstrate that SMaRT repair of CFTR can reduce the toxicity accompanying ectopic CFTR expression in airway stem cells. These findings suggest that SMaRT technology may increase stem cell reconstitution of CFTR by better regulating the quantity, or perhaps more importantly, the stage in cell development where CFTR is expressed. Supported by SBIR grant R44 DK 56526 (LGM) from the NIH to Intronn and RO1 DK47967 to JFE.
Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts Copyright ® The American Society of Gene Therapy
NAKED DNA: APPLICATIONS 1197. siRNA Transiently Inhibits Exogenous Gene Expression from Mouse Liver Qiuming Chu,1 Macey Joseph,1 Bradley Hodges,1 Seng H. Cheng,1 Ronald K. Scheule.1 1 Genzyme Corporation, Framingham, MA, United States. Hydrodynamic delivery of plasmid DNA (pDNA) to mammalian liver has been shown to result in relatively high levels of expression, predominantly from hepatocytes. For therapeutic reasons, it may become necessary to downregulate liver expression coming from either endogenous genes or gene therapy vectors. To ask whether small interfering RNA could act as a “gene switch” to turn off expression of a selected gene, we have delivered a reporter gene or a therapeutic gene to mouse models by hydrodynamic delivery and evaluated the kinetics and degree to which siRNA could inhibit expression from those genes. We show that expression from a hydrodynamically-delivered chloramphenicol acetyltransferase (CAT) plasmid can be inhibited by co-delivered siRNA in a dosedependent fashion, and that inhibition is specific and lasts for at least 2 weeks. We could also show that expression from a previouslyadministered CAT pDNA could be inhibited by hydrodynamic delivery of the corresponding siRNA a week later. This inhibition could not be replicated using the single-stranded sense or antisense components of the siRNA. We have also hydrodynamically-delivered pDNA encoding the a-galactosidase A (a-gal) gene to BALB/c and to Fabry mice, the latter model lacking the endogenous gene for agal. Several siRNAs were designed and tested by co-delivering them with an a-gal pDNA to mouse liver by hydrodynamics. The most potent siRNA demonstrated a rapid, dose-dependent inhibition of a-gal expression. Using a 200-fold molar excess of this siRNA relative to target pDNA, a-gal expression was inhibited by >99%. Importantly, expression returned to normal levels in ~3 weeks. This effect was demonstrated to be specific in that under the same conditions, (i) this siRNA had no effect on the expression kinetics of the irrelevant reporter gene, secreted alkaline phosphatase, and (ii) an siRNA designed against a chloramphenicol acetyltransferase message had essentially no effect on a-gal expression kinetics. These results suggest that siRNA can be used to turn off gene expression selectively and efficiently from exogenously administered therapeutic genes in the mammalian liver. All authors are employees of Genzyme Corporation
1198. Inhibition of Angiogenesis and Tumor Progression by Hydrodynamic Co-Transfection of Angiogenesis K1-3, Endostatin and Saxatilin Genes Keun Sik Kim,1 Hong Sung Kim,1 Myung Min Choi,1 Ji Eun Lee,1 Kwang Hoe Chung,2 Yong Serk Park.1 1 Biomedical Laboratory Science, Yonsei University, Wonju, Gangwon, Republic of Korea; 2Cardiovascular Research Institute, College of Medicine, Yonsei University, Seoul, Republic of Korea. Hydrofection (hydrodynamics-based transfection), which allows an efficient expression of exogenous genes in mice, is a convenient procedure for gene function studies. Previously, we have shown that administration of the angiostatin K1-3 (Kringle 1-3) and endostatin genes with cationic liposomes inhibited tumor growth. Angiostatin and endostatin are potent endothelial cell growth inhibitors that have been shown to inhibit angiogenesis in vivo and tumor growth in mice. Recently, it has been reported that saxatilin, a novel disintegrin purified from snake (Gloydius saxatilis) venom, strongly inhibits human platelet aggregation, bFGF-induced proliferation of HUVEC and vitronectin-induced smooth muscle S463
NAKED DNA: APPLICATIONS cell (SMC) migration. In this study, we have examined whether hydrofection of the mouse angiostatin K1-3 (pFLAG-Angio K1/3), endostatin (pFLAG-Endo) and saxatilin (pFLAG-Saxatilin) produces in vivo expression of the proteins and anti-angiogenic tumor inhibition. The hydrodynamically administered plasmids containing angiostatin, endostatin and saxatilin effectively reduced angiogenesis, as analyzed in Matrigel-implanted mice. Hydrodynamic coadministaration of the genes more effectively inhibited the B16BL6 melanoma growth and pulmonary metastasis in mice than administration of either gene alone. Compared with the untreated control group, co-treatment with pFLAG-Angio K1/3, pFLAGEndo and pFLAG-Saxatilin inhibited B16BL6 tumor growth by 82 % and B16BL6 pulmonary metastasis by 88 %. Theses results provide further evidence that systemic expression of novel combinations of the antiangiogenic factors, angiostatin K1-3, endostatin and saxatilin, are an alternative procedure for antiangiogenic cancer therapy.
1199. Intravascular Insulin Gene Delivery as Potential Therapeutic Intervention of the Diabetes Mellitus Kakei Yasutomi,1 Hidetsugu Asada,1 Tsunao Kishida,1 Feng-De Cui,1 Toshiyuki Takeuchi,2 Jiro Imanishi,1 Osam Mazda.1 1 Microbiology, Kyoto Prefectural University of Medicine, Kyoto, Kyoto, Japan; 2Molecular Medicine, Institute for Molecular and Cellular Regulation, Gunma University, Gunma, Gunma, Japan. Diabetes Mellitus (DM) is caused by insulin insufficiency and characterized by hyperglycemia and chronic complications including neural, retinal and renal lesions. As an attempt to genetically manage DM, intravascular transfection in vivo of insulin gene was conducted. Although naked DNA is low immunogenic and free of virusassociated side effects, the critical disadvantage of naked DNA transfer has been the poor transfection efficiency and the low expression level. Recently, however, it was shown that extremely high transgene expression is obtained in the liver of mice given an intravenous injection of naked DNA via tail vein under high pressure (the hydrodynamics-based transfection).The Epstein-Barr virus (EBV)based plasmid vectors equipped with the EBV nuclear antigen 1 (EBNA1) gene and oriP sequence enable high and prolonged expression when transfected into animals through the hydrodynamics-based procedure. The rat proinsulin-1 gene cDNA engineered to harbor furin consensus cleavage sequences was inserted into EBV-based plasmid vectors, which were subsequently injected into streptozotocin (STZ)-induced diabetic mice (C57/BL6 female mice, 8 wks) via their tail vein under high pressure. In case a powerful, ubiquitous promoter was employed to drive both the proinsulin and EBNA1 genes, the treatment markedly improved hyperglycemia of diabetic mice, accompanied by a considerable increase in serum insulin concentrations. The longevity of the therapeutic outcome was clearly dependent on the structure of the plasmid vectors To obtain blood glucose-dependent production of the transgenic insulin, the rat insulin promoter (RIP) was employed to regulate the EBNA1 gene and/or the rat proinsulin-1 gene. The glucose tolerance test (GTT) was performed after the intravascular transfection of the plasmids. Although the fasting blood glucose did not decline at a significant level, the GTT revealed that insulin production was regulated in the liver in a blood glucose level-dependent manner. From a standpoint of physiological secretion of insulin, the liver is more suitable than skeletal muscle for insulin gene therapy, because it is a major target organ of insulin action, playing critical roles in glucose homeostasis. In normal individuals, insulin is secreted from the pancreas to the portal vein according to the glucose levels monitored at the pancreatic circulation. As a result, the portal insulin concentration is higher than that in systemic circulation, suppressing S464
gluconeogenesis in the liver to down-regulate blood glucose. Because this feedback circuit is hampered in DM patients, local production of insulin in the liver may be of therapeutic value, effectively suppressing hepatic gluconeogenesis. The present results suggest that the intravascular insulin gene transfer may provide novel therapeutic means of controlling DM.
1200. Controlled Release of Plasmid DNA from Biodegradable Polymeric Scaffolds Induces Bone Formation In Vivo Pelled Gadi,1 Jang Jae-Hung,2 Zilberman Yoram,1 Shea D. Lonnie,2 Gazit Dan.1 1 Skeletal Biotechnology Laboratory, Hebrew UniversityHadassah Medical Center, Jerusalem, Israel; 2Department of Chemical Engineering, Northwestern University, Evanston, IL. Tissue replacements are required in orthopaedic medicine, which deals with various morbidities in the skeletal system. Most gene therapy-based approaches for bone repair involve the use of viral vectors encoding for osteogenic genes. Due to the risk of evoking the immune response reported in several viral-based gene therapy protocols, a huge effort is being conducted in order to develop a non-viral method for gene delivery. Current non-viral techniques are limited in efficiency of transduction. In addition, temporal or spatial control of DNA release from these vectors is impossible. We hypothesized that in vivo bone tissue formation could be achieved by a temporal and spatial controlled release of DNA from biodegradable scaffolds. A gas foaming process has been utilized to fabricate poly(lactide-co-glycolide) (PLG) porous scaffolds from polymeric microspheres. The microspheres, with diameters ranging from 2 to 200 um, were created using an emulsion process and were mixed with lyophilized plasmid DNA. These microspheres were further compressed molded into a cylindrical construct at 1500 psi. The DNA release kinetics from the scaffolds was analyzed using the Hoechst dye-binding assay. The ability of DNA released from the polymer scaffolds to transfect cells in vitro was subsequently examined using NIH/3T3 cells grown on scaffolds containing the reporter genes luciferase, b-galactosidase, and GFP. For the in vivo experiments, scaffolds containing 50 ugr of hBMP2 plasmid, the osteogenic gene, were prepared. In order to non-invasively detect osteogenic activity in vivo, we implanted the scaffolds, subcutaneously and in a calavrial bone defect, in transgenic mice, which express Luciferase under human Osteocalcin promoter. Since Osteocalcin is a late osteogenic marker, the detection of Luciferase bioluminescence by a CCCD system can be quantitatively correlated to an osteogenic process. The implants were harvested after 30 days and analyzed using uCT and immunohistochemistry. We were able to control the kinetics of plasmid DNA release from the scaffold, by varying the molecular weight of the polymer or the size of microspheres. In addition, NIH/3T3 cells grown on DNA loaded scaffolds, expressed the marker genes as demonstrated by Luciferase assay and X-Gal staining. Intense bioluminescence was detected in the site of implantation in the transgenic mice indicating osteogenic activity. The uCT 3-D analysis demonstrated the presence of calcified foci within the implantation site. Histochemistry and Immunohistochemistry anti Collagen II, performed on the samples revealed the presence of bone trabecules adjacent to PLG scaffold degradation material. We conclude that the controlled release of plasmid DNA from PLG scaffolds could serve as: 1) An efficient way for gene transfer to cells in vitro and in vivo, and 2) A therapeutic tool for bone regeneration.
Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts
Copyright ® The American Society of Gene Therapy
Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts Copyright ® The American Society of Gene Therapy
NAKED DNA: APPLICATIONS 1197. siRNA Transiently Inhibits Exogenous Gene Expression from Mouse Liver Qiuming Chu,1 Macey Joseph,1 Bradley Hodges,1 Seng H. Cheng,1 Ronald K. Scheule.1 1 Genzyme Corporation, Framingham, MA, United States. Hydrodynamic delivery of plasmid DNA (pDNA) to mammalian liver has been shown to result in relatively high levels of expression, predominantly from hepatocytes. For therapeutic reasons, it may become necessary to downregulate liver expression coming from either endogenous genes or gene therapy vectors. To ask whether small interfering RNA could act as a “gene switch” to turn off expression of a selected gene, we have delivered a reporter gene or a therapeutic gene to mouse models by hydrodynamic delivery and evaluated the kinetics and degree to which siRNA could inhibit expression from those genes. We show that expression from a hydrodynamically-delivered chloramphenicol acetyltransferase (CAT) plasmid can be inhibited by co-delivered siRNA in a dosedependent fashion, and that inhibition is specific and lasts for at least 2 weeks. We could also show that expression from a previouslyadministered CAT pDNA could be inhibited by hydrodynamic delivery of the corresponding siRNA a week later. This inhibition could not be replicated using the single-stranded sense or antisense components of the siRNA. We have also hydrodynamically-delivered pDNA encoding the a-galactosidase A (a-gal) gene to BALB/c and to Fabry mice, the latter model lacking the endogenous gene for agal. Several siRNAs were designed and tested by co-delivering them with an a-gal pDNA to mouse liver by hydrodynamics. The most potent siRNA demonstrated a rapid, dose-dependent inhibition of a-gal expression. Using a 200-fold molar excess of this siRNA relative to target pDNA, a-gal expression was inhibited by >99%. Importantly, expression returned to normal levels in ~3 weeks. This effect was demonstrated to be specific in that under the same conditions, (i) this siRNA had no effect on the expression kinetics of the irrelevant reporter gene, secreted alkaline phosphatase, and (ii) an siRNA designed against a chloramphenicol acetyltransferase message had essentially no effect on a-gal expression kinetics. These results suggest that siRNA can be used to turn off gene expression selectively and efficiently from exogenously administered therapeutic genes in the mammalian liver. All authors are employees of Genzyme Corporation
1198. Inhibition of Angiogenesis and Tumor Progression by Hydrodynamic Co-Transfection of Angiogenesis K1-3, Endostatin and Saxatilin Genes Keun Sik Kim,1 Hong Sung Kim,1 Myung Min Choi,1 Ji Eun Lee,1 Kwang Hoe Chung,2 Yong Serk Park.1 1 Biomedical Laboratory Science, Yonsei University, Wonju, Gangwon, Republic of Korea; 2Cardiovascular Research Institute, College of Medicine, Yonsei University, Seoul, Republic of Korea. Hydrofection (hydrodynamics-based transfection), which allows an efficient expression of exogenous genes in mice, is a convenient procedure for gene function studies. Previously, we have shown that administration of the angiostatin K1-3 (Kringle 1-3) and endostatin genes with cationic liposomes inhibited tumor growth. Angiostatin and endostatin are potent endothelial cell growth inhibitors that have been shown to inhibit angiogenesis in vivo and tumor growth in mice. Recently, it has been reported that saxatilin, a novel disintegrin purified from snake (Gloydius saxatilis) venom, strongly inhibits human platelet aggregation, bFGF-induced proliferation of HUVEC and vitronectin-induced smooth muscle S463
NAKED DNA: APPLICATIONS cell (SMC) migration. In this study, we have examined whether hydrofection of the mouse angiostatin K1-3 (pFLAG-Angio K1/3), endostatin (pFLAG-Endo) and saxatilin (pFLAG-Saxatilin) produces in vivo expression of the proteins and anti-angiogenic tumor inhibition. The hydrodynamically administered plasmids containing angiostatin, endostatin and saxatilin effectively reduced angiogenesis, as analyzed in Matrigel-implanted mice. Hydrodynamic coadministaration of the genes more effectively inhibited the B16BL6 melanoma growth and pulmonary metastasis in mice than administration of either gene alone. Compared with the untreated control group, co-treatment with pFLAG-Angio K1/3, pFLAGEndo and pFLAG-Saxatilin inhibited B16BL6 tumor growth by 82 % and B16BL6 pulmonary metastasis by 88 %. Theses results provide further evidence that systemic expression of novel combinations of the antiangiogenic factors, angiostatin K1-3, endostatin and saxatilin, are an alternative procedure for antiangiogenic cancer therapy.
1199. Intravascular Insulin Gene Delivery as Potential Therapeutic Intervention of the Diabetes Mellitus Kakei Yasutomi,1 Hidetsugu Asada,1 Tsunao Kishida,1 Feng-De Cui,1 Toshiyuki Takeuchi,2 Jiro Imanishi,1 Osam Mazda.1 1 Microbiology, Kyoto Prefectural University of Medicine, Kyoto, Kyoto, Japan; 2Molecular Medicine, Institute for Molecular and Cellular Regulation, Gunma University, Gunma, Gunma, Japan. Diabetes Mellitus (DM) is caused by insulin insufficiency and characterized by hyperglycemia and chronic complications including neural, retinal and renal lesions. As an attempt to genetically manage DM, intravascular transfection in vivo of insulin gene was conducted. Although naked DNA is low immunogenic and free of virusassociated side effects, the critical disadvantage of naked DNA transfer has been the poor transfection efficiency and the low expression level. Recently, however, it was shown that extremely high transgene expression is obtained in the liver of mice given an intravenous injection of naked DNA via tail vein under high pressure (the hydrodynamics-based transfection).The Epstein-Barr virus (EBV)based plasmid vectors equipped with the EBV nuclear antigen 1 (EBNA1) gene and oriP sequence enable high and prolonged expression when transfected into animals through the hydrodynamics-based procedure. The rat proinsulin-1 gene cDNA engineered to harbor furin consensus cleavage sequences was inserted into EBV-based plasmid vectors, which were subsequently injected into streptozotocin (STZ)-induced diabetic mice (C57/BL6 female mice, 8 wks) via their tail vein under high pressure. In case a powerful, ubiquitous promoter was employed to drive both the proinsulin and EBNA1 genes, the treatment markedly improved hyperglycemia of diabetic mice, accompanied by a considerable increase in serum insulin concentrations. The longevity of the therapeutic outcome was clearly dependent on the structure of the plasmid vectors To obtain blood glucose-dependent production of the transgenic insulin, the rat insulin promoter (RIP) was employed to regulate the EBNA1 gene and/or the rat proinsulin-1 gene. The glucose tolerance test (GTT) was performed after the intravascular transfection of the plasmids. Although the fasting blood glucose did not decline at a significant level, the GTT revealed that insulin production was regulated in the liver in a blood glucose level-dependent manner. From a standpoint of physiological secretion of insulin, the liver is more suitable than skeletal muscle for insulin gene therapy, because it is a major target organ of insulin action, playing critical roles in glucose homeostasis. In normal individuals, insulin is secreted from the pancreas to the portal vein according to the glucose levels monitored at the pancreatic circulation. As a result, the portal insulin concentration is higher than that in systemic circulation, suppressing S464
gluconeogenesis in the liver to down-regulate blood glucose. Because this feedback circuit is hampered in DM patients, local production of insulin in the liver may be of therapeutic value, effectively suppressing hepatic gluconeogenesis. The present results suggest that the intravascular insulin gene transfer may provide novel therapeutic means of controlling DM.
1200. Controlled Release of Plasmid DNA from Biodegradable Polymeric Scaffolds Induces Bone Formation In Vivo Pelled Gadi,1 Jang Jae-Hung,2 Zilberman Yoram,1 Shea D. Lonnie,2 Gazit Dan.1 1 Skeletal Biotechnology Laboratory, Hebrew UniversityHadassah Medical Center, Jerusalem, Israel; 2Department of Chemical Engineering, Northwestern University, Evanston, IL. Tissue replacements are required in orthopaedic medicine, which deals with various morbidities in the skeletal system. Most gene therapy-based approaches for bone repair involve the use of viral vectors encoding for osteogenic genes. Due to the risk of evoking the immune response reported in several viral-based gene therapy protocols, a huge effort is being conducted in order to develop a non-viral method for gene delivery. Current non-viral techniques are limited in efficiency of transduction. In addition, temporal or spatial control of DNA release from these vectors is impossible. We hypothesized that in vivo bone tissue formation could be achieved by a temporal and spatial controlled release of DNA from biodegradable scaffolds. A gas foaming process has been utilized to fabricate poly(lactide-co-glycolide) (PLG) porous scaffolds from polymeric microspheres. The microspheres, with diameters ranging from 2 to 200 um, were created using an emulsion process and were mixed with lyophilized plasmid DNA. These microspheres were further compressed molded into a cylindrical construct at 1500 psi. The DNA release kinetics from the scaffolds was analyzed using the Hoechst dye-binding assay. The ability of DNA released from the polymer scaffolds to transfect cells in vitro was subsequently examined using NIH/3T3 cells grown on scaffolds containing the reporter genes luciferase, b-galactosidase, and GFP. For the in vivo experiments, scaffolds containing 50 ugr of hBMP2 plasmid, the osteogenic gene, were prepared. In order to non-invasively detect osteogenic activity in vivo, we implanted the scaffolds, subcutaneously and in a calavrial bone defect, in transgenic mice, which express Luciferase under human Osteocalcin promoter. Since Osteocalcin is a late osteogenic marker, the detection of Luciferase bioluminescence by a CCCD system can be quantitatively correlated to an osteogenic process. The implants were harvested after 30 days and analyzed using uCT and immunohistochemistry. We were able to control the kinetics of plasmid DNA release from the scaffold, by varying the molecular weight of the polymer or the size of microspheres. In addition, NIH/3T3 cells grown on DNA loaded scaffolds, expressed the marker genes as demonstrated by Luciferase assay and X-Gal staining. Intense bioluminescence was detected in the site of implantation in the transgenic mice indicating osteogenic activity. The uCT 3-D analysis demonstrated the presence of calcified foci within the implantation site. Histochemistry and Immunohistochemistry anti Collagen II, performed on the samples revealed the presence of bone trabecules adjacent to PLG scaffold degradation material. We conclude that the controlled release of plasmid DNA from PLG scaffolds could serve as: 1) An efficient way for gene transfer to cells in vitro and in vivo, and 2) A therapeutic tool for bone regeneration.
Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts
Copyright ® The American Society of Gene Therapy