- Email: [email protected]
Pseudotyping of Lentiviral Vectors with New Envelope G Proteins from Vesiculoviruses VSV_New Jersey, Chandipura and Piry
RNA VIRUS VECTORS III terminus of the murine leukemia virus (MLV) amphotropic and ecotropic envelope proteins. We found that retroviruses pseudotyped with the ScFv anti-TAC amphotropic fusion protein, but not those pseudotyped with the ecotropic fusion protein, were able to transduce HeLa cells that expressed the chimeric TAC receptors. Gene transfer was significantly less efficient in HeLa cells that expressed TAC-DKQTLL, a receptor that is rapidly internalized and transported to the lysosomes, than in HeLa cells that expressed TAC, TAC-CD16, or in the parent cell line that did not express any of the chimeric TAC receptors. Gene transfer was most efficient to HeLa cells that expressed TAC, a receptor with a long residence time on the cell surface (half-life > 40 hours). Our findings suggest that the efficiency of retrovirus transduction is a function of the intracellular trafficking itinerary of cell-surface retrovirus-binding proteins. The implications of our findings to the development of improved gene transfer technologies for the purposes of human gene therapy will be discussed.
730. Lentiviral Vectors Pseudotyped with Envelope Glycoproteins Derived from Human Parainfluenza Virus Type 3 Cindy Jung,1 Bradley N. Grzybowski,1 Suxiang Tong,2 LiTeng Cheng,2 Richard W. Compans,2 Joseph M. Le Doux.1 1 The Wallace H. Coulter Department of Biomedical Engineering, Georgia Tech and Emory University, Atlanta, GA; 2Department of Microbiology and Immunology, Emory University, Atlanta, GA. We describe the generation of lentiviruses pseudotyped with human parainfluenza type 3 envelope glycoproteins. Lentivirus particles, expressed in 293T/17 cells, incorporate HPIV3 hemagglutinin-neuraminidase (HN) and fusion (F) proteins into their lipid bilayers and are able to transduce 293T/17 and MDCK cells. Neuraminidase, AZT, and anti-HPIV3 antisera block transduction, which is consistent with lentiviral-mediated transduction via the sialated receptors for HPIV3. Our findings show that LV(HPIV3) pseudotyped viruses can be formed and may have a number of useful properties for human gene transfer. Although gene transfer efficiencies are low, we hypothesize they are most likely due to low levels of envelope protein incorporation. We have therefore initiated a number of independent approaches to increase envelope protein incorporation levels. The results of the envelope incorporation studies, and their implications for the construction of pseudotyped viruses, will be discussed.
731. Pseudotyping of Lentiviral Vectors with New Envelope G Proteins from Vesiculoviruses VSV-New Jersey, Chandipura and Piry Akhil Banerjea,1 Ramesh Akkina.1 1 Microbiology, Immunology & Pathology, Colorado State University, Fort Collins, CO. Retroviral and lentiviral vectors are highly efficient in gene transfer. Their tropism, stability and ability to be concentrated by ultracentrifugation were greatly increased by pseudotyping with the vesicular stomatitis virus G (VSV-G) protein. This allowed expanded use of these vectors for many purposes. Although most current gene therapy studies focused on ex vivo gene transductions with these vectors, the next desirable step is to be able to inject them directly for systemic delivery. While vectors pseudotyped with VSV-G from Indiana serotype yielded high levels of gene transfer, successive rounds of their in vivo administration will have immunological limitations because this protein is known to elicit strong neutralizing antibodies. Hence, for the purpose of repeated administrations, additional non-crossreactive envelope proteins that have comparable value in pseudotyping retro- and lentiviral vectors are needed. Furthermore, as VSV-G is highly fusogenic, there is S278
cytotoxicity associated with its use via concentrated vector stocks. Accordingly, a less cytotoxic envelope protein for pseudotyping is also highly desirable. Taking these criteria into consideration, we tested envelope proteins from three different vesiculoviruses, namely VSV-New Jersey, Chandipura and Piry. A third generation lentiviral vector encoding GFP reporter gene was pseudotyped with each of these envelope G proteins in a transient transfection packaging system. The respective pseudotyped lentiviral vectors were used to transduce different mammalian cells including BHK-21, HeLa, Cos, Prostate cancer cell lines, as well as primary human CD34+ hematopoietic cells and macrophages derived in vitro. The transfection efficiencies were compared to the vector pseudotyped with VSV-G Indiana. Our results showed that the three new G proteins used were effective in pseudotyping the lentiviral vectors as assayed by GFP marker expression from transduced cells. Differential transduction levels were found between different cell types indicating variation in cell tropism between different envelopes. These studies established that envelope proteins from related vesiculoviruses are useful for pseudotyping lentiviral vectors and provide alternative non-crossreactive envelope proteins for in vivo administration of gene transfer vectors.
732. A Novel Double-Pseudotyped Viral Vector for Gene Therapy Ning Chai,1 Gary Kobinger,2 Giancarlo Tanzi,1 James Wilson,2 Paul Bates.1 1 Department of Microbiology, University of Pennsylvania, Philadelphia, PA; 2Division of Medical Genetics, University of Pennsylvania, Philadelphia, PA. A major goal of gene therapy is to achieve highly efficient and specific delivery of therapeutic genes into target cells. While current gene delivery vectors have been successful in many cases, there are still some barriers. These include low titer, low specificity, shortterm expression, inconvenient vector design for different targets and inefficient targeting of certain tissues due to the lack of known receptors. For retroviral-based vectors, one limitation is the low flexibility in modifying the viral envelope glycoprotein to achieve the desired tropism since the receptor binding domain (SU) and the fusion domain (TM) are coupled. In an attempt to improve some of these aspects, we have designed a novel two-molecule targeting system that employs a lentiviral-based double-pseudotyped vector in which two different proteins are incorporated into the viral membrane to separate the receptor binding and membrane fusion functions. The first molecule, a mutant form of influenza hemagglutinin (HA) deficient for intrinsic receptor-binding function, supplies the membrane fusion machinery. The second molecule, a modified cellular surface protein that encodes two copies of the IgG-binding domain from staphylococcal protein A, binds to specific receptors on target cells through antibodies used to coat the virus. Specificity of the vector can be readily altered by employing a different antibody. The validity of this system was confirmed by experiments in which vectors coated with specific antibodies are used to infect 293T cells expressing the cognate antigens. Using several different pairs of targeting antibody and surface receptor, the titer achieved with the double pseudotypes is high (106-108) and at least three logs over background infection of cells lacking receptor. Antibodymediated infection shows a dose dependence on the level of antigen expressed on the target cell surface. FACS analysis of mixed cell populations demonstrates a roughly 100-fold preferential infection of antigen-expressing cells by antibody-coated viruses. Application of this system is demonstrated by specific infection of tumor cells expressing EGFRvIII through an EGFRvIII-specific antibody, and by efficient targeting of airway epithelial cells in vitro and in vivo through a monoclonal antibody that binds to CD46 on the apical surface of differentiated airway epithelial cells. Comparison with Molecular Therapy Volume 9, Supplement 1, May 2004 Copyright © The American Society of Gene Therapy
730. Lentiviral Vectors Pseudotyped with Envelope Glycoproteins Derived from Human Parainfluenza Virus Type 3 Cindy Jung,1 Bradley N. Grzybowski,1 Suxiang Tong,2 LiTeng Cheng,2 Richard W. Compans,2 Joseph M. Le Doux.1 1 The Wallace H. Coulter Department of Biomedical Engineering, Georgia Tech and Emory University, Atlanta, GA; 2Department of Microbiology and Immunology, Emory University, Atlanta, GA. We describe the generation of lentiviruses pseudotyped with human parainfluenza type 3 envelope glycoproteins. Lentivirus particles, expressed in 293T/17 cells, incorporate HPIV3 hemagglutinin-neuraminidase (HN) and fusion (F) proteins into their lipid bilayers and are able to transduce 293T/17 and MDCK cells. Neuraminidase, AZT, and anti-HPIV3 antisera block transduction, which is consistent with lentiviral-mediated transduction via the sialated receptors for HPIV3. Our findings show that LV(HPIV3) pseudotyped viruses can be formed and may have a number of useful properties for human gene transfer. Although gene transfer efficiencies are low, we hypothesize they are most likely due to low levels of envelope protein incorporation. We have therefore initiated a number of independent approaches to increase envelope protein incorporation levels. The results of the envelope incorporation studies, and their implications for the construction of pseudotyped viruses, will be discussed.
731. Pseudotyping of Lentiviral Vectors with New Envelope G Proteins from Vesiculoviruses VSV-New Jersey, Chandipura and Piry Akhil Banerjea,1 Ramesh Akkina.1 1 Microbiology, Immunology & Pathology, Colorado State University, Fort Collins, CO. Retroviral and lentiviral vectors are highly efficient in gene transfer. Their tropism, stability and ability to be concentrated by ultracentrifugation were greatly increased by pseudotyping with the vesicular stomatitis virus G (VSV-G) protein. This allowed expanded use of these vectors for many purposes. Although most current gene therapy studies focused on ex vivo gene transductions with these vectors, the next desirable step is to be able to inject them directly for systemic delivery. While vectors pseudotyped with VSV-G from Indiana serotype yielded high levels of gene transfer, successive rounds of their in vivo administration will have immunological limitations because this protein is known to elicit strong neutralizing antibodies. Hence, for the purpose of repeated administrations, additional non-crossreactive envelope proteins that have comparable value in pseudotyping retro- and lentiviral vectors are needed. Furthermore, as VSV-G is highly fusogenic, there is S278
cytotoxicity associated with its use via concentrated vector stocks. Accordingly, a less cytotoxic envelope protein for pseudotyping is also highly desirable. Taking these criteria into consideration, we tested envelope proteins from three different vesiculoviruses, namely VSV-New Jersey, Chandipura and Piry. A third generation lentiviral vector encoding GFP reporter gene was pseudotyped with each of these envelope G proteins in a transient transfection packaging system. The respective pseudotyped lentiviral vectors were used to transduce different mammalian cells including BHK-21, HeLa, Cos, Prostate cancer cell lines, as well as primary human CD34+ hematopoietic cells and macrophages derived in vitro. The transfection efficiencies were compared to the vector pseudotyped with VSV-G Indiana. Our results showed that the three new G proteins used were effective in pseudotyping the lentiviral vectors as assayed by GFP marker expression from transduced cells. Differential transduction levels were found between different cell types indicating variation in cell tropism between different envelopes. These studies established that envelope proteins from related vesiculoviruses are useful for pseudotyping lentiviral vectors and provide alternative non-crossreactive envelope proteins for in vivo administration of gene transfer vectors.
732. A Novel Double-Pseudotyped Viral Vector for Gene Therapy Ning Chai,1 Gary Kobinger,2 Giancarlo Tanzi,1 James Wilson,2 Paul Bates.1 1 Department of Microbiology, University of Pennsylvania, Philadelphia, PA; 2Division of Medical Genetics, University of Pennsylvania, Philadelphia, PA. A major goal of gene therapy is to achieve highly efficient and specific delivery of therapeutic genes into target cells. While current gene delivery vectors have been successful in many cases, there are still some barriers. These include low titer, low specificity, shortterm expression, inconvenient vector design for different targets and inefficient targeting of certain tissues due to the lack of known receptors. For retroviral-based vectors, one limitation is the low flexibility in modifying the viral envelope glycoprotein to achieve the desired tropism since the receptor binding domain (SU) and the fusion domain (TM) are coupled. In an attempt to improve some of these aspects, we have designed a novel two-molecule targeting system that employs a lentiviral-based double-pseudotyped vector in which two different proteins are incorporated into the viral membrane to separate the receptor binding and membrane fusion functions. The first molecule, a mutant form of influenza hemagglutinin (HA) deficient for intrinsic receptor-binding function, supplies the membrane fusion machinery. The second molecule, a modified cellular surface protein that encodes two copies of the IgG-binding domain from staphylococcal protein A, binds to specific receptors on target cells through antibodies used to coat the virus. Specificity of the vector can be readily altered by employing a different antibody. The validity of this system was confirmed by experiments in which vectors coated with specific antibodies are used to infect 293T cells expressing the cognate antigens. Using several different pairs of targeting antibody and surface receptor, the titer achieved with the double pseudotypes is high (106-108) and at least three logs over background infection of cells lacking receptor. Antibodymediated infection shows a dose dependence on the level of antigen expressed on the target cell surface. FACS analysis of mixed cell populations demonstrates a roughly 100-fold preferential infection of antigen-expressing cells by antibody-coated viruses. Application of this system is demonstrated by specific infection of tumor cells expressing EGFRvIII through an EGFRvIII-specific antibody, and by efficient targeting of airway epithelial cells in vitro and in vivo through a monoclonal antibody that binds to CD46 on the apical surface of differentiated airway epithelial cells. Comparison with Molecular Therapy Volume 9, Supplement 1, May 2004 Copyright © The American Society of Gene Therapy