Selective Transduction of Primary Glioblastoma Multiforme Endothelial Cells
Selective Transduction of Primary Glioblastoma Multiforme Endothelial Cells
Object: Adenovirus transduction in gene therapy is dependent on the expression of the coxsackie virusadenovirus receptor (CAR) for initial binding and on the integrin receptors (αvβ3, αvβ5) for viral internalization. Low and variable expression of CAR may be responsible for the low transduction rates seen with native adenoviral vectors. The goal of this study was to demonstrate increased transduction efficiency by retargeting the adenovirus with a fibroblast growth factor (FGF) ligand, FGF-2.
Methods: The retargeted adenoviruses were used to transduce human glioblastoma multiforme (GBM)derived ECs (tumor-associated brain endothelial cells [TuBECs]), in which there is minimal CAR expression but a high expression of FGF receptor (FGFR). The results demonstrate that the transduction efficiency of TuBECs can reach as high as 80% when one uses an FGF2-conjugated adenovirus containing green fluorescent protein (FGF2-AdGFP) yet be only 5% when one uses the native adenovirus (AdGFP). The TuBECs were transduced with either a native adenovirus (AdHSV-TK) or a retargeted adenovirus (FGF2-AdHSV-TK), both of which carry the suicide herpes simplex virusthymidine kinase (HSV-TK) gene. Administered as a cytotoxic prodrug, ganciclovir induced a significant decline in the proliferation rate and increased apoptosis in TuBECs treated with the retargeted adenovirus, compared with its effect on TuBECs treated with the native adenovirus. Increased transduction efficiency was determined by performing GFP-based flow cytometry, and the expression of the TK protein by the retargeted adenovirus was assessed by performing an immunohistochemical analysis focused on HSV-TK. The mechanism of cytotoxicity was determined to be apoptosis by performing a terminal deoxynucleotidyl transferasemediated deoxyuridine triphosphate nick-end labeling assay.
Conclusions: Fibroblast growth factor2-retargeted adenoviral vectors may be used to increase the transduction of GBM-derived endothelial cells, enabling a new and efficient antiangiogenesis strategy for the treatment of malignant gliomas.
Adenoviral vectors have been extensively used in laboratory experiments and clinical trials for GBM gene therapy. They can be easily produced in high titers and purity, can accommodate large transgenes (< 7.5 kb), and can achieve high levels of transgene expression. Cellular transduction by a native adenovirus involves initial binding to CAR via the C-terminal knob domain of the viral fiber protein. Subsequent binding to the integrin receptors (αvβ3, αvβ5) via the arginyl-glycyl-aspartic acid motif within the adenoviral penton base protein is responsible for viral internalization in the form of clathrin-coated vesicles through receptor-mediated endocytosis. Additionally, the heparin sulfate proteoglycans are also responsible for viral entry into the cell. On systemic delivery, the native adenoviral vector primarily localizes in the liver due to its high CAR expression. Other sites of high CAR expression include the heart, prostate, pancreas, lung, and kidney. Subsequently, within 24 hours, most of the native adenovirus is cleared from the circulation, primarily through biliary excretion.
There have been recent published reports of gene therapy clinical trials in which adenoviral vectors have been used in the treatment of GBM, including the multicenter trial of ONYX-015 conducted by Chiocca, et al. Overall, these trials have produced mixed levels of therapeutic success, the reasons for which are manifold. The low and variable expression of CAR in GBMs causes a low rate of transduction by adenoviruses in these tumors, even when the vector is delivered regionally. Localized delivery of the vector into the tumor bed is not always clinically feasible, and systemic delivery of the adenoviral vector is even less effective because of sequestration of the vector in the liver and rapid clearance from the body. In such a case, it is necessary to deliver a higher dose of the adenoviral agent to achieve therapeutic levels; this can result in toxic damage to the liver, which manifests as elevated serum transaminases, inflammation, hepatocellular necrosis, and other systemic effects. The broad tropism of the native adenovirus and the lack of expression of CAR on the ECs preclude the systemic delivery of the adenovirus to target tumor vasculature. Other problems associated with the use of native adenoviruses include systemic toxicity, which manifests as an immune response that is both humoral and cellular in nature, particularly in the face of repeated administration.
Glioblastomas multiforme are characterized by a high degree of angiogenesis, which is a driving force in tumor growth and proliferation. Targeting tumor vasculature is therefore an attractive option for the treatment of these tumors. The stable and high expression of certain endothelial proteins provides important foci that can be targeted to disrupt tumor vasculature and to induce tumor regression. Such an approach to targeted delivery of therapeutic agents can lead to enhanced efficacy and safety. For this to succeed, it is necessary to develop a vector that can be systemically delivered to target the tumor vasculature. An ideal adenoviral vector is one that enables delivery of the gene of interest in a targeted manner and provides efficient trans-gene expression so therapy can be initiated. Also needed for success is the assurance that the vector is safe and is not associated with immune responses associated with the delivery of an exogenous agent.
The delivery of an adenovirus to specific cells can be enhanced by reducing its broad vector tropism by using various techniques such as a modification of the adenoviral fiber protein to alter its binding specificity or linking of ligands to cell-surface receptors via bispecific antibodies or bispecific chemical conjugates, for example, FGF-2, epidermal growth factor receptor, and epithelial cell adhesion molecule, biotinavidin bridges, or polymers such as polyethylene glycol or poly-(N-[2-hydroxypropyl]) methacrylamide.
Toward this goal, we used a gene therapy strategy based on retargeting adenovirus to tumor ECs by using FGFR-2, which is abundantly expressed on tumor ECs. We followed the transductional targeting technique by using FGF2 fragments conjugated to the native adenovirus. The addition of the FGF2-bispecific molecule enabled binding of the adenoviral capsid to the cell-surface FGFR, blocking uptake via CAR and redirecting viral tropism from the CAR receptor to high-affinity FGFRs. The FGF2-retargeted adenovirus was constructed by the chemical conjugation of a neutralizing antiadenoviral antibody to basic FGF-2. The binding affinity (Kd ~ 10 M) of FGF-2 to FGFR is at least three to four logs higher than other ligandreceptor interactions such as that between antigen and antibodies. The FGF2-conjugated adenoviral transduction is independent of CAR expression and dependent on the expression of FGFRs. On systemic delivery of the adenovirus, this property causes reduced delivery to nontargeted organs, especially the liver, which is the chief site of localization of the native adenovirus. Printz and colleagues demonstrated a 10- to 20-fold decrease in the amount of FGF2-conjugated adenovirus delivered to the liver, compared with the amount of native adenovirus. These authors also reported reduced liver toxicity, as measured by serum transaminase levels as well as by histopathological findings, and increased trans-gene expression in the tumor tissue following delivery of the retargeted adenovirus compared with delivery of the native adenovirus. Other diseases for which the FGF2-retargeted adenovirus was used for cancer therapy include ovarian cancer and Kaposi sarcoma.
A high level of FGF-2 expression has been reported in the tumor neovasculature. Therefore, retargeted adenoviruses can be used systemically to deliver genes of interest to tumor ECs. Such an approach enables not only the localization and enhancement of the viral transduction efficiency and transgene expression but also limits immune responses associated with adenoviruses.
In this study, we used an FGF2-retargeted adenovirus to transduce human GBMderived ECs (TuBECs) in vitro. We compared the transduction efficiency and the ability to induce transgene expression in both native and retargeted adenoviruses. We also examined rates of proliferation and apoptosis in cells transduced with the retargeted adenovirus and in cells transduced with the native adenovirus during treatment with the prodrug ganciclovir.
Abstract and Introduction
Abstract
Object: Adenovirus transduction in gene therapy is dependent on the expression of the coxsackie virusadenovirus receptor (CAR) for initial binding and on the integrin receptors (αvβ3, αvβ5) for viral internalization. Low and variable expression of CAR may be responsible for the low transduction rates seen with native adenoviral vectors. The goal of this study was to demonstrate increased transduction efficiency by retargeting the adenovirus with a fibroblast growth factor (FGF) ligand, FGF-2.
Methods: The retargeted adenoviruses were used to transduce human glioblastoma multiforme (GBM)derived ECs (tumor-associated brain endothelial cells [TuBECs]), in which there is minimal CAR expression but a high expression of FGF receptor (FGFR). The results demonstrate that the transduction efficiency of TuBECs can reach as high as 80% when one uses an FGF2-conjugated adenovirus containing green fluorescent protein (FGF2-AdGFP) yet be only 5% when one uses the native adenovirus (AdGFP). The TuBECs were transduced with either a native adenovirus (AdHSV-TK) or a retargeted adenovirus (FGF2-AdHSV-TK), both of which carry the suicide herpes simplex virusthymidine kinase (HSV-TK) gene. Administered as a cytotoxic prodrug, ganciclovir induced a significant decline in the proliferation rate and increased apoptosis in TuBECs treated with the retargeted adenovirus, compared with its effect on TuBECs treated with the native adenovirus. Increased transduction efficiency was determined by performing GFP-based flow cytometry, and the expression of the TK protein by the retargeted adenovirus was assessed by performing an immunohistochemical analysis focused on HSV-TK. The mechanism of cytotoxicity was determined to be apoptosis by performing a terminal deoxynucleotidyl transferasemediated deoxyuridine triphosphate nick-end labeling assay.
Conclusions: Fibroblast growth factor2-retargeted adenoviral vectors may be used to increase the transduction of GBM-derived endothelial cells, enabling a new and efficient antiangiogenesis strategy for the treatment of malignant gliomas.
Introduction
Adenoviral vectors have been extensively used in laboratory experiments and clinical trials for GBM gene therapy. They can be easily produced in high titers and purity, can accommodate large transgenes (< 7.5 kb), and can achieve high levels of transgene expression. Cellular transduction by a native adenovirus involves initial binding to CAR via the C-terminal knob domain of the viral fiber protein. Subsequent binding to the integrin receptors (αvβ3, αvβ5) via the arginyl-glycyl-aspartic acid motif within the adenoviral penton base protein is responsible for viral internalization in the form of clathrin-coated vesicles through receptor-mediated endocytosis. Additionally, the heparin sulfate proteoglycans are also responsible for viral entry into the cell. On systemic delivery, the native adenoviral vector primarily localizes in the liver due to its high CAR expression. Other sites of high CAR expression include the heart, prostate, pancreas, lung, and kidney. Subsequently, within 24 hours, most of the native adenovirus is cleared from the circulation, primarily through biliary excretion.
There have been recent published reports of gene therapy clinical trials in which adenoviral vectors have been used in the treatment of GBM, including the multicenter trial of ONYX-015 conducted by Chiocca, et al. Overall, these trials have produced mixed levels of therapeutic success, the reasons for which are manifold. The low and variable expression of CAR in GBMs causes a low rate of transduction by adenoviruses in these tumors, even when the vector is delivered regionally. Localized delivery of the vector into the tumor bed is not always clinically feasible, and systemic delivery of the adenoviral vector is even less effective because of sequestration of the vector in the liver and rapid clearance from the body. In such a case, it is necessary to deliver a higher dose of the adenoviral agent to achieve therapeutic levels; this can result in toxic damage to the liver, which manifests as elevated serum transaminases, inflammation, hepatocellular necrosis, and other systemic effects. The broad tropism of the native adenovirus and the lack of expression of CAR on the ECs preclude the systemic delivery of the adenovirus to target tumor vasculature. Other problems associated with the use of native adenoviruses include systemic toxicity, which manifests as an immune response that is both humoral and cellular in nature, particularly in the face of repeated administration.
Glioblastomas multiforme are characterized by a high degree of angiogenesis, which is a driving force in tumor growth and proliferation. Targeting tumor vasculature is therefore an attractive option for the treatment of these tumors. The stable and high expression of certain endothelial proteins provides important foci that can be targeted to disrupt tumor vasculature and to induce tumor regression. Such an approach to targeted delivery of therapeutic agents can lead to enhanced efficacy and safety. For this to succeed, it is necessary to develop a vector that can be systemically delivered to target the tumor vasculature. An ideal adenoviral vector is one that enables delivery of the gene of interest in a targeted manner and provides efficient trans-gene expression so therapy can be initiated. Also needed for success is the assurance that the vector is safe and is not associated with immune responses associated with the delivery of an exogenous agent.
The delivery of an adenovirus to specific cells can be enhanced by reducing its broad vector tropism by using various techniques such as a modification of the adenoviral fiber protein to alter its binding specificity or linking of ligands to cell-surface receptors via bispecific antibodies or bispecific chemical conjugates, for example, FGF-2, epidermal growth factor receptor, and epithelial cell adhesion molecule, biotinavidin bridges, or polymers such as polyethylene glycol or poly-(N-[2-hydroxypropyl]) methacrylamide.
Toward this goal, we used a gene therapy strategy based on retargeting adenovirus to tumor ECs by using FGFR-2, which is abundantly expressed on tumor ECs. We followed the transductional targeting technique by using FGF2 fragments conjugated to the native adenovirus. The addition of the FGF2-bispecific molecule enabled binding of the adenoviral capsid to the cell-surface FGFR, blocking uptake via CAR and redirecting viral tropism from the CAR receptor to high-affinity FGFRs. The FGF2-retargeted adenovirus was constructed by the chemical conjugation of a neutralizing antiadenoviral antibody to basic FGF-2. The binding affinity (Kd ~ 10 M) of FGF-2 to FGFR is at least three to four logs higher than other ligandreceptor interactions such as that between antigen and antibodies. The FGF2-conjugated adenoviral transduction is independent of CAR expression and dependent on the expression of FGFRs. On systemic delivery of the adenovirus, this property causes reduced delivery to nontargeted organs, especially the liver, which is the chief site of localization of the native adenovirus. Printz and colleagues demonstrated a 10- to 20-fold decrease in the amount of FGF2-conjugated adenovirus delivered to the liver, compared with the amount of native adenovirus. These authors also reported reduced liver toxicity, as measured by serum transaminase levels as well as by histopathological findings, and increased trans-gene expression in the tumor tissue following delivery of the retargeted adenovirus compared with delivery of the native adenovirus. Other diseases for which the FGF2-retargeted adenovirus was used for cancer therapy include ovarian cancer and Kaposi sarcoma.
A high level of FGF-2 expression has been reported in the tumor neovasculature. Therefore, retargeted adenoviruses can be used systemically to deliver genes of interest to tumor ECs. Such an approach enables not only the localization and enhancement of the viral transduction efficiency and transgene expression but also limits immune responses associated with adenoviruses.
In this study, we used an FGF2-retargeted adenovirus to transduce human GBMderived ECs (TuBECs) in vitro. We compared the transduction efficiency and the ability to induce transgene expression in both native and retargeted adenoviruses. We also examined rates of proliferation and apoptosis in cells transduced with the retargeted adenovirus and in cells transduced with the native adenovirus during treatment with the prodrug ganciclovir.
Source...