Generation of Expression Plasmids for Angiostatin, Endostatin and Timp-2 for Cancer Gene Therapy

Antiangiogenic therapy may represent a promising approach to cancer treatment. Indeed, the efficacy of endogenous angiogenesis inhibitors, including angiostatin, endostatin and TIMPs, has been demonstrated in many types of solid tumors in animal models. In view of the possible problems associated with long-term administration of inhibitors as recombinant proteins, we propose their delivery as nucleic acids through a gene therapy approach. To this end, eukaryotic expression constructs for murine angiostatin and endostatin as well as human TIMP-2 were generated, and characterized in vitro. All constructs carry the relevant cDNAs under the control of the strong HCMV promoter/enhancer, and cleavable leader signals to allow protein secretion. Expression of the angiogenesis inhibitors was detected by in vitro transcription/translation experiments as well as transfection of 293T cells, followed by Western blotting (WB) or radioimmunoprecipitation analysis of both cell lysates and supernatants (SNs). These constructs might be used for in vivo intramuscular delivery of plasmid DNA and as a set of reagents for the development of retroviral as well as adeno-associated viral (AAV) vectors expressing angiogenesis inhibitors.

Gene therapy of ovarian cancer with IFN-alpha-producing fibroblasts: comparison of constitutive and inducible vectors

Ovarian cancer represents a malignancy suitable for cell and gene therapy approaches owing to its containment within the peritoneal cavity, even at advanced tumor stages. As regulation of transgene expression would be preferable for conducting clinical trials for reasons of safety, we investigated whether intraperitoneal (i.p.) administration of retroviral vector-transduced fibroblasts encoding murine interferon-alpha (IFN-alpha) could have therapeutic activity, and compared its effect with the antitumor effects of fibroblasts producing IFN-alpha under a rapamycin analogue (AP21967)-inducible promoter. Human and murine fibroblasts were recruited into the solid component of transplantable ovarian cancer-grown i.p. in severe combined immunodeficiency mice. Multiple administrations of fibroblasts producing IFN-alpha in a constitutive manner showed therapeutic efficacy, leading to significant prolongation of survival in the majority of animals, associated with inhibition of tumor angiogenesis. Compared to cells transduced by the constitutive vector, fibroblasts transduced by the inducible vector released twofold higher IFN-alpha levels in vitro, following induction by AP21967, and production of the cytokine was under pharmacologic control both in vitro and in vivo. However, these cells elicited only modest therapeutic effects in vivo. Overall, these findings indicate that intracavitary IFN-alpha gene therapy using engineered fibroblasts requires sustained production of IFN-alpha to achieve durable antitumor effects.

Differential effects of angiostatin, endostatin and interferon-alpha(1) gene transfer on in vivo growth of human breast cancer cells

The administration of different angiogenesis inhibitors by gene transfer has been shown to result in inhibition of tumor growth in animal tumor models, but the potency of these genes has been only partially evaluated in comparative studies to date. To identify the most effective anti-angiogenic molecule for delivery by retroviral vectors, we investigated the effects of angiostatin, endostatin and interferon(IFN)-alpha(1) gene transfer in in vivo models of breast cancer induced neovascularization and tumor growth. Moloney leukemia virus-based retroviral vectors for expression of murine angiostatin, endostatin and IFN-alpha(1) were generated, characterized, and used to transduce human breast cancer cell lines (MCF7 and MDA-MB435). Secretion of the recombinant proteins was confirmed by biological and Western blotting assays. Their production did not impair in vitro growth of these breast cancer cells nor their viability, and did not interfere with the expression of angiogenic factors. However, primary endothelial cell proliferation and migration in vitro were inhibited by supernatants of the transduced cells containing angiostatin, endostatin, and IFN-alpha(1). Stable gene transfer of the IFN-alpha(1) cDNA by retroviral vectors in both MCF7 and MDA-MB435 cells resulted in a marked and long-lasting inhibition of tumor growth in nude mice that was associated with reduced vascularization. Endostatin reduced the in vivo growth of MDA-MB435, but not MCF7 cells, despite similar levels of in vivo production, and angiostatin did not impair the in vivo growth of either cell line. These findings indicate heterogeneity in the therapeutic efficacy of angiostatic molecules delivered by viral vectors and suggest that gene therapy with IFN-alpha(1) and endostatin might be useful for treatment of breast cancer.

Effects of angiostatin gene transfer on functional properties and in vivo growth of Kaposi's sarcoma cells

Gene transfer delivery of endogenous angiogenesis inhibitors such as angiostatin would circumvent problems associated with long-term administration of proteins. Kaposi's sarcoma (KS), a highly vascular neoplasm, is an excellent model for studying tumor angiogenesis and antiangiogenic agent efficacy. We investigated the effects of angiostatin gene transfer in in vitro and in vivo models of KS-induced neovascularization and tumor growth. A eukaryotic expression plasmid and a Moloney leukemia virus-based retroviral vector for expression of murine angiostatin were generated harboring the angiostatin cDNA with cleavable leader signals under the control of either the strong cytomegalovirus promoter/enhancer or the Moloney leukemia virus long terminal repeat. Angiostatin secretion was confirmed by radioimmunoprecipitation and Western blot analysis. Supernatants of angiostatin-transfected cells inhibited endothelial cell migration in vitro. Stable gene transfer of the angiostatin cDNA by retroviral vectors in KS-IMM cells resulted in sustained angiostatin expression and delayed tumor growth in nude mice, which was associated with reduced vascularization. These findings suggest that gene therapy with angiostatin might be useful for treatment of KS and possibly other highly angiogenic tumors.

Targeting of calsequestrin to sarcoplasmic reticulum after deletions of its acidic carboxy terminus

Calsequestrin (CS) is the Ca(2+) binding protein of the junctional sarcoplasmic reticulum (jSR) lumen. Recently, a chimeric CS-HA1, obtained by adding the nine-amino-acid viral epitope hemagglutinin (HA1) to the COOH terminus of CS, was shown to be correctly segregated to the sarcoplasmic reticulum [A. Nori, K. A. Nadalini, A. Martini, R. Rizzuto, A. Villa, and P. Volpe. Am. J. Physiol. 272 (Cell Physiol. 41): C1420-C1428, 1997]. A putative targeting mechanism of CS to jSR implies electrostatic interactions between negative charges on CS and positive charges on intraluminal domains of jSR integral proteins, such as triadin and junctin. To test this hypothesis, 2 deletion mutants of chimeric CS were engineered: CS-HA1DeltaGlu-Asp, in which the 14 acidic residues [-Glu-(Asp)(5)-Glu-(Asp)(7)-] of the COOH-terminal tail were removed, and CS-HA1Delta49(COOH), in which the last, mostly acidic, 49 residues of the COOH terminus were removed. Both mutant cDNAs were transiently transfected in HeLa cells, myoblasts of rat skeletal muscle primary cultures, or regenerating soleus muscle fibers of adult rats. The expression and intracellular localization of CS-HA1 mutants were studied by epifluorescence microscopy with use of antibodies against CS or HA1. CS-HA1 mutants were shown to be expressed, sorted, and correctly segregated to jSR. Thus short or long deletions of the COOH-terminal acidic tail do not influence the targeting mechanism of CS.