Many strategies have been designed to kill cancer cells [
16‐
18]. Because of its restricted and abundant surface expression on prostate cancer cells, PSMA constitutes an attractive target for immunotherapies against prostate cancer. Many papers have already reported that the combination of gene therapy and anti-PSMA antibody receives an ideal therapeutic result [
19‐
24]. Indeed, the antibody used in this study, J591, has been previously used for the treatment of prostate cancer cells [
23,
25‐
29].
Wild-type caspase-3 consists of an NH2-terminal prodomain, a large subunit, and a small subunit. However, active caspase-3, which is constructed with the reverse order of the subunits [
30], can induce apoptosis of tumor cells without apoptotic signals; these properties make active caspase-3 an attractive candidate molecule for gene therapy. In this study, the caspase 3 gene was generated by reversing the order of the coding sequences for the large and small subunits [
31]. To achieve successful proapoptotic gene therapy for prostate cancer cells, it is necessary to devise a gene construct that expresses the proapoptotic gene selectively in the prostate cells. Therefore, we generated a novel immunocasp-3 gene by fusing a leader sequence, an anti-PSMA antibody (J591) and the furin cleavage sequences of diphtheria toxin (Fdt) to the active revcaspase-3 [
14]. The results of our in vitro and in vivo studies revealed that the resultant protein, immunocasp-3, killed PSMA-overexpressing tumor cells, but not PSMA-negative cells. PC-3 cells are PSMA-negative tumor cells. Immunocasp-3 fusion proteins secreted in the culture media cannot enter PC-3 cells via receptor-mediated endocytosis, and thus cause no damage. Our results showed that PC-3 cells expressing recombinant Immunocasp-3 fusion proteins proliferate normally. Once internalized by PSMA-overexpressing tumor cells such as LNCaP cells, immunocasp-3 proteins are exposed to a low-pH environment in the endosome, where the peptide bond in the Fdt domain is cleaved by furin. Then immunocasp-3 proteins release COOH-terminal fragments, which consequently translocate to the cytosol and induce PSMA-overexpressing tumor cells to apoptosis. Both injection of lipofectamine-encapsulated immunocasp-3 and infusion of immunocasp-3 gene-modified Jurkat cells could suppress tumors due to continuous secretion of the killer protein and its diffusion through lymph fluid and blood. However, compared with direct injections of lipofectamine-encapsulated immunocasp-3, the applyment of immunocasp-3 gene-modified Jurkat cells may be a more effective and convenient therapeutic method, which simplified the procedure like protein purification. Moreover, caspase-3 proteins were endogenous human proteins that not only kill prostate cancer cells in a physiological manner, but also resulted in relatively weak immunogenicity and minor general toxicity over repeated administrations. However, because of immunogenicity of murine antibodies, fully human antibodies could become necessary.