Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), a promising and novel anticancer cytokine, specifically kills numerous tumor cells by apoptosis. However, some malignancies are resistant to TRAIL treatment in clinical trials, thus limiting its therapeutic potential. In the present study, the TRAIL-resistant murine hepatocellular carcinoma cell line Hepa1-6 was used to elucidate the physiological significance of TRAIL resistance, especially with respect to the immune regulatory function of TRAIL. Hepa1-6 cells were resistant to TRAIL-induced apoptosis in vitro; however, intratumoral injection of recombinant soluble TRAIL inhibited tumor growth and prolonged survival time in tumor-bearing mice. Local TRAIL treatment decreased the number of intratumoral CD4(+)CD25(+)Foxp3(+) regulatory T cells (Tregs) but did not affect CD4(+)CD25(+)Foxp3(+) Tregs in the draining lymph nodes and spleen. Further investigation showed that TRAIL induced apoptosis of tumor-activated CD4(+)CD25(+)Foxp3(+) Tregs, but not of CD4(+)CD25(-) T cells. Moreover, mouse TRAIL receptor DR5 expression was detected on the surface of the tumor-infiltrating CD4(+)CD25(+)Foxp3(+) Tregs, but not on naïve CD4(+)CD25(+)Foxp3(+) Tregs. Interestingly, intratumoral injection of TRAIL not only decreased the number of CD4(+)CD25(+)Foxp3(+) Tregs but also increased the number of tumor-specific CD8(+) CTL and augmented their cytotoxicity to the tumor cells. These data provide the novel evidence for an immune regulatory function of TRAIL and may shed light on the clinical application of TRAIL.

The recombinant sTRAIL has been in clinical trial for various human malignancies. However, the half-life time of sTRAIL is very short, which might be an important factor influencing its clinical efficacy for cancer therapy. We previously reported the recombinant adeno-associated virus (AAV)-encoding sTRAIL 95-281 -mediated sTRAIL expression in vivo up to 8 months and suppressed tumor growth markedly in mouse xenografts. In the present study, we further evaluated the clinical potency for cancer gene therapy and the safety in mouse and non-human primates. The mouse models with HCT-116, NCI-H460 and BEL-7402 cancers were injected intraperitoneally with a single dose of 1.0 × 10 11 , 1.0 × 10 10 and 1.0 × 10 9 vg of rAAV2-sTRAIL 95-281 virus, respectively. The cynomolgus monkeys were injected (i.m.) with a single dose of rAAV2-sTRAIL 95-281 of 1 × 10 11 , 3 × 10 11 and 1 × 10 12 vg, corresponding to 6-, 20-and 60-fold of intended use dosage for humans, respectively. The efficacy, pharmacology and toxicity of rAAV-sTRAIL in the animals were analyzed accordingly. The tumor inhibitory rates reached 44-76%, 48-52% and 55-74% in the three tumor models, respectively, and they had no influence on mouse spontaneous activity. Administration (s.c.) of a single dose of rAAV2-sTRAIL 95-281 virus of 1.0 × 10 9 or 1.0 × 10 10 vg in mice with implanted tumor led to mainly distribution in the spleen, liver, implanted tumor, blood, injected site of muscle and bone marrow. Two weeks later, there was no rAAV2-sTRAIL 95-281 detected in blood and bone marrow, and it significantly decreased in other tissues and organs and then gradually cleared away in 4-12 weeks after administration. There was no rAAV2-sTRAIL accumulation in the animal's body and no influence on the body weights. Administration (i.v.) did not cause animal death, and no dose-related abnormal clinical symptoms were found in the mice. There were no abnormal tissue and organ found in all animals. Long-term toxicity test in cynomolgus monkeys did not cause rAAV2-sTRAIL 95-281 -related toxic and side effects, except that anti-AAV and anti-sTRAIL antibodies were generated. In conclusion, these data demonstrated that administration of rAAV2-sTRAIL 95-281 in mice and in cynomolgus monkeys is safe without obvious toxic and side effects to the animals, and throw light on pharmacokinetics and safety in human clinical trials for cancer gene therapy.