Suppression of human colon tumor growth by adenoviral vector-mediated NK4 expression in an athymic mouse model

AIM: To investigate the suppressive effects of adenoviral vector-mediated expression of NK4, an antagonist of hepatocyte growth factor (HGF), on human colon cancer in an athymic mouse model to explore the possibility of applying NK4 to cancer gene therapy. METHODS: A human colon tumor model was developed by subcutaneous implantation of tumor tissue formed by LS174T cells grown in athymic mice. Fifteen tumor-bearing mice were randomized into three groups (n = 5 in each group) at d 3 after tumor implantation and mice were injected intratumorally with phosphate-buffered saline (PBS) or with recombinant adenovirus expressing beta-galactosidase (Ad-LacZ) or NK4 (rvAdCMV/NK4) at a 6-d interval for total 5 injections in each mouse. Tumor sizes were measured during treatment to draw a tumor growth curve. At d 26 after the first treatment, all animals were sacrificed and the tumors were removed to immunohistochemically examine proliferating cell nuclear antigen (PCNA), microvessel density (represented by CD31), and apoptotic cells. In a separate experiment, 15 additional athymic mice were employed to develop a tumor metastasis model by intraperitoneal injection (ip) of LS174T cells. These mice were randomized into 3 groups (n = 5 in each group) at d 1 after injection and were treated by ip injection of PBS, or Ad-LacZ, or rvAdCMV/NK4 at a 6-d interval for total two injections in each mouse. All animals were sacrificed at d 14 and the numbers and weights of disseminated tumors within the abdominal cavity were measured. RESULTS: Growth of human colon tumors were significantly suppressed in the athymic mice treated with rvAdCMV/NK4 (2537.4 +/- 892.3 mm(3)) compared to those treated by either PBS (5175.2 +/- 1228.6 mm(3)) or Ad-LacZ (5578.8 +/- 1955.7 mm(3)) (P < 0.05). The tumor growth inhibition rate was as high as 51%. Immunohistochemical staining revealed a similar PCNA labeling index (75.1% +/- 11.2% in PBS group vs 72.8% +/- 7.6% in Ad-LacZ group vs 69.3% +/- 9.4% in rvAdCMV/NK4 group) in all groups, but significantly lower microvessel density (10.7 +/- 2.4 in rvAdCMV/NK4 group vs 25.6 +/- 3.8 in PBS group or 21.3 +/- 3.5 in Ad-LacZ group, P < 0.05), and a markedly higher apoptotic index (7.3% +/- 2.4% in rvAdCMV/NK4 group vs 2.6 +/- 1.1% in PBS group or 2.1% +/- 1.5% in Ad-LacZ group, P < 0.05) in the rvAdCMV/NK4 group compared to the two control groups. In the tumor metastasis model, the number and weight of disseminated tumors of mice treated with rvAdCMV/NK4 were much lower than those of the control groups (tumor number: 6.2 +/- 3.3 in rvAdCMV/NK4 group vs 22.9 +/- 7.6 in PBS group or 19.8 +/- 8.5 in Ad-LacZ group, P < 0.05; tumor weight: 324 +/- 176 mg in rvAdCMV/NK4 group vs 962 +/- 382 mg in PBS group or 1116 +/- 484 mg in Ad-LacZ group, P < 0.05). CONCLUSION: The recombinant adenovirus, rvAdCMV/NK4, can attenuate the growth of colon cancer in vivo, probably by suppressing angiogenesis and inducing tumor cell apoptosis, but not by direct suppression of tumor cell proliferation. Moreover, rvAdCMV/NK4 may inhibit peritoneal dissemination of colon cancer cells in a murine tumor metastasis model. These findings indicate that NK4 gene transfer may be an effective tool for the treatment of colon cancer.