Upon androgen ablation, androgen-sensitive prostatecancer cells die by apoptosis (programmed cell death) (Isaacs et al., 1992). However, androgen-sensitive tumor cells can become androgen-resistant during the progression of prostate cancer. finally causing the death of patients. Determining the mechanism of the cell death of prostate tumors is a significant issue which could lead to successful therapy of androgenresistant prostate cancer. FasiAPO-1 is a cell-surface protein, a member of the TNF-receptor family, and it potentially induces apoptosis in cells (Itoh et al., 1991). Suda and Nagata (1994) have purified the Fas ligand, which is a member of the TNF family that mediates the cell-death signal by binding to Fas/APO-1, and have isolated its gene . In this study, we examined the expression and potential inducibility of apoptosis by Fas signals in prostate-cancer cell lines and the potential application to tumor therapy. MATERIAL AND METHODS Constntction of Fas expression vectorThe mammalian expression vector pCGAAS (Niwa et al., 1991) was provided by Dr. J. Miyazaki (University of Tokyo). The plasmid pCGAAS deprived of CMV enhancer [pCGAAS-CMV( -)] was obtained by digestion with SalI and SnaBI and blunting with Klenow fragment, followed by separation by agarose-gel electrophoresis, purification with a Gene Clean kit I1 (Funakoshi, Tokyo, Japan) and re-ligation with T4 ligase. Plasmid pBL58-I including human Fas cDNA was provided by Dr. S. Nagata (Osaka Bioscience Institute). A 1.8-kb EcoRI fragment of pBL58-1 including the coding region of human Fas cDNA was transferred into the EcoRI site of pCGAAS-CMV( -), then pCFas22 including human Fas cDNA in the transcribable direction under the control of the chicken p-actin promoter of pCGAAS-CMV(-) was obtained.Transfection of Fas cDNA PC-3 and LNCaP human prostate-cancer cells were first cloned by limiting dilution to avoid clonal heterogeneity of parental tumor cells for transfection. PC-3 and LNCaP cells of a single clone were co-transfected with PstI-digested pCFas22, forming linear DNA, and EcoRI-digested PUC19 inserted with the tyg-neomycin-resistant gene at a 20:l molar ratio by electroporation. As controls, parental PC-3 and LNCaP cells were co-transfected with PstI-digested pCGAAS-CMV( -) and EcoRI-digested PUC19 containing the tyg-neomycin resistance gene. Fas-transfected PC-3 and LNCaP cells as well as control PC-3 and LNCaP cells were cloned with 400 and 800 kg/ml, respectively, of G418 in RPMI-1640 containing 10% FCS, 100 kg/ml of streptomycin and 100 IU/ml of penicillin. Fas/APO-1 expression was confirmed by immunocytochemistry and flow cytometry, as described below. linmunocytochemistry. Cells were cultured in RPMI-1640 containing 10% FCS in 3-cm-diameter flasks containing autoclaved glass cover slips. Cells on cover slips were fixed in 4% formaldehyde in PBS (pH 7.4) for 10 min at room temperature, then in acetone for 3 min at -20°C. The specimens were incubated with 5% normal goat serum for 10 min, then with non-apoptosis-inducible anti-human mo...

Tumor regression in experimental systems has been linked to the activities of Th1 cells. It is, therefore, conceivable that Th2 cells interrupt the expression of tumor immunity since interleukin-4 (IL-4) and IL-10 inhibit the generation of Th1 from precursors and modulate the competence of antigen-presenting cells to activate this lymphocyte subpopulation. Naive murine renal cell carcinoma (renca) cells (1 x 10(5)) were implanted into the subcapsule of the left kidney of Balb/c and Balb/c nude mice at 6-8 weeks of age. After 14 days, Th2 cytokine (IL-4 and IL-10) mRNAs as well as transforming growth factor beta1 mRNA, assessed by reverse transcriptase/polymerase chain reaction were upregulated in the spleen of hosts upon naive renca tumor acceptance, while Th1 cytokine (IL-2 and interferon gamma) mRNAs were almost undetectable. In the renca tumor, IL-10 mRNA was detected but IL-2, interferon gamma, and IL-4 were not. Intraperitoneal administration of anti-(mouse IL-4) mAb (11B11) reduced the renca tumor size (P = 0.018) and prolonged host survival (P = 0.03), but did not reduce the acceptance rate of the tumor (P = 0.18). However, prior depletion of CD4+ or CD8+ cells with monoclonal antibodies abrogated the antitumor effects of anti-IL-4 mAb. In addition, the significant antitumor effect of anti-IL-4 mAb was not observed in Balb/c nude hosts. Renca cells were transfected with the mammalian expression vector pCAGGS containing murine IL-4 cDNA or vector alone, then stable IL-4 transfectants (RencaL or RencaH, low- or high-IL-4-producing respectively) and control renca cells (RencaC) were obtained. RencaL cells, RencaH cells, or RencaC cells (1 x 10(5) each) were implanted into the subcapsule of the left kidney of Balb/c, Balb/c nude, and allogenic C3H/HeJ mice, then tumor formation was evaluated 14 days later. When RencaH cells were innoculated into syngeneic Balb/c hosts, tumor volume was marginally suppressed (P = 0.03) and tumors tended to be rejected (P = 0.06) compared with RencaC cells. However, those effects were not observed in Balb/c nude mice. RencaC, RencaL, and RencaH cells were not accepted by allogeneic C3H mice with or without FK506 administration or donor-specific transfusion. The administration of anti-(mouse IL-4) mAb to Balb/c mice significantly suppressed renca tumor growth by a CD4+ and CD8+ T-cell-dependent mechanism. By contrast, relatively high levels of IL-4 production by renca cells and T cells seemed to be required to induce the rejection and growth suppression of IL-4-producing renca cells in syngeneic hosts.

A Th2 bias may contribute to allograft acceptance in part by inducing the down-regulation of Th1-cytokine mRNAs, but it may not be sufficient to induce indefinite graft survival.

No abstract