To investigate the effects of donor treatment with combination of recombinant human interleukin-11 (rhIL-11) and recombinant human granulocyte-colony stimulating factor (rhG-CSF) on incidence of lethal graft versus host disease (GVHD) after allogeneic bone marrow transplantation (BMT). Forty C57BL/6 mice, as donors, were treated with phosphate buffered saline (PBS), or optimal doses of rhG-CSF, rhIL-11, or rhG-CSF + rhIL-11 for five consecutive days. Fifty BALB/C mice, as recipients, received a lethal dose of 8.0 Gy total body irradiation (TBI) 4-6 h before the transplantation. Ten C57BL/6 mice without transplantation were used as pure radiation control group (group A). The other 40 C57BL/6 mice received the allogeneic bone marrow from the donors treated with PBS (group B), rhG-CSF (group C), rhIL-11 (group D), or rhG-CSF + rgIL-11 (group E) 14 and 30 days after the transplantation, suspensions of bone marrow cells were prepared from the 4 groups. Monoclone antibody of PE-H-2D(b) was added. The percentage of the H-2D(b) positive cells was observed by flow cytometry to detect the chimerism to determine the proportion of the cells of donor origin. The liver, spleen, intestine, and skin tissues of the mice presenting symptoms of GVHD were taken out before the diseased mice died. The long-surviving mice were killed 60 days after the transplantation and the above-mentioned tissues were taken out. Light microscopy was used to observe the pathological changes. 14 days after transplantation, the spleen cells of the different groups were obtained to undergo mixed leucocyte reaction (MLR), and 14 days after transplantation another spleen cells of groups B, C, D, and E were obtained to prepare suspension of mononuclear cells to test the levels of IL-4, interferon (IFN)-gamma and tumor necrosis factor (TNF)-alpha by ELISA. The survival time of group E was 38 d +/- 19 d, significantly longer than the other groups (all P < 0.01). The survival times of groups C and D were 23 d +/- 8 d and 25 d +/- 15 d respectively, both significantly longer than that of group B (14 d +/- 4 d, both P < 0.01). The mice of group A all died within 8 days with a survival time of 6.2 d +/- 1.3 d. The GVHD pathological changes were milder in group E than in the other groups. 14 and 30 days after transplantation, the percentage of H-2D(b) antibody positive cells was over 95% in all surviving transplanted mice. All the mice of group B developed GVHD, and the earlier GVHD appeared the severer the pathological changes. In group E 3 mice developed GVHD more than 20 days after transplantation, 7 mice survived more than 30 days, and 4 survived 60 days without any sign of GVHD. The pathological changes in the tissues of the mice that died within 20 days were more remarkable than those of the mice that died 20 days or more after transplantation. The levels of IFN-gamma of group C, D, and E were significantly lower than that of group B, and the levels of IL-4 of group C, D, and E were significantly higher than that of group B (all P < 0.01). The level of IFN-gamma of group E was significantly lower than those of groups C and D, and the level of IL-4 of group E was significantly higher than those of groups C and D (all P < 0.05). The levels of group D and E were significantly lower than those of group B and C (all P < 0.01). The MLR 14 days after transplantation showed that the proliferation to host alloantigen of group E was 0.22 +/- 0.08, significantly lower than those of groups B, C, and D (0.87 +/- 0.14, 0.54 +/- 0.21, and 0.49 +/- 0.18 respectively, all P < 0.01). Donor treatment with the combination of rhIL-11 and rhG-CSF shows a synergic function in inducing immune tolerance, and significantly reduces the incidence of lethal GVHD after BMT.