The study was aimed to explore the distribution and interaction mechanism of human bone marrow mesenchymal stem cells (MSC) and cord blood cytokine-induced killer (CIK)/natural killer (NK) cells infused via different ways at different times in NOD/SCID mice. 5 microl 1, 1'-dioctadecyl-3, 3, 3', 3'-tetramethylindocarbocyanine perchlorate (DiI) dye(red) was added in suspension of MSC per ml, and 1 microl carboxyfluorescein diacetate, succinimidyl ester(CFDA SE) dye(green) was added in suspension of CIK/NK cells per ml. The amounts of MSC and CIK/NK cells infused in each 6 NOD/SCID mouse were 1 x 10(6) (0.1 ml) and 1 x 10(7) (0.1 ml) respectively. All mice were divided into 4 groups, each group consisted of 6 mice. Group A: MSC (intravenous infusion, iv) + CIK/NK cells (iv) at the same time, group B: MSC (iv) + CIK/NK cells (iv) at 48 hours after infusion of MSC; group C: MSC (intramedullary infusion, im) + CIK/NK cells (iv) at the same time; group D: MSC (im) + CIK/NK cells (iv) at 48 hours after infusion of MSC. 3 NOD/SCID mice were sacrificed per batch at 24 hours and 48 hours after infused CIK/NK cells. Frozen sections of liver, spleen, lung and kidney were prepared, and then followed by counting the amounts of red and green fluorescence cells under fluorescence microscope, and calculating the ratio of MSC to CIK/NK cells for reflecting the interaction of MSC and CIK/NK cells in mice, and for showing the suppressive intensity of MSCs on CIK/NK cells. The results showed that the sums of average ratios of MSC to CIK/NK cells in lung, liver and spleen of group A and B were higher than that in group C and D at 24 hours and 48 hours respectively after infusing CIK/NK cells. The sum of average ratios of MSC to CIK/NK cells in group A was slightly higher than that in group B at 24 hours and 48 hours after infusing CIK/NK cells, but there was no significant difference between them. The sum of average ratios of MSC to CIK/NK cells in lung, liver and spleen in group C was slightly lower than that in group D at 24 hours after infusing CIK/NK cells, but reversed at 48 hours later and there was no significant difference between them. The sums of average ratios of MSC to CIK/NK cells in lung, liver and spleen in group A, B, C and D were all higher than those in kidney at 24 and 48 hours respectively after infusing CIK/NK cells. It is concluded, the MSC and CIK/NK cells may interact if they are infused via the same way and at the same time, the location where the suppression of MSC on CIK/NK cells occur in vivo may be reticulo-endothelial systems in lungs and livers.