Ses within order BTZ043 Hu-NOG mice reflected interspecies differences in benzene-induced hematotoxicity. The toxicity of benzene in leukocytes in the peripheral blood is induced mainly by benzene metabolites produced in organs such as the liver [45,46]. Because Hu-NOG and Mo-NOG 
mice obviously possess the same organs, we predicted that the degree of peripheral blood leukocyte toxicity would be almost the same in both. However, there was a significant difference in 18325633 the number of peripheral blood leukocytes between Hu-NOG and Mo-NOG mice in response to low levels of benzene. This difference may be attributed to differences in the amounts of cells supplied from the bone marrow, spleen, and thymus. In fact, the difference in the number of leukocytes in Hu-NOG and Mo-NOG mice was most significant in lymphoid organs (Fig. 5B). Moreover, in analyses targeting the bone marrow and peripheral blood, differences inIn Vivo Tool for Assessing Hematotoxicity in HumanIn Vivo Tool for Assessing Hematotoxicity in HumanFigure 5. Comparison of benzene toxicity in Hu-NOG and Mo-NOG mice. (A) Ratios of donor cell-derived human or mouse leukocytes in HuNOG (Hu) and Mo-NOG (Mo) mice after benzene administration. Each ratio was calculated based on the mean number of leukocytes in untreated HuNOG or Mo-NOG mice. (B) Ratios of myeloid (upper) and lymphoid (lower) cells in the bone marrow and peripheral blood of Hu-NOG (Hu) and MoNOG (Mo) mice after benzene administration. Each ratio was calculated based on the mean number of myeloid and lymphoid cell in untreated HuNOG or Mo-NOG mice. Mouse myeloid cells in Mo-NOG mice were identified as mCD45.2+mCD45.12mLy6C/6Ghi/mid. Mouse lymphoid cells in MoNOG mice were identified as mCD45.2+mCD45.12mLy6C/6Glo/2. The box plot shows the maximum (top of the vertical line), 75th percentile (top of the box), median (middle line in the box), 25th percentile (bottom of the box), and minimum (bottom of vertical line) values of data (n = 6?). * p,0.10 represents marginally significant differences between Hu-NOG and Mo-NOG mice, as determined by Mann-Whitney U tests. ** p,0.05 and *** p,0.01 represent significant differences. doi:10.1371/journal.pone.0050448.gsusceptibilities to benzene tended to be greater in lymphoid cells than in myeloid cells. These results suggested that interspecies differences in benzene-induced hematotoxicity are mainly due to differences in toxic responses in lymphoid cells, in the regulation of benzene in lymphoid development, or both. We speculate that there may be interspecies differences in the regulation of MEF2c expression by benzene on the basis of the reasons stated above. In conclusion, a human-like hematopoietic lineage established in NOG mice by transplanting human hematopoietic stem/ progenitor cells exhibited human-like susceptibility to at least 1 hematotoxicant, benzene. Hu-NOG and Mo-NOG mice offer a well-defined, reproducible, and easy-to-manipulate in vivo system for performing species-specific biochemical analyses of benzene metabolism. We think it is reasonable to assume that Hu-NOG mice will provide a powerful in vivo tool for assessing the hematotoxicity of chemical and physical agents on human hematopoietic cells. In the future, the similarities of SC66 thehematotoxic responses induced in Hu-NOG mice and humans should be evaluated more carefully by analyzing the detailed toxic response mechanism in Hu-NOG mice. Our strategy may be applicable to the study of other organs [47] and other toxicants as wel.Ses within Hu-NOG mice reflected interspecies differences in benzene-induced hematotoxicity. The toxicity of benzene in leukocytes in the peripheral blood is induced mainly by benzene metabolites produced in organs such as the liver [45,46]. Because Hu-NOG and Mo-NOG mice obviously possess the same organs, we predicted that the degree of peripheral blood leukocyte toxicity would be almost the same in both. However, there was a significant difference in 18325633 the number of peripheral blood leukocytes between Hu-NOG and Mo-NOG mice in response to low levels of benzene. This difference may be attributed to differences in the amounts of cells supplied from the bone marrow, spleen, and thymus. In fact, the difference in the number of leukocytes in Hu-NOG and Mo-NOG mice was most significant in lymphoid organs (Fig. 5B). Moreover, in analyses targeting the bone marrow and peripheral blood, differences inIn Vivo Tool for Assessing Hematotoxicity in HumanIn Vivo Tool for Assessing Hematotoxicity in HumanFigure 5. Comparison of benzene toxicity in Hu-NOG and Mo-NOG mice. (A) Ratios of donor cell-derived human or mouse leukocytes in HuNOG (Hu) and Mo-NOG (Mo) mice after benzene administration. Each ratio was calculated based on the mean number of leukocytes in untreated HuNOG or Mo-NOG mice. (B) Ratios of myeloid (upper) and lymphoid (lower) cells in the bone marrow and peripheral blood of Hu-NOG (Hu) and MoNOG (Mo) mice after benzene administration. Each ratio was calculated based on the mean number of myeloid and lymphoid cell in untreated HuNOG or Mo-NOG mice. Mouse myeloid cells in Mo-NOG mice were identified as mCD45.2+mCD45.12mLy6C/6Ghi/mid. Mouse lymphoid cells in MoNOG mice were identified as mCD45.2+mCD45.12mLy6C/6Glo/2. The box plot shows the maximum (top of the vertical line), 75th percentile (top of the box), median (middle line in the box), 25th percentile (bottom of the box), and minimum (bottom of vertical line) values of data (n = 6?). * p,0.10 represents marginally significant differences between Hu-NOG and Mo-NOG mice, as determined by Mann-Whitney U tests. ** p,0.05 and *** p,0.01 represent significant differences. doi:10.1371/journal.pone.0050448.gsusceptibilities to benzene tended to be greater in lymphoid cells than in myeloid cells. These results suggested that interspecies differences in benzene-induced hematotoxicity are mainly due to differences in toxic responses in lymphoid cells, in the regulation of benzene in lymphoid development, or both. We speculate that there may be interspecies differences in the regulation of MEF2c expression by benzene on the basis of the reasons stated above. In conclusion, a human-like hematopoietic lineage established in NOG mice by transplanting human hematopoietic stem/ progenitor cells exhibited human-like susceptibility to at least 1 hematotoxicant, benzene. Hu-NOG and Mo-NOG mice offer a well-defined, reproducible, and easy-to-manipulate 
in vivo system for performing species-specific biochemical analyses of benzene metabolism. We think it is reasonable to assume that Hu-NOG mice will provide a powerful in vivo tool for assessing the hematotoxicity of chemical and physical agents on human hematopoietic cells. In the future, the similarities of thehematotoxic responses induced in Hu-NOG mice and humans should be evaluated more carefully by analyzing the detailed toxic response mechanism in Hu-NOG mice. Our strategy may be applicable to the study of other organs [47] and other toxicants as wel.