However, in cell culture, it is extremely difficult to establish cell networks that mimic in vivo systems. As a result, a safe margin has been applied to health risk assessments to take into consideration the possibility of insufficient evaluation, particularly regarding interspecies differences, though such extrapolation to humans using safe margins occasionally results in overestimation of risks. However, the underestimation of risks by a small safety margin exposes humans to significant danger. Therefore, to perform more accurate health risk assessments, the development of an in vivo evaluation system that can reproduce human responses to toxic factors would be an important breakthrough. For many years, mouse models transgenically expressing human genes or harboring transplanted human cells, tissues, and organs, called humanized mice, have been developed to reproduce the responses of human cells in vivo. Mice that are humanized by transplantation of human cells are able to establish networks of human cells in their bodies. The available diverse mouse models were developed by transplantation of various types of cells to immunodeficient strains of mice. In cancer research, the biology of human tumor growth, metastasis, and angiogenesis has been evaluated in these mouse models. More recently, by transplanting human hepatocytes into liver-failure immunodeficient mice, mice with human livers have been developed for the study of human infectious diseases and metabolism. Moreover, allowing for the establishment of a functional human-like hematopoietic lineage. These techniques have proven valuable for the in vivo study of human hematopoietic stem cell function, infectious disease, and drug discovery, among other research questions. Interspecies differences in responses to toxicants are influenced greatly by the specificity and expression pattern of receptors, metabolic enzymes, and many other Rapamycin molecules. A human-like hematopoietic lineage may mimic the response to toxicants by human cells, and such humanized mice may therefore prove to be powerful tools for health assessment and aid in our evaluation of the hematotoxicity of various factors, while accounting for interspecies differences. Hematotoxicity is evaluated according to many factors, including decreased hematopoietic cell counts, abnormal blood coagulation, aberrant myelopoiesis, and induction of leukemia, all of which can be caused by diverse risk factors. Toxicants, such as benzene, can differentially affect human or animal hematopoietic lineages. Here, we took advantage of mice harboring a human-like hematopoietic lineage as a tool for assessing human hematotoxicity in vivo. These mice were established by transplanting NOG mice with human CD34+ cells. The response to benzene, a model toxicant, was measured by determining decreases in the number of leukocytes. Furthermore, we established chimeric mice by transplanting C57BL/6 mouse-derived bone marrow cells into NOG mice. To evaluate whether the response to benzene by Hu-NOG mice reflected interspecies differences, the degrees of benzene-induced hematotoxicities in Mo-NOG and Hu-NOG mice were compared. Here, we evaluated the toxic response of a human-like hematopoietic lineage established in NOG mice using the hematotoxicant benzene. Benzene-induced hematotoxicity is known to be transmitted by the aryl hydrocarbon receptor. Benzene metabolism is mediated by signals transmitted through interactions between AhR and benzene, benzene metabolites, or both, and the resulting benzene metabolites and reactive oxygen species induce cell damage. In hematopoietic cells, the AhR is expressed selectively by immature cells, such as hematopoietic stem/progenitor cells. Therefore, the toxic response of immature cells is the main cause of benzene-induced hematotoxicity.