Chemical suppression of a subpopulation of primitive hematopoietic progenitor cells: 1,3-butadiene produces a hematopoietic defect similar to steel or white spotted mutations in mice

The effects of benzene and other leukaemogenic agents on haematopoietic stem and progenitor cell differentiation

A characteristic shared by a diverse group of myelotoxic compounds and leukaemogens is the ability to act synergistically with granulocyte-macrophage colony stimulating factor (GM-CSF) in increasing clonogenic response. Pretreatment of murine or human bone marrow cells with the benzene metabolite, hydroquinone, but not phenol, catechol or trans, trans-muconaldehyde, results in a selective enhancement of GM-CSF but not an interleukin-3 (IL-3)-mediated clonogenic response. Clonal enhancement is preserved and magnified in enriched populations of CD34+ cells (> 95% purity), suggesting an intrinsic effect on haematopoietic progenitor cell (HPC) recruitment rather than a secondary effect involving accessory cytokines. Clonogenic enhancement of murine HPCs is not accompanied by alterations in GM-CSF receptor expression or ligand affinity and appears to be mediated via a p53-independent mechanism. These observations suggest that hydroquinone treatment alters recruitment and differentiation in a primitive subpopulation of CD34+ cells and are consistent with a role for altered stem cell differentiation in the development of chemically induced myelodysplasias.

PU.1 phosphorylation correlates with hydroquinone-induced alterations in myeloid differentiation and cytokine-dependent clonogenic response in human CD34+ hematopoietic progenitor cells

The transcriptional regulatory factor PU.1 is important for the regulation of a diverse group of hematopoietic and myeloid genes. Posttranslational phosphorylation of PU.1 has been demonstrated in the regulation of a variety of promoters in normal cells. In leukemia cells, differing patterns of PU.1 phosphorylation have been described among acute myelogenous leukemia (AML) subtypes. Therefore, we hypothesized that modulation of PU.1-dependent gene expression might be a molecular mediator of alterations in myeloid cell growth and differentiation that have been demonstrated to be early events in benzene-induced leukemogenesis. We found that freshly isolated human CD34(+) hematopoietic progenitor cells (HPC) exhibit multiple PU.1-DNA binding species that represent PU.1 proteins in varying degrees of phosphorylation states as determined by phosphatase treatment in combination with electrophoretic mobility shift assay (EMSA). Maturation of granulocyte and monocyte lineages is also accompanied by distinct changes in PU.1-DNA binding patterns. Experiments reveal that increasing doses of the benzene metabolite, hydroquinone (HQ) induce a time-and dose-dependent alteration in the pattern of PU.1-DNA binding in cultured human CD34(+) cells, corresponding to hyperphosphorylation of the PU.1 protein. HQ-induced alterations in PU.1-DNA binding are concomitant with a sustained immature CD34(+) phenotype and cytokine-dependent enhanced clonogenic activity in cultured human HPC. These results suggest that HQ induces a dysregulation in the external signals modulating PU.1 protein phosphorylation and this dysregulation may be an early event in the generation of benzene-induced AML.

Comparative toxicity of dithiocarbamates and butadiene metabolites in human lymphoid and bone marrow cells

Apparent differences in the pattern of leukemia risk have been observed between workers employed in 1,3-butadiene (BD) monomer production and those working in styrene-butadiene rubber production (SBR). There are a number of possible explanations for these discrepancies, including differences in disease classification and diagnosis as well as possible quantitative and qualitative differences in occupational exposure between these two industries. This led us to evaluate the possibility that the pattern of disease observed in SBR might be influenced by the presence of an important class of biologically reactive chemicals, dithiocarbamates (DTC), that were present in SBR but not BD monomer production. Therefore, we compared the immunotoxic and hematotoxic activities of DTC and BD metabolites in human immune and hematopoietic cells. Relative to the mouse, human CD34+ bone marrow cells are relatively resistant to the direct effects of BD metabolites, with only the bis-oxide producing any evidence of suppression of clonogenic response at concentrations between 1 and 10 microM. Similarly, treatment of human CD4+ lymphocytes with known (2,3-epoxybutene) and putative BD metabolites (D,L-butane-bis-oxide, (2S,3R)-3-epoxybutane-1,2-diol) does not result in appreciable T-cell toxicity at concentrations likely to be encountered in vivo. In contrast, treatment of human cells with DTC at concentrations as low as 100 nM results in significant suppression of hematopoietic clonogenic response and T-lymphocyte function. Additional studies in our laboratory and others suggest a role for copper in DTC toxicity in both human lymphocytes and bone marrow cells, although the pattern of altered transcriptional regulation observed is markedly different in these two cell populations. These results are consistent with the pattern of DTC toxicity previously observed in clinical and molecular studies.

Comparative toxicity of known and putative metabolites of 1, 3-butadiene in human CD34(+) bone marrow cells

Species-specific susceptibility to the hematotoxic effects of 1, 3-butadiene (BD) is well known. Previous studies have revealed that murine bone marrow is uniquely susceptible to toxicity following exposure to the parent compound in vivo or exposure of bone marrow cells to the monoepoxide metabolite, 3,4-epoxybutane, in vitro. Studies described herein compare the relative ability of putative and known BD metabolites to produce concentration dependent suppression of colony formation and cytotoxicity in human CD34(+) bone marrow cells. Compounds evaluated included 3,4-epoxybutane, D, L-butane-bis-oxide, meso-butane-bis-oxide and (2S, 3R)-3-epoxybutane-1,2-diol. In contrast to results previously observed in mice, only the bis-oxides produced significant suppression of colony formation at potentially relevant concentrations (10(-8) to 10(-3) M). No enantiospecific differences were observed between the meso- and D,L-bis-oxides and no significant lineage-specific differences in susceptibility to inhibition of clonogenic response were observed among early multi-potential myeloid and erythroid hematopoietic progenitor cells. The relative potencies of the bis-oxides were found to be comparable to that of the prototype hematotoxic compound, hydroquinone. These results confirm previous studies that reveal marked species-specific differences in the susceptibility of bone marrow cells to 3,4-epoxybutane. Moreover, these results suggest that the bis-oxides of BD are capable of suppressing the clonogenic function of human hematopoietic progenitor cells, if, in fact, they are produced in human bone marrow in significant concentration. Further interpretation of these findings requires a better understanding of the metabolism of BD in humans.

Characterization and phenotypic analysis of differentiating CD34+ human bone marrow cells in liquid culture

Our current understanding of human haematopoietic stem cell biology is based in part on the characterization of human CD34+ bone marrow cell differentiation in vitro. CD34 is highly expressed on early stem cells and haematopoietic progenitor cells with clonogenic potential and is gradually lost during differentiation and commitment. However, CD71 (transferrin receptor) is expressed at low levels on early stem cells and generally increases during haematopoietic progenitor cell proliferation. We reasoned that the combination of these surface markers would provide a useful framework for the simultaneous analysis of multiple lineage differentiation of CD34+ haematopoietic progenitor cells in liquid culture. In this report, we identify the phenotype of distinct subpopulations of myeloid, erythroid and lymphoid cells in liquid suspension culture using differential expression of CD34 vs. CD71 in combination with specific lineage markers. Freshly isolated human CD34+ bone marrow cells were introduced into suspension culture and monitored over a 6-d period using 3-colour flow cytometry. This is the first demonstration that differential expression of CD34 vs. CD71 can be used to simultaneously monitor differentiation of multiple haematopoietic cell lineages in liquid suspension culture, facilitating the study of cytokine-, drug- or chemical-induced alterations in haematopoietic progenitor cell differentiation in vitro.

2'3'-Dideoxycytidine-induced thymic lymphoma correlates with species-specific suppression of a subpopulation of primitive hematopoietic progenitor cells in mouse but not rat or human bone marrow

The nucleoside analogue, 2',3'-dideoxycytidine (ddC), is a potent inhibitor of HIV replication, and AIDS patients receiving ddC experience clinical improvement without significant hematologic toxicity. Repeated ddC administration (1,000 mg/kg per day) for 13 wk produces an increased incidence of thymic lymphoma in B6C3F1 mice. Previous studies reveal a common link between chemically induced and genetically associated models of mouse thymic lymphoma that involves a defect in a subpopulation of primitive hematopoietic progenitor cells. This defect is characterized by suppression of a subpopulation of IL-3-responsive cells and ablation of stem cell factor synergy with GM-CSF. The present study was undertaken to ascertain whether ddC produces the same pattern of bone marrow toxicity in mice, and whether this effect is observed in rat and human bone marrow. ddC exposure in vivo and in vitro produced a select suppression of murine CFU identical to that previously described for other models of mouse thymic lymphoma. In contrast, this selective CFU suppression was not observed in rat and human bone marrow or in CD34+ cells. These studies suggest that the mouse may not be a good predictive model for ddC hematotoxicity in humans and that susceptibility to the development of thymic lymphoma may be unique to the mouse.