No abstract

Previous reports have suggested that the adenosine triphosphate-binding cassette protein ABCG2 (breast cancer resistance protein [BCRP], mitoxantrone resistance [MXR]) is associated with drug resistance in acute myeloid leukemia (AML). The aims of this study were to determine the level of ABCG2 mRNA expression necessary to produce drug resistance and to define the ABCG2 levels in normal bone marrow (BM), peripheral blood (PB), cord blood (CB), and adult AML blast cell populations. First, using transduced clonal cell lines expressing varying levels of ABCG2, we found that ABCG2 expression conferred resistance to mitoxantrone and topotecan, but not to idarubicin. Next, we developed a real-time reverse transcription-polymerase chain reaction assay for measuring ABCG2 mRNA expression levels in clinical samples. Normal BM and PB contained low levels of ABCG2 mRNA, while higher levels were measured in CB mononuclear cells, CD34 ؉ , and Ac133 ؉ populations, consistent with the known stem cell enrichment in these populations. Next, we studied the ABCG2 mRNA levels in 40 specimens from newly diagnosed adult AML patients. Only 7% of these samples contained ABCG2 mRNA levels within the range of our drugresistant clone, although another 78% were higher than normal blood and bone marrow. Flow cytometry revealed very small subpopulations of ABCG2-expressing cells in the cases we examined. Our data suggest that high levels of ABCG2 mRNA expression in adult AML blast specimens are relatively uncommon and that ABCG2 expression may be limited to a small cell subpopulation in some cases. (Blood. 2002;100:4594-4601)

6-Mercaptopurine (6-MP) and its prodrug azathioprine (AZA) are well known for their lymphocytotoxic and bone marrow suppressive effects in the management of patients with leukemia. Although their immunosuppressive properties are mediated by the active AZA antimetabolite 6-thioguanine (6-TG), its mechanism of action is largely unknown. In IBD, a significant inverse correlation has been shown between erythrocyte 6-TG metabolite levels and disease activity, further supporting the proposed immunosuppressive role for 6-TG. Since leukocytes possess quantitatively different purine metabolic pathways compared to erythrocytes, this study aims to measure lymphocyte DNA 6-TG metabolites and correlate levels with the INF-gamma and IL-10 cytokine profile in patients with Crohn's disease (CD). Forty-six adult patients with CD, either naive (17) or on long-term (>4-month) AZA therapy (29), had erythrocyte and lymphocyte DNA 6-TG levels measured by reverse-phase HPLC under UV detection (6-TG, 340 nm). Lymphocyte DNA 6-TG was expressed as picomoles per milligram of DNA. Lymphocyte DNA 6-TG metabolite levels were correlated with INF-gamma and IL-10 cytokine profiles using the OptEIA kit (Pharmigen). Lymphocyte DNA 6-TG metabolite levels correlate with erythrocyte 6-TG levels (P < 0.03) but not total patient leukocyte levels. Erythrocyte 6-TG metabolite levels correlated (P < 0.01) inversely with INF-gamma but not IL-10 cytokine levels. This study suggests a preferential dampening of the TH1 response on exposure to 6-TG and a possible immunosuppressive mechanism of action for AZA. Future studies are needed to determine if cytokine profiles can be used to predict recalcitrant CD to AZA therapy.