The sensitivity of human cells expressing RUNX1-RUNX1T1 to chemotherapeutic agents

Abstract

The translocation (8;21)(q22;q22) is observed in approximately 12–15% of patients with acute myeloid leukaemia (AML), making it one of the most frequently observed translocations in AML.1 The translocation involves DNA rearrangement of the RUNX1 (aka. AML1, core binding factor (CBF)) gene on chromosome 21 with the RUNX1T1 (aka. ETO) gene on chromosome 8. This abnormality leads to the expression of the RUNX1-RUNX1T1 (aka. AML1-ETO) fusion gene, which promotes self-renewal of haematopoietic cells and also inhibits their subsequent differentiation.2 Leukaemias expressing this abnormality are generally associated with a good prognosis in terms of complete remission, relapse risk and overall survival compared with other subtypes and tend to respond favourably to treatment particularly with high-dose cytarabine and an anthracycline.3 It is currently not known why patients expressing the t(8;21) have a good prognosis. The chemosensitivity of patients with AML has previously been suggested to be associated with multi-drug resistance (MDR) gene expression.4, 5, 6 The MDR-1 gene encodes P-glycoprotein, a cell membrane drug efflux pump. It would be envisaged that patients who are considered to have a good prognosis (such as t(8;21)) would not overexpress MDR-1 (as demonstrated by Lutterbach et al.4), otherwise it is likely those individuals would show chemoresistance and have a more adverse prognoses. Surprisingly, previous studies have found positive correlations between the t(8;21) karyotype and MDR-1 gene expression,5, 6 suggesting that this transcription factor fusion gene may promote the expression of MDR-1. We therefore tested this hypothesis directly by expressing the RUNX1-RUNX1T1 fusion as a single abnormality in human haematopoietic cell subsets and performed Affymetrix microarray analysis (as described previously)2, 7 to determine whether this fusion had any effect on the transcription of MDR genes. Using this approach, we generated independent replicate sets of data from control and RUNX1-RUNX1T1-matched CD34+ cultures as well as matched sets constituting granulocytic (CD14lo, CD36lo, CD15hi) and monocytic (CD14hi) unilineage populations (isolated from day 6 cultures by immunomagnetic sorting). cRNA was prepared from each sample and hybridized to Affymetrix human 133A oligonucleotide arrays, which allowed the simultaneous analysis of six MDR family gene members. In each of these populations, the expression of MDR genes was not significantly different from controls (Figure 1a–c). In addition, using our cohort of French–American–British (FAB)-M2 patients, there was little difference in MDR gene expression between those individuals with a t(8;21) and those without this abnormality (Figure 1d). We could therefore find no evidence that RUNX1-RUNX1T1 expression directly influences MDR gene expression as a single abnormality or in t(8;21) patients. One alternative explanation for the aforementioned observations in AML patients is that other coexisting abnormalities may be influencing the expression of MDR, as suggested by Schaich et al.6