Multidrug resistance in acute leukemia: a conserved physiologic function

ABCB1 Modulation Does Not Circumvent Drug Extrusion from Primitive Leukemic Progenitor Cells and May Preferentially Target Residual Normal Cells in Acute Myelogenous Leukemia

PURPOSE

Acute myelogenous leukemia (AML) is a disease originating from normal hematopoietic CD34+ CD38- progenitor cells. Modulation of the multidrug ATP-binding cassette transporter ABCB1 has not resulted in improved outcome in AML, raising the question whether leukemic CD34+ CD38- cells are targeted by this strategy.

EXPERIMENTAL DESIGN

ABCB1-mediated transport in leukemic CD34+ CD38- cells compared with their normal counterparts was assessed by quantitating the effect of specific ABCB1 modulators (verapamil and PSC-833) on mitoxantrone retention [defined as efflux index (EI), intracellular mitoxantrone fluorescence intensity in the presence/absence of inhibitor].

RESULTS

ABCB1 was the major drug transporter in CD34+ CD38- cells in normal bone marrow (n = 16), as shown by the abrogation of mitoxantrone extrusion by ABCB1 modulators (EI, 1.99 +/- 0.08). Surprisingly, ABCB1-mediated drug extrusion was invariably reduced in CD34+ CD38- cells in AML (n = 15; EI, 1.21 +/- 0.05; P < 0.001), which resulted in increased intracellular mitoxantrone retention in these cells (mitoxantrone fluorescence intensity, 4.54 +/- 0.46 versus 3.08 +/- 0.23; P = 0.004). Active drug extrusion from these cells occurred in the presence of ABCB1 modulators in the majority of samples, pointing in the direction of redundant drug extrusion mechanisms. Residual normal CD34+ CD38- cells could be identified by their conserved ABCB1-mediated extrusion capacity.

CONCLUSION

ABCB1-mediated drug extrusion is reduced in leukemic CD34+ CD38- progenitor cells compared with their residual normal counterparts. Redundant drug transport mechanisms confer mitoxantrone transport from leukemic progenitors. These data argue that ABCB1 modulation is not an effective strategy to circumvent drug extrusion from primitive leukemic progenitor cells and may preferentially target residual normal progenitors in AML.

Retrovirus insertion and transcriptional activation of the multidrug-resistance gene in leukemias treated by a chemotherapeutic agent in vivo

To understand the molecular basis for multidrug-resistant (MDR) cancer cells in vivo, this study analyzed molecular changes of the mdr1a gene region in leukemia cells in mice during continuous treatment with vincristine. An inverse insertion of murine leukemia retrovirus (MuLV) into the 5'-flanking region of the mdr1a gene was found. This insertion was concomitantly accompanied by up-regulation of the mdr1a gene and the loss of chemosensitivity. Deletion of long-terminal repeat (LTR) sequences dramatically decreased the mdr1a promoter-driven reporter activity. The MuLV LTR insertion appears to exert its enhancer activity on mdr1a transcription during the appearance of MDR leukemia cells. Two mechanisms were postulated to explain the mdr1a gene activation by retrovirus insertion during in vivo chemotreatment: de novo insertion of MuLV induced by vincristine treatment and selection of a small fraction of pre-existing cells carrying MuLV insertion during vincristine treatment. No rearranged sequence was detected by polymerase chain reaction in parental cells. This result argued for the first mechanism. The randomly altered distribution of MuLV during repetitive chemotreatment might also be consistent with this hypothesis. On the other hand, the retrovirus insertion was detected at the same site of the mdr1a promoter region in 2 independent experiments, which suggests the second mechanism. It should be noted that in vivo chemotreatment using vincristine could generate the mdr1a-overexpressing cells through retrovirus insertion and the enhancer effect of the LTR.

Multidrug resistance in leukemias and its reversal

Drug resistance often results in failure of anticancer chemotherapy in leukemias. Several mechanisms of drug resistance are known with multidrug resistance (MDR) being the best characterized one. MDR can be due to enhanced expression of certain genes (MDR1, MRP or LRP), alterations in glutathione-S-transferase activity or GSH levels and to reduction of the amount or the activity of topoisomerase II. Here we review the current status of the clinical significance of the various mechanisms of MDR in leukemias and also discuss possibilities for the reversal of MDR. MDR1 gene expression has been seen in many leukemias, notably in acute myeloid leukemia (AML) and blast crisis of chronic myeloid leukemia. Both MDR1 RNA and P-glycoprotein expression of the leukemic cells have been shown to correlate with poor clinical outcome in AML. However, preliminary results indicate that the MRP gene as well as the LRP gene can be expressed in AML. Thus, drug resistance in leukemias appears to be multifactorial. P-glycoprotein-mediated MDR can be reversed by several drugs. These resistance modifiers are currently evaluated with regard to their clinical efficacy. Despite some encouraging results, reversal of drug resistance and subsequent improvement in clinical outcome remains to be shown.

New flow cytometric method for detection of minimally expressed multidrug resistance P-glycoprotein on normal and acute leukemia cells using biotinylated MRK16 and streptavidin-RED670 conjugate

To evaluate the expression of multidrug resistance (MDR) on normal and leukemia cells, we examined P-glycoprotein (P-gp) by a newly devised flow cytometric method, utilizing a biotinylated monoclonal antibody (mAb) against P-gp (MRK16), a streptavidin-RED670 conjugate (SA-RED670) and appropriate emission filters. The combination of biotinylated MRK16 (b-MRK16) and SA-RED670 resulted in higher sensitivity as compared with standard methods such as the use of streptavidin-phycoerythrin (SA-PE) conjugate. The sensitivity was examined in K562, K562/ADR, NOMO-1, NOMO-1/ADR and HL60 cells, and compared with the data obtained from reverse transcription polymerase chain reaction (RT-PCR) of mdr-1 gene. P-gp positivity on flow cytometry was 10.4%, 99.9%, 1.4%, 90.4% and 0%, respectively. Mdr-1 mRNA was well expressed in K562/ADR and NOMO-1/ADR cells, but not in NOMO-1 and HL60 cells. In K562 cells, mdr-1 was found after 40 cycles of PCR, but not 25 cycles. These data are well correlated with those from the flow cytometry. We then studied the P-gp expression on normal peripheral blood cells and acute leukemia cells. P-gp was little expressed on peripheral lymphocytes, monocytes and granulocytes. It was also little expressed on blast cells from 5 patients with acute promyelocytic leukemia (AML) and 5 acute lymphocytic leukemia (ALL) expressed P-gp at diagnosis, ranging from 8.5% to 34.5% (16.9 +/- 11.8%) and from 2.3% to 45.6% (24.0 +/- 17.8%), respectively. All 9 relapsed or refractory cases expressed P-gp, ranging from 21.1% to 99.8% (52.2 +/- 29.9%). Significant differences were found in APL, CD34-positive and relapse and refractory cases (P = 0.0006, 0.0007 and 0.0088, respectively). These results indicate that this flow cytometric analysis is useful for the evaluation of clinical MDR status and can identify a group of patients with resistant leukemia.