Level of Myeloperoxidase Expression as a Crucial Determinant of Sensitivity of Myeloid Leukemia Cells to Parthenolide-Induced Apoptosis

Parthenolide (PTL) is a sesquiterpene lactone found as the major active component in Feverfew (Tanacetum parthenium). PTL is a strong inhibitor of NF-kB activation and STAT transcriptional activity, resulting in a downregulation of antiapoptotic gene transcription. Recently, PTL was demonstrated to have promising anti-cancer effects through inhibition of DNA synthesis and reactive oxygen species (ROS)-associated intrinsic apoptosis. However, the sensitivity to PTL-induced cell death was different according to leukemia cells. Therefore, it will be important to identify parameters predicting response to PTL-induced cell death. Myeloperoxidase (MPO), a typical lineage marker for acute myeloid leukemia (AML), was also shown to have a prognostic significance in this disorder. Since it was shown that MPO is a potential regulator of oxidative stress, we examined the effect of MPO expression upon the PTL-induced leukemia cell death. First we compared the extent and mechanism of PTL-induced cell death between parental K562 and MPO-overexpressing K562 (K562/MPO) cells. K562/MPO cells were kindly provided by Dr. Sawayama (Nagasaki University, Japan). Annexin V labeling evaluation revealed that PTL induced apoptosis in K562/MPO cells in a dosedependent manner. The fraction of apoptotic cells after 24 hours treatment with 10mM of PTL was significantly higher in K562/MPO cells (52.2 ± 0.4%) compared to parental K562 cells (13.1 ± 4.8%, p<0.001). When the cells were treated with PTL, the population that lost mitochondrial membrane disruption (MMP) was 45.1 ± 12.4% in K562/MPO cells, which was significantly higher than K562 cells (0.7 ± 0.1%, p<0.001). Cleavage of caspase-3, -8, -9, and PARP was observed in K562/MPO cells after PTL treatment, whereas it was not shown in K562 cells. We next examined the possible involvement of ROS in PTL-induced cell death by flow cytometric analysis using dihydroethidium fluorescent probe. PTL drastically increased relative ROS levels in K562/MPO cells (4.1 ± 0.1). However, the increase in ROS levels induced by PTL was not demonstrated in K562 cells (1.5 ± 0.1, p<0.01). Marked downregulation of Bcl-2, Bcl-xL, and NF-kB was demonstrated preferentially in K562/MPO cells. c-Jun N-terminal kinase phosphorylation was remarkably increased only in K562/MPO cells. We next evaluated the PTL-induced cell death in primary leukemic blasts obtained from patients with AML. Quantitative measurement of MPO expression was done using flow cytometry analysis. Interestingly, the fraction of apoptotic cells induced by 24 hours treatment of PTL was significantly higher in AML specimens consisting of higher than 50% of MPO-positive cells (MPO^hi cases; 44.5 ± 1.6%, n = 6) compared to AML specimens consisting of lower than 20% of MPO-positive cells (MPO^lo cases; 1.4 ± 1.1%, n = 4, p<0.001). Our findings indicate that PTL induces apoptosis in myeloid leukemia cells is associated with increased ROS and the extent of MPO expression is a crucial determinant of their sensitivity to PTL-induced cell death.