The oxidative state is significantly perturbed in leukemia cells, compared to normal cells, due to generation of increased reactive oxygen species (ROS) and/or dysregulated antioxidant mechanisms. In a molecularly heterogeneous disease such as acute myeloid leukemia (AML), targeting the oxidative state, a fundamental physiological property, is an attractive strategy for developing novel anti-leukemic agents. We hypothesized that regardless of oncogenic mutations, therapeutic augmentation of ROS in AML cells with dysregulated antioxidants would kill leukemia cells and leukemia stem cells, while normal stem cells would remain relatively intact. Quinones can initiate and propagate free radical reactions. The electron-accepting nature of quinones could in principle be tuned to yield selective cytotoxicity for cells with a particular redox signature. Dimeric naphthoquinones (BiQs), with known ability to generate ROS in cancer cells, are a novel class of compounds with unique characteristics that make them excellent candidates to be tested against leukemia cells. The primary objective of this study was to determine the potency of BiQs in leukemia cells and to assess the therapeutic index of BiQs in AML cells, in relation to normal hematopoietic stem cells. The secondary objective was to determine whether BiQs induce apoptosis, increase ROS, target mitochondrial membrane potential, and/or affect antioxidant proteins.


We tested two BiQs (E6a and B1a) and one mononaphthoquinone in two AML cell lines, MOLM-14 (FLT3-ITD) and THP-1 (FLT3-WT), two normal karyotype primary AML samples (AML15, AML16), both with FLT3-WT, and fresh normal bone marrow cells. IC50 values were determined by exposing cell lines and bone marrow cells to BiQs for 72 h and 48 h, respectively. Each experiment was terminated with Alamar Blue (Sigma). Cell viability assays were carried out similarly except that the endpoint was trypan blue exclusion. For clonogenic assays, cells were plated in methylcellulose and exposed to BiQs either prior to plating or both prior to plating and during culture. Colonies were counted using an automated colony counter (Microbiology Intl.). Apoptosis was measured using the FITC Annexin V Kit (BD Pharmingen), and mitochondrial potential was assessed using the MitoRed Kit (Millipore); cells were analyzed by FACScan (BD Biosciences) and Flow Jo Software (Tree Flo). Standard Western blot techniques were used for measurement of caspase 3 and Nrf2 protein. Numbers are mean ± SD.


The IC50 values (µM) of E6a for MOLM-14, THP1, AML15 and AML16 were 5.5 ± 0.8, 4.2 ± 1.9, 5.1, and 0.4, respectively. The IC50 values (µM) of B1a for MOLM-14, THP1, AML15 and AML16 were 6.2 ± 0.7, 5.6 ± 0.1, 5.4, and 3.0, respectively. The IC50 value of E6a for normal bone marrow cells was 14.5 µM. The mononaphthoquinone did not show anti-leukemic activity. Viability tests showed a 52 ± 8% and 33 ± 9% increase in MOLM-14 cell kill after exposure to 5 µM E6a and 5 µM B1a, compared to vehicle, versus 69 ± 12% and 59 ± 2% in THP-1. Treatment with E6a, compared to vehicle, resulted in a concentration-dependent decrease in colony formation in MOLM-14 (52 ± 17 [10 µM] vs 228 ± 170 [5 µM] vs 338 ± 72 colonies). Interestingly, when MOLM-14 cells were treated with E6a for 24 h then placed in methylcellulose in continuous presence of E6a for 9 days, colony formation was completely inhibited with 5 µM E6a compared to vehicle (2.5 ± 1 vs 294 ± 43 colonies). Treatment of AML16 with 5 µM E6a for 24 h resulted in a 1.5-fold increase in apoptosis, and a 1.6-fold increase in loss of mitochondrial membrane potential compared to vehicle. After 6 h exposure of MOLM-14 and AML16 to 5 µM E6a, cleavage of caspase 3 was detected. After 2 h exposure of AML16 to 5 µM E6a and 5 µM B1a, Nrf2 proteins was significantly increased in response to oxidative stress.


We have demonstrated that BiQs possess anti-leukemia activity with a reasonable therapeutic index in relation to normal bone marrow cells. Measurements of effects on intracellular ROS and on proteins involved in oxidative stress are ongoing, and drug transport by multidrug resistance-associated ATP-binding cassette proteins including P-glycoprotein will be studied. The more active BiQ will be studied further in an AML primagraft model.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.