Arsenic Trioxide Resistance: More to It Than Mutations in PML-RARα

There has been a recent concern of arsenic trioxide (ATO) resistance in patients with acute promyelocytic leukemia (APL) treated with ATO as upfront therapy. The focus of ATO resistance has centred on mutations in PML-RARA gene (Blood 2011, NEJM 2014). NB4 cells in our laboratory were exposed to serial increasing concentrations of ATO. We subsequently generated 3 ATO resistant clones, NB4EV-ASR1, ASR2 and ASR3. In addition we have also evaluated an established ATRA resistant APL cell line UF1 (Gift from Dr. Chomienne. C).

In a viability assay, we observed that these cell lines are resistant to ATO and had an IC50 above 2µm (Table1). These resistant cell lines also had a higher IC50 to other therapeutic drugs such as Daunorubicin and Cytosine arabinoside and a reduced differentiation effect on exposure to ATRA (summarized in Table 1). To identify the differences between the naïve cells and the resistant cells, we did whole exome sequencing (NGS) by Iontorrent and found that only NB4 EV-ASR1 clone had ATO resistance causing mutation (A216V) in the PML B2 domain (VAF=91.7%) while the other two cell lines (NB4EV-ASR2 and ASR3) did not have a mutation in PML-RARA. Next, we did an expression array to find the differential regulated genes between the naïve cell line and the parent resistant cell line from where all the 3 cell lines had been derived. We found that 1490 genes were differentially regulated (> 2 fold). The pathways significantly enriched for differentially expressed genes were cell survival, ABC transporters, Glutathione synthesis, Ubiquitin- proteasome degradation system and signalling pathways like PI3-AKT and PTEN.

Validating the micro-array data, we found that there is an increased expression of ABC transporters such as MRP4, AQP9 (n=3; Figure1A) which are known to efflux ATO from the cells and a decreased expression of ABCA1 (known to efflux glutathione). The up regulation of these transporters also correlated with decreased levels of intracellular ATO (IC-ATO; measured using AAS, see Figure 1B for details) in the resistant cell lines. We also noted that there is a varying reduction in the basal reactive oxygen species levels and a varying increase in the amount of basal reduced glutathione (GSH) levels in the resistant cell lines (n=3, Table 1). We have noted that adding Buthionine sulphoximine (BSO - GSH inhibitor) along with ATO was able to restore the sensitivity of ATO in the resistant cells lines, however there was significant variation in the sensitivity of ATO among the cell lines when treated with the same concentration of BSO (Figure 1C).

At the transcript levels we did not find any difference in expression of PML-RARA but at the protein level we noted a significant reduction in the levels of PML-RARA in the resistant cell lines (Figure 1D). We also observed an increase in the proteasome activity in the resistant cell lines compared to naïve cells (data not shown). In an immunofluorescence assay probing for PML, we found an absence of micro-speckled pattern in the resistant cell lines and UF1 cell lines compared to naive cells (Figure 1D).

In conclusion, we have observed that in addition to PML-RARA mutations, variations in the Redox system, ABC transporters, intracellular ATO concentration and anti-apoptosis pathways are likely to be altered in ATO resistance. It is likely that ATO resistance is multi-factorial and that the dominant mechanism can vary between different resistant cell lines and potentially the same variation could be seen in relapsed patients. Importantly in the presence of ATO resistance there was also a decrease in sensitivity to other conventional agents used to treat APL. Novel agents and strategies based on these observations are required to address the issue of ATO resistance in patients with relapsed APL.

View largeDownload slide

View largeDownload slide

Close modal