Different mechanisms of drug resistance to hypomethylating agents in the treatment of myelodysplastic syndromes and acute myeloid leukemia.

Resistance to the hypomethylating agents (HMAs) 5-azacytidine (AZA) and 5-aza-2'-deoxycytidine (DAC) represents a major obstacle in the treatment of elderly patients with myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) which are not suitable for hematopoietic stem cells transplantation. Approximately 50 % of patients do not respond to HMA treatment because of intrinsic (primary) resistance, while others could acquire drug resistance during the repeated cycles of the treatment. To prevent, delay or surmount resistance development, the molecular mechanisms underlying drug resistance must be first identified. This is crucial as no further standard therapeutic opportunities are available for these patients who failed hypomethylating agents-based treatment. The current review provides an updated information about the different mechanisms that may contribute to the development of resistance to HMAs. Despite the similar structure and mechanism of action of HMA, several studies did not report the expected development of cross-resistance. It is clear that in addition to the common modalities of chemoresistance, there must be some specific mechanisms of drug resistance. Changes in transport and metabolism of HMAs are among the most studied mechanisms of resistance. Drug uptake provided by two solute carrier (SLC) families: SLC28 and SLC29 (also known as the concentrative and equilibrative nucleoside transporter families, respectively), could represent one of the mechanisms of cross-resistance. Changes in the metabolism of these drugs are the most likely mechanism responsible for the unique mode of resistance to AZA and DAC. Deoxycytidine kinase and uridine-cytidine kinase due to their necessity for drug activation, each could represent one of the response markers to treatment with DAC and AZA, respectively. Other mechanisms involved in the development of resistance common for both drugs involved: i. increased DNA repair (caused for example by constitutive activation of the ATM/BRCA1 pathway and inhibition of p53-dependent apoptosis); ii. changes in the regulation of apoptosis/disrupted apoptotic pathways (specifically increased levels of the anti-apoptotic protein BCL2) and iii. increased resilience of leukemic stem cells to multiple drugs including HMAs. Despite intense research on the resistance of MDS and AML patients to HMAs, the mechanisms that may reduce the response of these cells to HMAs are not known in detail. We herein highlight the most important directions that future research should take.

Changes in Apoptotic Pathways in MOLM-13 Cell Lines after Induction of Resistance to Hypomethylating Agents.

We established the following two variants of the MOLM-13 human acute myeloid leukemia (AML) cell line: (i) MOLM-13/DAC cells are resistant to 5-aza-2'-deoxycytidine (DAC), and (ii) MOLM-13/AZA are resistant to 5-azacytidine (AZA). Both cell variants were obtained through a six-month selection/adaptation procedure with a stepwise increase in the concentration of either DAC or AZA. MOLM-13/DAC cells are resistant to DAC, and MOLM-13/AZA cells are resistant to AZA (approximately 50-fold and 20-fold, respectively), but cross-resistance of MOLM-13/DAC to AZA and of MOLM-13/AZA to DAC was not detected. By measuring the cell retention of fluorescein-linked annexin V and propidium iodide, we showed an apoptotic mode of death for MOLM-13 cells after treatment with either DAC or AZA, for MOLM-13/DAC cells after treatment with AZA, and for MOLM-13/AZA cells after treatment with DAC. When cells progressed to apoptosis, via JC-1 (5,5',6,6'-tetrachloro-1,1',3,3'-tetraethyl-imidacarbocyanine iodide) assay, we detected a reduction in the mitochondrial membrane potential. Furthermore, we characterized promoter methylation levels for some genes encoding proteins regulating apoptosis and the relation of this methylation to the expression of the respective genes. In addition, we focused on determining the expression levels and activity of intrinsic and extrinsic apoptosis pathway proteins.

Resistance of Leukemia Cells to 5-Azacytidine: Different Responses to the Same Induction Protocol.

Three AML cell variants (M/A, M/A* from MOLM-13 and S/A from SKM-1) were established for resistance by the same protocol using 5-azacytidine (AZA) as a selection agent. These AZA-resistant variants differ in their responses to other cytosine nucleoside analogs, including 5-aza-2'-deoxycytidine (DAC), as well as in some molecular features. Differences in global DNA methylation, protein levels of DNA methyltransferases, and phosphorylation of histone H2AX were observed in response to AZA and DAC treatment in these cell variants. This could be due to changes in the expression of uridine-cytidine kinases 1 and 2 (UCK1 and UCK2) demonstrated in our cell variants. In the M/A variant that retained sensitivity to DAC, we detected a homozygous point mutation in UCK2 resulting in an amino acid substitution (L220R) that is likely responsible for AZA resistance. Cells administered AZA treatment can switch to de novo synthesis of pyrimidine nucleotides, which could be blocked by inhibition of dihydroorotate dehydrogenase by teriflunomide (TFN). This is shown by the synergistic effect of AZA and TFN in those variants that were cross-resistant to DAC and did not have a mutation in UCK2.

A decrease in cellular microRNA-27a content is involved in azacytidine-induced P-glycoprotein expression in SKM-1 cells.

We established an azacytidine (AzaC)-resistant human acute myeloid leukemia (AML) cell line (SKM-1/AzaC) by culturing SKM-1 cells in the presence of increasing amounts of AzaC for six months. Because AzaC is not a substrate of P-glycoprotein (a product of the ABCB1 gene; ABCB1), ABCB1 was not responsible for AzaC resistance; nevertheless, it was notably upregulated in SKM-1/AzaC cells. In addition, the transcription of the Nfkb1 gene, which encodes a member of the canonical NF-kappaB regulatory pathway, was downregulated, and the transcription of the Nfkb2 gene, which encodes a member of the non-canonical NF-kappaB regulatory pathway, was upregulated in SKM-1/AzaC cells. Here, we investigate whether miRNA-27a and miRNA-138 (both of which are known to be regulators of ABCB1 expression) are involved in the regulation of ABCB1 expression in SKM-1/AzaC cells. We observed decreased levels of miRNA-27a but of not miRNA-138 in SKM-1/AzaC cells compared with SKM-1 cells. The transfection of SKM-1/AzaC cells with a miRNA-27a mimic induced the downregulation of the ABCB1 mRNA. This was associated with an increase in Nfkb1 and a decrease in Nfkb2 transcript levels in SKM-1/AzaC cells. Taken together, these data indicate that the downregulation of miRNA-27a is involved in the upregulation of ABCB1 expression in SKM-1/AzaC cells, and this effect is associated with a switch between the canonical and non-canonical NF-kappaB pathways.

Acute myeloid leukemia cells MOLM-13 and SKM-1 established for resistance by azacytidine are crossresistant to P-glycoprotein substrates.

Establishment of the acute myeloid leukemia cells SKM-1 and MOLM-13 for resistance by azacytidine (AzaC) resulted in SKM-1/AzaC and MOLM-13/AzaC cell variants with reduced sensitivity to AzaC. Despite the fact that AzaC is not substrate of P-glycoprotein (P-gp), the adaptation procedure resulted in an induction in P-gp expression/efflux activity that confers crossresistance to P-gp substrates in both resistant cell variants. While the resistance to P-gp substrates in SKM-1/AzaC and MOLM-13/AzaC cells could be reversed by the P-gp inhibitors, resistance to AzaC was insensitive to these inhibitors in both resistant cell variants. In addition, NF-κB and the antiapoptotic protein Bcl-2 were downregulated and the proapoptotic proteins Bax and p53 were upregulated in both resistant cell variants when compared with their sensitive counterparts. Moreover, at least five times the elevation in overall glutathione S-transferase activity was measured with 1-chloro-2, 5-dinitrobenzene as a substrate in the resistant variant of both cell lines. Taken together, the findings of the present study indicate that the treatment of AML cells with AzaC might lead to a drug resistance phenotype that may be associated with cross resistance to P-gp substrates and substrates of glutathione S-transferases.

Targeting lectin activity into inclusion bodies for the characterisation of glycoproteins.

Physiological aggregation of lectin functional domains into active inclusion bodies would provide a simple tool for glycocode reading by well-established agglutination assays.