Establishment and molecular characterization of decitabine‐resistant K562 cells

Abstract The clinical activity of decitabine (5‐aza‐2‐deoxycytidine, DAC), a hypomethylating agent, has been demonstrated in acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) patients. However, secondary resistance to this agent often occurs during treatment and leads to treatment failure. It is important to clarify the mechanisms underlying the resistance for improving the efficacy. In this study, by gradually increasing concentration after a continuous induction of DAC, we established the DAC‐resistant K562 cell line (K562/DAC) from its parental cell line K562. The proliferation and survival rate of K562/DAC was significantly increased, whereas the apoptosis rate was remarkably decreased than that of K562 after DAC treatment. In K562/DAC, a total of 108 genes were upregulated and 118 genes were downregulated by RNA‐Seq. In addition, we also observed aberrant expression of DDX43/H19/miR‐186 axis (increased DDX43/H19 and decreased miR‐186) in K562/DAC cells. Ectopic expression of DDX43 in parental K562 cells rendered cells resistant to the DAC. Taken together, we successfully established DAC‐resistant K562 cell line which can serve as a good model for investigating DAC resistance mechanisms, and DDX43/H19/miR‐186 may be involved in DAC resistance in K562.


| INTRODUC TI ON
DNA methylation is a major contributor to epigenetics involved in carcinogenesis especially in leukemogenesis. The balance is needed to be precisely maintained between DNA hypermethylation and hypomethylation, and dysregulation of the balance may give rise to human diseases. 1 Abnormal DNA methylation changes, associated with DNA methyltransferases (DNMTs), are frequently observed in leukemia and supposedly contribute to disease occurrence and progression. 2,3 Therapy targeting DNA methylation modifiers has been regarded as a success in the treatment of hematopoietic malignancies. 4,5 Gene silencing caused by DNA hypermethylation can be reversed pharmacologically by prototypical DNMT inhibitors decitabine (5-aza-2-deoxycytidine, DAC) and 5-azacytidine (AZA), which have been recommended as one of the primary treatments for older acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) patients. [6][7][8] The DAC is transported into the cell and then phosphorylated by deoxycytidine kinase (DCK) to the active metabolite 5-aza-dCTP, which incorporates into DNA during DNA replication to form a covalent complex with DNMTs, thereby inhibiting their activities followed by a reduction of DNA methylation, and consequently inducing antileukemia effects. 9 However, increasing clinical studies have found that resistance to such drug can develop during treatment and lead to treatment failure. Drug resistance was the major clinical obstacle to successful treatment of leukemia patients compared to patients with relatively sensitive cells. The clinical outcome of patients after failure with hypomethylating therapy was poor. 10,11 Insufficient incorporation into DNA was suggested to explain in vitro DAC resistance. 12 It was reported that DNMT3b was upregulated in hypomethylating agent-resistance cell lines. 13 Also, high cytidine deaminase (CDA)/DCK ratio could be a mechanism of primary resistance to DAC in some patients. 14 Nevertheless, the detailed mechanisms leading to DAC resistance still remains obscure. In this study, we induced K562 cell line for long periods of time using DAC to obtain the DAC-resistant K562 cell line and investigated the potential mechanisms of DAC resistance.

| Morphology and measurement of drug sensitivity
An inverted light microscope (Nikon) and Wright-Giemsa's compound stain were used to observe K562 and K562/DAC cells during the exponential phase. The nuclear to cytoplasm ratio of the cells was measured, which was the ratio of the diameter of the nucleus to the thickness of the cytoplasm on both sides.
K562 and K562/DAC cells were collected and placed in 6-well plates at a density of 1 × 10 5 /mL with 2 mL medium. Fresh medium containing DAC at final concentration ranging from 0 to 2 µmol/L was added immediately, then fresh DAC was supplemented every 24 hours. After 96 hours, the surviving cells were calculated by trypan blue exclusion. The concentration of DAC required for 50% growth inhibition was scored as half maximal (50%) inhibitory concentration (IC50) value. The degree of resistance was evaluated by IC50 value.
Each experiment was repeated three times. IC50 value of DAC was analyzed by the method of probit analysis in SPSS21.0 (SPSS Inc, USA).

| Cell survival and proliferation assays
Cell viability of the K562 and K562/DAC cells were assessed. Briefly, cells were seeded in 6-well plates at a density of 1 × 10 5 cells/well with growth medium containing 0% FBS (cell survival assay) or 10% FBS (cell proliferation assay). DAC was added with the final concentration of 1 µmol/L for 96 hours. The results were presented from three independent experiments.

| Cell apoptosis
To study cell apoptosis, cells were treated in 25 cm 2 tissue cul-

| RNA-Seq analysis
Total RNA was extracted from the cell samples by Trizol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions. RNA was subjected to RNA-Seq analysis by Beijing BerryGenomics Institute, China.

| RQ-PCR
cDNA was reverse transcribed from the RNA. Real-time quantitative PCR (RQ-PCR) was conducted to evaluate the mRNA and miRNA expression levels in the DAC resistant cells as previously described using the primer sets (Table S1). [15][16][17][18][19]

| DNA isolation, chemical modification, RQ-MSP and BSP
Genomic DNA isolation, chemical modification, real-time quantitative methylation-specific PCR (RQ-MSP) and bisulfite sequencing PCR (BSP) were performed as our previous study. 15

| Statistical analysis
All experiments were performed in triplicate (n ≥ 3) and the data were presented as mean ± SD. The Student's t test for independent samples was applied to define differences in the experiments. The differences of results were determined statistically significant if P was less than 0.05. To further explore biological property of K562/DAC cells, we found that after treatment with DAC, K562/DAC cells had significantly higher proliferation and survival rates and lower apoptosis rate as compared to K562 cells ( Figure 1C-E). Meanwhile, the ratio of G0/G1 phase in K562/DAC increased ( Figure 1F).

| Gene expression alterations identified in K562/DAC cells
To recognize genes associated with DAC resistance, the candidate genes differentially expressed in K562 and K562/DAC cells were identified. RNA-Seq analysis was used to screen the candidates ( Figure 2A, Figure S1). Then RQ-PCR was performed to validate the up-regulated oncogene in K562/DAC cells. Four up-regulated oncogenes in K562/DAC cells were validated by RQ-PCR. The levels of H19, ID1, ID3 and ITGA2 expressions dramatically increased in K562/ DAC cells ( Figure 2B). We also performed gene ontology (GO) enrichment analysis to classify differential genes into the categories of cellular component, molecular function and biological process, including extracellular space, protein binding and system development ( Figure 2C). To gain deeper understanding the roles of these differentially expressed genes in K562/DAC, we further carried out KEGG pathway enrichment analysis. It was found that these genes were mostly enriched in hematopoietic cell lineage, NF-kappa B signaling pathway and many other pathways ( Figure 2D).

| The role of DDX43/H19/miR-186 in DAC resistance
Our previous study had reported that overexpression of DDX43 in K562 cell line upregulated H19 through demethylation. 20 Also, miR-186 was found to target DDX43, and miR-186 was downregulated in DDX43-transfected cells. 20 Here, the density of H19 and DDX43 methylation was greatly decreased in K562/DAC cells ( Figure 3A and B).

| D ISCUSS I ON
The clinical outcome of patients after treatment failure with the DNA methylation inhibitors is poor in the clinics. 11,21 Therefore, it is important to illuminate the resistance mechanism and to overcome H19 is a long chain non-coding RNA with length of 2.3 kb. Several studies have reported that overexpression of H19 was correlated with drug resistance in many tumors such as lung adenocarcinoma, ovarian cancer, human glioma, liver cancer. [31][32][33][34] Our previous study revealed that H19 expression level, associated with its promoter methylation status, was significantly upregulated in CML patients involving in disease progression. 35 Also, H19 was identified to be upregulated by DDX43 through demethylation related to CML progression. 20 Among AML, H19 overexpression correlated with poor chemotherapy response and shorter overall survival 36 Taken together, we deduced that H19 may play a role in drug resistance during leukemogenesis. Thus, we detected the expression and methylation level of H19 and DDX43 in K562/ DAC cells, and showed positive results. However, the relevance of methylation-associated DDX43/H19 with DAC resistance need to be further explored. The axis of DDX43/H19/miR-186 may be an attractive candidate for overcoming drug resistance in leukemia therapy.
In conclusion, a good in vitro model was successfully established, which can be used for elucidating the molecular mechanisms related to DAC resistance, and DDX43/H19/miR-186 axis may be associated with DAC resistance.

CO N FLI C T S O F I NTE R E S T
The authors stated that there are no conflicts of interest regarding the publication of this article. Research support played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.