Azacitidine regulates DNA methylation of GADD45γ in myelodysplastic syndromes

Abstract Background Myelodysplastic syndrome (MDS) is a heterogeneous clonal disease originated from hematopoietic stem cells. Epigenetic studies had demonstrated that DNA methylation and histone acetylation were abnormal in MDS. Azacitidine is an effective drug in the treatment of demethylation. Methods RT‐PCR was performed to determine GADD45γ in 15 MDS clinical samples. Myelodysplastic syndrome cell lines SKM‐1 and HS‐5 were transfected with GADD45γ eukaryotic expression vector and/or GADD45γ shRNA interference plasmid, and treated with azacitidine. Proliferation and apoptosis were examined by CCK‐8 and Western blot analysis to confirm the function role of GADD45γ and azacitidine. The methylation level of GADD45γ gene was detected by bisulfite conversion and PCR. Results This study found that GADD45γ gene was down‐expressed in MDS patients' bone marrow and MDS cell lines, and the down‐regulation of GADD45γ in MDS could inhibit MDS cell apoptosis and promote proliferation. Azacitidine, a demethylation drug, could restore the expression of GADD45γ in MDS cells and inhibit the proliferation of MDS cells by inducing apoptosis, which was related to prognosis and transformation. Conclusion This study indicated that GADD45γ was expected to become a new target of MDS‐targeted therapy. The findings of this study provided a new direction for the research and development of new MDS clinical drugs, and gave a new idea for the development of MDS demethylation drug to realize precise treatment.

the lack of new MDS drugs confirmation methods, and the lack of long-term prospective randomized MDS clinical control to guide allogeneic blood and bone marrow transplantation. 1  in MDS. 2 The developments of small molecular drugs targeting specific molecular and the strategy to minimize their adverse reactions are the future development trend. 3 DNA methylation, or the covalent addition of a methyl group to cytosine within the context of the CpG dinucleotide, had been widely found in mammalian genome. These effects included transcriptional repression and chromatin remodeling, X chromosome inactivation, and parasitic DNA suppression. Normal methylation patterns were frequently disrupted in tumor cells, and the promoter of a tumor suppressor gene was hypermethylation within gene silence and cell deletions or mutations. 4 MDS-specific therapies included drugs targeting abnormal DNA methylation and chromatin remodeling, regulating/activating the immune system to enhance tumor-specific cellular immune response and reduce abnormal cytokine signaling, and blocking abnormal interactions between hematopoietic progenitor cells and stromal cells. 5 For MDS treatment, research began to deepen to the gene level, such as a mutated cluster affecting three inositol-specific genes, which was significantly related to the loss of response to azacitidine and linedoxamine treatment in high-risk MDS patients. 6 With the exploration of the pathogenesis of MDS, new potential therapeutic methods had been proposed, including hypomethylating agents with longer half-life and exposure time, regulatory proteins such as antiapoptotic BCL2 protein, inhibition of PD-1 or CTLA-4, natural immunity, and targeted therapy with CD33/ CD3 polyclonal antibody. 7 Azacitidine (AZA) was the first drug approved by the US Food and Drug Administration (FDA) (May 2004) for the treatment of MDS, and had been granted the status of orphan drug. Azacitidine (Vidaza) had been approved by the European Union for use in patients with high-risk MDS and AML. 8 Azacitidine was a DNA methylation inhibitor, targeting at epigenetic gene silencing, which was used by cancer cells to inhibit gene expression against malignant phenotypes. 9 At the same time, azacitidine was a valuable choice for the first-line treatment of high-risk MDS/AML patients. 10 The pathogenesis of MDS was supposed to hypermethylation of specific DNA sequences. 5-azacitidine and decitabine, which reactivate tumor suppressor gene transcription by DNA methylation, were the promising new agents. 11 Abnormal DNA methylation was related to gene silencing. Hypomethylation drugs worked by inducing the re-expression of epigenetic silencing genes. 12 Hypermethylation of tumor suppressor gene had been considered as an important pathogenesis of MDS. Azacitidine, a pyrimidine nucleoside analogue, had the function of inhibiting DNA methyltransferase. It could be used as a treatment to prolong the survival of MDS patients, thus changing the natural history of these malignant tumors. The activity of azacitidine in MDS promotes its combination with other epigenetic modified to treat MDS and AML. 13 In preclinical studies, azacitidine had low methylation/differentiation activity at low concentration, while high concentration was related to cytotoxic effect. In clinical trials, azacitidine not only improved MDS-related cell reduction, but also delayed the transformation of leukemia, improved the quality of life, and improved the overall survival rate of many patients receiving AZA treatment. 14 AZA could relieve the high-risk MDS, which had become the frontline therapy for MDS that did not meet the conditions of allogeneic stem cell transplantation. AZA therapy, combination of AZA and other drugs, and prognosis therapy with AZA in specific cases could treat myeloproliferative neoplasms (MPN), chronic myelomonocytic leukemia (CMML), and AML. 15 Epigenetic dysregulation was related to the pathogenesis of many malignant tumors, including MDS and AML. DNA methylation could lead to transcriptional silencing of tumor suppressor genes. By inhibiting DNA methyltransferase to re-express these genes, it could treat benign and malignant diseases. In hematology, azacitidine and decitabine were widely used in clinic as demethylation drugs. 16 5-azacitidine and decitabine had therapeutic effect on high-risk MDS, and 5-azacitidine could improve the survival rate of high-risk MDS patients. 17 The two structures were slightly different, and they were finally metabolized into 5-aza-CTP and 5-aza-dCTP through different metabolic pathways. Meanwhile, azacitidine was more effective in the treatment of MDS caused by mutation of DNA methylation pathway. 18 AZA was the most effective drug in the treatment of MDS, but was only suitable for 50% of patients. The cause of resistance to AZA was debatable. Recent studies had found that AZA responders had more hematopoietic progenitor cells (HPCS) in cell cycle. 19 Azacitidine changed the DNA methylation level of DLK1-DIO3 region in the treatment of MDS and myelodysplastic-related AML. 20 Systematic evaluation of azacitidine in the treatment of MDS and AML had also been paid more and more attention. 21 According to the latest research, the therapeutic effect of 5-azacitidine in MDS was analyzed by detecting DNA methylation in peripheral blood. 22 In the treatment of high-risk MDS, the relative dose intensity of azacitidine had a significant impact on the survival rate of patients. 23 After entering DNA, 5-azacitidine based form covalent compounds with methyltransferase, and then, proteasome was recruited to degrade methyltransferase, so that highly methylated genes could be re-expressed due to demethylation. 24 Reactivate genes that were inactivated by DNA over methylation allowed cells to return to normal terminal differentiation, senescence, or apoptosis. 25 At present, the prognosis evaluation system used in MDS is composed of morphology, clinical characteristics, and cytogenetics, but did not include molecular genetic data. The molecular mechanism of 5-aza-2-deoxycytidine (5-aza-2-deoxycytidine) in multiple myeloma (MM) was to upregulate a large number of tumor-related genes silenced by epigenetic genes. 36 The multi-targets and the lack of specificity made the treatment of MDS difficult. 25 The frequency of GADD45γ methylation and the epigenetic change of GADD45γ might be related to the progress of diffuse large B-cell lymphoma. 37

| Cell line culture
The cells were cultured in RPMI 1640 medium containing 10% fetal bovine serum and placed in a cell incubator at 37°C and 5% CO 2 saturated humidity until the logarithmic growth period.

| Construction of p3XFLAG-GADD45γ eukaryotic expression vector
Referring to the GADD45γ base sequence of GenBank mice, the  Plasmid extraction was used to obtain a pure lentivirus plasmid.

| Plasmid transfection into SKM-1 cells
The MDS cell line SKM-1 was inoculated into a 6-well plate for cell culture. When the cell fusion rate reached 70%-80% in 24 hours, the cells were transfected with p3XFLAG-GADD45γ, p3XFLAG (as negative control), pLKO.1-shRNA, and pLKO.1-shGFP at a concentration of 2.5:1 by Lipofectamine TM 2000. After incubation for 6 hours, the protein and RNA were extracted, respectively. Lipofectamine TM 2000 was replaced by DMEM containing 10% fetal bovine serum for further culture.

| Detection of GADD45γ mRNA expression by RT-PCR
MDS cells were inoculated on the 6-well plate. When the cell fusion rate reached 70%-80%, p3XFLAG-GADD45γ plasmid and p3X-FLAG plasmid were transfected with Lipofectamine TM 2000, and the cells were cultured for 72 hours. The third-generation HS-5 cells were inoculated on the 6-well plate, and the cell fusion rate reached 60%-70%. The qPCR primer sequences were listed in Table 1. When the lentivirus was infected, the culture solution of polyamine was changed, and 500 μL virus solution was added into each hole, the final concentration of polyamine was 8 μg/mL 48 hours later, and the minimum killing concentration of puromycin was used to screen the stable interfering cell line and continue to culture for 2 generations. The total RNA was extracted by Trizol method, and the cDNA was inversely obtained. RT-PCR was used to calculate the mRNA expression of the target gene after standardization by the internal reference gene.

| Detection of apoptosis-related protein expression by Western blot
The MDS cell line SKM-1 was inoculated on the 6-well plate. When the cell confluence was 70%-80%, the p3XFLAG-GADD45γ plasmid and p3XFLAG plasmid were transfected with Lipofectamine TM 2000. Seventy-two hours later, the third-generation FDC-P1 was inoculated on the 6-well plate, and the fusion degree of the cells was 60%-70%. When the lentivirus was infected, the culture medium containing polyamine was changed, and 500 μL lentivirus solution was added into each hole, so that the final concentration of polyamine was 8 μg/mL, 48 hours, and the minimum killing concentration of puromycin was selected to stabilize the interfering cell lines. The 2nd generation was continuously cultured, and the protein lysate was dissolved in 30 minutes by ice bath and centrifugated, and then, the 50 g protein thereof was extracted by dodecyl sulfonate-polyacrylamide gel electrophoresis. The wet spinning method was transferred to the polyvinylidene fluoride membrane.
The 5% skimmed milk powder was sealed at room temperature, and the 1H was added at room temperature (Caspase-3, Caspase-7, and Caspase-9). Caspase-3, Caspase-7, and Caspase-9 were diluted with 1:1000 and incubated overnight at 4°C. After TBST washing, HRPlabeled second antibody (diluted with 1:5000) was added. After incubation for 1H at room temperature, TBST membrane was rinsed for three times, the positive bands were displayed with chemiluminescent reagent, and the images were processed and analyzed with gray-scale analysis software.   SPSS 22.0 software was used to statistical analyze all data. All data were expressed as mean ± SD of at least three separate experiments. For the RT-PCR or Western blot analyses, the mean ± SD was derived from triplicate measurements of one experiment. A t test or one-way ANOVA was used to compare the groups. P < .05 means significant difference.

| Expression of GADD45γ gene in MDS patients' bone marrow cells and MDS cell lines
Based on the fact that GADD45γ could inhibit tumor growth by promoting apoptosis, and it had been reported that GADD45γ has biological activity in AML treatment, this study first analyzed the expression level of GADD45γ in MDS patients' bone marrow cells.

| Effection of GADD45γ on proliferation and apoptosis of MDS cells
After that, we studied the biological activity of GADD45γ in MDS showed that GADD45γ was successfully expressed in SKM-1 cell line, and the expression level was significantly higher than that in the control group. The expression of GADD45γ was successfully inhibited in FDC-P1 cells (Figure 2A Figure 2D). So far, we found that GADD45γ can inhibit or promote apoptosis by changing the expression level of apoptosis-related proteins in MDS cell line.

| Effection of azacitidine on the expression of GADD45γ in MDS cells
Azacitidine was the first drug to treat MDS, but its mechanism needed further study. After confirming the effect of GADD45γ on MDS cell line by inducing apoptosis, we began to try to detect the change of

| Mechanism of azacitidine targeting GADD45γ gene therapy for MDS
In order to study how azacitidine acted on MDS cell line through GADD45γ, we further overexpressed GADD45γ in SKM-1 cell line. The results showed that GADD45γ was overexpressed in SKM-1 cell line, and with the increase of GADD45γ expression level, the protein expression level related to apoptosis was also increased, and the proliferation of SKM-1 cell line was significantly inhibited (Figure 4).
When azacitidine was added to SKM-1 cell line, the expression level of GADD45γ was further increased, and the effect of apoptosis was more obvious.
Azacitidine is a classical demethylation drug. GADD45γ DNA was extracted from the experimental group and the control group.
Compared with the control group, the methylation level of GADD45γ in the experimental group with azacitidine was significantly lower.

ACK N OWLED G M ENTS
This work was supported by the Key Natural Science Research Projects of Anhui Education Department (KJ2019A0375).

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interests.