Background
Currently, the treatment for acute myeloid leukemia (AML) includes standard chemotherapy with anthracycline and cytarabine [
1,
2]. Although treatment outcomes in adult AML patients have improved over the past decade, up to 30% of adult patients fail to achieve complete remission (CR) following 2 cycles of intensive chemotherapy. In addition, a large number of patients experienced relapse despite having achieved complete remission [
3‐
5]. Therefore, clinicians continue to find it challenging to manage these two types of patients [
6‐
9]. Furthermore, as the exact mechanism for drug resistance remains largely unclear, there is an urgent need to discover the mechanisms for drug resistance and develop more effective treatment regimens for patients with refractory/relapsed AML.
As a promising class of anticancer drugs, histone deacetylase (HDAC) inhibitors induce apoptosis and suppress tumor cell growth [
10,
11]. Chidamide is a novel oral HDAC inhibitor designed to inhibit the activity of HDAC1, 2, 3 and 10. Chidamide is currently being used in multiple clinical trials as monotherapy or combination therapy for the treatment of various hematological and solid cancers [
12‐
16]. In these trials, chidamide inhibited cell growth in a variety of cancers, such as lung cancer, pancreatic cancer, leukemia, and lymphoma [
15‐
18]. Chidamide has also inhibited the viability of MDS and AML cells [
19]. However, the possibility and ways of using chidamide as a treatment option for relapsed/refractory AML following anthracycline therapy remain to be explored. This study shown that chidamide-based regimen improves the overall remision rate of patients with relapsed/refractory AML, as well as prolongs disease-free survival time and overall survival rates. Moreover, this study is the first to find out chidamide increases the sensitivity of anthracycline-resistant cells to anthracycline drugs, and these effects are associated with the inhibition of the HDAC3-AKT-P21-CDK2 signaling pathway, thus demonstrating the potential for application.
Materials and methods
Analysis of clinical cases
Our dataset consists of 27 patients with relapsed/refractory AML, who had received anthracycline-based treatment regimen as induction therapy between Jan. 01, 2018 and Jan. 01, 2019 at the Chinese PLA General Hospital (Supplementary Table 1). According to NCCN (The National Comprehensive Cancer Network) clinical practice guideline for AML, relapse following CR is defined as the reappearance of leukemic blasts in the peripheral blood or the finding of more than 5% blasts in the bone marrow, not attributable to any other cause (bone marrow regeneration after consolidation therapy) or extramedullary relapse. Primary refractory or resistant disease is defined by being unable to achieve complete remission after 1 to 2 cycles of intensive induction therapy.
The earliest time at which AML was diagnosed was on Jun. 01, 2016. Patients who relapsed or were resistant to anthracycline were treated with chidamide (30 mg on days 1, 4, 8, and 11) in combination with an anthracycline-based regimen as salvage therapy. The salvage therapy was as follows: ① chidamide, 30 mg, oral, days 1, 4, 8, and 11; DAC, 20 mg/m2/d, i.v., days 1–5; Acla, 20 mg/d, i.v., days 1, 3, 5; Ara-C, 100 mg/m2, q12h, i.h., days 1–5; G-CSF, 300 μg/d, i.h., day 0 until neutrophil levels return to normal. ② chidamide, 30 mg, days 1, 4, 8, and 11; Acla, 20 mg/d, i.v., days 1, 3, 5; Ara-C, 100 mg/m2, q12h, i.v., days 1–5; G-CSF, 300 μg/d, i.h., day 0 until neutrophil levels return to normal. ③ chidamide, 30 mg, days 1, 4, 8, and 11; mitoxantrone, 5 mg/d, i.v., days 1–5; Ara-C, 100 mg/m2, i.h., days 1–5; VP-16, 100 mg/d, i.v., days 3–5; sorafenib, 400 mg, oral, q12h. This study was approved by the ethics committee of the Chinese PLA General Hospital. Patients have given written informed consent for the collection of clinical data.
RNA-sequencing and data analysis
Total RNA was extracted using TRIzol Reagent (Invitrogen, USA) according to the manufacturer’s instructions. RNA-seq was performed with a Genome Analyzer IIx (Illumina, San Diego, CA, USA). Expression levels were measured using the RPKM method, and the distribution of gene expression was analyzed.
TCGA database analysis of HDAC3 prognosis
AML patient datasets were obtained from The Cancer Genome Atlas database (TCGA) [
20], which had integrated clinical and RNA-seq information. Written informed consent was obtained from all patients and approved by the Human Studies Committee at Washington University [
20]. Patients in the top quartile of HDAC3 expression were classified as high-HDAC3, and patients in the bottom quartile of HDAC3 expression were classified as low-HDAC3.
Drugs and cell lines
Chidamide was received at no cost from Shenzhen Chipscreen Biosciences Ltd. (Shenzhen, China). K562, K562/A02 (multi-drug resistant leukemic cells, chronic myelogenous leukemia transferred to acute myelogenous leukemia), HL60 and its parallel anthracycline-resistant cell line, HL60/ADR, were kindly gifted by Tianjin Institute of Hematology. The human acute myeloid leukemia were also from Tianjin Institute of Hematology. THP-1 cells were purchased from ATCC (American type culture collection) and THP-1/ADR were its parallel anthracycline-resistant cell line. HEK293T was obtained from ATCC.
The cells were maintained at 37 °C in a humidified atmosphere containing 5% CO
2. K562/A02 cells were cultured in RPMI-1640 media, supplemented with 10% fetal bovine serum (FBS), 100 μg/mL penicillin, and 10 μg/mL streptomycin. HEK293T cells were cultured in DMEM with 10% fetal bovine serum, 2 mM L-glutamine and antibiotics. To maintain drug resistance, doxorubicin was added to the media (final concentration of 0.5 μg/mL) for at least 1 week every 2 months. One week prior to experiments, cells were re-cultured without doxorubicin [
21].
Peripheral blood mononuclear cells (PBMCs) from 22 patients, who had resistance to anthracyclines-based treatment regimens, were isolated with Ficoll-Hypaque centrifugation, and cultured in RPMI-1640 media supplemented with 10% fetal bovine serum (FBS), 100 μg/mL penicillin, and 10 μg/mL streptomycin. PBMCs were treated with different doses of chidamide as monotherapy or the combination of chidamide + doxorubicin for 24 h and 48 h, respectively. The basic characteristics are shown in Table
1. All patients have given informed consent for the use of their cells in this study.
Table 1
Patients characteristics
1 | AML-M2 | Relapse | PB | 91.2% | MA, DCAG |
2 | AML-M2 | Relapse | PB | 87.5% | CAG, IA, |
3 | AML-M4 | Relapse | PB | 85.6% | DCAG, MA |
4 | AML-M5 | refractory | PB | 98.7% | DA, DCAG |
Cell proliferation assays
Cell viability was measured using CCK-8 assay (Dojindo Molecular Technologies, Inc) according to the manufacturer’s instructions. Cells were treated with different concentrations of chidamide as monotherapy or the combination of chidamide + doxorubicin for 24 and 48 h, respectively. Subsequently, 10 μL of CCK-8 was added to each well and incubated for 2–3 h. OD values were measured with a microplate reader at 450 nm.
Cell-cycle analysis
Following 24 and 48 h, cells were harvested and washed once or twice with PBS and fixed with ethanol overnight. After fixation, the cells were washed with PBS, treated with 100 g/mL RNase A, and stained with 100 g/mL PI. Cell-cycle data were collected on a flow cytometer with 488 nm laser and analyzed with MoFlo MLS sorter (Dako, FortCollins, CO).
Apoptosis assays
At 24 and 48 h after drug treatment, the cells were harvested, washed twice with ice-cold PBS, and resuspended in binding buffer containing 10 uL PI and 5 uL Annexin-V-FITC (YEASEN) for 15 min at room temperature in a light-protected chamber. All specimens were analyzed on a FACS Calibur.
Real-time PCR
Total RNA was extracted by TRIzol (Invitrogen, Carlsbad, CA, USA) and cDNA was synthesized by PrimeScript™ RT reagent Kit (Takara) according to the manufacturer’s instructions. Real-time PCR was then performed using KAPA SYBR FAST q-PCR Master Mix (2x) Kit using the primers specified in Table
2. We used the 2-ΔΔCt formula to examine the relative quantification of the target genes.
Table 2
Primers sequences (5′-3′)
HDAC1 | AACTgCTAAAgTATCACCAgAggg | TggCCTCATAggACTCgTCA |
HDAC2 | TCCTgAggTggTTTggTggC | ATATCACCgTCgTAgTAgTAgCAgA |
HDAC3 | ATgCCTTCAACgTAggCgATg | CgAgggTggTACTTgAgCAgC |
HDAC10 | TCggCAggATTTgACTCAgC | TggACTCTAgggCACTCTgAC |
AKT | AAgTCATCgTggCCAAggAC | AggTggAAgAACAgCTCgC |
P21 | gAgCTgCgCCAgCTgAggTgT | gACATggCgCCTCCTCTgAgTgCC |
CDK2 | TggACACgCTgCTggATg | AATggCAgAAAgCTAggCCC |
GAPDH | CTCTggTAAAgTggATATTgT | ggTggAATCATATTggAACA |
Western blot analysis
After cells were harvested, proteins were separated by SDS polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene difluoride membrane. The blots were blocked with 5% non-fat dry milk at 37 °C for 2 h and incubated with a specific primary antibody and secondary antibody according to the manufacturer’s instructions.
HDAC activity assay
HDAC activity was detected using a Colorimetric HDAC Activity Assay Kit (BioVision). Each reaction was conducted with 100 μL, which contained 50 μg of proteins extracted from cells. HDAC activity was measured with a microplate reader (SpectramMaxM5) at 405 nm.
Plasmid and transfection
HDAC3 shRNA plasmid (sc-35,538-SH, Santa Cruz Biotechnology, City, State, Country) was transfected into cells using Superfect reagent (Qiagen) according to the manufacturer’s protocol. Transfected cells with non-specific shRNA plasmid were used as blank control. HA-HDAC3 plasmid (NM_003883-HA) and its corresponding control plasmid were obtain from Genechem. Flag-AKT and its control plasmid were purchase from Hanbio. Plasmids and their respective control were transfected into cells using Lipofectamine™ 3000 reagent (ThermoFisher #L3000008). After transfection, cells were subjected to RT-PCR and Western blot to verify HDAC3 or ATK expression.
Co-immunoprecipitation
HEK293 cells were transfected with Flag-AKT (wild type) alone or plus HA-HDAC3 for 48 h. And lysates immunoprecipitated with Flag were immunoblotted for Acetylated-lysine, HA, p-AKT and Flag. Co-immunoprecipitation reagents (ThermoFisher, #88804) were obtain from ThermoFisher.
In vivo study
The xenograft experiments were performed in NOD/SCID immunodeficient mice (aged 6–8 weeks). The mice were maintained in an air-conditioned pathogen-free room under conditions of controlled lighting (12 h light and 12 h dark per day) and fed a standard diet of laboratory rodent food and water. 1*10
7 cells K562/A02 cells in 200 μL PBS were injected subcutaneously into the lateral flanks of mice. Tumors were observed and measured every other day. The tumor volumes were determined by the formula V = 0.5*Length*width
2. The tumor-bearing mice were randomized into 4 groups (
n = 5, Figure
S3A). The mice in doxorubicin-treated group were injected doxorubicin at the dose of 5 mg/kg intraperitoneally once a week and intragastrically instillated normal saline at the same time. The mice in chidamide-treated group were intragastrically instillated at the dose of 5 mg/kg chidamide three times a week and injected with PBS. And the mice in group control were treated both with PBS and normal saline as a control. The mice in the combined-therapy treated group received both intraperitoneal injection of doxorubicin and intragastric instillation of chidamide. Mice were sacrificed at the 21st day after inoculation and tumors were harvested for molecular characterizations. All animal experiments complied with the international and institutional rules and all animal protocols were approved by the Experimental Animal Ethics Committee of Chinese PLA General Hospital.
Statistical analysis
To determine gene expressions, the threshold of q-value was set at 0.05, and the absolute value of log2 ratios at KEGG and GO enrichment using the database for Annotation. Modified Fisher’s exact test was used to analyze the significance of GO and KEGG enrichment. Data obtained from triplicated experiments were reported in mean ± SD and analyzed using SPSS18.0 software. An appropriate Mann–Whitney test or Student’s t-test was adopted to perform comparisons. Kaplan–Meier method and log-rank test were used to analyze the relationship between HDAC3 expression levels and OS or EFS. Cox proportional hazard models were performed to test the associations between HDAC3 expression levels, OS and EFS using multivariate analysis. All data were analyzed using the R 3.1.1 software, and P-values below 0.05 were considered to be statistically significant. (*P < 0.05, **P < 0.01, ***P < 0.001).
Discussion
Currently, the treatment for relapsed/refractory AML remains a major challenge. A drug or treatment regimen is required to improve the remission rates for R/R to prolong overall survival time. It is necessary to discover the mechanism behind AML resistance, and find target drugs or regimens to overcome drug resistance, so as to improve the CR rate of patients with relapsed/refractory AML, prolong disease-free survival time and overall survival rates.
This study shown that the chidamide-based regimen improves the overall remision rate of patients with relapsed/refractory AML, so we want to further analyze the mechanism of bring such good clinical therapeutic results. To obtain a better understanding of the mechanism of drug resistance, we performed RNA-sequencing in drug-resistant and control cells. From the results of RNA-sequencing, we found that the PI3K-AKT signaling pathway was activated and HDAC3 was upregulated in drug-resistant cells.
Activated AKT regulated apoptosis, cell growth, and the cell-cycle through phosphorylation of numerous downstream targets [
27]. Inhibition of the PI3K-AKT signaling pathway decreased leukemia stem cell growth and increased apoptosis [
28,
29]. Moreover, the PI3K/AKT signaling pathway is often hyperactive in AML patients with poor prognoses [
28‐
30]. In several types of cancer, including AML, activation of the PI3K/AKT signaling pathway reduces sensitivity to chemotherapeutic drugs [
31].
It was shown that the aberrant recruitment of HDACs plays an important role in leukemogenesis. Changes in activity and/or expression levels of HDACs were also observed in leukemia and solid tumors [
32,
33]. Overexpression of HDACs in tumor cells was shown to protect cells from genotoxic insults [
23]. For example, the expression of HDAC3 was shown to be vital to the maintenance of genome stability and DNA damage control. HDAC3 knockdown impaired DNA repair [
34] and resulted in growth inhibition of human colon cancer cell lines [
35]. Moreover, depletion or pharmacological inhibition of HDAC3 triggered apoptosis in cutaneous T-cell lymphoma and multiple myeloma [
36,
37].
Previous studies have confirmed the close relationship between HDAC3 and AKT [
24,
25]. HDAC3 binds to AKT in several cell lines, and regulates the phosphorylation and acetylation levels of AKT. Overexpression of HDAC3 reduces AKT acetylation levels but increases AKT phosphorylation levels.
Taking earlier research and this study into account, it is evident that HDAC3 and P-AKT expressions are positively correlated in drug-resistant K562/A02 cells. In addition, activation of the HDAC3-AKT-P21-CDK2 pathway represents one of the mechanisms through which K562/A02 cells achieve resistance to anthracycline. Therefore, there is an urgent need for a drug to inhibit this signaling pathway and reverse drug resistance. Here, we found that chidamide inhibits cell proliferation, increases cell apoptosis, and induces cell-cycle arrest in a time- and dose-dependent manner in anthracycline-resistant AML cells. These effects are associated with the inhibition of the HDAC3 -AKT-P21-CDK2 signaling pathway.
Our study showed that expression levels of HDAC3 had increased in anthracycline-resistant cells. Univariate and multivariate analyses confirmed that HDAC3 had an adverse effect on OS and EFS. Our research showed that the HDAC inhibitor, chidamide also exhibited a significant inhibitory effect on anthracycline-resistant AML cell growth by suppressing HDAC3 expression. This result is consistent with another study that showed that chidamide inhibits pancreatic tumor growth by suppressing the expression of HDACs [
38].
Cyclin-dependent kinases (CDKs) are a family of serine/threonine protein kinases that regulate cell cycle progression [
39]. CDK2 activity is necessary for cells to progress through the S phase [
40,
41]. CDKs are inhibited by CDK inhibitors (CKIs) [
42], such as P21, which suppresses CDK2 activity and blocks cell cycle transition from G1 phase to S phase [
43‐
45]. Moreover, the expression of P21 has always been regulated by HDAC inhibitors [
38,
46,
47] and P21 is upregulated following HDAC3 depletion [
48]. In this study, the HDAC inhibitor, chidamide induced cell-cycle arrest in a time- and dose-dependent manner in anthracycline-resistant AML cells. The mechanism of action may be through the inhibition of CDK2 expression and increase in P21 expression.
Patients who are resistant to primary chemotherapy or have relapsed have poor prognoses due to the inability to control disease progression and therapeutic complications. Histone deacetylases (HDACs) are responsible for the regulation of gene transcription, protein function, and stability [
49,
50]. HDAC inhibitors are able to potently induce cell cycle arrest, differentiation, and apoptosis of malignant cells [
51]. HDAC inhibitors have been used in the treatment of several hematologic tumors, such as acute myeloid leukemia and T-cell lymphomas [
52‐
62]. In the present study, 27 patients who experienced R/R after receiving anthracycline therapy were given a combination of chidamide and anthracycline-based regimen. The combination of chidamide and anthracycline increases the sensitivity of leukemia cells to anthracyclines and is able to reverse drug resistance.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.