Methods
Subjects
Consecutive 102 CLL patients untreated were enrolled between January 2004 and December 2013. All subjects were given written informed consent complying the requirements of the Declaration of Helsinki. The research was approved by the Institutional Review Boards of Nanjing Medical University. CLL diagnosis was based on the criteria specified in the National Cancer Institution [
11]. Variables ascertained in diagnosis included age, gender, Binet staging, blood absolute lymphocyte count (ALC), lactate dehydrogenase (LDH), β
2-microglobulin (β
2-MG), CD38, ZAP-70, mutation status of IGHV gene,
TP53 mutation (exon 4–9) status, and cytogenetics (studied by fluorescence in situ hybridization; FISH). In addition, 40 age-matched healthy controls (HC) were recruited between January 2012 and December 2013 from the center of Health Examination.
Primary cell cultures
CLL cells from 11 subjects were isolated from venous blood by density-gradient centrifugation. > 90% of cells were B-cells, as determined by staining for CD5+CD19+ coexpression (Becton Dickinson, USA), were cultured in RPMI-1640 media. Ten percent fetal calf serum (Gibco, USA) was added to the media. CLL cells were cultured at 37 °C with 5% CO2 in humidified atmosphere.
NDRG2 mRNA detection
Whole RNAs was extracted using Trizol reagent (Invitrogen). RNA was reversely transcribed using random hexamers, and amplified with fluorescent dye SYBR Green MasterMix and qRT-PCR specific reverse primers (Additional file
1). β-actin was used as an internal control. Reaction conditions for
NDRG2 and β-actin were one cycle at 95 °C for 5 min followed by 35 cycles at 95 °C for 30 s, 60 °C for 30 s and 72 °C for 30 s. A final extention was run at 72 °C for 5 min. Relative expression was analyzed using the comparative cycle threshold (Ct) method. ΔCt was calculated by subtracting the Ct of β-actin from those for the gene of interest. Relative amounts of the interest gene were calculated using eq. 2
-Δ Ct. Reactions for qRT-PCR were performed in triplicate, using the Applied Biosystems ABI 7500 Real-time PCR system (Applied Biosystems, USA). Sequences of amplifed production were confirmed via DNA sequencing.
Dual luciferase reporter assay
NDRG2-related miRNAs were initially identified in bioinformatics miRNA databases such as target scan (
http:/www.targetscan.org/) and miR-Base (
http://www.mirbase.org/). Dual luciferase reporter assay (Promega, USA) was used to evaluate whether the conserved miRNAs with higher scores binds directly to
NDRG2. Renilla-luciferase assay was performed with a modified expression pEZX vector containing the complete 3′ untranslated regions (UTR) region of
NDRG2 cloned in the 3’UTR region of dual luciferase gene. The miRNAs or negative controls (NC) were transfected in the HEK293T cell line together with pEZX vector using lipofectamine 2000 for dual-luciferase assay. HEK293T cell line was obtained from ATCC (American Type Culture Collection, Manassas, VA, USA, ATCC@-ACS 4500) in 2012. The cell line has been authenticated by using Single Tandem Repeat (STR) profiling method and there is no mycoplasma contamination. All cells were transfected for 24 h and assayed using a Luciferase Assay Kit (Promega, USA).
Transient transfection
CLL cells were transferred in 6-well plates by density of 5.0 × 106 cells/well and transiently transfected with 100 nM of mature miRNA inhibitors of miR-28-5p and miR-650 and 100 nM as random negative-control miRNA (miR-NC) (GenePharma Company, China) using Lipofectamine 2000 Transfection Reagent (Invitrogen, USA) according to the manufacturer’s protocol.
Western blotting
Cells were harvested with lysis buffer 24 h after transfection with miRNA inhibitors (including miR-28-5p, miR-650 and negative control miRNA inhibitors). Protein concentration was calculated using BCA (Beyotime, China). Total protein was separated using 10% sodium dodecyl sulfate–polyacrylamide gel, transferred to PVDF membranes and incubated with goat monoclonal antibody against NDRG2 (Santa Cruz Biotechnology, USA) in 1:200 dilutions. The secondary antibody was rabbit anti-goat IgG (Santa Cruz Biotechnology, USA) in 1:2500 dilutions. The normalized control used was GAPDH.
Apoptosis assay
The apoptotic ratio of CLL cells was detected using Annexin V/propidium iodide (PI) flow cytometric assay (Becton, Dickinson and Company). CLL cells after 24 h-transfection were washed twice with cold phosphate-buffered saline (PBS), and then re-suspended in 500 μL binding buffer (Bestbio, Shanghai, China). Annexin V and PI were added to the transferred cells and the plate was incubated in the dark for 15 min at room temperature. Apoptosis cells analysis was performed using a FACSCalibur flow cytometer and CellQuest software (BD Biosciences, USA).
Statistical analyses
Statistical analyses were performed using SPSS software for Windows (version 20.0). The difference of target gene mRNA expression between groups with different prognostic factors was described using the Mann–Whitney U test. The difference of miRNA and NDRG2 expression and apoptosis rate between groups was calculated using the Paired-Samples t- Test. Survival was calculated as time from diagnosis until death or loss to follow-up or to May 2017. Time to first treatment (TTT) was calculated as interval from diagnosis until first CLL-specific treatment or the last follow-up. Survival and TTT were estimated by the Kaplan-Meier method and results were compared using the log-rank test. Prognostic influence of variables was tested using the Cox proportional hazards model in univariate and multivariate analyses. Protein bands of Western blot were quantified through the Image J program for Windows after normalizing the data for GAPDH. 2-sided P-values < 0.05 were considered statistically significant.
Discussion
NDRG2, located at chromosome 14q11.2, is an important member of the NDRG family and recognized as tumor suppressor. Series of studies have been done to explore the expression and the clinical significance of
NDRG2 in cancers including hematological malignancies [
5,
6,
9]. Nakahata S et al. [
6] found that
NDRG2 expression was significantly reduced in adult T-cell leukemia lymphoma (ATLL) cell lines and primary acute-type ATLL samples, and downregulation of
NDRG2 can activate the PI3K-AKT signaling pathway. Lchikawa T et al. [
9] reported that loss of
NDRG2 also enhanced activation of the NF-κB pathway in ATLL. In current work, we tentatively investigated the expression level of
NDRG2 and its relation with prognostic factors of CLL and found marked decrease of
NDRG2 mRNA expression in CLL patients compared to HC. The observations showed that
NDRG2 was significantly downregulated in CLL patients with Binet B/C, high LDH level, IGHV un-mutated and p53 aberrations. TP53 mutation and p53 deletion, which are the powerful prognostic factors in CLL, were defined as “p53 aberrations” in this study. The result revealed that patients with lower level of
NDRG2 presented aggressive characteristics and
NDRG2 may play a critical role in the pathogenesis and development of CLL. Importantly, we found a strong correlation between
NDRG2 expression and TTT as well as OS. Patients with a lower expression level of
NDRG2 had a significantly shorter TTT and inferior OS than those with a higher
NDRG2 expression by univariate analysis. Multivariate analysis further demonstrated that
NDRG2 mRNA was prognostic value for TTT and OS independent of IGHV mutation status as well as p53 abnormalities. These findings proved that
NDRG2 expression should be a new prognostic factor for CLL. FISH analysis failed to detect del(11q22.3), del(13q14) and + 12 in part of CLL patients, for which those factors were not included in our survival analysis.
Based on the above results, we further investigated the molecular mechanisms underlying
NDRG2 regulatory pathways. MiRNAs are a family of approximate 22 nucleotides small and single-stranded non-coding RNAs that negatively regulate gene expression [
12,
13]. In recent years, there is growing attention to the role of miRNAs in cancers, and abnormal miRNAs expression is extensively reported in hematologic neoplasms [
14‐
17]. MiRNAs affect the stability of targeted oncogenes or tumor suppressors, thus leading to the impact on cellular physiology in certain malignancies. However, limited studies have ever examined miRNAs targeting
NDRG2 in CLL. Therefore, we investigated miRNAs targeting
NDRG2 and evaluate the effect on CLL cell apoptosis. Using computational analyses, four conserved miRNAs were identified as
NDRG2-related miRNAs. The dual-luciferase assay confirmed that the mimics of miR-28-5p and miR-650 obviously suppressed the activity of
NDRG2. In order to understand whether the
NDRG2 is a direct target of miR-28-5p and miR-650 in CLL, qRT-PCR was performed to determine the expression of
NDRG2, miR-28-5p and miR-650 in 30 CLL patients and 10 HC. Interestingly, we found that the expression level of miR-28-5p and miR-650 were significantly increased in CLL patients compared to HC, and were negatively associated with
NDRG2. Furthermore, we examined the effects of antisense oligonucleotides targeting miR-28-5p and miR-650 on the primary CLL cells. We investigate transfection efficiency and found that the expression of miR-28-5p and miR-650 were significantly under-expressed after transfection with those miRNA inhibitors, yet the levels of
NDRG2 mRNA and proteins were up-regulated in transfected cells with miR-28-5p and miR-650 inhibitors compared with the NC. This observation supplied strong evidence that
NDRG2 expression can be regulated by miR-28-5p and miR-650 in CLL. In other words, miR-28-5p and miR-650 can function as oncogene negatively regulating the expression of tumor suppressor in CLL.
Previous studies indicated significantly increased miR-650 expression and its association with the progression in gastric cancer, colorectal cancer, hepatocellular cancer and glioma [
10,
18‐
20]. Further, Feng et al. [
10] described regulation of
NDRG2 by miR-650 in human colorectal cancer cells, which is consistent with our findings. However, Mraz M et al. [
21] found that CLL patients with higher expression of miR-650 had significantly longer OS and TTT as well as transfection with miR-650 resulted in a reduction in the proliferative capacity of B cells, which partly argued against our results. This disparity can be associated with the diverse biological characteristics of CLL in Chinese population from those of European origins. Almeida et al. [
22] reported that miR-28-5p was downregulated in colorectal cancer cells compared with normal colon samples, and that overexpression of miR-28-5p reduced colorectal cancer cell proliferation, migration and invasion in vitro. This discrepancy from ours may be associated with miR-28-5p exerting different functions in diverse types of tumors. In addition, it is worth mentioning that the level of
NDRG2 mRNA and protein were both up-regulated in transfected cells with miR-28-5p and miR-650 inhibitors, which is consistent with our previous work that demonstrated up-regulated levels of PTEN mRNA and proteins in transfected cells with miR-26a and miR-214 inhibitors as compared with the controls [
23]. It is generally accepted that most animal miRNAs exert their regulatory effects through incomplete matching with 3′-untranslated region (3’-UTR) of their mRNA targets, repress target-gene expression at the level of translation, reducing the protein levels of their target genes, yet the mRNA levels of these genes are barely affected [
24]. Nevertheless, some findings indicate that miRNAs share only partial complementarity with their targets, and can also induce mRNA degradation and reduce the mRNA levels of their target genes in mammals [
25,
26], which may explain the results in current study.
We observed the knockdown of
NDRG2 with miR-28-5p and miR-650 inhibitors inducing CLL cell apoptosis, yet found no increased apoptosis rates in patients with p53 aberrations following transfection with the above miRNAs inhibitors. Contrarily, significant apoptosis increase was seen in patients without p53 aberrations after transfection. Previous studies showed that overexpression of
NDRG2 markedly promoted tumor cell apoptosis in renal cell carcinoma [
27], esophageal carcinoma [
28] and breast cancer [
29]. Although we found the similar results, yet overexpression of
NDRG2 promoting CLL cell apoptosis was only seen in patients without p53 aberrations. Liu et al. studied the association of
NDRG2 with p53, and reported that
NDRG2 was p53-inducible target gene that is transactivated by p53 and is required for the full p53-mediated apoptotic response [
7]. Cao et al. also found that adenoviruses carrying
NDRG2 enhanced p53-mediated apoptosis of hepatocarcinoma cells [
30]. In our study, we found that
NDRG2 expression was significantly reduced in CLL patients with p53 aberrations, yet increased by transfecting with related miRNAs-inhibitors promoting CLL cells apoptosis without p53 aberrations. However, association between
NDRG2 and p53 pathway in CLL needs verification by expanding population samples due to limited cases included in current work.