Abstract

Long non-coding RNAs (lncRNAs) play important roles in hematological malignancies. We have previously identified several differentially expressed lncRNAs in myelodysplastic syndromes (MDS) by microarray analysis. In the present study, we explored the regulatory circuitry, potential functions, clinical and prognostic relevance of these lncRNAs in MDS by developing a lncRNA regulation network. We identified a novel lncRNA, LOC101928834, which was significantly up-regulated in the bone marrow of patients with MDS and acute myeloid leukemia (AML). We further evaluated the clinical relevance of LOC101928834 in 89 MDS and 110 AML patients and found that higher level of LOC101928834 expression was associated with higher white blood cell count, higher blast percentage, the subtype of refractory cytopenia with excess blasts (RAEB) and shorter overall survival in MDS patients. Receiver operating characteristic (ROC) curve analysis showed that LOC101928834 expression could discriminate MDS-RAEB patients from control with an area under the receiver-operating curve (AUC) of 0.9048. Moreover, functional analysis showed that LOC101928834 promoted cell proliferation and cell cycle progression, and activated Wnt/β-catenin signaling pathway in vitro. In conclusion, LOC101928834 expression is correlated with clinical and biological features of MDS and may serve as a novel diagnostic and prognostic biomarker.

Introduction

Myelodysplastic syndromes (MDS) comprise a heterogeneous group of clonal hematopoietic disorders and are the most common class of acquired bone marrow (BM) failure syndromes in adults [1,2]. Myelodysplastic syndromes are characterized by BM dysplasia with ineffective hematopoiesis, peripheral blood cytopenia, and an increased risk of transformation to acute myeloid leukemia (AML) in 30–40% patients [3–5]. The majority of MDS patients are diagnosed at a late stage with poor prognosis due to lack of effective diagnostic biomarkers [5–7]. Therefore, elucidating the molecular mechanisms of MDS progression is essential for identification of key biomarkers, as well as development of effective targeted therapies [7,8].

Although the pathogenesis of MDS remains poorly understood, it is generally accepted that genetic factors play pivotal roles in the initiation and progression of MDS [1,3]. Recently, long non-coding RNAs (lncRNAs) have been recognized as key factors in the regulation of genetic imprinting, immune response, tumorigenesis, cellular development and metabolism rather than simply ‘transcription noise’ [9]. Meanwhile, they have been suggested to participate in not only normal hematopoiesis but also the pathogenesis of multiple hematological cancers [10]. Long non-coding RNAs are a class of RNA molecules with more than 200 nucleotides in length and make up the largest portion of the mammalian non-coding transcriptome [9]. It is estimated that lncRNAs dysregulation is closely associated with a variety of human diseases including cancer, Huntington’s and Alzheimer’s diseases [11,12]. Since HOTAIR was discovered in breast cancer in 2007 [13], more and more lncRNAs have been found in different tumors, but only a few numbers of lncRNAs have been reported in MDS [7,10,11,14], including MEG3 that is the first lncRNA reported in MDS [15]. In 2010, Benetatos et al. found abnormal methylation of the MEG3 promoter region in one-third of MDS and half of AML patients [15]. In 2013, Yildirim et al. found that the penetrability of MDS/myeloproliferative neoplasms (MPN) was as high as 100% in lncRNA-Xist knockout female mice [16]. In 2017, Liu et al. performed the first whole-genome sequencing analysis of lncRNAs in MDS and found linc-ARFIP1-4, linc-TAAR9-1, lincC2orf85, linc-RNFT2-1 and linc-RPIA, etc. were differentially expressed in refractory anemia with excess blasts-2 (RAEB2) [17]. At the same time, Yao et al. analyzed the expression of lncRNAs in 176 MDS patients and found that higher levels of four lncRNAs (TC07000551.hg.1, TC08000489.hg.1, TC02004770.hg.1 and TC03000701) were correlated with shorter overall survival [7]. Although more and more differentially expressed lncRNAs have been found in MDS, the role of only a few lncRNAs in MDS pathogenesis have been explored [14]. The clinical significance of lncRNAs in MDS and whether they can be used as biomarkers for the diagnosis and prognosis of MDS remain to be investigated.

In the present study, we performed a comprehensive expression profiling analysis of lncRNAs in patients with MDS and AML and identified a novel lncRNA, LOC101928834, which may serve as a biomarker in the diagnosis and prognostic stratification of these patients. Moreover, we performed functional analysis and demonstrated that LOC101928834 could promote cell proliferation and cell cycle progression possibly through activating Wnt/β-catenin signaling pathway.

Materials and methods

Patients

In total, 129 adult patients with primary MDS, 120 adult patients with primary AML, and 57 adult patients with benign hematological diseases seen at eight hospitals in Shanghai between June 2003 and April 2007 were included as the validation cohort. The diagnoses were made based on the 2008 World Health Organization (WHO) classification guidelines. Bone marrow (BM) samples from all patients were collected at the time of diagnosis and were stored at −80°C for RNA extraction. Informed consents were obtained and the research procedure was approved by the Ethics Committee of Huashan Hospital, Fudan University.

RNA extraction and quantitative real-time PCR

Total RNA was isolated from primary BM cells using Trizol reagent (Invitrogen, Carlsbad, CA, U.S.A.) according to the manufacturer’s instructions. For mRNA and lncRNA detection, the extracted RNA was reverse transcribed into cDNAs using Takara PrimeScript RT Master Mix (Perfect Real Time; Takara Bio Inc., Otsu, Shiga, Japan). The quantitative real-time polymerase chain reaction (qRT-PCR) was carried out using SYBR® Premix Ex Taq™ (Tli RNaseH Plus; Takara Bio Inc.) on an ABI ViiA 7 Real-Time PCR System (Applied Biosystems, Foster City, CA, U.S.A.). GAPDH was used as the internal control for normalizing the expression of target genes. The primer sequences were listed in Supplemental Table S1.

LncRNA microarray

Total RNA (200 ng) was labeled with the mRNA Complete Labeling and Hyb Kit (Agilent Technologies, Santa Clara, CA, U.S.A.) and hybridized on the SurePrint G3 Human Gene Expression 8x60K Microarray (Agilent Technologies). The microarray contains 27,958 human mRNA and 7,419 human lncRNAs, which were derived from authoritative databases, including RefSeq, Ensemble, GenBank, and the Broad Institute. The microarray also contains probes for small noncoding RNAs, including small nucleolar RNAs (snoRNAs), small nuclear RNAs (snRNAs), small Cajal body-specific RNAs (scaRNAs), ribosomal RNAs (rRNAs), telomerase RNAs, and valine transfer RNA, but not microRNAs, with snoRNAs and snRNAs being the majority components. After hybridization and wash, processed slides were scanned with the Agilent G2505C microarray scanner (Agilent Technologies). Raw data were extracted using Feature Extraction (version10.7.1.1; Agilent Technologies). Quantile normalization and subsequent data processing were then performed using Genespring software (version 12.0; Agilent Technologies). The microarray profiling was conducted in the laboratory of the OE Biotechnology Company in Shanghai, China.

Construction of the lncRNA–mRNA regulation network and functional prediction

An absolute >2-fold change was used to filter differentially expressed lncRNAs and mRNAs. In the present study, we have developed a lncRNA–miRNA–target approach to predict potential targets of lncRNAs based on the mechanism that lncRNAs can be processed into short ncRNAs (e.g. miRNA) which can further silence certain gene expression by degrading the mRNA transcripts. For each differentially expressed lncRNA, miRcode database [PMID:22718787] was used to obtain potential miRNA products. Database miRbase v21, mirTarbase, miRanda 2010, miRbase v18, Tarbase5.0 and targetscan v6.2 were used to predict potential targets of miRNAs. Targets that have been annotated by less than three databases were filtered out in the first step. The remaining targets were further compared with differently expressed gene sets in the microarray experiment. Each predicted lncRNA–mRNA pair was then assigned to a different regulation group based on their expression profile: the positive regulation group was defined as that the lncRNA and its mRNA target were both over-expressed and inhibited simultaneously; the negative regulated group was defined as the opposite expression profile between lncRNA and mRNA. Moreover, chipBase [Nucleic Acids Res. 2013 Jan;41] was used to analyze the transcriptional factors of the lncRNAs. Information of protein–protein interaction was obtained via the STRING database (version 9.0) [18] by the default setting with combined score > 0.4. A text mining was excluded. The build-in one-step expanding algorithm on the STRING web-server is used to discover the potential indirect interactions. DAVID [Nature Protocols 2009; 4(1):44 & Genome Biology 2003; 4(5): P3] bioinformatic tools were used for the functional analysis. Enrichment was calculated using the default setting (EASE score < 0.1, counts ≥ 2). Enrichment of signaling pathways was based on KEGG pathway database. To evaluate the potential relationship of the differentially expressed mRNAs and lncRNAs, a network based on the regulation model was constructed with information of lncRNA expression, mRNA expression, lncRNA–miRNA–mRNA relationship, protein–protein interaction, and KEGG pathway/biological processes linked with genes. The Networks were visualized with the cytoscape 3.0 [19]. Functions of lncRNAs were predicted based on the GO enrichment analysis of their potential mRNA targets.

Cell culture and shRNA transfection

Human myeloid leukemia cell lines THP-1 and K562 were obtained from the Chinese Academy of Sciences, and SKM-1 was obtained from the Japanese Collection of Research Bioresources. All cell lines used in the present study were grown and maintained in RPMI 1640 (Hyclone; GE Healthcare www.impactjournals.com/oncotarget 36528 Oncotarget Life Sciences, Logan, UT, U.S.A.) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientic, Inc., Waltham, MA, U.S.A.), at 37°C with 5% CO2.

For LOC101928834 silencing, K562 and THP-1 cell lines were transfected with three LOC101928834-shRNA lentiviruses and a negative control shRNA lentiviruse, according to the manufacturer’s instructions (GeneChem, Shanghai, China). All shRNAs were purchased from GeneChem Co. Ltd. The shRNA sequences were listed in Supplementary Table S2. Stably transfected cell lines were selected by puromycin, and the expression level of LOC101928834 was confirmed by qRT-PCR.

Cell proliferation assay

Cell proliferation was measured using the Cell Counting Kit-8 assay (Dojindo Molecular Technologies, Gaithersburg, MD, U.S.A.) according to the manufacturer’s instructions. Briefly, aliquots (200 μl) of the cell suspension were dispensed into 96-well plates and then placed in a humidified incubator for 24, 48, 72, 96 or 120 h at 37°C with 5% CO2. Next, 10 μl of CCK-8 reagent was added and incubated for 4 h and then the plates were further incubated until wells with the maximum absorbance at 450 nm reached values of approximately 1 optical density (OD). Cell viability was expressed as the percentage of the experimental to the control value.

Flow cytometric analysis

An annexin V apoptosis detection kit (BD Biosciences, Franklin Lakes, NJ, U.S.A.) was used for measuring cell apoptosis. Approximately 1 × 105 cells were harvested, washed with cold phosphate-buffered saline (PBS), and then stained with Annexin V-FITC/7-aminoactinomycin D (7-AAD) solution according to the manufacturer’s instructions. After incubation, flow cytometry analysis was performed immediately using a BD FACS Calibur flow cytometer with FCS Express 3.0 software.

For cell cycle analysis, cells were harvested, washed twice with ice-cold PBS, and fixed in 70% ethanol at −20°C overnight. Prior to the analysis, the fixed cells were washed twice with ice-cold PBS and resuspended in 50 μg/ml PI in the dark at 4°C for 30 min. Cell-cycle distribution was analyzed using a BD FACS Calibur flow cytometer with FCS Express 3.0 software.

Western blot analysis

Total proteins were isolated and extracted using RIPA buffer supplemented with 1% phenylmethylsulfonyl fluoride (PMSF). Proteins were separated on 10% SDS-PAGE, transferred to polyvinylidene fluoride membranes (Bio-Rad Laboratories, Inc., Hercules, CA, U.S.A.), then blotted with primary antibodies (anti-cyclin D1, anti-cyclin D3, anti-TCF-1, anti-β-catenin, anti-FRAT2, and anti-GAPDH, all from Cell Signaling Technology, Danvers, MA, U.S.A.), followed by HRP-labeled anti-rabbit secondary antibody. The protein bands were detected by enhanced chemiluminescence system (Beyotime Institute of Biotechnology, Shanghai, China). GAPDH was used as the internal control.

Statistical analysis

The statistical analyses were performed by Stata version 14.0 software (StataCorp LP, College Station, TX, U.S.A.) and GraphPad Prism version 7.00 (GraphPad Software, La Jolla, CA, U.S.A.). Kruskal–Wallis test, t test, and one-way ANOVA were conducted to compare lncRNAs and mRNAs expression in clinical samples and cell lines. The chi-square test or the Fisher exact test were used to compare the difference of categorical variables and the Mann–Whitney U test was used to compare continuous variables between patients with lower and higher lncRNA expression groups. Receiver-operating characteristic (ROC) curve analyses and the area under the ROC curve (AUC) were constructed to evaluate the diagnostic value. The survival analysis was plotted using Kaplan–Meier curves and the statistical significance was obtained using log-rank test. The Cox proportional hazards model was used for multivariate analysis and hazard ratios (HR) and 95% confidence intervals (CI) were calculated. The level of statistical significance was set at P<0.05.

Results

lncRNAs expression profile in MDS

By analyzing the genome-wide mRNA and lncRNA expression profile, we have previously identified that, compared with the control group, MDS group had 2705 differentially expressed genes (DEGs), 515 methylated genes and 543 differentially expressed lncRNAs that achieved statistically significance [8,20]. The top 20 up-regulated and top 20 down-regulated lncRNAs were shown in Figure 1A. To validate the lncRNAs of microarray analysis, we selected 10 lncRNAs from the list and analyzed their expression by qRT-PCR in 10 controls, 40 MDS and 10 AML samples. The results showed that the levels of 8 lncRNAs were consistent with the microarray data (Figure 1B). To further explore the roles of these differentially expressed lncRNAs, we then built a lncRNA regulatory network by integrating a large number of hypermethylated genes, DEGs, miRNAs and transcription factors (TFs), which are connected with these differentially expressed lncRNAs and identified 214 lncRNAs, 72 miRNAs and 1746 mRNAs in this network. Among the network, 98 TFs were found to correspond to 28 lncRNAs by the threshold of P<0.01 and FDR<0.01 (Supplementary Table S3). About 1384 target genes were predicted for 72 miRNAs, which were all included in the DEG. Through GO and KEGG analysis by inputing the protein-coding genes in the lncRNAs-regulation network, we obtained the functional framework module of the network (Figure 2), which can evaluate the MDS-specific lncRNAs biological regulatory functions. We also found that the top 10 molecular functions of this lncRNA-centric regulatory network included nucleotide binding, purine nucleotide binding, ribonucleotide binding, purine ribonucleotide binding, nucleoside binding, transcription regulator activity, purine nucleoside binding, adenyl nucleotide binding, adenyl ribonucleotide binding and ATP binding. The top 10 cellular components included plasma membrane part, membrane-enclosed lumen, organelle lumen, intracellular organelle lumen, nuclear lumen, cytosol, intrinsic to plasma membrane, integral to plasma membrane, cell fraction, and mitochondrion. The mostly enriched KEGG pathways included: pathways in cancers, salivary secretion, small cell lung cancer, bile secretion, VEGF signaling pathway, transcriptional misregulation in cancer, T-cell receptor signaling pathway, bladder cancer, insulin signaling pathway and mTOR signaling pathway. These results were consistent with our previous analysis of all DEGs [8], indicating that differentially expressed lncRNAs play an important role in MDS pathogenesis.

LncRNAs expression level of microarray

Figure 1
LncRNAs expression level of microarray

(A) The top 20 up-regulated and top 20 down-regulated lncRNAs in our microarray. (B) The qRT-PCR results of the 10 chosen lncRNAs to validate those of the microarray. (*:P<0.05; **:P<0.01; ***:P<0.001; ****:P<0.0001)

Figure 1
LncRNAs expression level of microarray

(A) The top 20 up-regulated and top 20 down-regulated lncRNAs in our microarray. (B) The qRT-PCR results of the 10 chosen lncRNAs to validate those of the microarray. (*:P<0.05; **:P<0.01; ***:P<0.001; ****:P<0.0001)

The functional framework module of the lncRNA regulation network

Figure 2
The functional framework module of the lncRNA regulation network

It included all of biological process of the GO enrichment results of the network.

Figure 2
The functional framework module of the lncRNA regulation network

It included all of biological process of the GO enrichment results of the network.

Wnt/β-catenin signaling pathway in the lncRNAs-regulation network

In our previous study, we found that methylation of Wnt antagonist genes, such as SFRP1, SFRP4 and SFRP5, were correlated with poor prognosis and were predictive of shorter survival or higher risk of leukemia evolution in MDS [21]. Thus, we constructed a sub-regulation-network in Wnt pathway using these elements, such as mRNAs, lncRNAs and miRNAs that are associated with Wnt/β-catenin signaling pathway in the lncRNAs-regulation network. We selected 18 out of 1937 differentially expressed genes associated with Wnt pathway based on KEGG database and our previous study. We constructed the protein–protein interaction network of these 18 Wnt related genes using String database and annotated each gene with KEGG pathway (Figure 3A). In addition, we identified 5 novel lncRNAs with significant correlation with some of these 18 genes, suggesting a role in Wnt pathway (Table 1 and Figure 3B). Therefore, lncRNAs were also added to the network. To identify candidate lncRNA and predict its regulated transcript for further investigation, we extracted the lncRNA sequence by the probe information and blast against any transcript sequence in Ensembl database. Next, we performed qRT-PCR to analyze the expression of these 5 lncRNAs in the BM of 15 control, 40 MDS and 10 AML and found that all of the 5 lncRNAs were new lncRNAs (Figure 3C). The expression of most lncRNAs was consistent with the microarray analysis except LOC105379175.

The Sub-Regulation-Network in Wnt pathway

Figure 3
The Sub-Regulation-Network in Wnt pathway

(A) The protein–protein interaction network of the 18 Wnt related genes using String database and annotate each gene with KEGG pathway. The red notes were the points which directly connected with Wnt pathway. (B) The Sub-Regulation-Network in Wnt pathway visualized with the cytoscape 3.0. The green notes, down-regulated mRNAs; The red notes, up-regulated mRNAs; The triangle, lncRNAs; The rectangle, KEGG pathway. (C) The qRT-PCR results of the 5 novel lncRNAs in the BM of 15 control, 40 MDS and 10 AML. (*:P<0.05; **:P<0.01; ***:P<0.001; ****:P<0.0001). (D) The qRT-PCR results of overlapping sequence between the transcripts and transcript specific sequence of LOC101928834 in the BM of 10 control (0021A, 0024A, 0070, 0074A, 0105, 0107, 0110, 0112, 0115, 0116), 8 AML (A1, A2, A3, A5, A6, A7, A8, A9) and 13 MDS (M1-M13).

Figure 3
The Sub-Regulation-Network in Wnt pathway

(A) The protein–protein interaction network of the 18 Wnt related genes using String database and annotate each gene with KEGG pathway. The red notes were the points which directly connected with Wnt pathway. (B) The Sub-Regulation-Network in Wnt pathway visualized with the cytoscape 3.0. The green notes, down-regulated mRNAs; The red notes, up-regulated mRNAs; The triangle, lncRNAs; The rectangle, KEGG pathway. (C) The qRT-PCR results of the 5 novel lncRNAs in the BM of 15 control, 40 MDS and 10 AML. (*:P<0.05; **:P<0.01; ***:P<0.001; ****:P<0.0001). (D) The qRT-PCR results of overlapping sequence between the transcripts and transcript specific sequence of LOC101928834 in the BM of 10 control (0021A, 0024A, 0070, 0074A, 0105, 0107, 0110, 0112, 0115, 0116), 8 AML (A1, A2, A3, A5, A6, A7, A8, A9) and 13 MDS (M1-M13).

Table 1
Five novel lncRNAs included in the sub-regulation-network in Wnt pathway
lncRNA descriptionPFC (abs)Regulation
lincRNA: chr4:89637527-89646627 forward strand 0.002411715 3.227302 Down 
lincRNA: chr5:114534926-114542826 reverse strand 0.002432028 3.3346636 Down 
lincRNA: chr5:131798076-131807951 forward strand 0.005231156 2.3785024 Down 
lincRNA: chr19:28269547-28283939 reverse strand 0.007668024 2.90322 Down 
lincRNA: chr10:19930169-20099994 reverse strand 0.006807031 6.4264836 UP 
lncRNA descriptionPFC (abs)Regulation
lincRNA: chr4:89637527-89646627 forward strand 0.002411715 3.227302 Down 
lincRNA: chr5:114534926-114542826 reverse strand 0.002432028 3.3346636 Down 
lincRNA: chr5:131798076-131807951 forward strand 0.005231156 2.3785024 Down 
lincRNA: chr19:28269547-28283939 reverse strand 0.007668024 2.90322 Down 
lincRNA: chr10:19930169-20099994 reverse strand 0.006807031 6.4264836 UP 

Identification of a novel lncRNA LOC101928834

Interestingly, LincRNA: chr10:19930169-20099994 reverse strand was found to match three transcripts, and two of them, ENST00000431157.6 and ENST00000423551.1, could be detected by our oligoprobe. To validate the prediction, we designed specific PCR primers using overlapping sequence between the transcripts and transcript specific sequences to assess the transcripts in patient samples. We found that the actual differentially expressed transcript was ENST00000431157.6, whereas there was no difference in the expression of ENST00000423551.1 between MDS group and control group (Figure 3D). ENST00000431157.6 (Gene ID: LOC101928834) is located at Chromosome 10: 19,721,588-19,728,550 reverse strand and has 4 exons and no coding exon with a transcript length of 932 bps. By searching for its sequence in LNCipedia (https://lncipedia.org), we found that the PhyloCSF21 score of LOC101928834 was -15.4992, and the CPAT22 coding probability was 14.77%. The analysis result of PRIDE reprocessing 2.023 was zero. Additionally, LNCipedia failed to identify any Bazzini small open reading frames24 or Lee translation initiation sites25 in LOC101928834. In GeneCards database, we found that it was expressed as an enhancer in various cells, such as astrocytes, common myeloid progenitor CD34+ cells, fibroblasts of dermis, skeletal muscle myoblasts, muscle leg (fetal), normal human lung fibroblasts, osteoblasts, SK-N-SH (a human neuroblastoma cell line) and muscle trunk (fetal). Taken together, these data suggest that LOC101928834 is a novel lncRNA.

LOC101928834 is up-regulated in the bone marrow of MDS patients

We expanded the sample size and further verified LOC101928834 expression in 42 cases of control, 89 cases of MDS (64 cases of refractory cytopenia with multilineage dysplasia [RCMD], 25 cases of refractory anemia with excess blasts [RAEB]) and 110 cases of AML. The expression level of LOC101928834 was increased in a stepwise manner from control, to RCMD, to RAEB, and to AML groups (Figure 4A and Table 2). To determine the clinical significance of increased expression of LOC101928834, we divided the MDS patients into two groups based on the cutoff RQ value of LOC101928834 determined by maximally selected log-rank statistics, which was found to be 1.33 (Figure 4B and Table 3). We found that higher LOC101928834 expression was more frequently seen in patients with higher white blood cell (WBC) count, higher blast percentage and the MDS-RAEB subtype, whereas there was no significant correlation between LOC101928834 expression and age, gender, hemoglobin level, absolute neutrophil count or karyotype. We further assessed whether LOC101928834 could be a biomarker for the diagnosis of MDS using receiver operating characteristic (ROC) curve and found that using the area under the curve (AUC) value of 0.9048 (95% confidence interval [CI], 0.80371–1.00000), this novel lncRNA could serve as a useful diagnostic marker for MDS-RAEB (Figure 4C). To explore the prognostic value of LOC101928834 in MDS and AML, we investigated the expression of LOC101928834 with survival using Kaplan–Meier survival analysis (Table 4). We found that patients with higher level of LOC101928834 expression had a significantly poorer prognosis compared with those with lower level of LOC101928834 expression in the MDS group (P=0.0019, log-rank test; Figure 4D), but not in AML patients. Multivariate analyses revealed that the level of LOC101928834 expression is an independent prognostic factor for overall survival in MDS patients (HR: 1.09; 95%; CI: 1.01–1.19; P=0.038). These data suggest that LOC101928834 has an important role in MDS pathogenesis.

The expression of LOC101928834 in Patients

Figure 4
The expression of LOC101928834 in Patients

(A) The qRT-PCR result of LOC101928834 in the BM of 42 control, 64 of MDS-RCMD, 25 of MDS-RAEB, and 110 of AML. (*:P<0.05; ****:P<0.0001). (B) The cutoff value for baseline RQ value of LOC101928834 was 1.33 for the 89 MDS patients. (C) Receiver-operating characteristic (ROC) curve analysis of LOC101928834 signature to discriminate MDS-RAEB patients from controls. (D) Kaplan–Meier survival curve for association of LOC101928834 with survival.

Figure 4
The expression of LOC101928834 in Patients

(A) The qRT-PCR result of LOC101928834 in the BM of 42 control, 64 of MDS-RCMD, 25 of MDS-RAEB, and 110 of AML. (*:P<0.05; ****:P<0.0001). (B) The cutoff value for baseline RQ value of LOC101928834 was 1.33 for the 89 MDS patients. (C) Receiver-operating characteristic (ROC) curve analysis of LOC101928834 signature to discriminate MDS-RAEB patients from controls. (D) Kaplan–Meier survival curve for association of LOC101928834 with survival.

Table 2
Relative expression level of LOC101928834 of different patient groups
VariablesNo. of patientsRelative lncRNA level (95% CI)P-value
Control 42 0.9240 (0.7213–1.127)  
MDS 89 3.366 (2.162–4.570) 0.6154 
RCMD 64 2.427 (1.003–3.851) 0.0369 
RAEB 25 5.770 (3.665–7.874) < 0.0001 
AML 110 7.877 (5.193–10.56) 0.0368 
VariablesNo. of patientsRelative lncRNA level (95% CI)P-value
Control 42 0.9240 (0.7213–1.127)  
MDS 89 3.366 (2.162–4.570) 0.6154 
RCMD 64 2.427 (1.003–3.851) 0.0369 
RAEB 25 5.770 (3.665–7.874) < 0.0001 
AML 110 7.877 (5.193–10.56) 0.0368 
Table 3
Clinical characteristics of 78 MDS patients according to LOC101928834 expression
LOC101928834 normal and lowLOC101928834 highP-value
Age at diagnosis     
  ≥60 Num. (%) 24 (50.0%) 17 (%) 0.566 
  <60  24 (50.0%) 13 (%)  
Gender*     
  Male  26 (54.2%) 20 (66.7%) 0.275 
  Female  22 (45.8%) 10 (33.3%)  
Hgb     
  <10 g/dl Num. (%) 43 (89.6%) 26 (92.9%) 0.695 
  ≥10 g/dl  5 (10.4%) 4 (7.1%)  
WBC     
  <4 × 109L Num. (%) 44 (91.7%) 21 (70%) 0.012 
  ≥4 × 109L  4 (8.3%) 9 (30%)  
ANC     
  <1.8 × 109 L Num. (%) 39 (81.3%) 24 (80%) 0.892 
  ≥1.8 × 109 L  9 (18.7%) 6 (20%)  
Platelet     
  <100 × 109 L Num. (%) 35 (72.9%) 22 (73.3%) 0.968 
  ≥100 × 109 L  13 (27.1%) 8 (26.7%)  
BLAST%     
  <5 Num. (%) 43 (89.6%) 10 (33.3%) 0.000 
  5–10  3 (6.3%) 11 (36.7%)  
  >10  2 (4.1%) 9 (30%)  
Type of MDS Num. (%)    
  RCMD  45 (93.8%) 8 (26.7%) 0.000 
  RAEB  3 (6.2%) 22 (73.3%)  
IPSS Karyotype Num. (%)    
  Good  38 (80.9%) 23 (79.4%) 0.523 
  Intermediate  7 (14.9%) 3 (10.3%)  
  Poor  2 (4.2%) 3 (10.3%)  
LOC101928834 normal and lowLOC101928834 highP-value
Age at diagnosis     
  ≥60 Num. (%) 24 (50.0%) 17 (%) 0.566 
  <60  24 (50.0%) 13 (%)  
Gender*     
  Male  26 (54.2%) 20 (66.7%) 0.275 
  Female  22 (45.8%) 10 (33.3%)  
Hgb     
  <10 g/dl Num. (%) 43 (89.6%) 26 (92.9%) 0.695 
  ≥10 g/dl  5 (10.4%) 4 (7.1%)  
WBC     
  <4 × 109L Num. (%) 44 (91.7%) 21 (70%) 0.012 
  ≥4 × 109L  4 (8.3%) 9 (30%)  
ANC     
  <1.8 × 109 L Num. (%) 39 (81.3%) 24 (80%) 0.892 
  ≥1.8 × 109 L  9 (18.7%) 6 (20%)  
Platelet     
  <100 × 109 L Num. (%) 35 (72.9%) 22 (73.3%) 0.968 
  ≥100 × 109 L  13 (27.1%) 8 (26.7%)  
BLAST%     
  <5 Num. (%) 43 (89.6%) 10 (33.3%) 0.000 
  5–10  3 (6.3%) 11 (36.7%)  
  >10  2 (4.1%) 9 (30%)  
Type of MDS Num. (%)    
  RCMD  45 (93.8%) 8 (26.7%) 0.000 
  RAEB  3 (6.2%) 22 (73.3%)  
IPSS Karyotype Num. (%)    
  Good  38 (80.9%) 23 (79.4%) 0.523 
  Intermediate  7 (14.9%) 3 (10.3%)  
  Poor  2 (4.2%) 3 (10.3%)  
Table 4
Outcome of 78 MDS patients according to LOC101928834 expression
VariablesNo. of patientsOverall survival (OS)
No. of deathP-value
LncRNA level    
High (>1.33) 30 22 0.0019 
Normal and low 48 17  
(≤1.33)    
Age    
<60 37 16 0.0674 
≥60 41 23  
Gender    
Male 46 26 0.1663 
Female 32 13  
WBC(×109 L−1)    
<4 65 34 0.3498 
≥4 13  
ANC(×109 L−1)    
≥1.8 15 0.0572 
<1.8 63 35  
Hemoglobin (g/dl)    
≥10 0.4316 
<10 69 36  
PLT (×109 L−1)    
≥100 21 0.0497 
<100 57 32  
BM blast (%)    
<5 53 18 0.0004 
5–10 16  
11–19 12  
IPSS karyotype    
Good 61 32 0.0412 
Intermediate 10  
Poor  
VariablesNo. of patientsOverall survival (OS)
No. of deathP-value
LncRNA level    
High (>1.33) 30 22 0.0019 
Normal and low 48 17  
(≤1.33)    
Age    
<60 37 16 0.0674 
≥60 41 23  
Gender    
Male 46 26 0.1663 
Female 32 13  
WBC(×109 L−1)    
<4 65 34 0.3498 
≥4 13  
ANC(×109 L−1)    
≥1.8 15 0.0572 
<1.8 63 35  
Hemoglobin (g/dl)    
≥10 0.4316 
<10 69 36  
PLT (×109 L−1)    
≥100 21 0.0497 
<100 57 32  
BM blast (%)    
<5 53 18 0.0004 
5–10 16  
11–19 12  
IPSS karyotype    
Good 61 32 0.0412 
Intermediate 10  
Poor  

LCO101928834 regulates cell proliferation and cell cycles in K562 and THP-1 cell lines

To evaluate the effects of LOC101928834 on cell biological functions, we constructed K562 and THP-1 cell lines with down-regulation of LOC01928834. We performed cell counting kit-8 assays that showed that knockdown of LOC101928834 in K562 and THP-1 cells reduced cell proliferation (Figure 5A), indicating that LOC101928834 promotes cell proliferation in vitro. Next, we examined whether LOC101928834 accelerated cell proliferation by promoting cell cycle using fluorescence-activated cell sorting (FACS) analysis of propidium-iodide–stained cells, and found that LOC101928834 knockdown caused a significant increase in the percentage of cells in the G2 phase (Figure 5B,D). However, LOC101928834 down-regulation had no significant effect on cell apoptosis in vitro (Figure 5C). In keeping with these findings, down-regulation of LOC101928834 resulted in a reduction of the expression of known proliferation-related gene Ki67 and cell cycle regulation-related genes CCND1 and CCND3, without affecting the expression of apoptosis-related genes, including CASP3, CASP8 and CASP9 (Figure 5E). In addition, the cyclin D1 and cyclin D3 proteins were down-regulated in K562 cells after down-regulation of LOC101928834 (Figure 6C). These results indicate that LOC101928834 can promote cell cycle progression and accelerate cell proliferation in vitro.

LOC101928834 regulates the cell proliferation and cell cycles in K562 and THP-1 cell lines

Figure 5
LOC101928834 regulates the cell proliferation and cell cycles in K562 and THP-1 cell lines

(A) The cell counting kit-8 assays indicated that cell proliferations were reduced by the knockdown of LOC101928834 in K562 and THP-1 cells. (B and D) Knockdown of LOC101928834 in K562 and THP-1 cells increased the cell numbers in the G2 phase as analyzed using flow cytometry (**, P<0.01). (C) LOC101928834 down-regulation had no significant effect on cell apoptosis as analyzed using flow cytometry. (E) Down expression of LOC101928834 resulted in reduction of proliferation related gene Ki67 and cell cycle regulation related gene CCND1 and CCND3. Apoptotic genes, including CASP3, CASP8 and CASP9 had no significant change (*, P<0.05).

Figure 5
LOC101928834 regulates the cell proliferation and cell cycles in K562 and THP-1 cell lines

(A) The cell counting kit-8 assays indicated that cell proliferations were reduced by the knockdown of LOC101928834 in K562 and THP-1 cells. (B and D) Knockdown of LOC101928834 in K562 and THP-1 cells increased the cell numbers in the G2 phase as analyzed using flow cytometry (**, P<0.01). (C) LOC101928834 down-regulation had no significant effect on cell apoptosis as analyzed using flow cytometry. (E) Down expression of LOC101928834 resulted in reduction of proliferation related gene Ki67 and cell cycle regulation related gene CCND1 and CCND3. Apoptotic genes, including CASP3, CASP8 and CASP9 had no significant change (*, P<0.05).

The changes of expression level of genes in the Sub-Regulation-Network in Wnt pathway in LOC101928834-down-regulated K562 cell

Figure 6
The changes of expression level of genes in the Sub-Regulation-Network in Wnt pathway in LOC101928834-down-regulated K562 cell

(A) Seven of the expression levels of genes which had direct connection in the sub-regulation-network, and mir-9, which was predicted to be a downstream target of LOC101928834 in the network, were decreased. (B) The changes of expression levels of some typical components and target genes of Wnt/ β-catenin signaling pathway. (*:P<0.05; **:P<0.01; ***:P<0.001; ****:P<0.0001). (C) The changes in the protein levels of FRAT2, β-catenin, TCF-1, cyclin D1 and cyclin D3.

Figure 6
The changes of expression level of genes in the Sub-Regulation-Network in Wnt pathway in LOC101928834-down-regulated K562 cell

(A) Seven of the expression levels of genes which had direct connection in the sub-regulation-network, and mir-9, which was predicted to be a downstream target of LOC101928834 in the network, were decreased. (B) The changes of expression levels of some typical components and target genes of Wnt/ β-catenin signaling pathway. (*:P<0.05; **:P<0.01; ***:P<0.001; ****:P<0.0001). (C) The changes in the protein levels of FRAT2, β-catenin, TCF-1, cyclin D1 and cyclin D3.

LOC101928834 regulates the Wnt/β-catenin signaling pathway

According to the results of the lncRNA-regulation network and pathway analysis, LOC01928834 is associated with the Wnt/β-catenin signaling pathway. Thus, we focused on the alteration of genes in the sub-regulation-network in Wnt pathway to further explore the molecular mechanisms of LOC101928834. First, we analyzed mRNA expression of 12 genes that had direct connection to LCO101928834 in our prediction model. The expression levels of seven of these genes and mir-9, which was predicted to be a downstream target of LOC101928834 in our network, were decreased in K562 cells after down-regulation of LOC101928834 (Figure 6A), indicating that decreased expression of LOC101928834 could affect many components in this network, and thus LOC101928834 might occupy an important position in the regulatory network. Notably, FRAT2, a gene in the upstream of Wnt/β-catenin signaling pathway, was among the seven genes with decreased expression. We further assessed the expression of several well-studied genes in Wnt/ β-catenin signaling pathway and found that the expression of FRAT2, β-catenin and TCF-1 were decreased after down-regulation of LOC101928834 at the mRNA level (Figure 6B), and the expression of cyclin D1, cyclin D3, β-catenin and TCF-1 were decreased after down-regulation of LCO101928834 at the protein level (Figure 6C). These results indicate that LOC101928834 may have a role in regulating Wnt/ β-catenin signaling pathway.

Discussion

Our study showed that differentially expressed lncRNAs in MDS were involved in a wide range of biological functions. Specifically, we identified a novel lncRNA, LOC101928834, which had a significant impact on MDS diagnosis and prognosis. By applying loss-of-function approaches, we identified that LOC101928834 played a key role in the regulation of cell-cycle and leukemia cell proliferation likely via Wnt/β-catenin signaling pathway. To our best knowledge, this is the first study on the clinical value and biologic functions of LOC101928834 in MDS pathogenesis.

With the aging of the general population and the emergence of side effects of new chemotherapeutic agents, the incidence of MDS is increasing. However, the etiology and pathogenesis of MDS is still unclear [2,4]. Although an increasing number of IncRNAs has been found to be associated with MDS, the roles of only a small fraction of these lncRNAs have been elucidated [7,11,17]. We constructed a LncRNA-centric MDS regulation network to explore the functions of these unknown lncRNAs by interrogating their potential target protein-coding genes and related biological pathways. This ‘guilt by association’ approach represents one of the most widely used methods in the exploration of IncRNA functions and may guide the direction of future research in this field [22–24].

By clustering these lncRNAs into GO annotations, we found that the top enriched terms, such as regulation of apoptosis, participating in immune response, regulation of cell proliferation, positive regulation of biosynthetic process, were closely related to the etiology and pathogenesis of MDS [25]. The results of KEGG enrichments, including cancer pathway, MAPK signal transduction pathway, chemokine signal transduction pathway, actin cell skeletal regulation, etc., were also critical in the initiation and development of MDS [26]. Taken together, all these data suggest that these lncRNAs play central roles in MDS development.

We have previously shown that methylation of Wnt antagonist genes, such as SFRP1, SFRP4, and SFRP5, were predictive of shorter survival or higher risk of leukemia evolution in MDS [21]. By clustering these lncRNAs into Wnt pathway, we found five novel lncRNAs without any functional annotations. Notably, lncRNA LOC101928834 was highly expressed in MDS and its expression profile followed a stepwise increase from control, RCMD, RAEB, to AML. In other words, we found a correlation between LOC101928834 expression and the blast percentage in the BM with a correlation coefficient of 0.5350 and a P-value of <0.001 (Supplementary Figure S1). We speculate that LOC101928834 is likely to be preferably expressed in leukemic blasts, and this may explain why LOC101928834 expression was extremely low in normal tissues and organs (Supplementary Figure S2), but high in high-grade MDS and AML. We do not have a good explanation on why LOC101928834 expression was only correlated with poorer prognosis in MDS, not AML. One possibility is that there is a ‘dosage saturation’ effect and the prognostic impact of LOC101928834 is more pronounced at a certain range of blast percentage, such as <20%. Unfortunately, we do not have follow up data of these MDS patients, which precludes further study on the effects of LOC101928834 on leukemic transformation. Nevertheless, our study shows that LOC101928834 has a diagnostic value in MDS-RAEB, is an independent prognostic factor in MDS, and may serve as a potential marker for targeted therapy.

To investigate the effects of LOC101928834 on biological functions of leukemic cells, we constructed cell lines with stable down-expression of LOC101928834 and found that decreased LOC101928834 expression inhibited K562 and THP-1 cell proliferation and resulted in a significant increase in the percentage of cells in the G2 phase. There are many causes for cell cycle arrest in the G2/M phase [27], the best known mechanism is checkpoint regulation, which is the arrest of cell cycle in the G2 phase when the DNA is damaged [28], and thus a prohibition of cell proliferation [29]. To further explore the molecular mechanisms of LOC101928834 in promoting cell proliferation, we investigated potential target genes that are involved in Wnt/β-catenin signaling pathway sub-regulation network, which included 18 differentially expressed mRNAs and 5 lncRNAs such as LOC101928834. Several lncRNAs have been implicated in the activation of Wnt/β-catenin signaling pathway that has a critical role in the regulation of cell proliferation in cancers [30–33]. lnc-TCF7 was found to activate the Wnt pathway by activating TCF7 expression, thereby regulating the self-renewal of liver cancer stem cells [34]. MEG3 was down-regulated in dental follicle cells and activated Wnt/β-catenin signaling pathway via epigenetically regulating the H3K27me3 level on Wnt gene promoters [35]. lncRNA CASC2c, on the other hand, inactivated ERK1/2 and Wnt/β-catenin signaling pathway and thereby inhibited the proliferation of hepatocellular carcinoma cells [36]. In our bioinformatics prediction model, LOC101928834 was connected to Wnt pathway through FRAT2 (frequently rearranged in advanced T-cell lymphomas-2). FRAT2 has a 77% homology with FRAT1, which has been shown to be overexpressed in a nude mouse leukemia model [37]. They belong to the GSK-3 binding protein family and have been shown to activate Wnt/β-catenin signaling pathway through regulating GSK-3β and Disheveled in several tumors, especially in T-cell lymphomas [38]. Decreased expression of FRAT2 will lead to decreased Wnt/β-catenin signaling pathway activation [38], which is in keeping with our results. After down-regulation of LOC101928834, the expression levels of FRAT2 and other key proteins in the Wnt/β-catenin signaling pathway, β-catenin and TCF-1, were decreased. However, the expression of β-catenin was not decreased significantly at the mRNA level, suggesting the regulation of LOC101928834 on β-catenin might occur at the post-transcriptional level. Gallagher et al. found that β-catenin played an important role in the metastasis of melanoma through post-transcriptional modification [39]. The regulation of β-catenin by the Axin–GSK-3–APC complex was accomplished by the phosphorylation of serine residues at the N-terminal S45 site and T41-S37-S33, which was followed by degradation and thus reduction of β-catenin concentration in the cytoplasm [40]. Furthermore, we found that the expression of TCF-1, CyclinD1 and Cyclin D3 was decreased at both the mRNA and protein levels. T-cell factor (TCF) and lymphocyte-enhancer binding factor (LEF) are the ultimate downstream effectors of the Wnt pathway [22]. Cyclin D1 is a known target of Wnt/β-catenin signaling that regulates cell cycle and is involved in multiple cancers [30,41]. Cyclin D3 is another member of the cyclin family, and the CKD4 activity associated with Cyclin D3 has been reported to be necessary for cell cycle through G2 phase into mitosis after UV radiation [42,43]. Thus, we hypothesize that down-regulation of expression of cyclin D1 and cyclin D3 may have contributed, at least in part, to the G2 phase arrest. Moreover, mir-9, which was predicted to be a downstream target of LOC101928834 in our network, was also decreased significantly after down-regulation of LOC101928834. MiR-9 dysregulation has been found in many types of leukemias and a large number of its target genes, such as CCND1, were overexpressed and played essential roles in leukemogenesis [44]. Taken together, we speculate that LOC101928834 activates Wnt/β-catenin signaling pathway through up-regulating the expression of FRAT2, which in turn up-regulating cyclin D1 expression, and promoting cell cycle progression and cell proliferation. However, further studies are required to confirm this hypothesis.

In summary, we identified a novel lncRNA, LOC101928834, which was associated with distinct clinical features and poor prognosis in MDS. To our best knowledge, this is the first report on LOC101928834 in human diseases. We further demonstrated that overexpression of LOC101928834 could promote cell cycle progression and cell proliferation via Wnt/β-catenin signaling pathway. Our results show that LOC101928834 may serve as a useful biomarker for the diagnosis, prognosis and targeted therapy of MDS.

Clinical perspectives

  • Although a large number of lncRNAs have been found in MDS, the clinical significance of lncRNAs in MDS and whether they can be used as biomarkers for MDS diagnosis and prognosis of MDS remain to be investigated.

  • In the present study, we identified a novel lncRNA, LOC101928834, which had a significant impact on MDS diagnosis and prognosis, and plays a key role in the regulation of cell-cycle and leukemia cell proliferation likely via Wnt/β-catenin signaling pathway.

  • To our knowledge, this is the first report on the investigation of LOC101928834. Our results suggested that LOC101928834 may serve as a novel diagnostic and prognostic biomarker.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was supported by Shanghai Municipal Commission of Health and Family Planning [grant number 201640132] and National Natural Science Foundation of China (NSFC) [grant number 81500099].

Author Contribution

Nianyi Li and Xiaoqin Wang designed the research. Nianyi Li, Wanlin Wu, Jieguang Zhao and Shuang Li performed the experiments. Nianyi Li, Jieguang Zhao, Wei Wang and Yan Ma contributed patient tissues and case selection. Nianyi Li and Ruichen Sun analyzed the data and drafted the work. Chen Cameron Yin and Ruichen Sun revised it critically for important intellectual content. Xiaoqin Wang approved the final version of the manuscript.

Acknowledgements

We thank Zhihe Shi (East China University of Science and Technology) for writing program and making graphs.

Abbreviations

     
  • AML

    acute myeloid leukemia

  •  
  • DEG

    differentially expressed gene

  •  
  • lncRNA

    long non-coding RNA

  •  
  • MDS

    myelodysplastic syndromes

  •  
  • RAEB

    refractory anemia with excess blasts

  •  
  • ROC

    receiver operating characteristic

References

References
1.
Greenberg
P.L.
,
Attar
E.
,
Bennett
J.M.
,
Bloomfield
C.D.
,
De Castro
C.M.
,
Deeg
H.J.
et al.
(
2011
)
NCCN Clinical Practice Guidelines in Oncology: myelodysplastic syndromes
.
J. Natl. Compr. Canc. Netw.
9
,
30
56
[PubMed]
2.
Greenberg
P.L.
,
Stone
R.M.
,
Al-Kali
A.
,
Barta
S.K.
,
Bejar
R.
,
Bennett
J.M.
et al.
(
2017
)
Myelodysplastic Syndromes, Version 2.2017, NCCN Clinical Practice Guidelines in Oncology
.
J. Natl. Compr. Canc. Netw.
15
,
60
87
[PubMed]
3.
Arber
D.A.
,
Orazi
A.
,
Hasserjian
R.
,
Thiele
J.
,
Borowitz
M.J.
,
Le Beau
M.M.
et al.
(
2016
)
The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia
.
Blood
127
,
2391
2405
[PubMed]
4.
Ma
X.
,
Does
M.
,
Raza
A.
and
Mayne
S.T.
(
2007
)
Myelodysplastic syndromes: incidence and survival in the United States
.
Cancer
109
,
1536
1542
[PubMed]
5.
Neukirchen
J.
,
Schoonen
W.M.
,
Strupp
C.
,
Gattermann
N.
,
Aul
C.
,
Haas
R.
et al.
(
2011
)
Incidence and prevalence of myelodysplastic syndromes: data from the Dusseldorf MDS-registry
.
Leuk. Res.
35
,
1591
1596
[PubMed]
6.
Zhao
G.
,
Li
N.
,
Li
S.
,
Wu
W.
,
Wang
X.
and
Gu
J.
(
2019
)
High methylation of the 4-aminobutyrate aminotransferase gene predicts a poor prognosis in patients with myelodysplastic syndrome
.
Int. J. Oncol.
54
,
491
504
[PubMed]
7.
Yao
C.Y.
,
Chen
C.H.
,
Huang
H.H.
,
Hou
H.A.
,
Lin
C.C.
,
Tseng
M.H.
et al.
(
2017
)
A 4-lncRNA scoring system for prognostication of adult myelodysplastic syndromes
.
Blood Adv.
1
,
1505
1516
[PubMed]
8.
Zhao
X.
,
Yin
H.
,
Li
N.
,
Zhu
Y.
,
Shen
W.
,
Qian
S.
et al.
(
2019
)
An Integrated Regulatory Network Based on Comprehensive Analysis of mRNA Expression, Gene Methylation and Expression of Long Non-coding RNAs (lncRNAs) in Myelodysplastic Syndromes
.
Front. Oncol.
9
,
200
[PubMed]
9.
Bergmann
J.H.
and
Spector
D.L.
(
2014
)
Long non-coding RNAs: modulators of nuclear structure and function
.
Curr. Opin. Cell Biol.
26
,
10
18
[PubMed]
10.
Esteller
M.
(
2011
)
Non-coding RNAs in human disease
.
Nat. Rev. Genet.
12
,
861
874
[PubMed]
11.
Kopp
F.
(
2019
)
Molecular functions and biological roles of long non-coding RNAs in human physiology and disease
.
J. Gene Med.
21
,
e3104
[PubMed]
12.
Fan
Y.
,
Li
J.
,
Yang
Q.
,
Gong
C.
,
Gao
H.
,
Mao
Z.
et al.
(
2019
)
Dysregulated Long Non-coding RNAs in Parkinson's Disease Contribute to the Apoptosis of Human Neuroblastoma Cells
.
Front. Neurosci.
13
,
1320
[PubMed]
13.
Woo
C.J.
and
Kingston
R.E.
(
2007
)
HOTAIR lifts noncoding RNAs to new levels
.
Cell
129
,
1257
1259
[PubMed]
14.
Hruštincová
A.
,
Szikszai
K.
,
Krejčík
Z.
,
Loudová
N.
and
Merkerová
M.D.
(
2018
)
Noncoding RNAs in Myelodysplastic Syndromes, Recent Developments in Myelodysplastic Syndromes
.
Chapter 4: 12
14
,
15.
Bendetatos
L.
,
Hatzimichael
E.
,
Dasoula
A.
,
Dranitsaris
G.
,
Tsiara
S.
,
Syrrou
M.
et al.
(
2010
)
CpG methylation analysis of the MEG3 and SNRPN imprinted genes in acute myeloid leukemia and myelodysplastic syndromes
.
Leuk. Res.
34
,
148
153
[PubMed]
16.
Yildirim
E.
,
Kirby
J.E.
,
Brown
D.E.
,
Mercier
F.E.
,
Sadreyev
R.I.
,
Scadden
D.T.
et al.
(
2013
)
Xist RNA is a potent suppressor of hematologic cancer in mice
.
Cell
152
,
727
742
[PubMed]
17.
Liu
K.
,
Beck
D.
,
Thoms
J.A.I.
,
Liu
L.
,
Zhao
W.
,
Pimanda
J.E.
et al.
(
2017
)
Annotating function to differentially expressed LincRNAs in myelodysplastic syndrome using a network-based method
.
Bioinformatics
33
,
2622
2630
[PubMed]
18.
Franceschini
A.
,
Szklarczyk
D.
,
Frankild
S.
,
Kuhn
M.
,
Simonovic
M.
,
Roth
A.
et al.
(
2013
)
STRING v9.1: protein-protein interaction networks, with increased coverage and integration
.
Nucleic. Acids. Res.
41
,
D808
D815
[PubMed]
19.
(
2016
)
Ten years of Nature Protocols
.
Nat. Protoc.
11
,
1327
[PubMed]
20.
Chen
Y.
,
Zhao
G.
,
Li
N.
,
Luo
Z.
,
Wang
X.
and
Gu
J.
(
2019
)
Role of 4aminobutyrate aminotransferase (ABAT) and the lncRNA coexpression network in the development of myelodysplastic syndrome
.
Oncol. Rep.
42
,
509
520
[PubMed]
21.
Wang
H.
,
Wang
X.Q.
,
Xu
X.P.
and
Lin
G.W.
(
2010
)
ID4 methylation predicts high risk of leukemic transformation in patients with myelodysplastic syndrome
.
Leuk. Res.
34
,
598
604
[PubMed]
22.
Sun
P.R.
,
Jia
S.Z.
,
Lin
H.
,
Leng
J.H.
and
Lang
J.H.
(
2014
)
Genome-wide profiling of long noncoding ribonucleic acid expression patterns in ovarian endometriosis by microarray
.
Fertil. Steril.
101
,
1038e1037
1046e1037
23.
Rinn
J.L.
and
Chang
H.Y.
(
2012
)
Genome regulation by long noncoding RNAs
.
Annu. Rev. Biochem.
81
,
145
166
[PubMed]
24.
Zheng
J.
,
Huang
X.
,
Tan
W.
,
Yu
D.
,
Du
Z.
,
Chang
J.
et al.
(
2016
)
Pancreatic cancer risk variant in LINC00673 creates a miR-1231 binding site and interferes with PTPN11 degradation
.
Nat. Genet.
48
,
747
757
[PubMed]
25.
Aoyama
Y.
,
Sakai
K.
,
Kodaka
T.
,
Tsunemine
H.
,
Nishio
K.
,
Itoh
T.
et al.
(
2019
)
Myelodysplastic/myeloproliferative neoplasm with ring sideroblasts and thrombocytosis (MDS/MPN with RS-T) complicated by hyperleukocytosis and gene analysis in relation to leukocytosis
.
J. Clin. Exp. Hematop.
59
,
29
33
[PubMed]
26.
Yoshida
K.
,
Sanada
M.
,
Shiraishi
Y.
,
Nowak
D.
,
Nagata
Y.
,
Yamamoto
R.
et al.
(
2011
)
Frequent pathway mutations of splicing machinery in myelodysplasia
.
Nature
478
,
64
69
[PubMed]
27.
Arber
D.A.
,
Orazi
A.
,
Hasserjian
R.
et al.
(
2016
)
The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia
.
Blood
127
,
2391
2405
,
Blood. 128 (2016) 462-463.
[PubMed]
28.
Grana
X.
and
Reddy
E.P.
(
1995
)
Cell cycle control in mammalian cells: role of cyclins, cyclin dependent kinases (CDKs), growth suppressor genes and cyclin-dependent kinase inhibitors (CKIs)
.
Oncogene
11
,
211
219
[PubMed]
29.
Houghton
J.A.
,
Harwood
F.G.
and
Houghton
P.J.
(
1994
)
Cell cycle control processes determine cytostasis or cytotoxicity in thymineless death of colon cancer cells
.
Cancer Res.
54
,
4967
4973
[PubMed]
30.
MacDonald
B.T.
,
Tamai
K.
and
He
X.
(
2009
)
Wnt/beta-catenin signaling: components, mechanisms, and diseases
.
Dev. Cell
17
,
9
26
[PubMed]
31.
Simon
M.
,
Grandage
V.L.
,
Linch
D.C.
and
Khwaja
A.
(
2005
)
Constitutive activation of the Wnt/beta-catenin signalling pathway in acute myeloid leukaemia
.
Oncogene
24
,
2410
2420
[PubMed]
32.
Yao
X.
,
Liu
C.
,
Liu
C.
,
Xi
W.
,
Sun
S.
and
Gao
Z.
(
2019
)
lncRNA SNHG7 sponges miR-425 to promote proliferation, migration, and invasion of hepatic carcinoma cells via Wnt/beta-catenin/EMT signalling pathway
.
Cell Biochem. Funct.
37
,
525
533
[PubMed]
33.
Liu
R.
,
Li
Z.
,
Song
E.
,
Hu
P.
,
Yang
Q.
,
Hu
Y.
et al.
(
2020
)
LncRNA HOTTIP enhances human osteogenic BMSCs differentiation via interaction with WDR5 and activation of Wnt/beta-catenin signalling pathway
.
Biochem. Biophys. Res. Commun.
524
,
1037
1043
34.
Wang
Y.
,
He
L.
,
Du
Y.
,
Zhu
P.
,
Huang
G.
,
Luo
J.
et al.
(
2015
)
The long noncoding RNA lncTCF7 promotes self-renewal of human liver cancer stem cells through activation of Wnt signaling
.
Cell Stem Cell
16
,
413
425
[PubMed]
35.
Deng
L.
,
Hong
H.
,
Zhang
X.
,
Chen
D.
,
Chen
Z.
,
Ling
J.
et al.
(
2018
)
Down-regulated lncRNA MEG3 promotes osteogenic differentiation of human dental follicle stem cells by epigenetically regulating Wnt pathway
.
Biochem. Biophys. Res. Commun.
503
,
2061
2067
[PubMed]
36.
Li
Q.Y.
,
Yang
K.
,
Liu
F.G.
,
Sun
X.G.
,
Chen
L.
,
Xiu
H.
et al.
(
2020
)
Long noncoding RNA CASC2c inhibited cell proliferation in hepatocellular carcinoma by inactivated ERK1/2 and Wnt/beta-catenin signaling pathway
.
Clin. Transl. Oncol.
22
,
302
310
[PubMed]
37.
Freemantle
S.J.
,
Portland
H.B.
,
Ewings
K.
,
Dmitrovsky
F.
,
DiPetrillo
K.
,
Spinella
M.J.
et al.
(
2002
)
Characterization and tissue-specific expression of human GSK-3-binding proteins FRAT1 and FRAT2
.
Gene
291
,
17
27
[PubMed]
38.
McCubrey
J.A.
,
Steelman
L.S.
,
Bertrand
F.E.
,
Davis
N.M.
,
Abrams
S.L.
,
Montalto
G.
et al.
(
2014
)
Multifaceted roles of GSK-3 and Wnt/beta-catenin in hematopoiesis and leukemogenesis: opportunities for therapeutic intervention
.
Leukemia
28
,
15
33
[PubMed]
39.
Gallagher
S.J.
,
Rambow
F.
,
Kumasaka
M.
,
Champeval
D.
,
Bellacosa
A.
,
Delmas
V.
et al.
(
2013
)
Beta-catenin inhibits melanocyte migration but induces melanoma metastasis
.
Oncogene
32
,
2230
2238
[PubMed]
40.
Liu
C.
,
Li
Y.
,
Semenov
M.
,
Han
C.
,
Baeg
G.H.
,
Tan
Y.
et al.
(
2002
)
Control of beta-catenin phosphorylation/degradation by a dual-kinase mechanism
.
Cell
108
,
837
847
[PubMed]
41.
MacLachlan
T.K.
,
Sang
N.
and
Giordano
A.
(
1995
)
Cyclins, cyclin-dependent kinases and cdk inhibitors: implications in cell cycle control and cancer
.
Crit. Rev. Eukaryot. Gene Expr.
5
,
127
156
[PubMed]
42.
Davy
C.
and
Doorbar
J.
(
2007
)
G2/M cell cycle arrest in the life cycle of viruses
.
Virology
368
,
219
226
[PubMed]
43.
Dash
B.C.
and
El-Deiry
W.S.
(
2004
)
Cell cycle checkpoint control mechanisms that can be disrupted in cancer
.
Methods Mol. Biol.
280
,
99
161
[PubMed]
44.
Khosravi
A.
,
Alizadeh
S.
,
Jalili
A.
,
Shirzad
R.
and
Saki
N.
(
2018
)
The impact of Mir-9 regulation in normal and malignant hematopoiesis
.
Oncol Rev.
12
,
348
[PubMed]