LncRNA CCDC26 Interacts with CELF2 Protein to Enhance Myeloid Leukemia Cell Proliferation and Invasion via the circRNA_ANKIB1/miR-195-5p/PRR11 Axis.

LncRNA CCDC26 is aberrantly expressed in myeloid leukemia (ML) and promotes myeloid leukemia progression, but the potential mechanism of CCDC26 in regulating ML progression is unclear. In this study, we observed that lncRNA CCDC26 was upregulated in both chronic and acute ML cell lines. LncRNA CCDC26 promoted the proliferation and invasion of K562 and HL-60 cells, which was determined by cell counting kit-8 test and Transwell invasion assay. Flow cytometry showed that lncRNA CCDC26 inhibited cell apoptosis. Bioinformatics and expression correlation analyses revealed that there was a potential interaction between CCDC26 and CUGBP Elav-like family member 2 (CELF2) protein, an RNA bind protein (RBP). Then the relationship between CCDC26 and the RBP CELF2 was identified by using RNA pull-down and RNA immunoprecipitation (RNA-IP) assays. Further analysis showed that overexpression of CCDC26 could noticeably upregulate circRNA_ANKIB1 expression via sponging CELF2. Subsequently, we found that overexpressed circRNA_ANKIB1 could significantly promote proline rich 11 (PRR11) protein expression by sponging miR-195a-5p. Moreover, PRR11 was also upregulated by CCDC26 and downregulated by CELF2. Mechanically, we uncovered that the miR-195a-5p inhibitor activated the phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathways through upregulating PRR11 protein expression. Furthermore, the inhibitors of AKT, p65-NF-κB, or Bcl-2 could inhibit the effect of the miR-195a-5p inhibitor on ML cell behaviors. In conclusion, lncRNA CCDC26 could upregulate PRR11 protein expression by sponging miR-195a-5p, thereby activating the PI3K/AKT and NF-κB pathways to enhance ML cell proliferation and invasion and suppress cell apoptosis.

Introduction

Myeloid leukemia (ML) is a hematological malignancy, which can be divided into acute myeloid leukemia (AML) and chronic myeloid leukemia (CML)[1,2]. There are many risk factors associated with increased ML, including older age, susceptible genes, exposure to ionizing radiation, antecedent hematological disease, chemicals such as benzene, and certain chemicals[3]. Although there are drugs to treat and control ML, ML is still a major disease threatening human life and health and has been the research object of many scholars. In particular, the diagnosis and prognostic indicators of ML have attracted wide attention.

Noncoding RNA is a kind of RNA without coding ability, which has been proved to be involved in various cancer processes and is often used as a biomarker for diagnosis and prognosis[4,5]. Long noncoding RNA (LncRNA) coiled-coil domain-containing 26 (CCDC26), also known as RAM, is one of the noncoding RNA, which has been reported to be involved in many cancer processes[6,7]. LncRNA CCDC26 expression was upregulated in pancreatic cancer and associated with tumor number, tumor size, and reduced overall survival. Further analysis showed that lncRNA CCDC26 expression was an independent prognostic factor of overall survival in patients with pancreatic cancer and could be a diagnostic marker for distinguishing pancreatic cancer from normal[8]. In gastrointestinal stromal tumors, lncRNA CCDC26 was related to imatinib resistance through regulating IGF-1R expression[9]. Silencing lncRNA CCDC26 could inhibit gliomas cell proliferation and promote apoptosis. In vivo experiments, knockdown of lncRNA CCDC26 could inhibit glioma growth and metastasis[10]. Moreover, lncRNA CCDC26 was also acted as a biomarker in AML. Knockdown of lncRNA CCDC26 significantly reduced cell growth rate through upregulating tyrosine kinase receptor expression[11]. Subsequently, researchers found that lncRNA CCDC26 level in patients with AML was significantly associated with age, anemia, risk stratification, and remission. Furthermore, the overall survival of AML patients with a high expression level of lncRNA CCDC26 was poor (P = 0.0105)[12].

The interaction of noncoding RNAs has long been a question of great interest in a wide range of fields[13,14]. Wu et al.[15] constructed lncRNA-miRNA-mRNA and circRNA-miRNA-mRNA by co-expressing lncRNA/circRNA and mRNAs in atrial fibrillation. In intervertebral disc degeneration, multiple competitive endogenous RNA (ceRNA) networks were obtained, such as the lncRNA metastasis-associated lung adenocarcinoma transcript 1 (MALAT1)/circRNA_102348/miR-185-5p/transforming growth factor-beta 1 (TGFB1) axis, the circRNA_102399/miR-302a-3p/hypoxia-inducible factor 1 subunit alpha (HIF1A) axis, and the circRNA_100086/miR-509-3p/ mitogen-activated protein kinase 1 (MAPK1) axis, etc[16]. In pulmonary fibrosis, circRNA_949 (chromosome 14: 30346797-30350949) and circRNA_057 (chromosome 6:99003199-99100057) form a regulatory network with lncRNA NONMMUT039865 and lncRNA NONMMUT039556, simultaneously regulating miR-29b-2-5p targeting signal transducer and activator of transcription 3 (STAT3) phosphorylation in a bleomycin-induced mouse model[17].

In total, previous research had shown that lncRNA CCDC26 played a key role in ML. However, the potential mechanism of lncRNA CCDC26 affected on ML progression was not still clear, especially the intermolecular interaction. In this article, we uncovered that lncRNA CCDC26 not only aberrantly expressed in AML cell lines but also in CML cell lines. In ML cell lines, lncRNA CCDC26 could influence cell proliferation, invasion, and apoptosis. Further analysis found the interaction of lncRNA CCDC26, RNA-binding protein (RBP) CUGBP Elav-like family member 2 (CELF2), and circRNA_ankyrin repeat and IBR domain containing 1 (ANKIB1). We have then focused on the phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) which were affected by the interaction of CCDC26, CELF2, and circRNA_ANKIB1. Collectively, our data revealed the potential molecular mechanism of lncRNA CCDC26 in ML and provided several prognostic and/or therapeutic targets for ML patients.

Materials and Methods

Cell Culture

All cell lines (Normal: HS-27A, AML: HL-60 and ML-1, and CML: K562 and MYLR) used in this article were all purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were propagated in 85% Roswell Park Memorial Institute (RPMI)-1640 medium supplemented with 15% fetal bovine serum (FBS; Gibco, Rockville, MD, USA). The cell culture condition was 5% carbon dioxide (CO2) at 37 °C in a CO2 incubator (Thermo Fisher Scientific, Waltham, MA, USA).

RNA Extraction and Quantitative Polymerase Chain Reaction (qPCR)

The total RNA of cells was extracted utilizing Trizol Reagent (Invitrogen, Carlsbad, CA, USA) and treated with RNase-free DNase I (Invitrogen) to avoid DNA contamination. Then, according to the manufacturer’s protocol, the complementary DNA (cDNA) was synthesized by using the PrimeScriptTMII 1st Strand cDNA Synthesis Kit (Takara Biotechnology, Dalian, China) and stored at −40 °C. qPCR was used to analyze targeted RNA expression and was performed in the QuantStudio Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). According to the SYBR Premix Ex Taq II instructions (Takara Biotechnology, Dalian, China), a reaction system of 30 µL was prepared (temple: 1 μL, primers: 0.6 μL, respectively, 2× buffer: 15 μL, RNA-free water: 12.8 μL). The reactive condition of qPCR was performed the following: incubation at 95 °C for 90 s, followed by 40 cycles of 96 °C for 20 s, 60 °C for 55 s. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was utilized as an endogenous control. Each test was performed 3 times independently. Relative quantification was calculated with the 2−ΔΔCT formula.

Plasmid Construction and Cell Transfection

The short interfering RNA (siRNA) against lncRNA CCDC26 (CCDC26 siRNA), CELF2 siRNA, miR-195-5p, scramble, mimic-negative control (NC), pcDNA3.1 overexpression vector of lncRNA CCDC26 (pcDNA-CCDC26), CELF2 (pcDNA-CELF2), circRNA_ANKIB1 (pcDNA-ANKIB1), and pcDNA3.1 were all designed, synthesized, and validated by the company (Thermo Fisher Scientific). The cells were subgrown at a density of approximately 2 × 105 cells/well in 6-well plates. According to the manufacturer’s protocol, when cells were reached about 70% confluence, the siRNAs or vectors were transfected into cells by using the Lipofectamine3000 reagent (Thermo Fisher Scientific).

Cell Counting kit-8 (CCK-8) Test

According to the manufacturer’s instructions, K562 and HL-60 cells were grown to subconfluence and seeded in triplicate wells of 6-well plates (bacteria-free) at a concentration of 1 × 104 cells/well in a final volume of 200 µL and allowed to adhere overnight at 37 °C. Then, 10 µL of thawed CCK-8 solution was added into 100 µL fresh Dulbecco’s modified Eagle’s medium into each well, and the plates were incubated for 2 h at the same incubator conditions. The graph was prepared according to the absorbance value, which was read at 450 nm.

Transwell Invasion Assay

Cells were suspended in RPMI-1640 containing 15% FBS and placed in Transwell plates with an 8-µm pore size as well as with Matrigel. The upper insert was filled with a serum-free medium at a concentration of about 1 × 105 cells per insert, and the final volume of medium is 200 µL for each insert. The lower insert was filled with a culture medium containing 10% FBS as the chemoattractant. After incubation of 24 h at 37 °C and 5% CO2, the cells moved to the bottom of the membrane. The cells in the inner side of the membrane were eliminated with a swab. Then the cells were fixed and stained with 0.1% crystal violet. The invading cells were imaged at least random 5 fields/well under a light microscope. An average cell count of the 5 images was statistically analyzed.

RNA Immunoprecipitation (RIP) Assay

RIP experiments were performed using a Magna RIP RBP Immunoprecipitation kit (Millipore, Billerica, MA, USA) according to the manufacturer’s instructions. Briefly, the HEK293 cells were washed twice with precooled phosphate-buffered saline (PBS) and lysed with an equal volume of RIP lysis buffer. RIP wash buffer was used to prepare the magnetic beads. Then, 100 µL of cell lysate was added to the magnetic beads and resuspended in 900 µL of RIP buffer, and incubated overnight at 4 °C. The magnetic bead and antibody (or specific probes) complex were resuspended with proteinase K buffer incubated for 30 min at 55 °C and washed in RIP buffer, phenol, and chloroform. Finally, salt solution and precipitate enhancer were added, and anhydrous ethanol was also added and incubated at 80 °C for 1 h. After centrifuging, the precipitate was dissolved in diethylpyrocarbonate for qPCR analysis

RNA Pull-Down Assay With a Biotinylated RNA Probe

The RNA pull-down assay was used to identify the CCDC26 interaction with CELF2. Briefly, before harvest, cells were transfected with 50 nM biotinylated RNA probe for 48 h. Then, the cells were washed with PBS and incubated for 10 min in an RNA pull-down lysis buffer (Ambion, Austin, TX, USA) on ice. The lysates were precleared by centrifugation, and the samples (20 μL) were aliquoted for input. The remaining lysates were incubated with M-280 streptavidin magnetic beads precoated with RNase-free bovine serum albumin and yeast transfer RNA (Sigma, St. Louis, MO, USA) at 4 °C for 3 h. After that, the beads were washed 2 times with ice-cold lysis buffer and 3 times with a sodium dodecyl sulfate (SDS)-Tris low salt buffer (pH 8.0 containing 150 mM sodium chloride [NaCl]), and once with a high salt buffer containing 500 mM NaCl. The bound complexes were purified for the following analysis.

Luciferase Reporter Gene Assay

HEK293 cells were inoculated to a 24-well plate (5 × 105 cells/well). Wild type (WT) or mutant type (Mut) PRR11 was cloned to pGL3-Basic vector (Promega, Madison, WI, USA) and transfected into HEK293 cells. miR-195-5p mimics or negative control mimic was co-transfected together with the cells above, respectively. Then, luciferase activity was determined according to the manufacturer’s instructions 48 h after transfection (Promega).

Western Blotting

Western blotting was used to identify and semiquantitative analysis of targeted protein. β-actin was utilized as an endogenous reference. Briefly, samples were extracted and separated by SDS-polyacrylamide gel electrophoresis (PAGE). Then, the samples were transferred onto polyvinylidene fluoride (PVDF) membranes to identify and semiquantitate targeted protein expression. The specific primary antibodies (cleaved PRRP, cleaved caspase3, CELF2, PRR11, p-PI3 K, p-AKT, p65-NF-κB, Bcl2, cleaved caspase9, cyclin D, E-cadherin, N-cadherin, vimentin, and β-actin) were all purchased from Cell Signaling Technology (CST) company (Boston, MA, USA). The specific primary antibodies were diluted to the appropriate concentration and incubated for 1 h at 37 °C, and then washed 3 times with PBS and incubated with corresponding secondary antibody (CST, Boston, MA, USA) for 1 h at 37 °C. Finally, the bands of membranes were detected using enhanced chemiluminescence Western blotting substrate kit (Thermo Fisher Scientific).

Statistical Analysis

Herein, the measurement data were expressed by mean ± SEM in triplicate samples. A comparison between the 2 groups was conducted using the Student’s t-test. Comparison among 3 or more groups was performed using one-way or two-way analysis of variance. Data analysis was performed with SPSS 22.0 software, and graphs were performed with GraphPad Prism 5.0 software. It was considered to be statistically significant when the P-value <0.05.

Results

LncRNA CCDC26 was Aberrantly Expressed in ML cells and Promoted Cell Proliferation and Invasion

First, we compared the expression of lncRNA CCDC26 in ML cells and normal human bone marrow cells. We found that CCDC26 was upregulated in both AML and CML cells (Fig. 1A). In the following experiments, we used K562 and HL-60 cells as a representative. We transfected siRNA CCDC26 to knockdown CCDC26. The results showed that CCDC26 expression was significantly downregulated (Fig. 1B). Subsequently, we examined the effects of CCDC26 on cell proliferation, invasion, and apoptosis and found that knockdown of CCDC26 inhibited cell proliferation, invasion, and promoted apoptosis (Fig. 1C–F).

Fig. 1.

Knockdown of lncRNA CCDC26 reduced myeloid leukemia cell proliferation, invasion, and promoted apoptosis. (A) The expression of lncRNA CCDC26 was detected in a normal human bone marrow cell line (HS-27A), AML cell lines (HL-60 and ML-1), and CML cell lines (K562 and MYLR) by qPCR. ***P < 0.001 versus HS-27A group. (B) qPCR analysis of CCDC26 expression in K562and HL-60 cells that were transfected with the CCDC26 siRNA or scramble after 24 h. (C) Cell counting kit-8 test analysis of cell proliferation in K562 and HL-60 cells, which were transfected with the CCDC26 siRNA or scramble after 24 h. (D) Transwell invasion analysis of cell invasion in K562 and HL-60 cells, which were transfected with the CCDC26 siRNA or scramble after 24 h. (E) Flow cytometry analysis of the apoptosis in K562 and HL-60 cells, which were transfected with the CCDC26 siRNA or scramble after 24 h. (F) Western blotting analysis of cell apoptosis marker proteins in K562 cells, which were transfected with the CCDC26 siRNA or scramble after 24 h. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. *P < 0.05, **P < 0.01 versus control or scramble group. AML: acute myeloid leukemia; lncRNA: long noncoding RNA; qPCR: quantitative polymerase chain reaction; siRNA: short interfering RNA.

To further verify the effects of CCDC26, we overexpressed lncRNA CCDC26 in K562 and HL-60 cells. The results showed that CCDC26 successfully was overexpressed in K562 and HL-60 cells (Fig. 2A). Furthermore, we found that overexpression of CCDC26 promoted K562 and HL-60 cell proliferation and invasion and inhibited cell apoptosis (Fig. 2B–E). These results suggested that lncRNA CCDC26 could affect the behavior of ML cells.

Fig. 2.

Overexpression of CCDC26 could enhance cell proliferation, invasion, and promote apoptosis. (A) qPCR analysis of CCDC26 expression in K562and HL-60 cells that were transfected with the pcDNA CCDC26 or empty vector after 24 h. (B) Cell counting kit-8 test analysis of cell proliferation in K562 and HL-60 cells, which were transfected with the pcDNA CCDC26 or empty vector after 24 h. (C) Transwell invasion analysis of cell invasion in K562 and HL-60 cells, which were transfected with the pcDNA CCDC26 or empty vector after 24 h. (D) Flow cytometry analysis of the apoptosis in K562 and HL-60 cells, which were transfected with the pcDNA CCDC26 or empty vector after 24 h. (E) Western blotting analysis of cell apoptosis marker proteins in K562 cells, which were transfected with the pcDNA CCDC26 or empty vector 24 h. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. *P < 0.05 or **P < 0.01 versus scramble group. qPCR: quantitative polymerase chain reaction.

LncRNA CCDC26 Promoted cell Proliferation and Invasion via Binding with RBP CELF2

Here, we predicted that RBP CELF2 is a potential target of lncRNA CCDC26 through the StarBase database (http://starbase.sysu.edu.cn/index.php). The result suggested that there is a CELF2-binding motif (TCCTCTG) in the sequence of lncRNA CCDC26 (Fig. 3A). Then, we used RIP assay (Fig. 3B) and RNA pull-down assay (Fig. 3C and D) to determine whether CCDC26 binds with CELF2. The results indicated that CCDC26 bound with CELF2. We overexpressed lncRNA CCDC26 in K562 cells and found that CELF2 protein expression was declined (Fig. 3E, F).

Fig. 3.

CCDC26 could bind with CELF2 protein. (A) Starbase online (http://starbase.sysu.edu.cn/) was used to predict the relationship between lncRNA CCDC26 and CELF2 protein. Schema of CELF2-binding motif in the CCDC26 sequence. Red fonts represent the binding motifs. (B) qPCR analysis of CCDC26 expression in HEK293 cells, which was incubated with a CELF2 specific antibody. (C and D) Binding of CCDC26 and CELF2 was validated by CCDC26 probe-based RNA pull-down assay. (E and F) Western blotting analysis of CELF2 expression in K562 cells after transfection of the pcDNA CCDC26 or empty vector for 24 h. GAPDH served as the control. Statistical significance was assessed using a one-way variation analysis or Student’s t-test. Values are exhibited as mean ± SEM. *P < 0.05 versus control or pcDNA3.1 group, # P < 0.05 versus CCDC26 (0.5 μg/ml) group. CELF2: CUGBP Elav-like family member 2; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; qPCR: quantitative polymerase chain reaction.

We performed the analysis of CELF2 expression in K562 cells which has transfected with pcDNA CELF2, and the results showed that CELF2 was significantly upregulated (Fig. 4A ). CCK-8 test showed that overexpression of CELF2 inhibited K562 and HL-60 cell proliferation (Fig. 4B). Transwell invasion assay showed that overexpressed CELF2 inhibited K562 and HL-60 cell invasion (Fig. 4C). Flow cytometry and Western blotting showed that overexpressed CELF2 promoted apoptosis of K562 and HL-60 cells (Fig. 4D, E). Moreover, we knocked down CELF2 in ML cells (Fig. 5A). We uncovered that knockdown of CELF2 promoted K562 and HL-60 cell proliferation, invasion, and reduced apoptosis (Fig. 5B–E). These results showed that CCDC26 could bind with CELF2, inhibit CELF2 expression, promote cell proliferation and invasion, and inhibit apoptosis.

Fig. 4.

CELF2 inhibited cell proliferation, invasion, and apoptosis and was negatively regulated by lncRNA CCDC26. (A) Western blotting analysis of CELF2 expression in K562 cells that were transfected with the pcDNA CELF2 or empty vector after 24 h. (B) Cell counting kit-8 test analysis of cell proliferation in K562 and HL-60 cells, which were transfected with the pcDNA CELF2 or empty vector after 24 h. (C) Transwell invasion analysis of cell invasion in K562 and HL-60 cells, which were transfected with the pcDNA CELF2 or empty vector after 24 h. (D) Flow cytometry analysis of the apoptosis in K562 and HL-60 cells, which were transfected with the pcDNA CELF2 or empty vector after 24 h. (E) Western blotting analysis of cell apoptosis marker protein in K562 cells, which were transfected with the pcDNA CELF2 or empty vector after 24 h. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. *P < 0.05 or **P < 0.01 versus pcDNA3.1 group. CELF2: CUGBP Elav-like family member 2; lncRNA: long noncoding RNA.

Fig. 5.

Silencing CELF2 could promote cell proliferation, invasion, and suppress cell apoptosis. (A) Western blotting analysis of CELF2 expression in K562 and HL-60 cells that were transfected with the CELF2 siRNA or scramble vector after 24 h. (B) Cell counting kit-8 test analysis of cell proliferation in K562 and HL-60 cells, which were transfected with the CELF2 siRNA or scramble after 24 h. (C) Transwell invasion analysis of cell invasion in K562and HL-60 cells, which were transfected with the CELF2 siRNA or scramble after 24 h. (D) Flow cytometry analysis of the apoptosis in K562 and HL-60 cells, which were transfected with the CELF2 siRNA or scramble after 24 h. (E) Western blotting analysis of cell apoptosis marker proteins in K562 cells, which were transfected with the CELF2 siRNA or scramble 24 h. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. *P < 0.05 or **P < 0.01 versus scramble group. CELF2: CUGBP Elav-like family member 2; siRNA: short interfering RNA.

LncRNA CCDC26 upregulated circRNA_ANKIB1 Expression via Sponge of RBP CELF2

Herein, we predicted that there is a CELF2-binding motif (CTGGAAT) in the sequence of circRNA ANKIB1 (Fig. 6A) through the StarBase database (http://starbase.sysu.edu.cn/index.php). Then, we used RIP to confirm the relationship between circRNA_ANKIB1 and CELF2 (Fig. 6B) and the relationship between circRNA_ANKIB1 and CCDC26 (Fig. 6C). Furthermore, we detected the circRNA_ANKIB1 expression level in K562 cells transfected with pcDNA ANKIB1 or pcDNA CCDC26 or pcDNA CELF2. We found that the expression of circRNA_ANKIB1 was positively regulated by pcDNA ANKIB1 or pcDNA CCDC26 and negatively regulated by pcDNA CELF2 (Fig. 6D).

Fig. 6.

LncRNA CCDC26 could regulate circRNA_ANKIB1 expression via the sponge of CELF2. (A) Starbase online (http://starbase.sysu.edu.cn/) was used to predict the relationship between circRNA_ANKIB1 and CELF2 protein. Schema of CELF2 binding motif in the circRNA_ANKIB1 sequence. Red fonts represent the binding motifs. (B) qPCR analysis of circRNA_ANKIB1 expression in HEK293 cells which was incubated with CELF2 specific antibody. (C) qPCR analysis of circRNA_ANKIB1 expression in HEK293 cells, which were transfected with pcDNA CCDC26 and incubated with CELF2 specific antibody. (D) qPCR analysis of circRNA_ANKIB1 expression in K562 cells, which were transfected with pcDNA ANKIB1, pcDNA CCDC26, pcDNA CELF2, or empty vector. (E) Starbase database showed miR-195-5p binding sites in the circRNA_ANKIB1 sequence. (F) qPCR analysis of ANKIB1 expression in HEK293 cells, which were incubated with Bio-miR-195-5p mimic. (G) qPCR analysis of circRNA_ANKIB1 expression in nuclear and cytoplasm in K562 cells. (H) qPCR analysis of miR-195-5p expression in K562 cells, which were transfected with the pcDNA CCDC26, pcDNA CELF2, pcDNA ANKIB1, or empty vector after 24 h. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. **P < 0.01 versus control or pcDNA3.1 group. CELF2: CUGBP Elav-like family member 2; lncRNA: long noncoding RNA; NC: negative control; qPCR: quantitative polymerase chain reaction.

CircRNA_ANKIB1, as a ceRNA, Upregulated PRR11 Expression Through Sponging miR-195-5p

One of the functions of circRNA is sponged to miRNA[18,19]. According to the StarBase database (http://starbase.sysu.edu.cn/index.php), we found that circRNA_ANKIB1 sponged miR-195-5p (Fig. 6E), and RIP assay revealed that circRNA_ANKIB1 bound with miR-195-5p (Fig. 6F). First, we found that circRNA_ANKIB1 was present in the cytoplasm of K562 cells (Fig. 6G), and the miR-195-5p expression was downregulated by pcDNA CCDC26 or pcDNA ANKIB1 and upregulated by pcDNA CELF2 (Fig. 6H). Furthermore, PRR11, one of the miR-195-5p targeted protein, was identified (Fig. 7A). Luciferase reporter gene assay showed that miR-195-5p targeted PRR11 3′UTR but not the mutated PRR11 3′UTR (Fig. 7B). Then we detected the PRR11 expression and found that PRR11 expression was inhibited by miR-195-5p (Fig. 7C). Moreover, the PRR11 expression also was upregulated by pcDNA CCDC26 or pcDNA ANKIB1 and downregulated by pcDNA CELF2 (Fig. 7D). Finally, we detected the cell behaviors after transfected with pcDNA PRR11 or miR-195-5p mimic. We found that pcDNA PRR11promoted cell proliferation, invasion, and inhibited cell apoptosis while miR-195-5p inhibited cell proliferation, invasion, and enhanced apoptosis (Fig. 7E–G).

Fig. 7.

MiR-195-5p could regulate cell behaviors by targeting PRR11 3′UTR. (A) TargetScanHuman database showed miR-195-5p binding sites in the PRR11 3′UTR sequence. (B) Luciferase reporter gene assay showed that miR-195-5p targeted PRR11 3′UTR sequence. (C) Western blotting analysis of PRR11 protein expression in K562 cells, which were transfected with pcDNA PRR11 or miR-195-5p mimic or empty vector or mimic NC. (D) Western blotting analysis of PRR11 protein expression in K562 cells, which were transfected with pcDNA CCDC26, pcDNA CELF2, pcDNA ANKIB1, or empty vector. (E) Cell counting kit-8 test analysis of cell proliferation in K562 and HL-60 cells, which were transfected with pcDNA PRR11 or miR-195-5p mimic or empty vector or mimic NC. (F) Flow cytometry analysis of the apoptosis in K562 and HL-60 cells, which were transfected with pcDNA PRR11 or miR-195-5p mimic or empty vector or mimic NC. (G) Western blotting analysis of cell apoptosis marker proteins in K562 cells, which were transfected with pcDNA PRR11 or miR-195-5p mimic or empty vector or mimic NC. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. *P < 0.05; **P < 0.01 versus control, pcDNA3.1 or NC mimic group. NC: negative control; PPR11: proline rich 11; UTR: untranslated region.

PRR11 Promoted Cell Proliferation, Invasion, and Inhibited Apoptosis via the PI3K/AKT and NF-κB Signaling Pathways

In the end, we focused on the effects of PRR11 on the PI3K/AKT and NF-κB pathways. We found that the miR-195-5p inhibitor promoted PRR11 expression and activated the PI3K/AKT and NF-κB signaling pathways. However, the effect of miR-195-5p inhibitor could be inhibited by the inhibitors of AKT, p65-NF-κB, or Bcl2 (Fig. 8A, B). Furthermore, we detected the marker proteins of cell proliferation and invasion. The results showed that the miR-195-5p inhibitor upregulated cyclin D, N-cadherin, and vimentin protein expression and downregulated E-cadherin protein expression. The inhibitors of AKT or p65-NF-κB inhibited the effect of miR-195-5p inhibitor in marker protein expression (Fig. 9A). Finally, the effect of the miR-195-5p inhibitor on cell behaviors was also inhibited by the inhibitors of AKT or p65-NF-κB (Fig. 9B, D). These results showed that miR-195-5p could influence ML cell behaviors through inhibiting PRR11 protein expression thereby effect on the PI3K/AKT and NF-κB pathways.

Fig. 8.

MiR-195-5p inhibited activation of the PI3K/AKT and NF-κB signaling pathway through downregulating PRR11 expression. (A) Western blotting analysis of the PI3K/AKT signaling pathway-related proteins in K562 cells, which were transfected with miR-195-5p mimic or miR-195-5p inhibitor and incubated with the inhibitors of AKT, Bcl2, NC. (B) Western blotting analysis of the NF-κB signaling pathway-related proteins in K562 cells, which were transfected with miR-195-5p mimic or miR-195-5p inhibitor and incubated with the inhibitors of p65. Values are exhibited as mean ± SEM. *P < 0.05 versus NC mimic group, # P < 0.05 versus NC inhibitor, & P < 0.05 versus miR-195-5p inhibitor (50 nM). AKT: protein kinase B; NC: negative control; NF-κB: nuclear factor kappa-light-chain-enhancer of activated B cells; PI3K: phosphatidylinositol 3-kinase.

Fig. 9.

PRR11 regulated cell proliferation, invasion, and apoptosis via the PI3K/AKT signaling pathway. K562 cells were transfected with miR-195-5p mimic or miR-195-5p inhibitor and incubated with the inhibitors of AKT and p65 or NC. (A) Western blotting analysis of cell proliferation and invasion marker proteins in K562 cells. (B) The proliferation of K562 cells was analyzed by cell counting kit-8 test. (C) The invasion of K562 cells was analyzed by Transwell invasion assay. (D) The apoptosis of K562 cells was analyzed by flow cytometry. Statistical significance was analyzed utilizing the analysis of variance. Values are exhibited as mean ± SEM. *P < 0.05 versus NC mimic group, # P < 0.05 versus NC inhibitor, & P < 0.05 versus miR-195-5p inhibitor (50 nM). AKT: protein kinase B; NC: negative control; PPR11: proline rich 11; PI3K: phosphatidylinositol 3-kinase.

Discussion

Despite previously being ignored, noncoding RNAs are increasingly recognized as important regulators. The lncRNA CCDC26 has already been reported in multiple cancers including ML[20,21]. However, the potential mechanism of CCDC26 in ML was unclear. In this article, we described that CCDC26 could bind with the RBP CELF2, thereby upregulating circRNA_ANKIB1 expression. The upregulation of circRNA_ANKIB1 could promote PRR11 protein expression through sponging miR-195-5p. Finally, we focused on the PI3K/AKT and NF-κB signaling pathways, which may be the pathway affected by PRR11 protein, and it may also be the pathway affected by lncRNA CCDC26. We found that miR-195-5p inhibitor promoted cell proliferation, invasion, and inhibited cell apoptosis through activating the PI3K/AKT and NF-κB pathways. However, this effect of miR-195-5p inhibitor was inhibited by the inhibitors of AKT or p65-NF-κB. That means lncRNA CCDC26 regulated cell behaviors via the CELF2/circRNA_ANKIB1/miR-195-5p/PRR11/PI3K/AKT axis.

As an RBP, CELF2 can bind multiple RNAs to perform various functions[22,23]. CELF2 could bind with heme oxygenase-1 mRNA transcripts, thereby downregulating heme oxygenase-1 expression in hypertrophic cardiomyopathy[24]. CELF2 controlled alternative polyadenylation of its own message in a signal-dependent manner by competing with core enhancers of the polyadenylation machinery for binding to RNA[25]. In nonsmall cell lung cancer, CELF2 could inhibit the phosphatidylinositol-3,4,5-trisphosphate dependent Rac exchange factor 2 (PREX2)-phosphatase and tensin homolog (PTEN) interaction. Overexpression of CELF2 could significantly inhibit tumor growth in vivo or in vitro through inhibiting the PI3K/AKT signaling pathway phosphorylation[26].

In recent years, there has been growing interest in the effect of circRNA[27]. More and more researches indicated that circRNA had critical regulator roles in disease progression. CircRNA_0001445 inhibited hepatocellular carcinoma growth and metastasis by promoting TIMP metallopeptidase inhibitor 3 expression (TIMP3)[28]. CircRNA_forkhead box O3 (FOXO3) interacted with ID-1, E2F1, FAX, and HIF-1A resulting in increased cellular senescence[29]. With the deepening of research, the function of circRNA has been found by researchers. For instance, circRNA can compete with linear RNA production regulating the accumulation of full-length mRNA or circRNA can pleiotropically modulate gene expression by protein binding[30,31]. And one of the most frequent functions is a sponge of miRNA. Such as circRNA_ANKIB1 could suppress family 26, subfamily B, cytochrome P450, polypeptide 1 expression by sponging miR-423-5p, miR-666-3p, and miR-485-5p, leading to the induction of Schwann cell proliferation and nerve regeneration[32]. Furthermore, circRNA_ANKIB1 could sponge miR-19b, thereby activating the STAT3 pathway[33].

During the last few years, the link between PRR11 and cancer has been much attention. PRR11 was responsible for the maintenance of self-renewal and tumorigenicity of gastric cancer stem cells via the MAPK signaling pathway[34]. Overexpression of PRR11 could facilitate esophageal squamous cell carcinoma progression[35]. Silencing of PRR11 could suppress nonsmall cell lung cancer proliferation and induce cell autophagy through inactivating the AKT signaling pathway[36]. In ovarian carcinoma, PRR11 overexpression promoted cell proliferation, migration, and invasion via the PI3K/AKT pathway[37]. Moreover, in prostate cancer, miR-195 inhibited cell proliferation and angiogenesis by downregulating PRR11 expression[38]. All of these studies could provide support to our results in this research.

In summary, we observed aberrantly expression of lncRNA CCDC26 in ML cell lines and then used databases and experiments to explore the mechanism of lncRNA CCDC26 in ML cells. In the end, we acquired the relationship between lncRNA CCDC26 and PRR11 and found that lncRNA CCDC26 interacts with CELF2 protein to enhance ML cell proliferation and invasion via the circRNA_ANKIB1/miR-195-5p/PRR11 axis.