MiR-10a-5p regulates proliferation, apoptosis and cell cycle of NPM1-mutated acute myeloid leukemia cells by targeting SHANK3

Objective: Although NMP1 mutation plays a crucial role in regulating the occurrence and development of acute myeloid leukemia (AML), there is still a lack of effective means to improve the prognosis of AML. Studies showed that miR-10a-5p is signicantly highly expressed in leukemia and other cancers. However, the function and mechanism of miR-10a-5p in NPM1-mutated AML remain unclear. Methods: The differential expression of miRNAs and mRNAs related to AML from GEO database were analyzed by bioinformatics. NPM1-mutated AML cell lines were constructed, while miR-10a-5p and SHANK3 were overexpressed to analyze cell proliferation, apoptosis and cell cycle. The targeting relationship between SHANK3 and miR-10a-5p was predicted by bioinformatics and further validated by dual luciferase assay. Results: Bioinformatics analysis on NPM1-mutated AML samples revealed that miR-10a-5p was highly expressed while SHANK3 was poorly expressed, and miR-10a-5p might target to regulate SHANK3 expression. Overexpression of miR-10a-5p promoted the proliferation, inhibited the apoptosis and regulated the cell cycle of NPM1-mutated AML cells, while opposite results were observed when SHANK3 was overexpressed. Collectively, miR-10a-5p regulated the proliferation, apoptosis and cell cycle of NPM1-mutated AML cells partially by inhibiting SHANK3 expression. Conclusion: These results demonstrate the role of miR-10a-5p/SHANK3 in NPM1-mutated AML, which provides a potential method for prognosis prediction of AML patients.


Introduction
Leukemia is a malignant clonal disease originating from hematopoietic stem cells, and it is regarded as an acute and chronic disease according to the maturity of leukemia cells and the natural course of the disease. The differentiation of acute leukemia cells ceases at an early stage. Acute myeloid leukemia (AML), accounting for 60% of all leukemia cases, is the predominant type of leukemia in adults and a hematopoietic malignancy that severely threatens human life and health [1][2][3]. However, the etiology and pathogenesis of AML remain unclear. NPM1 (nucleophosmin, NPM1), which codes 12-exon-contained nucleophosmin protein, is the gene highly mutated in AML at present [4]. Up to date, NPM1 mutations have been identi ed to be present not only in AML patients with normal karyotype [5], but also in patients with myelodysplastic syndromes [6], or in patients with chronic myeloid leukemia in blast crisis [7]. Research has shown that NPM1 mutations can be used as a proto-oncogene synergistically working with E1A to induce cell malignant transformation [8]. Meanwhile, NPM1 mutations can also promote the malignant proliferation of leukemia cells by regulating the activity of p53 and p53-dependent pathways [9], enhancing the stability of c-Myc and its cytoplasmic accumulation, and cooperating with FLT3-ITD and other genes [10]. These ndings suggest that NPM1 mutations are bene cial for the occurrence and development of leukemia.
Nevertheless, studies on the speci c prognosis and pathogenesis of patients with NPM1-mutated AML remain scarce. MicroRNAs (miRNAs), known as small RNA molecules with a length of approximately 22 nucleotides, are widely present in eukaryotes and highly conserved in evolution, with molecular functions of regulating cell differentiation, proliferation and apoptosis [11,12]. Research indicated that the expression level of miRNAs in tumor tissues is signi cantly different from that in normal tissues, and its tissue-speci c characteristic might help to identify the origin of tumor [13]. Therefore, miRNAs play an essential role in the processes of tumor differentiation and metastasis. MiR-10a-5p, an important miRNA belonging to the miR-10 family, has drawn extensive attention from people and been used in the study of the prognosis and pathogenesis of AML patients [14,15] , [16,17]. Bryant et al. discovered that miR-10a-5p is overexpressed in NMP1-mutated AML, which provides a new evidence for the high expression of the miR-10 family in AML patients. Zhi et al. reported that miR-10a-5p expression in AML patients is higher than that in normal individuals and the nding illustrated that miR-10a-5p could act as a prognostic biomarker for AML patients [18]. However, the speci c functional role of miR-10a-5p in patients with NMP1-mutated AML remains unknown.
This study focused on miR-10a-5p and its predicted target gene SHANK3 to identify the regulatory role of their targeting relationship in the proliferation, apoptosis and cell cycle of NPM1-mutated AML, so as to search for a new targeted therapy against AML.
1 Materials And Methods

Bioinformatics analysis
MiRNA expression matrix data of GSE68467 (109 samples, including 11 mutant samples and 98 wildtype samples) and mRNA expression matrix data of GSE68466 (109 samples, including 11 mutant samples and 98 wild-type samples) were download from GEO database (https://www.ncbi.nlm.nih.gov/geo/). Limma package was used for differential analysis with the wildtype samples as control. MiRNAs with |logFC|>1.5 and padj < 0.05 were considered differentially expressed miRNAs and mRNAs with |logFC|>1.0, padj < 0.05 were considered differentially expressed mRNAs. The database mirDIP (http://ophid.utoronto.ca/mirDIP/index.jsp) was used to predict potential target genes of miR-10a-5p, the miRNA differentially up-regulated in mutant samples. The target gene of interest was obtained from intersection of the differentially down-regulated mRNAs in GSE68466 and the predicted potential target genes of miR-10a-5p.

Cell transfection
100 nmol/L miR-10a-5p mimic, 100 nmol/L oe-SHANK3 and their corresponding negative controls (NCmimic, oe-NC) were purchased from GenePharma (Shanghai, China). Cells (1 × 10 5 ) were seeded into a 12-well plate before cell transfection. Oe-SHANK3, miR-10a-5p mimic and negative controls were transfected into cells using the LipoFiter reagent kit (Hanbio, Shanghai, China) in accordance with the manufacturer's instructions. Total RNA and proteins were extracted 48 h after transfection. Primer sequences were listed in Supplementary Table 1. 1.4 Real-time uorescence quantitative PCR (qRT-PCR) Total RNA was extracted from treated cells using Trizol (Invitrogen, Carlsbad, USA) according to the manufacturer's instructions. Complementary DNA (cDNA) was synthesized by the reverse transcription reagent kit (Invitrogen, Carlsbad, USA). qRT-PCR was performed using the ABI 7900HT device (Applied Biosystems, USA) with the miScript SYBR Green PCR Kit (Qiagen, Germany) under the following thermal cycling conditions: 95 ℃ 10 min, followed by 40 cycles of 95 ℃ 5 s and 60 ℃ 30 s, nally 72 ℃ 2 min. SHANK3 and miR-10a-5p expression levels were normalized to GAPDH and U6, respectively. All the primers used were shown in Supplementary Table 1. The differences in the relative expression of target genes in the control and experimental groups were analyzed by 2 −ΔΔCt method. The experiment was performed in triplicate and repeated three times.

Western blot
After transfection for 48 h, cells in different treatment groups were washed with cold PBS for 3 times (Thermo sher, USA). Total proteins were extracted using whole cell lysate on ice for 10 min, and quantitation was performed using the BCA protein assay kit (Thermo sher, USA). Cell extracts were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) at 100 V after boiled for 10 min at 95 ℃ with 10 µl of loading buffer. Thereafter, the proteins were transferred onto nitrocellulose membrane at 100 mA within 120 min, and the membrane was incubated with primary antibodies overnight at 4 ℃ after blocked with 5% bovine serum albumin or Tris-Buffered Saline Tween-20 for 60 min. Next, the membrane was washed with 1 × TBST (Solarbio, Beijing, China) on a shaker for three times with 5 min per time, and sequentially incubated with horseradish peroxidase (HRP)conjugated secondary antibody goat anti-mouse IgG for 120 min at room temperature. The membrane was washed by TBST 3 times, 20 min each time. Protein bands were visualized using the electrochemiluminescence (ECL) reagent kit (Solarbio, Beijing, China), and images were captured for observation. The experiment was performed in triplicate. All antibodies were shown in Supplementary  Table 2. 1.6 MTT and colony formation assays Cell proliferation was detected by MTT assay. Cells were planted into 96-well plates (5 × 10 3 cells per 100 µl) and each treatment was run in triplicate. After culture for 1, 2, 3, 4 and 5 d, sterile MTT solution (Beyotime) was added into cells to evaluate cell proliferation according to the instructions. The absorbance at 490 nm was measured using the spectrophotometer (Molecular Devices, Sunnyvale, CA, USA). For colony formation assay, cells were seeded onto 6-well plates with a density of 1 × 10 3 cells/well. After culture for 14 d, cell colonies were xed with 30% formaldehyde for 15 min, and then stained with 0.1% crystal violet (Thermo sher, USA). The colonies were counted by an optical microscope (each colony was set to have over 50 cells). To analyze distribution of cell cycle, cells were xed in 70% cold ethanol for 12 h and then incubated with PI (50 µg/ml) and RNase A (Sigma-Aldrich) for 30 min. Flow cytometry was applied for further analysis.
1.9 Statistical analysis SPSS 22.0 statistical software (SPSS, Inc, Chicago, IL, USA) was used for statistical analysis, and measurement data were presented as mean ± standard deviation. Student's t-test was used for analyzing differences between two independent groups, and one-way ANOVA method was used for analyzing differences among multiple groups. P < 0.05 suggested a statistically signi cant difference, while p < 0.01 and p < 0.0001 suggested a highly statistically signi cant difference.

MiR-10a-5p expression is signi cantly up-regulated in NPM1-mutated AML tissue and cells
Differential analysis was performed on the miRNA expression matrix data of AML patients (including NPM1-mut/wt) download from GEO database, and 10 differentially expressed miRNAs were screened ( Fig. 1A and 1B). Since the expression of miR-10a-5p was highly remarkably up-regulated in NPM1mutated patients, miR-10a-5p was in turn selected as the target miRNA for this study. We compared the expression of miR-10a-5p in NPM1-mut, NPM1-wt and common MOLM-14 cells. The results showed that, the expression of miR-10a-5p in NPM1-mut cells was signi cantly higher than that in NPM1-wt cells and common MOLM-14 cells (Fig. 1C). Based on above results, we found that miR-10a-5p expression in AML with mutated NPM1 was higher than that in AML with wild NPM1.

Overexpression of miR-10a-5p modulates the proliferation, apoptosis and cell cycle of NPM1-mut MOLM-14 cells
In order to explore the regulatory effect of miR-10a-5p on biological functions of NPM1-mutcells, we transfected miR-10a-5p mimic and NC-mimic into the AML cells. Firstly, we detected miR-10a-5p expression of the two groups. The result showed that miR-10a-5p expression in the miR-10a-5p mimic group was markedly higher than that in the NC-mimic group ( Fig. 2A), demonstrating that miR-10a-5p was e ciently overexpressed in the cells. Afterwards, to study the effect of miR-10a-5p on the proliferation of NPM1-mut AML cells, the proliferation of NPM1-mut MOLM-14 cells upon miR-10a-5p overexpression was measured. The results of MTT and colony formation assays revealed that overexpression of miR-10a-5p elevated the viability of NPM1-mut MOLM-14 cells (Fig. 2B and 2C). The apoptosis and cell cycle of NPM1-mut MOLM-14 cells were further analyzed under the presence of overexpressed miR-10a-5p, and it was found that miR-10a-5p overexpression signi cantly decreased the apoptosis of NPM1-mut MOLM-14 cells, while the cell number in G0/G1 phase was decreased and that in S phase was increased ( Fig. 2D and 2E). These results demonstrated that miR-10a-5p could promote NPM1-mut MOLM-14 cell proliferation, regulate cell cycle and inhibit cell apoptosis.

MiR-10a-5p down-regulates SHANK3 expression in NPM1-mut MOLM-14 cells.
We further investigated the potential target gene of miR-10a-5p after con rming that miR-10a-5p could promote the proliferation of NPM1-mut MOLM-14 cells. Firstly, we differentially analyzed the mRNA expression matrix data of AML patients (including NPM1-mut/wt) downloaded from GEO database. Totally, 23 signi cantly down-regulated mRNAs were screened out (Fig. 3A and Supplementary Table 3). Then, we used the mirDIP database to predict the target mRNAs of miR-10a-5p, which were intersected with the 23 differentially down-regulated mRNAs in GEO, and SHANK3 was identi ed eventually (Fig. 3B). The expression of SHANK3 in NPM1-mut, NPM1-wt and common MOLM-14 cells was tested, showing that SHANK3 expression in NPM1-mut MOLM-14 cells was signi cantly lower than that in other two groups ( Fig. 3C and 3D). After that, binding site sequences of miR-10a-5p on SHANK3 3'UTR were predicted using the TargetScan database (http://www.targetscan.org/vert_72/) (Fig. 3E). The results of dual-luciferase assay showed that overexpression of miR-10a-5p inhibited the luciferase activity of SHANK3-wt while had no effect on SHANK3-mut (Fig. 3F). Further, western blot and qRT-PCR results demonstrated that overexpression of miR-10a-5p inhibited the expression of SHANK3 in cells (Fig. 3G and  3H). Based on these results, we could nd that SHANK3 was a target gene of miR-10a-5p in NPM1-mut MOLM-14 cells.
2.4 MiR-10a-5p affects the proliferation, apoptosis and cell cycle of NPM1-mut MOLM-14 cells by regulating SHANK3 In this part, we rstly detected SHANK3 expression in three groups: NC mimic + oe-NC, miR-10a-5p mimic + oe-NC, and miR-10a-5p mimic + oe-SHANK3. The results of qRT-PCR and western blot showed that SHANK3 expression in the miR-10a-5p mimic + oe-SHANK3 group was signi cantly increased in comparison with that in the miR-10a-5p mimic + oe-NC group (Fig. 4A and 4B). MTT and colony formation assays showed that miR-10a-5p overexpression remarkably promoted cell viability (p < 0.05) and increased cell colony number (p < 0.01). However, the promotive effect on cell viability and the number of colonies was decreased when SHANK3 and miR-10a-5p were simultaneously overexpressed ( Fig. 4C and 4D). The cell cycle and apoptosis were further measured and it was noted that the cell apoptosis was decreased obviously after miR-10a-5p was overexpressed (p < 0.05), and more cells were found to aggregate in S phase. Nevertheless, the cell apoptosis was reversely increased when SHANK3 and miR-10a-5p were simultaneously overexpressed, and cell cycle was then arrest in G0/G1 phase ( Fig. 4E and 4F). These ndings validated that miR-10a-5p regulated the proliferation, apoptosis and cell cycle of NPM1-mut MOLM-14 cells by targeting the expression of SHANK3.

Discussion
MiRNAs are involved in multiple processes during the initiation of AML, and it is essential to understand the role of miRNAs in regulating the biological function of oncogenes or tumor suppressor genes in AML [19,20]. MiR-10a-5p has been proved to have a regulatory effect on tumor growth in various cancers [21,22]. However, its biological function and molecular mechanism in NPM1-mutated AML have received little notice. Our research showed that miR-10a-5p was highly expressed in NPM1-mutated AML cells. Functional analysis also revealed that miR-10a-5p overexpression could evidently facilitate NPM1mutated AML cell proliferation. These ndings indicated that miR-10a-5p works as a promoter in NPM1mutated AML. Debernardl et al. observed that the expression levels of miR-10a-5p, miR-10b, and miR-196a are positively correlated with the expression of HOXA and HOXB in AML patients and actively associated with the occurrence of AML as well [23]. Bosman et al. noted that the expression of miR-10a-5p is signi cantly higher in adriamycin-resistant AML cell strain HL-60 than that in susceptible cell strain HL-60 [24]. In conclusion, we believed that miR-10a-5p serves as a promoter and plays a crucial role in growth of AML.
In the present study, bioinformatics analysis and luciferase reporter assay elucidated that miR-10a-5p could target and regulate SHANK3. Furthermore, miR-10a-5p was negatively correlated with SHANK3 in expression in NPM1-mutated AML cells. SHANK3 gene codes multidomain scaffold proteins containing 1,747 amino acids, which mainly express in post synaptic density (PSD) of excitatory neuron [25]. Besides, SHANK3 is a susceptibility gene of multiple mental disorders, such as schizophrenia and bipolar disorder, etc [26]. However, there are few studies on the role of SHANK3 gene in tumor and leukemia. Our research ndings revealed that SHANK3 overexpression could markedly reverse the promotive effect of overexpressed miR-10a-5p on the proliferation of NPM1-mutated AML cells. This nding showed that the regulatory effect of miR-10a-5p on NPM1-mutated AML cells may partially performed by targeting SHANK3.
All in all, our experiments veri ed the positive regulatory effect of miR-10a-5p on NPM1-mutated AML cells, speci cally, miR-10a-5p can promote the proliferation, inhibit the apoptosis and regulate the cell cycle of NPM1-mutated AML cells via targeting SHANK3. The nding above not only provides a deep understanding about the role of miR-10a-5p in AML, but also lays a foundation for exploring new targeted therapies against AML.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The data used to support the ndings of this study are included within the article. The data and materials in the current study are available from the corresponding author on reasonable request. MiR-10a-5p is signi cantly up-regulated in NPM1-mutated AML tissue and cells. (A) Volcano plot of the differentially expressed miRNAs in AML from GEO database (Red represents up-regulated miRNAs, and green represents down-regulated miRNAs); (B) Box plots of the 10 differential miRNAs in NPM1-mutated AML patients; (C) qRT-PCR was used to detect miR-10a-5p expression in NPM1-mut, NPM1-wt and common MOLM-14 cells; *** p<0.0001.

Figure 2
MiR-10a-5p overexpression regulates NPM1-mut MOLM-14 cell proliferation, apoptosis and cell cycle. (A) qRT-PCR was used to detect miR-10a-5p expression in NPM1-mut MOLM-14 cells transfected with miR-10a-5p mimic or NC mimic; (B) MTT was used to detect cell viability of NPM1-mut MOLM-14 cells in two groups; (C) Colony formation assay was used to detect the proliferation of NPM1-mut MOLM-14 cells in two groups; (D) Flow cytometry was used to detect the effect of miR-10a-5p overexpression on cell apoptosis; (E) Flow cytometry was used to detect the effect of miR-10a-5p overexpression on cell cycle; * p<0.05.

Figure 3
MiR-10a-5p targets to regulate SHANK3. (A) Volcano plot of the differentially expressed mRNAs of AML patients downloaded from GEO database (Red for up-regulated mRNAs and greed for down-regulated mRNAs); (B) Intersection of the differentially down-regulated mRNAs in GEO and the target genes of miR-10a-5p predicted by the TargetScan database; (C-D) qRT-PCR and western blot were used to detect SHANK3 expression in NPM1-mut, NPM1-wt and common MOLM-14 cells; (E) Putative binding sites between SHANK3 and miR-10a-5p; (F) Dual-luciferase assay was used to detect the targeting relationship