microRNA-660 Enhances Cisplatin Sensitivity via Decreasing SATB2 Expression in Lung Adenocarcinoma

Increasing evidence suggests that microRNAs’ (miRNAs) abnormal expression is one of the main factors of chemotherapy resistance in various cancers. However, the role of miRNAs in lung adenocarcinoma (LUAD) resistance to cisplatin is still unclear. In this study, we analyzed a microarray dataset to investigate miRNAs related to cisplatin resistance in LUAD. The expression of miRNAs in LUAD tissues and cell lines was detected using real-time quantitative polymerase chain reaction (RT-qPCR). Special AT-Rich Sequence-Binding Protein 2 (SATB2) in LUAD cell lines was detected using RT-qPCR and Western blot. Cell proliferation was measured by CCK8 and colony formation assays, while cell cycle and apoptosis were measured by flow cytometry. A dual-luciferase reporter assay was performed to confirm that SATB2 is a target gene of microRNA-660 (miR-660). We showed that the expression of miR-660 was not only decreased in LUAD cells and tissues but also further decreased in the cisplatin-resistant A549 cell line. The overexpression of miR-660 increased cisplatin sensitivity in LUAD cells. In addition, we identified SATB2 as a direct target gene of miR-660. We also revealed that miR-660 increased cisplatin sensitivity in LUAD cells via targeting SATB2. In conclusion, miR-660/SATB2 axis is a key regulator of cisplatin resistance in LUAD.


Introduction
Lung cancer is the leading cause of cancer-related deaths globally, leading to over 20% of cancer deaths [1]. Lung adenocarcinoma (LUAD) is a pathological subtype of lung cancer [2]. LUAD patients who have no indications for surgical interventions are often treated with medical therapy. In the past two decades, numerous biomarkers of LUAD have been found, resulting in targeted therapy and immunotherapy development. For example, gefitinib for EGFR mutation, clothotinib for ALK-EML4 fusion, and pembrolizumab for PD-1 molecular have greatly extended the survival of LUAD patients [3,4]. Despite the rapid development of multiple therapies, platinum-based chemotherapy remains the main therapy regimen for LUAD patients. Cisplatin, a typical platinum drug, is capable of killing tumor cells. Dual-drug chemotherapy based on cisplatin is not only the first-line treatment of LUAD but also the preferred drug of chemotherapy combined with targeted therapy or immunotherapy [5]. However, LUAD cells may develop resistance to cisplatin, which ultimately results in treatment failure [6]. Therefore, it is necessary to find a way to overcome cisplatin resistance in LUAD.
MicroRNAs (miRNAs), a kind of short (20-24 nucleotides) non-coding RNA, can regulate gene expression post-transcriptionally by repressing translation or degrading targeted messenger RNAs (mRNAs) [7,8]. MiRNA dysregulation affects a spectrum of physiological and pathological processes in various cancers, including proliferation, metastasis, and drug resistance [9]. It has been reported that miR-205 can enhance chemotherapy resistance in non-small cell lung cancer via targeting the phosphatase and tensin homolog [10]. Chen et al. showed that miR-181a overexpression degrades protein kinase C delta, thereby inducing cisplatin resistance in human cervical squamous cell carcinoma [11]. However, the role of miRNAs in regulating cisplatin sensitivity in LUAD is still not fully clear.
In this study, we screened potential miRNAs related to cisplatin resistance in LUAD by identifying differentially expressed miRNAs in cisplatin-resistant and cisplatin-sensitive LUAD cell lines. We found that the expression of microRNA-660 (miR-660) was significantly decreased in the cisplatin-resistant LUAD cell line. MiR-660 overexpression reduced cisplatin resistance in LUAD. Moreover, we revealed that miR-660 targeted Special AT-Rich Sequence-Binding Protein 2 (SATB2), thereby impairing cisplatin resistance in LUAD. This study suggests that miR-660 and SATB2 might potentially be new therapy targets for reversing cisplatin resistance in LUAD.

Clinical Tissue Samples
In this study, 30 paired LUAD and normal lung tissues were collected from our hospital (The Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University/Hunan Cancer Hospital). The tissues were snap frozen in liquid nitrogen and stored until RNA extraction. All samples were identified by 2 senior pathologists. All patients signed an informed consent form, and this study was approved by the Ethics Committee of Hunan Cancer Hospital (Ethical lot number: KYJJ-2021-242). The clinical information of patients is reported in Table S1.

Cell Culture
LUAD cell lines (A549, H1975, and CALU-3) were cultured in RPMI 1640 medium (Gibco, Waltham, MA, USA), and human bronchial epithelioid cell line (HBE) and HEK293T cell lines were cultured in Dulbecco's modified eagle medium (Gibco). The cell culture medium was supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin/streptomycin (New Cell & Molecular Biotech, Suzhou, China). All cells were obtained from the cell bank of the Cancer Research Institute of Central South University (Changsha, China) [21,22] and incubated at 37 • C with 5% CO 2 . The cisplatin-resistant A549 cell line (A549/CDDP) was derived from A549 cells using stepwise increased cisplatin (Qilu Pharmaceutical, Shandong, China) concentration incubation. All cells were passaged to the third generation to start the experiments.

RNA Isolation and Real-Time Quantitative Polymerase Chain Reaction (RT-qPCR) Assays
The RNA sample was extracted using TRIzol solution (Thermo Fisher Scientific, Waltham, MA, USA). The first-strand complementary DNA was synthesized using a miDETECT A Track™ miRNA qRT-PCR Starter Kit (RiboBio) or RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific), followed by PCR amplification in Bio-Rad CFX Connect™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) or 2x SYBR Green qPCR Master Mix (Bimake, Houston, TX, USA) according to the manufacturer's instructions. U6 or β-actin was used as reference genes. Gene expression was analyzed using the comparative threshold cycle (2 −∆∆Ct ) method.

Cisplatin IC 50 Detection
Cells (1000/well) were seeded into 96-well plates and incubated in 37 • C for 12 h. Then, the medium was replaced with a medium containing 0, 3, 12, 24, 36, and 48 µM cisplatin. After 48 h, Cell Counting Kit-8 (CCK-8, New Cell & Molecular Biotech) solution was added into the well, and the optical density values were measured at 450 nm (OD450). The relative cell viability (experimental group OD450/control group OD450×100%) with the corresponding cisplatin concentration was used to calculate the cell cisplatin IC 50 values using nonlinear fitting analysis in GraphPad Prism 7 software (version 7.0, GraphPad, La Jolla, CA, USA).

Cell Proliferation
Cells (1000/well) were seeded into 96-well plates and incubated in 37 • C for 12 h. Then, the medium was replaced with a medium containing 10 µM cisplatin. The CCK-8 solution was added at the same time for 0, 1, 2, 3, 4, and 5 days, and the OD450 was measured. The results were visualized using GraphPad Prism 7 software.

Colony Formation Assay
Cells (3000/well) were seeded into 6-well plates and cultured in medium containing 10 µM cisplatin for two weeks. The cells were fixed with methanol and stained with 0.1% crystal violet. The images were scanned, and the number of colonies was counted using Image J software (NIH Image J system, Bethesda, MD, USA, Version: 1.8.0).

Flow Cytometry
The cells were treated with 10 µM cisplatin for 48 h. Then, 2 × 10 5 cells were collected and washed twice with phosphate-buffered saline. A Cell Cycle Staining Kit (Multi sciences, Hangzhou, China) was used to analyze the cell cycle, and Annexin V-FITC/propidium iodide (PI) Apoptosis Kit (Multi sciences) was used to perform the apoptosis assay according to the manufacturer's instructions. The results were visualized using Modfit software (Verity Software House, Topsham, ME, USA, version LT3.1) or FlowJo software (BD Biosciences, Franklin Lakes, NJ, USA, version 10.0).

Luciferase Reporter Assay
HEK293T cells were co-transfected with miR-660 mimics and SATB2-WT/MUT, together with Renilla luciferase pRLTK vector (Tsinke) as a control. After 24 h, the cells were lysed using radio immunoprecipitation assay buffer (RIPA, Beyotime, Shanghai, China). The Dual-Glo dual luciferase reporter assay system (Promega, Beijing, China) was utilized here to analyze and calculate the ratio of luminescence intensity.

Western Blot
The proteins were extracted using RIPA (Beyotime) with a protease inhibitor (Beyotime). A BCA™ protein analysis kit (Thermo Fisher Scientific) was used to quantify the proteins. The SATB2 (Proteintech, Rosemont, IL, USA) was validated by Western blot analysis, and GAPDH (Proteintech) was used as a reference protein. HRP, Goat Anti-Rabbit IgG (Proteintech) was used as the secondary antibody. The signal was quantified using Image J software.

Statistical Analysis
All experiments were repeated at least three times. The data were analyzed using GraphPad Prism 7 software and were reported as the means ± standard deviation (SD). Student's t-test and one-way analysis of variance were used to compare the two groups or multiple groups. Differences were considered significant when p < 0.05. * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001.

miR-660 Expression Is Associated with Cisplatin Sensitivity in LUAD Cells
To identify the candidate miRNAs related to cisplatin resistance in LUAD, we reanalyzed the GSE84200 dataset, which contains miRNA expression profiling data of the A549 and A549/CDDP cell line. Then, we obtained 70 miRNAs with significant differences in expression, including 65 downregulated and 5 upregulated miRNAs in A549/CDDP cell line ( Figure 1A,B). We queried the dbDEMC and starBase databases for the expression and prognosis of 65 of the above-mentioned miRNAs in LUAD. We found 16 miRNAs that were significantly downregulated in LUAD tumor tissues ( Figure S1). In these 16 miRNAs, overexpression of miR-660, miR-140, and miR-195 was closely related to better prognosis in LUAD patients ( Figures 1C-F and S2).

miR-660 Enhances Cisplatin Sensitivity in LUAD Cells
In order to investigate the function of miR-660 in LUAD cisplatin resistance, the expression of miR-660 was upregulated using mimics in A549/CDDP cells and CALU-3 cells and knocked down using inhibitors in A549 cells, respectively ( Figure S3B). Subsequently, a series of experiments was executed in vitro. The cisplatin IC 50 detection showed that miR-660 overexpression decreased the cisplatin IC 50 value from 40.63 ± 0.87 µM to 14.13 ± 0.86 µM in A549/CDDP cells (Figure 2A). MiR-660 upregulation also decreased the proliferation and colony formation capacity in A549/CDDP cells treated with cisplatin. Meanwhile, the cell apoptosis rate and G1 cell cycle arrest increased, and S-G2 cell cycle arrest decreased in cisplatin-treated A549/CDDP cells ( Figure 2B-E). In contrast, cisplatin IC 50 detection showed that miR-660 downregulation increased cisplatin resistance in A549 cells ( Figure 2F). The miR-660 inhibitor notably increased the proliferation and colony formation capacity in A549 cells with cisplatin treatment. Meanwhile, the cell apoptosis rate and G1 cell cycle arrest decreased, and S-G2 cell cycle arrest increased in cisplatintreated A549 cells ( Figure 2G-J). In addition, miR-660 overexpression induced a decrease in cell proliferation and an decrease in G2 cell cycle arrest in cisplatin-treated CALU-3 cells ( Figure S3C-F). Taken together, these results showed that miR-660 could efficiently increase cisplatin sensitivity in LUAD cells.

SATB2 Is a Direct Target of miR-660
Using three target-prediction databases (miRWalk, Targetscan, and starBase), 481 predicted target genes of miR-660 were identified ( Figure 3A). Comparing differential genes from the GSE43493 dataset, which contains mRNA expression profiling data of the A549 and A549/CDDP cell lines, we found 297 upregulated genes in the A549/CDDP group compared with the A549 group ( Figure 3B). Interestingly, there were 9 potential miR-660 downstream target genes also upregulated in the A549/CDDP cell line, namely PTP4A1, HAS3, LFNG, ANKS1A, BCL2, GAS7, SATB2, FAM173B, and MKRN1 ( Figure 3C). We examined the expression and prognosis of these target genes in the starBase database and found that SATB2 was upregulated in LUAD tissues and overexpression of SATB2 was related to poor prognosis in patients with LUAD ( Figures 3D,E and S4). To verify the results of the bioinformatics analysis, we detected the expression of SATB2 in paired LUAD and normal lung tissues using RT-qPCR assay. We found that SATB2 was indeed overexpressed in most tumor tissues ( Figure 3F). We obtained the target sequences of miR-660 and SATB2. We found that miR-660 overexpression repressed the luciferase activity of reporter plasmids containing SATB2-WT in HEK293T cells, while the luciferase activity of SATB2-MUT plasmids was not altered ( Figure 3G,H). In A549/CDDP cells, upregulated miR-660 led to an obvious decrease in SATB2 expression. Conversely, suppressing miR-660 resulted in upregulation of SATB2 expression in A549 cells ( Figures 3I,J and S5A). These results suggest that SATB2 could be a direct target of miR-660 in LUAD cells. The expression relationship between miR-660 and SATB2 was determined by RT-qPCR and Western blot. Data are presented as mean ± standard deviation. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

miR-660 Regulates Cisplatin Sensitivity in LUAD Cells through SATB2
We detected SATB2 expression in A549 and A549/CDDP cells. After cisplatin treatment, compared with A549 cells, the results of RT-qPCR and Western blot showed that SATB2 was upregulated in A549/CDDP cells (Figures 4A,B and S5B). To observe the function of SATB2 in LUAD cisplatin resistance, we first transfected SATB2 overexpression plasmid into A549 cells and shSATB2 into A549/CDDP cells ( Figures 4C,D and S5C). The cisplatin IC 50 detection showed that SATB2 overexpression increased cisplatin resistance and promoted cell proliferation in A549 cells ( Figure 4E-G). In contrast, cisplatin IC 50 detection showed that SATB2 knockdown decreased cisplatin resistance and inhibited cell proliferation in A549/CDDP cells ( Figure 4H-J). These results proved that SATB2 could indeed promote cisplatin resistance in LUAD cells.
Finally, we conducted rescue experiments to verify whether miR-660 relied on SATB2 to influence cisplatin resistance in LUAD. In cisplatin-treated A549/CDDP cells, the decrease in cisplatin IC 50 value and proliferation capacity caused by miR-660 upregulation was completely rescued by SATB2 overexpression (Figure 5A-C). Moreover, co-transfection of miR-660 inhibitor and shSATB2 plasmid in cisplatin-treated A549 cells weakened the enhanced cisplatin resistance brought by miR-660 knockdown ( Figure 5D-F). Taken together, we clarified that miR-660 inhibits LUAD cisplatin resistance by decreasing SATB2 expression.  (E,F) With cisplatin treatment, the capacities of cell proliferation and colony formation of A549 cells in inhibitor CTR, miR-660 inhibitor, and miR-660 inhibitor + shSATB2 groups were examined using CCK8 and colony formation assays. Data are presented as the mean ± standard deviation. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Discussion
To obtain candidate miRNAs related to cisplatin resistance in LUAD, we analyzed the LUAD cisplatin-resistant/sensitive cell dataset GSE84200 and obtained 65 miRNAs with downregulation in LUAD cisplatin-resistant cells. After successful establishment of the A549/CDDP cell line, we found that miR-660 was not only lowly expressed in LUAD cells and tissues but also further decreased in the LUAD cisplatin-resistant cell line. In  -660 inhibitor, and miR-660 inhibitor + shSATB2 groups were examined using CCK8 and colony formation assays. Data are presented as the mean ± standard deviation. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Discussion
To obtain candidate miRNAs related to cisplatin resistance in LUAD, we analyzed the LUAD cisplatin-resistant/sensitive cell dataset GSE84200 and obtained 65 miRNAs with downregulation in LUAD cisplatin-resistant cells. After successful establishment of the A549/CDDP cell line, we found that miR-660 was not only lowly expressed in LUAD cells and tissues but also further decreased in the LUAD cisplatin-resistant cell line. In addition, miR-660 downregulation was associated with poor prognosis in LUAD patients.
miR-660 was upregulated in breast and liver cancer [25][26][27][28]. miR-660 had low expression in lung cancer and targeted MDM2, which induced destabilization of the p53 protein, thus promoting cell proliferation in lung cancer [29]. miR-660 could inhibit lung cancer progression and bone metastasis by regulating SMARCA5 expression [30]. CircFARP1 increased leukemia inhibitory factor (LIF) expression through sponging miR-660, which enhanced gemcitabine resistance in pancreatic ductal adenocarcinoma [31]. Until now, there has been no research report on the relationship between miR-660 and cisplatin resistance in LUAD. Cisplatin promotes tumor cell apoptosis by forming an adduct with the DNA of tumor cells, thereby achieving its anti-tumor effect [32]. Many studies have confirmed that the decrease in cell cycle arrest is the key factor in tumor cisplatin resistance [33][34][35]. We found that the overexpression of miR-660 decreased the cell proliferation capacity and increased the cell apoptosis rate and cell cycle arrest in cisplatin-treated A549/CDDP and CALU-3 cells. These results suggest that miR-660 may be a key regulator in LUAD cisplatin sensitivity.
To identify the potential mechanism of miR-660 in cisplatin resistance, we used multiple miRNA databases and analyzed upregulated mRNAs in cisplatin resistance A549 cells. We found that miR-660-targeted SATB2 was overexpressed in LUAD tissues and A549/CDDP cells, and the prognosis of LUAD patients with SATB2 overexpression was poor. SATB2 was found to be overexpressed in a variety of cancers, including colorectal cancer, hepatocellular carcinoma, and neuroendocrine tumors, and played an important role in the occurrence, development, and metastasis of cancers [36][37][38][39][40][41][42][43][44]. In head and neck cancer, SATB2 could bind with ∆Np63a to inhibit downstream Tap73β transcription which enhanced chemotherapy resistance of cancer cells [45]. In our research, a dual-luciferase reporter assay confirmed the combination of miR-660 and SATB2. RT-qPCR and Western blot simultaneously proved that the expression of miR-660 and SATB2 was negatively correlated. After cisplatin treatment, we found that SATB2 was overexpressed in A549/CDDP cells and SATB2 knockdown decreased the proliferation ability of A549/CDDP cells. SATB2 overexpression enhanced cisplatin resistance in A549 cells. Moreover, rescue experiments verified that miR-660 regulated cisplatin sensitivity in LUAD cells through SATB2. The expression regulation mechanism of miR-660 in A549/CDDP cells and how SATB2 regulated cisplatin resistance in LUAD will be studied in future studies.
In this study, we found that the downregulation of miR-660 increased the cisplatin resistance of LUAD cells by enhancing the expression of SATB2. miR-660/SATB2 axis may be a key regulator of cisplatin treatment effect in LUAD.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/genes14040911/s1, Table S1 Clinical information of 30 LUAD patients. Figure S1 Expression of 16 candidate miRNAs in multiple tumors, according to the dbDEMC database. Figure S2 The relationship between the expression of 13 remaining candidate miRNAs and prognosis of LUAD patients, according to the starbase database. Figure S3A The expression of miR-195 and miR-140 in LUAD cells was detected using RT-qPCR. Figure S3B The effects of miR-660 mimics and inhibitors were determined by RT-qPCR. Figure S3C-F With cisplatin treatment, the capacities of cell proliferation and colony formation in CALU-3 mimics miR-660 cells and the corresponding control cells were examined using CCK8 and colony formation assays. The apoptosis rate and cell cycle detection were determined by flow cytometry assays. Data are presented as the mean ± standard deviation. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Figure S4 The expression and survival analysis results of the remaining eight miR-660 potential target genes, according to the starbase database. Figure S5A The expression relationship between miR-660 and SATB2 was determined by Western blot. Figure S5B Expression levels of SATB2 in A549 and A549/CDDP cells were measured by Western blot. Figure S5C The effects of SATB2 overexpression and knockdown were determined by Western blot. Data are presented as the mean ± standard deviation. * p < 0.05, ** p < 0.01, *** p < 0.001.
Author Contributions: Z.W., Z.L. and Y.C. designed the project. Z.W. designed and performed most of the experiments and drafted the manuscript. L.Z., B.C., W.X. and A.W. collected tissue samples from the patients and assisted in completing some experiments. X.L., Q.Z. and L.F. provided technical assistance and material support. All authors have read and agreed to the published version of the manuscript. Data Availability Statement: Data used or analyzed during the current study are available from the corresponding author upon reasonable request.