The Molecular Pathogenesis of Tumor-Suppressive miR-486-5p and miR-486-3p Target Genes: GINS4 Facilitates Aggressiveness in Lung Adenocarcinoma

Simple Summary Two microRNAs (miRNAs) (miR-486-5p and miR-486-3p) derived from pre-miR-486 acted as tumor-suppressive miRNAs in lung adenocarcinoma (LUAD). We identified seven genes (MKI67, GINS4, RRM2, HELLS, MELK, TIMELESS, and SAPCD2) involved in the malignant phenotype of LUAD cells coordinately regulated by these miRNAs. It is possible to suppress the malignant transformation of LUAD by controlling these genes. Abstract The involvement of passenger strands of miRNAs in the molecular pathogenesis of human cancers is a recent concept in miRNA research, and it will broaden our understanding of the molecular mechanisms of miRNA-mediated cancer. The analysis of our miRNA signature of LUAD revealed that both strands of pre-miR-486 (miR-486-5p and miR-486-3p) were downregulated in LUAD tissues. Ectopic expression of both miRNAs induced cell cycle arrest in LUAD cells, suggesting both strands of miRNAs derived from pre-miR-486 were tumor suppressive. Our in silico analysis showed a total of 99 genes may be under the control of both miRNAs in LUAD cells. Importantly, among these targets, the high expression of seven genes (MKI67, GINS4, RRM2, HELLS, MELK, TIMELESS, and SAPCD2) predicted a poorer prognosis of LUAD patients (p < 0.05). We focused on GINS4, a DNA replication complex GINS protein that plays an essential role in the initiation of DNA replication. Our functional assays showed that GINS4 was directly controlled by both strands of pre-miR-486, and its aberrant expression facilitated the aggressive behavior of LUAD cells. GINS4 is attractive as a therapeutic target for this disease. MiRNA analysis, including passenger strands, will further improve our understanding of the molecular pathogenesis of LUAD.


Introduction
Lung cancer is the leading cause of cancer-related deaths in men and women worldwide, with approximately 2.3 million people diagnosed with lung cancer and approximately 1.8 million deaths from lung cancer each year [1]. Lung cancer is divided histologically into two groups: small cell lung cancer (SCLC), which accounts for 15% of patients, and non-SCLC (NSCLC), which accounts for 85% [2]. NSCLC are subdivided into squamous cell carcinoma (LUSQ), large cell carcinoma, and lung adenocarcinoma (LUAD); the latter accounts for approximately 60% of NSCLC [2]. 2 of 20 The curative treatment for lung cancer patients at an early stage of the disease (stage I or II) is surgical resection; however, even those who undergo radical surgery have a 5-year survival rate of only 65% [3]. However, fewer than 30% of lung cancer patients are diagnosed early in the course of disease and proceed to surgical treatment. Patients diagnosed with advanced-stage lung cancer have a very poor prognosis, with only 20% of patients surviving 5 years [2,4].
Therapeutic strategies for advanced stage LUAD are developing in a remarkable way. Molecular-targeted drugs that counteract driver gene alteration (e.g., EGFR mutation, ALK rearrangement, ROS1 rearrangement, BRAF mutation, MET mutation, and KRAS mutation), and immune checkpoint inhibitors are showing therapeutic effects [5,6]. However, the prognosis for lung cancer patients remains extremely poor, with only 20% of those diagnosed at an advanced stage surviving for five years [2,4]. Therefore, the search for new diagnostic biomarkers and therapeutic target molecules is an important research theme for improving the prognosis of patients with LUAD.
Numerous non-coding RNAs (ncRNAs) are involved in a wide variety of biologic functions, e.g., basic metabolism and cell differentiation. At present, ncRNAs are thought to play important roles for maintaining cellular homeostasis [7][8][9]. A large number of studies have shown that various ncRNAs are aberrantly expressed in diverse tumors, indicating that such RNAs play important roles in tumorigenesis and development [10][11][12]. Among ncRNAs, miRNAs are small, single-stranded ncRNAs (19)(20)(21)(22) nucleotides in length). They act as fine tuners of gene expression and modulate almost all biological processes [13,14]. Aberrant expressions of miRNAs are frequently detected in a wide range of cancers, including LUAD. Aberrant-expressed miRNAs are closely involved in the malignant transformation of human cancers, e.g., proliferation, metastasis, and resistance [15,16].
In the previous concept of miRNA biogenesis, only the guide strand of miRNAs derived from pre-miRNAs were actually functional miRNAs in cells. On the other hand, the passenger strand was thought to be broken down inside the cell and to have no function. In contrast to the miRNA theory, some passenger strands of miRNAs have been shown to regulate their target molecules [17,18]. These studies indicate that an miRNA analysis of gene regulation requires the inclusion of both the guide and passenger strands. For example, our recent studies on NSCLC cells demonstrated that both strands of miR-99a, miR-144, miR-145, and miR-150, had tumor-suppressive activity through their control of several oncogenic genes [19][20][21][22]. Based on these studies, we hypothesized that genes that are commonly regulated by both strands derived from pre-miRNA are highly involved in the oncogenesis of LUAD.
To identify dysregulated miRNAs in LUAD clinical tissues, we generated miRNA expression signatures by using small RNA sequencing technology. The analysis of the signatures revealed that both strands of pre-miR-486 (miR-486-5p, the guide strand, and miR-486-3p, the passenger strand) were downregulated in LUAD tissues. Moreover, they acted as antitumor miRNAs in our functional assays. Importantly, seven genes (MKI67, GINS4, RRM2, HELLS, MELK, TIMELESS, and SAPCD2) commonly regulated by miR-486-5p and miR-486-3p were closely involved in the molecular pathogenesis of LUAD. Furthermore, the aberrant expression of GINS4, a DNA replication complex GINS protein, facilitated LUAD cell aggressiveness.
Our signature-based miRNA analysis accelerates the discovery of genes closely involved in LUAD tumorigenesis. These genes are potential therapeutic targets for this disease.

Clinical Course of Patients with LUAD Cells
We obtained primary lesions and normal lung tissues from lung adenocarcinoma patients. The background and clinical characteristics of the patients are described in Table S1.

Cell Lines and Cell Culture
Two LUAD cell lines, A549 and H1299, were used in this study (American-Type Culture Collection (ATCC), Manassas, VA, USA). We have previously described the method of cell maintenance [21,23].

Construction of the miRNA Expression Signature in LUAD Based on RNA Sequencing
LUAD and normal lung specimens were sequenced using a the NextSeq 500 instrument (Illumina, Inc., San Diego, CA, USA) to evaluate miRNA expression. The raw sequencing data were registered in Gene Expression Omnibus (GEO; GEO accession number: GSE230229).
2.6. Transfection with siRNA and miRNA siRNA and miRNA were transfected into cell lines using Opti-MEM (catalog no.: 31985070, Gibco, Carlsbad, CA, USA) and LipofectamineTM RNAiMax Transfection Reagent (catalog no.: 13778150, Invitrogen, Carlsbad, CA, USA). Transfection protocols for siRNA and miRNA were described in our previous studies [21,23,24]. siRNA and miRNA used in this study are listed in Table S2.

RNA Extraction and Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR)
Total RNA obtained from LUAD cell lines was isolated using Isogen II (catalog no.: 311-07361, NIPPON GENE Co., Ltd., Tokyo, Japan). cDNA was synthesized using Prime-ScriptTM RT Master Mix (catalog no.: RR036A, Takara Bio Inc., Shiga, Japan). Gene expression was analyzed by real-time PCR using a SYBR green assay (ThermoFisher Scientific, Rockford, IL, USA) on a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The internal control used in the gene expression assays was glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The reagents used in this study are listed in Table S2 Pure Chemical Corporation, Osaka, Japan) in TBST. The signal was detected using Amersham ECL Prime Western Blotting Detection Reagent (Cytiva, Marlborough, MA, USA). The reagents used in this study are listed in Table S2.

Cell Proliferation and Cell Cycle Assays
Cell proliferation was evaluated with XTT assays using Cell Proliferation Kits (catalog no.: 20-300-1000, Biological Industries, Beit-Haemek, Israel). The cell cycle was evaluated using a BD CycletestTM Plus DNA Reagent Kit (catalog no.: 340242, BD Biosciences, Franklin Lakes, NJ, USA) and flow cytometry (BD FACSCelestaTM Flow Cytometer, BD Biosciences). The procedures for assessing cell proliferation and cell cycle behaviors were described previously [21,23,24].

Plasmid Construction and Dual-Luciferase Reporter Assays
The following two sequences were cloned into the psiCHECK-2 vector (C8021; Promega, Madison, WI, USA): the wild-type sequence of the 3 -untranslated regions (UTRs) of GINS4 and the deletion-type sequence, which lacked the miR-486-5p and miR-486-3p target sites of GINS4. The procedures for the transfection and dual-luciferase reporter assays were provided previously [21,23,24].

Immunohistochemical Staining
GINS4 expression was evaluated by immunohistochemical staining using tissue microarray slides (catalog no.: LC811a, US Biomax, Inc. Derwood, MD, USA). The VEC-TASTAIN Universal Elite ABC Kit (catalog no.: PK-6200, Vector Laboratories, Burlingame, CA, USA) was used for blocking, the primary antibody reaction, the secondary antibody reaction, and the binding of avidin to the biotin complex. Primary antibodies were diluted with Dako Real antibody diluent (catalog no.: K5007, Agilent, Santa Clara, CA, USA). Dako REALTM EnVisionTM Detection System Peroxidase/DAB+, Rabbit/Mouse (Agilent) was used to develop the chromogenic reaction. The primary antibody used in this study is described in Table S2. Clinical tissue information is presented in Table S3.

Putative miRNA Binding to GINS4 and miRNA Expression
We obtained the data for putative miRNA binding to GINS4 from TargetScanHuman database (release 8.0) and the data for miRNA expression levels from FIREBROWSE (http: //firebrowse.org/, accessed on 31 October 2022) and Genomic Data Commons (GDC) Data Portal (https://portal.gdc.cancer.gov/) (accessed on 31 October 2022).

Statistical Analysis
Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA) and R ver. 4.2.1 (R Core Team, Vienna, Austria; https://www.R-project. org/, accessed on 3 September 2022). The differences between 2 groups were analyzed by Student's tor Mann-Whitney U tests. Multiple group comparison was achieved using a one-way analysis of variance (ANOVA) and Tukey's tests for post hoc analysis. Survival rates were analyzed by Kaplan-Meier survival curves and the log-rank test.

Selection of Downregulated miRNAs in LUAD Clinical Specimens by Small RNA Sequencing
To create the miRNA expression signatures of LUAD, we prepared nine cDNA libraries obtained from clinical specimens (five LUAD tissues and four normal lung tissues) and performed RNA sequencing. The clinical information for the LUAD tissues is summarized in Table S1. The processing of the data based on the RNA sequencing analysis and details of analyzed small RNA taxonomies are presented in Table S4. We successfully mapped a sufficient number of miRNA reads to the human genome (Table S4). We analyzed RNA sequence data (a total of 41 miRNAs) in the LUAD tissues for comparison with normal lung tissues (Table 1, Figure 1A). We found they were significantly downregulated (log 2 fold-change < −2.0 and p-value < 0.05). Interestingly, our signature revealed that both the guide and passenger strands of four miRNAs (miR-34c, miR-486, miR-34b, and miR-144) were downregulated in the LUAD tissues ( Table 1). The involvement of passenger strands of miRNAs derived from pre-miRNAs in the molecular pathogenesis of human cancers is a recent concept in miRNA biology.

Expression Levels of miR-486-5p and miR-486-3p in LUAD Specimens and Cell Lines
To confirm our miRNA signature, we evaluated the expression levels of miR-486 and miR-486-3p in LUAD tissues and normal lung tissues.

Antitumor Functions of miR-486-5p and miR-486-3p in LUAD Cells
In order to prove that both strands of pre-miR-486 had antitumor functions in LU cells, we performed an ectopic expression of these miRNAs in LUAD cells (A549 a H1299), and then investigated the behavior of the cancer cells.

Expression Levels of miR-486-5p and miR-486-3p in LUAD Specimens and Cell Lines
To confirm our miRNA signature, we evaluated the expression levels of miR-486-5p and miR-486-3p in LUAD tissues and normal lung tissues.

Antitumor Functions of miR-486-5p and miR-486-3p in LUAD Cells
In order to prove that both strands of pre-miR-486 had antitumor functions in LUAD cells, we performed an ectopic expression of these miRNAs in LUAD cells (A549 and H1299), and then investigated the behavior of the cancer cells.
Based on these results, we conclude that the two types of miRNAs derived from pre-miR-486 are tumor suppressive miRNAs in LUAD cells.

Identification of Genes Controlled by miR-486-5p and miR-486-3p in LUAD Cells
The fact that both miRNA strands derived from pre-miR-486 acted as antitumor miR-NAs was quite interesting. The subsequent challenge is to identify the oncogenic targets controlled by these miRNAs in LUAD cells.
Our strategy for the search for miRNAs targets is shown in Figure 3. In this study, we obtained genome-wide gene expression data using miR-486-5p-or miR-486-3p-transfected A549 cells. Our gene expression data were deposited in the GEO database (accession number: GSE230056). Based on these results, we conclude that the two types of miRNAs derived from pre-miR-486 are tumor suppressive miRNAs in LUAD cells.

Identification of Genes Controlled by miR-486-5p and miR-486-3p in LUAD Cells
The fact that both miRNA strands derived from pre-miR-486 acted as antitumor miRNAs was quite interesting. The subsequent challenge is to identify the oncogenic targets controlled by these miRNAs in LUAD cells.
Our strategy for the search for miRNAs targets is shown in Figure 3. In this study, we obtained genome-wide gene expression data using miR-486-5por miR-486-3p-transfected A549 cells. Our gene expression data were deposited in the GEO database (accession number: GSE230056). Using a combination of the TargetScan database and miRNA-transfected LUAD cells expression data, we searched for putative targets controlled by miR-486-5p (635 genes) and miR-486-3p (2118 genes). Notably, 99 genes were shown to be potential targets of both miR-486-5p and miR-486-3p in LUAD cells (Table 2). To identify miR-486-5p or miR-486-3p targets, we used the TargetScanHuman (release 8.0) database and gene expression profiles generated after miR-486-5p or miR-486-3p were transfected into A549 cells. Our original gene expression array data were deposited in the GEO database (accession number: GSE230056). In total, 99 genes were identified as possibly being controlled by both miR-486-5p and miR-486-3p in A549 cells.

Expression and Clinical Significance of Both Strands of Pre-miR-486 Target Genes in LUAD
A further analysis of these 99 genes was performed to search for those that promoted cancer in LUAD cells.
We validated the expression levels of these 99 target genes using a large amount of clinical data (TCGA-LUAD). The expression of seven genes (MKI67, GINS4, RRM2, HELLS, MELK, TIMELESS, and SAPCD2) was upregulated in the LUAD tissues (n = 499) compared with normal lung tissues (n = 58) (Figure 4).    A clinicopathological analysis of the seven genes was performed using TCGA-LUAD datasets. Kaplan-Meier curve (5-year survival rates) analysis was performed according to the expression levels of the seven genes. The high expression of all genes significantly affected the poorer survival rates of the patients ( Figure 5).
In our previous analysis, we focused on the genes involved in DNA replication [25,26]. In this study, we focused on GINS4 and investigated the oncogenic roles of its aberrant expression in LUAD cells.
We confirmed that GINS4 expression in LUAD cells was suppressed at the mRNA and protein levels by the ectopic expression of miR-486-5p or miR-496-3p ( Figure 6A,B). Full-size images of Western blots are shown in Figure S2.
To demonstrate that both miRNAs directly bound to the 3 -UTR of the GINS4 gene in a sequence-dependent manner, we performed dual-luciferase reporter assays. The putative miR-486-5p binding site on the 3 -UTR of the GINS4 gene is shown in Figure 7A. Luciferase activity was significantly reduced when co-transfected with miR-486-5p and a vector containing binding sites for the 3 -UTR of GINS4 ( Figure 7C). However, luciferase activity did not change when co-transfected with miR-486-5p and a vector lacking the miR-486-5p binding site ( Figure 7C). Thus, miR-486-5p appeared to directly bind to the 3 -UTR of GINS4 in a sequence-dependent manner.
In our previous analysis, we focused on the genes involved in DNA replication [25,26]. In this study, we focused on GINS4 and investigated the oncogenic roles of it aberrant expression in LUAD cells.
We confirmed that GINS4 expression in LUAD cells was suppressed at the mRNA and protein levels by the ectopic expression of miR-486-5p or miR-496-3p ( Figure 6A,B) Full-size images of Western blots are shown in Figure S2.
To demonstrate that both miRNAs directly bound to the 3′-UTR of the GINS4 gen in a sequence-dependent manner, we performed dual-luciferase reporter assays. The pu tative miR-486-5p binding site on the 3′-UTR of the GINS4 gene is shown in Figure 7A Luciferase activity was significantly reduced when co-transfected with miR-486-5p and vector containing binding sites for the 3′-UTR of GINS4 ( Figure 7C). However, luciferas activity did not change when co-transfected with miR-486-5p and a vector lacking the miR 486-5p binding site ( Figure 7C). Thus, miR-486-5p appeared to directly bind to the 3′-UTR of GINS4 in a sequence-dependent manner.
We also investigated the sequence-dependent direct binding of miR-486-3p and th 3′-UTR of the GINS4 gene. The putative miR-486-3p binding site on the 3′-UTR of th GINS4 gene is shown in Figure 7B. Luciferase activity was significantly reduced when co transfected with miR-486-3p and a vector containing binding sites for the 3′-UTR of GINS ( Figure 7D). However, there was no change after the co-transfection of miR-486-3p and vector lacking the miR-486-3p binding site ( Figure 7D). These results indicate that miR 486-3p directly binds to the 3′-UTR of GINS4 in a sequence-dependent manner.

Functional Significance of GINS4 in LUAD Cells
To investigate the oncogenic function of GINS4 in LUAD cells, we made use of knockdown assays with siRNAs that were transfected into LUAD cells (A549 and H1299). We evaluated the knockdown efficiency of several siRNAs (siGINS4-1, siGINS4-2, and siGINS4-3) for GINS4. Transient transfection with three types of siRNAs significantly reduced GINS4 mRNA and protein expression in LUAD cells (Figures S3 and S4).
LUAD cell proliferation assays showed that cell growth was inhibited by suppressing the expression of GINS4 ( Figure 8A). Moreover, cell cycle assays demonstrated that cell cycle arrest in the G0/G1 phase after the expression of GINS4 was suppressed in LUAD cells ( Figure 8B). These results suggest that GINS4 is a cancer-promoting gene that modulates cell cycle progression. Dual-luciferase reporter assays showed reduced luminescence activity after co-transfection of the wild-type vector (containing the miR-486-5p binding site) with miR-486-5p in A549 cells. In contrast, no luminescence activity was seen after the co-transfection of the deletion-type vector (lacking the miR-486-5p binding site) with miR-486-5p in A549 cells. Similar analytic results were obtained for wild-or deletion-type vectors (with or without the miR-486-3p binding site) and miR-486-3p in A549 cells. ****, p < 0.001; N.S., not significant.
We also investigated the sequence-dependent direct binding of miR-486-3p and the 3 -UTR of the GINS4 gene. The putative miR-486-3p binding site on the 3 -UTR of the GINS4 gene is shown in Figure 7B. Luciferase activity was significantly reduced when co-transfected with miR-486-3p and a vector containing binding sites for the 3 -UTR of GINS4 ( Figure 7D). However, there was no change after the co-transfection of miR-486-3p and a vector lacking the miR-486-3p binding site ( Figure 7D). These results indicate that miR-486-3p directly binds to the 3 -UTR of GINS4 in a sequence-dependent manner.

Functional Significance of GINS4 in LUAD Cells
To investigate the oncogenic function of GINS4 in LUAD cells, we made use of knockdown assays with siRNAs that were transfected into LUAD cells (A549 and H1299). We evaluated the knockdown efficiency of several siRNAs (siGINS4-1, siGINS4-2, and siGINS4-3) for GINS4. Transient transfection with three types of siRNAs significantly reduced GINS4 mRNA and protein expression in LUAD cells (Figures S3 and S4).
LUAD cell proliferation assays showed that cell growth was inhibited by suppressing the expression of GINS4 ( Figure 8A). Moreover, cell cycle assays demonstrated that cell cycle arrest in the G0/G1 phase after the expression of GINS4 was suppressed in LUAD cells ( Figure 8B). These results suggest that GINS4 is a cancer-promoting gene that modulates cell cycle progression.
duced GINS4 mRNA and protein expression in LUAD cells (Figures S3 and S4).
LUAD cell proliferation assays showed that cell growth was inhibited by suppressing the expression of GINS4 ( Figure 8A). Moreover, cell cycle assays demonstrated that cell cycle arrest in the G0/G1 phase after the expression of GINS4 was suppressed in LUAD cells ( Figure 8B). These results suggest that GINS4 is a cancer-promoting gene that modulates cell cycle progression.

Immunostaining of GINS4 in LUAD Clinical Tissues
We examined the expression of GINS4 in tissue microarray studies. Compared with normal tissues, the GINS4 protein was overexpressed in LUAD tissues. In particular, cancer cells showed heavy cytoplasmic staining ( Figure 9).

Immunostaining of GINS4 in LUAD Clinical Tissues
We examined the expression of GINS4 in tissue microarray studies. Compared wi normal tissues, the GINS4 protein was overexpressed in LUAD tissues. In particular, ca cer cells showed heavy cytoplasmic staining (Figure 9).

GINS4-Mediated Pathways Determined by Gene Set Enrichment Analysis (GSEA)
To investigate GINS4-modulated molecular pathways in LUAD cells, we used t

GINS4-Mediated Pathways Determined by Gene Set Enrichment Analysis (GSEA)
To investigate GINS4-modulated molecular pathways in LUAD cells, we used the Gene Set Enrichment Analysis (GSEA) based on TCGA-LUAD RNA sequencing data. "Cell cycle", "DNA replication", "homologous recombination", and "oocyte meiosis" pathways were enriched in patients with high expression of GINS4 compared to lowexpression patients ( Figure 10, Table 3).

Discussion
During classic miRNA maturation, pre-miRNA is transported to the cytoplasm where it is cleaved by Dicer to become a miRNA duplex. One strand derived from the miRNA duplex is incorporated into the RNA-Induced Silencing Complex (RISC), where it regulates specific target genes. That strand is defined as the guide strand. The nonloaded strand (the passenger strand) is degraded in the cytoplasm, as it has no known function [27]. However, in recent studies, both strands of the miRNA duplex were shown to be functional [27,28]. In this study, we confirmed that both strands of pre-miR-486 had tumor suppressive functions by regulating their respective target genes.
miR-486 is transcribed from the intron of the host gene ANK1 (Ankyrin 1) on human chromosome 8p11.21 [29]. There have been many reports describing miR-486-5p (the guide strand) in various cancer types [30][31][32][33]. Previous studies showed that the expression of miR-486-5p was reduced in cancer tissues, and that it functions as a tumor suppressor in breast cancer, colorectal cancer, gastric cancer, hepatocellular carcinoma, and renal cell carcinoma [34]. In contrast to the preceding examples, the overexpression of miR-486-5p was observed in prostate cancer, and its expression is associated with the malignant transformation of these cancers [35].
In previous reports of lung cancer, the function of miR-486-5p was that of tumor suppression, which is consistent with our results [31]. For example, the expression of miR-486-5p blocked mTOR pathways through its targeting of ribosomal proteins S6 kinase A1 and B1 [36]. Moreover, the overexpression of miR-486-5p attenuated tumor growth and

Discussion
During classic miRNA maturation, pre-miRNA is transported to the cytoplasm where it is cleaved by Dicer to become a miRNA duplex. One strand derived from the miRNA duplex is incorporated into the RNA-Induced Silencing Complex (RISC), where it regulates specific target genes. That strand is defined as the guide strand. The non-loaded strand (the passenger strand) is degraded in the cytoplasm, as it has no known function [27]. However, in recent studies, both strands of the miRNA duplex were shown to be functional [27,28]. In this study, we confirmed that both strands of pre-miR-486 had tumor suppressive functions by regulating their respective target genes.
miR-486 is transcribed from the intron of the host gene ANK1 (Ankyrin 1) on human chromosome 8p11.21 [29]. There have been many reports describing miR-486-5p (the guide strand) in various cancer types [30][31][32][33]. Previous studies showed that the expression of miR-486-5p was reduced in cancer tissues, and that it functions as a tumor suppressor in breast cancer, colorectal cancer, gastric cancer, hepatocellular carcinoma, and renal cell carcinoma [34]. In contrast to the preceding examples, the overexpression of miR-486-5p was observed in prostate cancer, and its expression is associated with the malignant transformation of these cancers [35].
In previous reports of lung cancer, the function of miR-486-5p was that of tumor suppression, which is consistent with our results [31]. For example, the expression of miR-486-5p blocked mTOR pathways through its targeting of ribosomal proteins S6 kinase A1 and B1 [36]. Moreover, the overexpression of miR-486-5p attenuated tumor growth and inhibited metastasis according to in vivo assays [36,37]. A recent study showed that the anesthetic propofol induced the expression of miR-486-5p, resulting in the inhibition of the Ras-associated protein1-NF-κB pathway [38]. These events contributed to cisplatinsensitivity in NSCLC cells [38].
Since both strands of pre-miR-486 are tumor suppressive, our subsequent interest is to elucidate the molecular networks regulated by these miRNAs in LUAD cells. Numerous studies have characterized the target molecules of miRNAs, including miR-486-5p and miR-486-3p; however, none have searched for common targets of these miRNAs in LUAD cells. Our study revealed that 99 genes were putative targets of miR-486-5p and miR-486-3p regulations in LUAD cells. In fact, all 99 genes were regulated by both miR-486-5p and miR-486-3p in LUAD cells. It should be noted that high expression of seven genes (MKI67, GINS4, RRM2, HELLS, MELK, TIMELESS, and SAPCD2) had a negative impact on the prognosis of patients with LUAD. These genes are important for elucidating the molecular mechanisms of lung cancer malignancy.
For these genes, we referred to reports of genes under microRNA regulation in lung cancer cells. The RRM2 protein is one of the two subunits of the ribonucleotide reductase complex. This reductase is a key enzyme in DNA synthesis as it catalyzes the formation of deoxyribonucleotides from ribonucleotides [45]. Previous studies showed that the overexpression of RRM2 was detected in a wide range of cancers, including lung cancer [45,46]. A recent study showed that the expression of miR-203-3p was reduced in LUAD tissues, and its overexpression inhibited LUAD aggressive phenotypes through targeting RRM2 in LUAD cells [47]. Our previous study showed that miR-150-3p (the passenger strand) was significantly suppressed in LUSQ tissues, and performed a tumor-suppressive role in LUSQ cells via controlled several cell cycle-related genes, including HELLS [48]. HELLS belongs to the SNF2 family of chromatin-remodeling ATPases and is recruited to specific DNA sites to control the transcription of targeted genes [49,50]. Our data demonstrate that HELLS is directly regulated by miR-150-3p in LUSQ cells [48]. Previous reports revealed that SAPCD2 is highly expressed in various cancers and is highly involved in the malignant transformation of cancer cells [51]. Numerous studies have shown that SAPCD2 interacts with multiple proteins within the cell cycle interaction network and functions as a mitotic phase-promoting factor [51]. Notably, a recent study showed that miR-486-5p suppressed cell malignant progression in LUAD cells by targeting SAPCD2 [52]. This fact is consistent with our data and indicates that miR-486-5p-mediated molecular networks have pivotal effects on LUAD cell malignancy.
Among these targets, we focused on GINS4, and we showed that its expression facilitated the malignant transformation of LUAD cells. GINS4 is a member of the GINS complex, which consists of four different subunits, e.g., GINS1 to GINS4 [53]. Precisely maintained genomic DNA replication is critical for all forms of cellular life and requires a complex interplay of various protein factors. DNA helicases play a key role in unwinding double-stranded DNA during replication, recombination, and repair processes. The GINS complex forms the CMG (Cdc45-MCMs-GINS) complex with MCM (mini-chromosome maintenance) and CDC45. This complex functions as a replicative helicase that unwinds double-stranded DNA during chromosome replication [53,54].
The overexpression of GINS4 occurs in breast cancer, colorectal carcinoma, bladder cancer, pancreatic ductal adenocarcinoma, glioma, and gastric cancer [55]. In NSCLC cells, lymphoid-specific helicase binds to the 3 -UTR region of GINS4 and stabilizes GINS4 expression [56]. The overexpression of GINS4 facilitates lung cancer malignant transformation. The aberrant expression of other members of GINS (GINS1, GINS2, and GINS3) occurs in different types of human cancers [57,58]. The TCGA database analysis revealed that all members of GINS were upregulated in LUAD tissues and their high expression predicted the prognosis of the patients. Several studies demonstrated that the expression of a GINS member enhanced cancer cell aggressiveness, e.g., proliferation, drug resistance, and epithelial-mesenchymal transition [55,59,60]. Therefore, GINS members are closely involved with LUAD pathogenesis and may be potential therapeutic targets.

Conclusions
The analysis of the miRNA expression signature revealed that both strands of pre-miR-486 (miR-486-5p and miR-486-3p) were downregulated in LUAD tissues. From this study and previous reports, we confirmed that these miRNAs had tumor-suppressive functions in LUAD cells. A total of 99 genes were identified as cooperatively controlled by miR-486-5p and miR-486-3p in LUAD cells. Among these targets, seven genes (MKI67, GINS4, RRM2, HELLS, MELK, TIMELESS, and SAPCD2) were closely involved in the molecular pathogenesis of LUAD. Furthermore, GINS4 was directly regulated by these two miRNAs and the overexpression of GINS4 facilitated LUAD cell aggressiveness. Based on the tumor-suppressive miRNA analysis, it was possible to identify miRNA target genes closely involved in the molecular pathogenesis of LUAD.