Involvement of Dual Strands of miR-143 (miR-143-5p and miR-143-3p) and Their Target Oncogenes in the Molecular Pathogenesis of Lung Adenocarcinoma

Our analyses of tumor-suppressive microRNAs (miRNAs) and their target oncogenes have identified novel molecular networks in lung adenocarcinoma (LUAD). Moreover, our recent studies revealed that some passenger strands of miRNAs contribute to cancer cell malignant transformation. Downregulation of both strands of the miR-143 duplex was observed in LUAD clinical specimens. Ectopic expression of these miRNAs suppressed malignant phenotypes in cancer cells, suggesting that these miRNAs have tumor-suppressive activities in LUAD cells. Here, we evaluated miR-143-5p molecular networks in LUAD using genome-wide gene expression and miRNA database analyses. Twenty-two genes were identified as potential miR-143-5p-controlled genes in LUAD cells. Interestingly, the expression of 11 genes (MCM4, RAD51, FAM111B, CLGN, KRT80, GPC1, MTL5, NETO2, FANCA, MTFR1, and TTLL12) was a prognostic factor for the patients with LUAD. Furthermore, knockdown assays using siRNAs showed that downregulation of MCM4 suppressed cell growth, migration, and invasion in LUAD cells. Aberrant expression of MCM4 was confirmed in the clinical specimens of LUAD. Thus, we showed that miR-143-5p and its target genes were involved in the molecular pathogenesis of LUAD. Identification of tumor-suppressive miRNAs and their target oncogenes may be an effective strategy for elucidation of the molecular oncogenic networks of this disease.


Introduction
Lung cancer (non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC)) is the leading cause of cancer-related deaths among men and women in developed countries, including Japan. In 2018, approximately 2,300,000 people were diagnosed with lung cancer, and 1,800,000 patients died from this disease [1]. NSCLC (accounting for approximately 80% of lung cancer cases) is divided mainly into three subtypes: lung adenocarcinoma (LUAD), squamous cell carcinoma (LUSQ), and large cell carcinoma [2]. Approximately 40% of patients with lung cancer have LUAD, and this type of lung cancer is often observed in patients (especially young women) who have never smoked, in contrast to other types of lung cancers [3].
LUAD has been extensively studied using molecular biology analyses, and various treatment options have been developed as a result. Based on the mutation statuses of cancer cells, several molecular targeted drugs have been available (e.g., epidermal growth factor receptor tyrosine kinase inhibitors and anaplastic lymphoma kinase inhibitors), and these treatments have dramatically improved patient outcomes [4,5]. However, these treatment effects are limited because acquired resistance and cancer recurrence occur in almost all patients [6].
Human genome research has revealed that the human genome encodes a large number of protein-noncoding RNA molecules, which are transcribed and have various important functions [7]. microRNAs (miRNAs) are small, noncoding RNAs that act as fine-tuners of RNA expression in cells [8]. Notably, bioinformatics predictions have shown that a single miRNA can control a vast number of RNA transcripts in cells, and more than half of RNAs in the human genome are controlled by miRNAs [9]. The search for molecular networks controlled by miRNAs has provided us with novel knowledge of RNA networks in normal and diseased cells.
Identification of tumor-suppressive or oncogenic miRNAs and their target cancer-related genes has contributed to understanding of the molecular pathogenesis of human cancers [11]. Based on miRNA expression signatures, we have sequentially identified antitumor miRNAs and their target oncogenes in several cancers [12][13][14][15].
In the conventional miRNA biogenesis process, the guide strand of miRNA derived from the miRNA duplex functions to control the target molecules [16], whereas the passenger strand of miRNA is not incorporated into the RNA-induced silencing complex (RISC), and it is disassembled and has no function [17]. However, in lung cancer, including SCLC, LUAD, and LUSQ, our recent studies showed that both strands of the miRNA duplex (e.g., miR-144-5p/-3p in LUSQ, miR-145-5p/-3p in LUSQ and LUAD, and miR-150-5p/-3p in LUAD) are significantly downregulated in lung cancer tissues, and these miRNAs act as tumor-suppressive miRNAs by targeting several oncogenes [18][19][20][21]. Moreover, the expression levels of miRNAs and their target genes are closely associated with lung cancer pathogenesis [22].
In this study, we focused on both strands of the miR-143 duplex (miR-143-5p: the passenger strand, and miR-143-3p: the guide strand) in LUAD cells. Previous studies have shown that miR-143-3p functions as a tumor-suppressive miRNA in many types of cancers, and the oncogenes regulated by this miRNA duplex were identified [23,24]. On the other hand, functional analysis of miR-143-5p in cancer cells has not previously been reported. Here, we showed that ectopic expression of miR-143-5p attenuated malignant phenotypes in LUAD cells. Furthermore, 11 genes (MCM4, RAD51, FAM111B, CLGN, KRT80, GPC1, MTL5, NETO2, FANCA, MTFR1, and TTLL12) were found to be closely associated with the molecular pathogenesis of LUAD. Identification of tumor-suppressive miRNAs and their oncogenes may be an effective strategy for elucidation of the molecular pathogenesis of this disease.

Downregulation of the miR-143 Duplex (miR-143-5p and miR-143-3p) in LUAD Clinical Specimens and Cell Lines
Expression levels of miR-143-5p and miR-143-3p were markedly reduced in cancer tissues compared with those in normal tissues (p < 0.0001 and p < 0.0001, respectively; Figure 1A,B). The clinical features of LUAD specimens are listed in Table 1 and Table S1. In two LUAD cell lines (A549 and H1299), the expression levels of these miRNAs were very low compared with those in normal tissues ( Figure 1A,B). A positive correlation was detected between miR-143-5p and miR-143-3p expression levels by Spearman's rank analysis (r = 0.9215, p < 0.0001; Figure 1C).  12.5 Cohort analysis using data from The Cancer Genome Atlas (TCGA) database showed that there was no significant relationship between expression of miR-143-5p or miR-143-3p and prognosis of patients with LUAD ( Figure S1).

Ectopic Expression of miR-143-5p and miR-143-3p Blocked LUAD Aggressiveness
To investigate the functions of miR-143-5p and miR-143-3p in LUAD cells, we designed ectopic expression assays to assess cell proliferation, migration, and invasion after transfection of these miRNAs into A549 and H1299 cells. Gain-of-function assays showed that overexpression of miR-143-5p attenuated cell proliferation in A549 and H1299 cells (Figure 2A). However, miR-143-3p expression did not affect cell proliferation in A549 cells (Figure 2A). We counted LUAD cells for the antiproliferative effects of ectopic expression of miR-143-5p and miR-143-3p, and obtained similar  Cohort analysis using data from The Cancer Genome Atlas (TCGA) database showed that there was no significant relationship between expression of miR-143-5p or miR-143-3p and prognosis of patients with LUAD ( Figure S1).

Ectopic Expression of miR-143-5p and miR-143-3p Blocked LUAD Aggressiveness
To investigate the functions of miR-143-5p and miR-143-3p in LUAD cells, we designed ectopic expression assays to assess cell proliferation, migration, and invasion after transfection of these miRNAs into A549 and H1299 cells. Gain-of-function assays showed that overexpression of miR-143-5p attenuated cell proliferation in A549 and H1299 cells ( Figure 2A). However, miR-143-3p expression did not affect cell proliferation in A549 cells (Figure 2A). We counted LUAD cells for the antiproliferative effects of ectopic expression of miR-143-5p and miR-143-3p, and obtained similar results with the XTT assay ( Figure S2). Cell cycle distributions were analyzed by flow cytometry following transfection of LUAD cells with miR-143-5p and miR-143-3p (72 h after transfection). Cell cycle phase distributions (G 0 /G 1 , S, and G 2 /M) are shown in the bar chart. G 2 /M phase arrest was detected by miR-143-5p transfection into LUAD cells. In contrast to miR-143-5p transfection, G 0 /G 1 phase arrest was detected by miR-143-3p transfection ( Figure 2B,C). results with the XTT assay ( Figure S2). Cell cycle distributions were analyzed by flow cytometry following transfection of LUAD cells with miR-143-5p and miR-143-3p (72 h after transfection). Cell cycle phase distributions (G0/G1, S, and G2/M) are shown in the bar chart. G2/M phase arrest was detected by miR-143-5p transfection into LUAD cells. In contrast to miR-143-5p transfection, G0/G1 phase arrest was detected by miR-143-3p transfection ( Figure 2B,C). Moreover, as shown in Figure 3A and B, the migration and invasion of A549 and H1299 cells were significantly suppressed by miR-143-5p or miR-143-3p transfection. The photomicrographs are presented in Figure S3. and miR-143-3p. (A) Cell proliferation was assessed using XTT assays. Data were measured 72 h after transfection with miRNAs. Cell proliferation was suppressed by miR-143-5p and miR-143-3p in H1299 cells (* p < 0.001), but it was not affected by miR-143-3p transfection into A549 cells. (B,C) Cell cycle assays by flow cytometry showed that G 2 /M phase arrest was induced following miR-143-5p transfection into A549 and H1299 cells. In contrast to miR-143-5p transfection, G 0 /G 1 phase arrest was detected by miR-143-3p transfection.
Moreover, as shown in Figure 3A,B, the migration and invasion of A549 and H1299 cells were significantly suppressed by miR-143-5p or miR-143-3p transfection. The photomicrographs are presented in Figure S3. Due to the cell migration and invasive abilities being remarkably suppressed by miR-143-5p and miR-143-3p expression in LUAD cells, we investigated the regulation of epithelial-mesenchymal transition (EMT)-related genes (e.g., E-cadherin, Vimentin, N-cadherin, SNAIL, and SLUG) by these miRNAs. Interestingly, the expression of Vimentin, SNAIL, and SLUG was reduced by miR-143-5p in LUAD cells ( Figure S4). The elucidation of the molecular mechanism of suppression of these EMTrelated genes by miR-143-5p is an issue for the future.

Incorporation of miR-143-5p into the RISC in LUAD Cells
To verify that miR-143-5p had actual functions in LUAD cells, we performed immunoprecipitation assays using anti-Ago2 antibodies. Ago2 is an essential component of the RISC. After transfection of miR-143-5p into A549 cells, the amount of miR-143-5p was dramatically increased compared with that in untransfected cells ( Figure S5). These data suggested that transfected miR-143-5p was incorporated into the RISC in LUAD cells.

Screening of miR-143-5p-Controlled Oncogenes in LUAD Cells
To identify genes controlled by miR-143-5p in LUAD cells, we used three independent datasets: gene expression data from miR-143-5p-transfected A549 cells, gene expression data from lung cancer clinical specimens, and data from the TargetScanHuman database, showing annotated putative targets with miR-143-5p binding sites. Our strategy searching for miR-143-5p targets is shown in Figure S6. In this study, 22 genes were identified as putative miR-143-5p-controlled oncogenes in LUAD cells (Table 2). We also identified miR-143-3p-controlled oncogenic targets using a similar strategy ( Figure S6, Table S2).  Due to the cell migration and invasive abilities being remarkably suppressed by miR-143-5p and miR-143-3p expression in LUAD cells, we investigated the regulation of epithelial-mesenchymal transition (EMT)-related genes (e.g., E-cadherin, Vimentin, N-cadherin, SNAIL, and SLUG) by these miRNAs. Interestingly, the expression of Vimentin, SNAIL, and SLUG was reduced by miR-143-5p in LUAD cells ( Figure S4). The elucidation of the molecular mechanism of suppression of these EMT-related genes by miR-143-5p is an issue for the future.

Incorporation of miR-143-5p into the RISC in LUAD Cells
To verify that miR-143-5p had actual functions in LUAD cells, we performed immunoprecipitation assays using anti-Ago2 antibodies. Ago2 is an essential component of the RISC. After transfection of miR-143-5p into A549 cells, the amount of miR-143-5p was dramatically increased compared with that in untransfected cells ( Figure S5). These data suggested that transfected miR-143-5p was incorporated into the RISC in LUAD cells.

Screening of miR-143-5p-Controlled Oncogenes in LUAD Cells
To identify genes controlled by miR-143-5p in LUAD cells, we used three independent datasets: gene expression data from miR-143-5p-transfected A549 cells, gene expression data from lung cancer clinical specimens, and data from the TargetScanHuman database, showing annotated putative targets with miR-143-5p binding sites. Our strategy searching for miR-143-5p targets is shown in Figure S6. In this study, 22 genes were identified as putative miR-143-5p-controlled oncogenes in LUAD cells ( Table 2). We also identified miR-143-3p-controlled oncogenic targets using a similar strategy ( Figure  S6, Table S2).

Direct Regulation of MCM4 Expression by miR-143-5p in LUAD Cells
Next, we performed dual-luciferase reporter assays to determine whether MCM4 was directly regulated by miR-143-5p. We used vectors encoding the partial wild-type sequences of the 3′untranslated region (UTR) of MCM4 including the predicted miR-143-5p target sites (binding site 1: positions 448-454, binding site 2: positions 968-974 in the MCM4 3′-UTR) or deletion vectors lacking these sites ( Figure 7C). We found that the luciferase activity was significantly decreased by cotransfection with miR-143-5p and the vector carrying the wild-type 3′-UTR of MCM4, whereas transfection with the deletion vector blocked the decrease in luminescence in A549 cells ( Figure 7D). These data demonstrated that miR-143-5p directly bound to two sites in the 3′-UTR of MCM4.  We focused on further evaluation of MCM4 in LUAD cells. Expression levels of MCM4 mRNA and MCM4 protein were significantly reduced by miR-143-5p transfection in A549 and H1299 cells ( Figure 7A,B).

Direct Regulation of MCM4 Expression by miR-143-5p in LUAD Cells
Next, we performed dual-luciferase reporter assays to determine whether MCM4 was directly regulated by miR-143-5p. We used vectors encoding the partial wild-type sequences of the 3untranslated region (UTR) of MCM4 including the predicted miR-143-5p target sites (binding site 1: positions 448-454, binding site 2: positions 968-974 in the MCM4 3 -UTR) or deletion vectors lacking these sites ( Figure 7C). We found that the luciferase activity was significantly decreased by cotransfection with miR-143-5p and the vector carrying the wild-type 3 -UTR of MCM4, whereas transfection with the deletion vector blocked the decrease in luminescence in A549 cells ( Figure 7D). These data demonstrated that miR-143-5p directly bound to two sites in the 3 -UTR of MCM4.

Figure 7. Expression of MCM4/MCM4 was directly regulated by miR-143-5p in LUAD cells. (A) Expression of MCM4 mRNA was significantly reduced by miR-143-5p transfection into A549 and H1299 cells (72 h after transfection). Expression of GAPDH was used as an internal control. (B) Expression of MCM4 protein was reduced by miR-143-5p transfection into LUAD cells (72 h after transfection).
Expression of GAPDH was used as an internal control. (C) TargetScanHuman database analyses predicted two putative miR-143-5p binding sites in the 3′-UTR of MCM4. (D) Dual luciferase reporter assays showed that luminescence activities were reduced by cotransfection with wild-type (miR-143-5p binding site) vectors and miR-143-5p in LUAD cells. Normalized data were calculated as Renilla/firefly luciferase activity ratios (*p < 0.001).

Effects of MCM4 Knockdown on LUAD Cells Malignant Phenotypes
To investigate the oncogenic functions of MCM4 in LUAD cells, knockdown assays were conducted using small interfering RNAs (siRNAs). Both mRNA and protein expression levels were successfully suppressed by siMCM4 transfection in A549 and H1299 cells ( Figure 8A,B).
In functional assays, cell proliferation was significantly suppressed by siMCM4 transfection into A549 and H1299 cells ( Figure 8C). Cell migration and invasive abilities were significantly blocked by knockdown of MCM4 (siMCM4-1 and siMCM4-2) in LUAD cells ( Figure 8D,E). The photomicrographs were presented in Figure S7. Cell cycle assays showed that significant G0/G1 phase arrest was observed in LUAD cells ( Figure S8). In functional assays, cell proliferation was significantly suppressed by siMCM4 transfection into A549 and H1299 cells ( Figure 8C). Moreover, cell migration and cell invasion were significantly suppressed by siMCM4 transfection into A549 and H1299 cells (Figures 8D,E). Notably, cell cycle assays by MCM4 knockdown showed that significant G0/G1 phase arrest was observed in LUAD cells ( Figure S8). Ectopic expression of miR-143-5p induced the S phase arrest or delayed the S phase progression (Figure 2). MCM4 is one of the pivotal genes directly controlled by miR-143-5p in LUAD cells. This suggests that miR-143-5p also regulates other cell-cycle-related genes involved in S phase progression.

Effects of MCM4 Knockdown on LUAD Cells Malignant Phenotypes
To investigate the oncogenic functions of MCM4 in LUAD cells, knockdown assays were conducted using small interfering RNAs (siRNAs). Both mRNA and protein expression levels were successfully suppressed by siMCM4 transfection in A549 and H1299 cells ( Figure 8A,B).
In functional assays, cell proliferation was significantly suppressed by siMCM4 transfection into A549 and H1299 cells ( Figure 8C). Cell migration and invasive abilities were significantly blocked by knockdown of MCM4 (siMCM4-1 and siMCM4-2) in LUAD cells ( Figure 8D,E). The photomicrographs were presented in Figure S7. Cell cycle assays showed that significant G 0 /G 1 phase arrest was observed in LUAD cells ( Figure S8). In functional assays, cell proliferation was significantly suppressed by siMCM4 transfection into A549 and H1299 cells ( Figure 8C). Moreover, cell migration and cell invasion were significantly suppressed by siMCM4 transfection into A549 and H1299 cells ( Figure 8D,E). Notably, cell cycle assays by MCM4 knockdown showed that significant G 0 /G 1 phase arrest was observed in LUAD cells ( Figure S8). Ectopic expression of miR-143-5p induced the S phase arrest or delayed the S phase progression (Figure 2). MCM4 is one of the pivotal genes directly controlled by miR-143-5p in LUAD cells. This suggests that miR-143-5p also regulates other cell-cycle-related genes involved in S phase progression.

Aberrant Expression of MCM4 in LUAD Clinical Specimens by Immunohistochemistry
MCM4 expression at the protein levels were evaluated by using tissue microarray of 20 LUAD cases and 14 noncancerous cases (Table S3). Compared with normal lung tissues, MCM4 protein was strongly expressed in LUAD tissues ( Figure 9A-D).

Aberrant Expression of MCM4 in LUAD Clinical Specimens by Immunohistochemistry
MCM4 expression at the protein levels were evaluated by using tissue microarray of 20 LUAD cases and 14 noncancerous cases (Table S3). Compared with normal lung tissues, MCM4 protein was strongly expressed in LUAD tissues ( Figure 9A-D We also investigated the expression of MCM4 using miR-143-5p-measured clinical specimens. Overexpression of MCM4 was detected in LUAD clinical specimens (Table S1 and Figure S9).

Discussion
Several miRNAs exist in close proximity on the human genome (defined as miRNA clusters). The expression levels of miRNAs within miRNA clusters are controlled by the same molecular mechanisms [25]. In the human chromosome 5q32 region, pre-miR-143 and pre-miR-145 form the miRNA cluster, and their promoter regions contain p53 binding sites, resulting in p53-dependent expression [26]. Many studies have shown that miR-143/miR-145 cluster miRNAs are frequently downregulated in human cancers where miR-143-3p and miR-145-5p (the guide strand) act as pivotal tumor suppressors [27,28]. Extensive searches have been carried out to identify the oncogenic targets and molecular networks controlled by miR-143-3p and miR-145-5p in several cancers [27][28][29]. In contrast, few studies have evaluated functional roles and targets of miR-143-5p and miR-145-3p (the passenger strand) in cancer cells.
Recently, we found that the expression levels of both strands of pre-miR-145 (miR-145-5p and miR-145-3p) were reduced in LUAD clinical specimens and that cancer cell malignant phenotypes were attenuated by ectopic expression of these miRNAs [20]. Furthermore, target searches for miR-145-5p/miR-145-3p showed that aberrant expression of LMNB2 acted as an oncogene in LUAD cells [20]. Our previous studies revealed that tumor-suppressive miR-145-3p was closely involved in human cancer pathogenesis by targeting oncogenes, for example, MTDH in LUSQ; MELK, NCAPG, BUB1, and CDK1 in prostate cancer; MYO1B in head and neck cancer; and MYO1B and DHRS2 in esophageal cancer [19,[30][31][32]. These data suggest that some passenger strands of miRNAs are actually functional in human cancers.
In this study, we focused on miR-143-5p and demonstrated the tumor-suppressive functions of this miRNA, similar to those of miR-143-3p, in LUAD cells. Previous studies have shown that miR-143-5p is downregulated in gastric cancer and gallbladder cancer tissues and that forced miR-143-5p We also investigated the expression of MCM4 using miR-143-5p-measured clinical specimens. Overexpression of MCM4 was detected in LUAD clinical specimens (Table S1 and Figure S9).

Discussion
Several miRNAs exist in close proximity on the human genome (defined as miRNA clusters). The expression levels of miRNAs within miRNA clusters are controlled by the same molecular mechanisms [25]. In the human chromosome 5q32 region, pre-miR-143 and pre-miR-145 form the miRNA cluster, and their promoter regions contain p53 binding sites, resulting in p53-dependent expression [26]. Many studies have shown that miR-143/miR-145 cluster miRNAs are frequently downregulated in human cancers where miR-143-3p and miR-145-5p (the guide strand) act as pivotal tumor suppressors [27,28]. Extensive searches have been carried out to identify the oncogenic targets and molecular networks controlled by miR-143-3p and miR-145-5p in several cancers [27][28][29]. In contrast, few studies have evaluated functional roles and targets of miR-143-5p and miR-145-3p (the passenger strand) in cancer cells.
Recently, we found that the expression levels of both strands of pre-miR-145 (miR-145-5p and miR-145-3p) were reduced in LUAD clinical specimens and that cancer cell malignant phenotypes were attenuated by ectopic expression of these miRNAs [20]. Furthermore, target searches for miR-145-5p/miR-145-3p showed that aberrant expression of LMNB2 acted as an oncogene in LUAD cells [20]. Our previous studies revealed that tumor-suppressive miR-145-3p was closely involved in human cancer pathogenesis by targeting oncogenes, for example, MTDH in LUSQ; MELK, NCAPG, BUB1, and CDK1 in prostate cancer; MYO1B in head and neck cancer; and MYO1B and DHRS2 in esophageal cancer [19,[30][31][32]. These data suggest that some passenger strands of miRNAs are actually functional in human cancers.
Next, we identified cancer-promoting genes and oncogenic molecular networks regulated by tumor-suppressive miR-143-5p in LUAD cells. In this screening, we successfully identified 22 putative target oncogenes regulated by miR-143-5p in LUAD cells. Among these targets, high expression of 11 genes (MCM4, RAD51, FAM111B, CLGN, KRT80, GPC1, MTL5, NETO2, FANCA, MTFR1, and TTLL12) was significantly associated with poor prognosis in patients with LUAD. Moreover, multivariate analysis revealed the expression levels of six genes (MCM4, FAM111B, CLGN, MTL5, FANCA, and MTFR1) were independent predictive factors for overall survival in patients. Recently, overexpression of MTL5 was reported in NSCLC, and its expression was shown to predict prognosis [38]. Another study showed that knockdown of FAM111B induced G 2 /M phase arrest and apoptosis in LUAD cells [39]. These data suggested that miR-143-5p-controlled genes were closely involved in LUAD oncogenesis and molecular pathogenesis.
In this study, we further analyzed the roles of MCM4 in LUAD cells. Our data revealed that aberrant expression of MCM4 enhanced LUAD cell proliferation and that MCM4 expression was an independent prognostic biomarker in patients with LUAD. Our data were consistent with a previous report [20], strongly suggesting that aberrantly expressed MCM4 was an oncogene in LUAD.
Minichromosome maintenance (MCM) proteins (MCM2-MCM7) form a hexamer complex that functions as a DNA helicase in DNA replication [40]. Moreover, the MCM2-7 hexamer helicase is activated by forming a complex with CDC45 and GINS proteins, termed the CMG (CDC45/MCMs/GINS) complex [40]. Notably, TCGA database analyses showed that high expression of CDC45, GINS1, and GINS3 was associated with a poor prognosis compared to low expression of these genes in patients with LUAD. Moreover, aberrant expression of MCM proteins has been shown to contribute to cancer progression, and these proteins could be effective diagnostic markers in various cancers [41]. In lung cancer, high expression levels of MCM2, MCM5, MCM6, and MCM7 could be useful prognostic markers [41][42][43]. This is the first report demonstrating that tumor-suppressive miR-143-5p directly regulates MCM4 in LUAD cells. However, our understanding of the molecular mechanisms controlling the expression of MCM proteins is insufficient. Recent studies have reported that tumor-suppressive miRNAs negatively regulate the expression of MCM family genes (e.g., MCM2 is targeted by miR-31 and MCM5 is targeted by miR-885-5p and miR-362-3p) [44][45][46]. Interestingly, reduced expression of MCM2-7 in DNA replication stress is associated with p53-dependent miRNAs, including the miR-34-family [47]. Future studies are needed to determine the involvement of noncoding RNAs, including miRNAs, in the regulation of MCM/CMG expression.
Our results showed that all members of the miR-143/miR-145 cluster (miR-143-5p, miR-143-3p, miR-145-5p, and miR-145-3p) acted as tumor-suppressive miRNAs in LUAD. The involvement of passenger strands of miRNAs in LUAD oncogenesis is a novel concept in cancer research. The tumor-suppressive functions of miRNAs depend on controlling a vast number of oncogenic genes in the cancer cell. Our present study suggests that the miR-143-5p regulates cell-cycle-related genes and EMT-related genes in LUAD cells. The miR-143-5p may control many genes contributing to LUAD malignancy. Identification of oncogenes and oncogenic networks controlled by the miR-143/miR-145 cluster could help to establish novel prognostic and therapeutic targets in LUAD.

Clinical Human LUAD Specimens and LUAD Cell Lines
Forty-three clinical specimens (19 LUAD and 24 normal lung tissues) were used in this study. The clinical characteristics of these specimens are summarized in Table 1 and Table S1. Classification of sample stages was carried out according to the Association for the Study of Lung Cancer TNM classification (7th edition).
All specimens were obtained from lung cancer surgeries performed at Kagoshima University Hospital (2010-2013). Written informed consent was obtained from all patients before the use of their specimens. This study was approved by the Bioethics Committee of Kagoshima University (approval number: 26-164, 10th February 2015).
Two LUAD cell lines, A549 and H1299 (American Type Culture Collection, Manassas, VA, UAS), were used in this study.

RNA Preparation and Quantitative Real-Time Reverse Transcription Polymerase Chain Reaction (qRT-PCR)
RNA extraction from clinical specimens and cell lines was performed as previously described [21,48]. Expression levels of miRNAs were evaluated using qRT-PCR as described previously [21,48]. TaqMan probes and primers used in this study are listed in Table S4.

Transfection of miRNAs, siRNAs, and Plasmid Vectors into LUAD Cells
The procedures for transfection of miRNAs, siRNAs, and plasmid vectors into LUAD cells were described previously [21,48]. The reagents used in this study are listed in Table S4.

Functional Assays for LUAD Cells (i.e., Cell Proliferation, Migration, Invasion, and Cell Cycle Assays)
The procedures for functional assays in cancer cells (proliferation, migration, invasion, and cell cycle assays) were described in our previous studies [21,48].

Measurement of Amount of miR-143-5p Incorporated into the RISC
Immunoprecipitation using anti-Ago2 antibodies was used to determine whether miR-143-5p was incorporated into the RISC. The procedure for immunoprecipitation was described in a previous study [31]. The reagents used in this study are listed in Table S4.

Plasmid Construction and Dual-Luciferase Reporter Assay
Plasmid vectors, including vectors containing the wild-type sequence of miR-143-5p binding sites in the 3 -UTR of MCM4 or the deletion sequences of miR-143-5p binding sites in the 3 -UTR of MCM4, were prepared. The inserted sequences are shown in Figure S10. The procedures for transfection and dual luciferase reporter assays were described in our previous studies [21,48]. The reagents used in this study are listed in Table S4.

Western Blotting and Immunohistochemistry
The procedures for Western blotting and immunohistochemistry were described in our previous studies [21,48]. MCM4 expression at the protein levels were evaluated by immunohistochemical staining using tissue microarray (BC04002a; US Biomax, Inc. Derwood, MD, USA) of 20 LUAD cases and 14 noncancerous cases (Table S3). The expression of MCM4 using miR-143-5p-measured clinical specimens was investigated (Table S1 and Figure S9). The antibodies used in this study are listed in Table S4.

Statistical Analysis
Mann-Whitney U tests were applied for comparisons between two groups. For multiple groups, one-way analysis of variance and Tukey tests for post hoc analysis were applied. These analyses were performed with GraphPad Prism 7 (GraphPad Software, La Jolla, CA, USA) and JMP Pro 14 (SAS Institute Inc., Cary, NC, USA).

Conclusions
This is the first report revealing that miR-143-5p (the passenger strand of the miR-143 duplex) had tumor-suppressive functions in LUAD cells. Several genes controlled by miR-143-5p were closely involved in LUAD oncogenesis and pathogenesis. Notably, the expression of MCM4 was directly regulated by miR-143-5p, and its aberrant expression enhanced LUAD malignant transformation. Our studies demonstrated that all members of the miR-143/miR-145 cluster acted as tumor-suppressive miRNAs and that their targets may be novel prognostic and therapeutic targets in LUAD.