Interactome of Radiation-Induced Microrna-Predicted Target Genes

Hindawi Publishing Corporation Comparative and Functional Genomics Volume 2012, Article ID 569731, 12 pages doi:10.1155/2012/569731

Research Article Interactome of Radiation-Induced microRNA-Predicted Target Genes

Tenzin W. Lhakhang and M. Ahmad Chaudhry

Department of Medical Laboratory and Radiation Sciences, University of Vermont, 302 Rowell Building, Burlington, VT 05405, USA

Correspondence should be addressed to M. Ahmad Chaudhry, [email protected]

Received 26 March 2012; Accepted 19 April 2012

Academic Editor: Giulia Piaggio

Copyright © 2012 T. W. Lhakhang and M. A. Chaudhry. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The microRNAs (miRNAs) function as global negative regulators of gene expression and have been associated with a multitude of biological processes. The dysfunction of the microRNAome has been linked to various diseases including cancer. Our laboratory recently reported modulation in the expression of miRNA in a variety of cell types exposed to ionizing radiation (IR). To further understand miRNA role in IR-induced stress pathways, we catalogued a set of common miRNAs modulated in various irradiated cell lines and generated a list of predicted target genes. Using advanced bioinformatics tools we identified cellular pathways where miRNA predicted target genes function. The miRNA-targeted genes were found to play key roles in previously identified IR stress pathways such as cell cycle, p53 pathway, TGF-beta pathway, ubiquitin-mediated proteolysis, focal adhesion pathway, MAPK signaling, thyroid cancer pathway, adherens junction, insulin signaling pathway, oocyte meiosis, regulation of actin cytoskeleton, and renal cell carcinoma pathway. Interestingly, we were able to identify novel targeted pathways that have not been identified in cellular radiation response, such as aldosterone-regulated sodium reabsorption, long-term potentiation, and neutrotrophin signaling pathways. Our analysis indicates that the miRNA interactome in irradiated cells provides a platform for comprehensive modeling of the cellular stress response to IR exposure.

1. Introduction levels or complete deletion of key miRNAs have been reported in tumor cells [6]. MicroRNAs (miRNAs) are approximately 21 nucleotides in Cellular stress pathways protect cells from the deleterious length that do not code for proteins. miRNAs were discov- effects of genotoxic insult. Ionizing radiation disrupts cel- ered in 1993 but their significance was not realized until 2000 lular homeostasis through multiple mechanisms. The cells [1].miRNAsactasnegativeregulatorsofgeneexpressionby respond to stress induced by ionizing radiation exposure mRNA degradation and protein downregulation [2]. miRNA through complex processes by activating many pathways bind to the target mRNA and initiate mRNA degradation. ranging from DNA damage processing, signal transduction, Alternatively miRNAs inhibit the protein machinery from altered gene expression, cell-cycle arrest, and genomic insta- latching on to the mRNA. The interplay between the miRNA bility to cell death [7, 8]. The current data suggests that the and mRNA forms a highly complex regulatory network, exposure to radiation provokes cellular responses controlled mainly because a single miRNA can target hundreds of in part by gene expression networks [7, 9]. miRNAs regulate different mRNA molecules [3]. Higher production of miRNA gene expression and have been shown to control multiple leads to lower expression levels of its target proteins. The intracellular processes involved in the response to cellular miRNAsarereportedtobeinvolvedincelldifferentiation, stress [10, 11]. metabolic regulation, apoptosis, and many other biological Alterations in the miRNA expression levels occur fol- processes [4]. Dysfunction of miRNA has been associated lowing exposure to ionizing radiation [12–14]. The miRNA with numerous cancers [5] and alterations in the expression expression levels in primary human dermal microvascular 2 Comparative and Functional Genomics endothelial cells (HDMEC) after 2 Gy radiation treatment This search retrieved 236 research articles. PubMed was a indicated upregulation of hsa-let-7g, hsa-miR-16, hsa-miR- preferred choice over the other available article databases 20a, hsa-miR-21 and hsa-miR-29c, and downregulation of such as Web of Science, BIOSIS Previews (http://thomson- hsa-miR-18a, hsa-miR-125a, hsa-miR-127, hsa-miR-148b, reuters.com/) and Science Direct because PubMed provided hsa-miR-189,andhsa-miR-503 [15]. miRNA profiles of comprehensive results that overlapped and displayed more irradiated lung cancer cells indicated that the level of hsa- articles. The articles were imported into EndNote after deter- let-7g was higher in radiosensitive cells Caski, NCI-H460 mining its relevancy to miRNA and IR topic. The relevant (H460), and ME180 than in radioresistant cells A549, H1299, articles extracted from PubMed were subjected to addi- DLD1 [16]. Changes in expression patterns of hsa-miR-92b, tional refinement. The research studies which were selected, hsa-miR-137, hsa-miR-656, hsa-miR-558, hsa-miR-660,and performed ionizing radiation experiments on human cell hsa-miR-662 after low (0.1 Gy) and high (2.0 Gy) doses of lines only and recorded the miRNA expression levels. We X-ray in human fibroblasts were observed [17]. The hsa- identified the miRNA species, cell types, type of radiation, let-7 family miRNAs were upregulated in irradiated M059K radiation dose, and analysis time from these studies. We then cells and downregulated in M059J cells. The hsa-miR-17-3p, assembled a common list of miRNAs that were investigated hsa-miR-17-5p, hsa-miR-19a, hsa-miR-19b, hsa-miR-142-3p, among a group of cell types. and hsa-miR-142-5p were upregulated in both M059K and M059J cells. [14]. Radiation treatment of prostate cancer 2.2. Generation of IR-Induced miRNA Database. The miRNA cells changed the expression levels of hsa-miR-521 [18]. The expression data extracted from the published research arti- expression profiles of miRNAs in HCT116 human colon cles was used to assemble a database. Over 1000 records were carcinoma cells and its p53-null derivative correlated with generated from the data extraction procedure. Microsoft p53 status [19]. The expression of hsa-let-7 family miRNAs, Excel was utilized to tabulate the information from various which are negative regulators of the rat sarcoma, RAS articles. This “mastersheet” formed the platform for the sub- oncogene, was upregulated in irradiated p53 positive TK6 sequent analysis. The data was organized in the Excel “mas- cells but was downregulated in p53 negative WTK1 cells. The tersheet” with the following headings: cell type, cell line, radi- hsa-miR-15a and hsa-miR-16 were upregulated in 0.5 Gy- ation type, radiation dose (Gy), dose rate (Gy/min), analysis irradiated TK6 cells but were downregulated after a 2 Gy time (hours) after treatment, miRNA species, qualitative dose of X-rays [13]. The expression levels of hsa-let-7 family miRNA expression from base, qualitative miRNA expression miRNA and miRNA associated with MYC translocation were from base (numerical), quantitative miRNA expression modulated after gamma radiation treatment in Jurkat cells (fold change), and data source. The Pivot Table tool of [12]. While many studies have reported dose-dependent Microsoft Excel was employed to reorganize the master changes in the expression profiles of miRNAs in irradiated dataset. This “pivot table” was referenced to the original IM9 human B lymphoblastic cells [20, 21], human lung car- unchanged mastersheet. The pivot table allowed to generate a cinoma cell line A549 [22], and human fibroblasts [23]; some list of miRNA that were observed to show altered expression studies did not observe any significant alterations in miRNA in 5 or more cell lines after exposure to ionizing radiation. expression in cells treated with gamma-irradiation [24]. We were interested to examine the role of miRNAs in 2.3. Prediction of miRNA Target Genes. The miRNAs of ionizing radiation- (IR-)induced stress pathways. Although interest were assembled in Microsoft Access database miRNAs have been implicated as crucial posttranscriptional then searched against the mirDB dataset (http://mirdb.org/ gene regulators, their role in the cellular response to IR is miRDB/)[25] to predict target genes. Microsoft Access not comprehensively examined. We asked the question: can allowed for the quick creation of local database from we use microRNAome and their target genes to corroborate which high-level queries could be placed. To obtain gene previously identified IR responsive pathways? We also argued ontology (GO) terms for each target gene, we equipped the if the miRNAome would allow us to discover new perspective Access database with other genomic datasets, such as Entrez to radiation exposure. This study was undertaken (1) to Gene, GaRNeT (Genomics and Randomized Trials Net- assemble miRNA species that are modulated after radiation work), and KEGG (Kyoto Encyclopedia of Genes and Gen- exposure in many human cell lines, (2) to identify the omes) Pathways. The Entrez Gene dataset provided infor- genes that are targeted by these miRNA using bioinformatics mation regarding gene symbol and description, which was approaches, and (3) to determine the role of miRNA target linked by the Entrez ID. The GaRNeT dataset allowed linking genes in radiation-induced cellular pathways. the mirDB dataset to the Entrez Gene dataset.

2. Materials and Methods 2.4. Identification and Visualization of miRNA-Predicted Genes and Biological Pathways. The output dataset queried 2.1. Selection of IR-Induced miRNA. We collected the data through the Access database provided the predicted target from our published work on radiation-induced miRNA genes for the miRNA species of interest. The Entrez ID and also searched the PubMed database to collect articles list was imported in DAVID (Database for Annotation, that investigated the modulation of miRNA after IR exposure. Visualization andIntegrated Discovery) (http://david.abcc The keywords miRNA, microRNA, ionizing radiation and .ncifcrf.gov/), an online database that can link genes to radiation were used in performing the literature search. its biological pathways through the KEGG database. This Comparative and Functional Genomics 3 strategy enabled an enriched analysis of genes for GO terms, around 66 chromosomes. The radiation-induced miRNA using the miRNA target genes with respect to the category expression profiles differ considerably among all these cell of biological process. The miRNA-predicted target genes and types. We looked for miRNA expression similarities across biological pathways were visualized with Cytoscape software these cell lines to find common miRNA responses after rad- (http://cytoscape.org/). Cytoscape is an open source software iation exposure. Table 1 shows information regarding some package that allows for powerful visual mappings of the of these cell lines from where the miRNA dataset was selected provided datasets. Cytoscape provided a visualization that in this study. helped connect miRNA to their respective target genes and associated pathways. Cytoscape enabled us to perform more detailed functional analysis to identify miRNA-mediated net- 3.2. Identification of miRNA Modulated in IR-Exposed Cell works, biological functions and canonical signaling pathways Lines. We built an miRNA database to warehouse the IR- for both miRNAs and the target mRNAs. induced miRNA expression data extracted from various published studies. All of these studies employed low LET radiation and the doses ranged from 0.1 to 40 Gy. The analy- 3. Results and Discussion sis methodology ranged from real-time PCR to microarray techniques. The number of radiation modulated miRNA Exposure to ionizing radiation induces various physiological ranged from 8 to 36 in individual cell lines. The Microsoft responses including DNA repair, cell cycle arrest, signal Excel’s pivot table from the original “mastersheet” permitted transduction, cell death, and cell differentiation [7, 8]. It is to create a list of miRNA that were modulated in 5 or more becoming clear that changes in the expression profile of cell lines. 20 miRNA species were identified that were modu- many genes play a significant role in these processes [7, latedin5ormorecelllinesaftertreatmentwithIR.Thepivot 9]. The discovery of miRNA and the establishment of table feature was further exploited to build a heat map of their involvement in regulating the gene expression have the IR-modulated miRNAs. This heat map displays 3 dimen- prompted the need to interrogate their participation in the sional information which accounts for cell line, analysis time radiation response. It is not completely understood how after IR treatment and the miRNA species (Figure 1). We miRNAs function in the cellular response to radiation expo- first looked at the modulation of eight let-7 family miRNA sure. Our laboratory is engaged in assessing the miRNA hsa-let-7a, hsa-let-7b, hsa-let-7c, hsa-let-7d, hsa-let-7e, hsa- responses in irradiated human cells. We have collected data let-7f, hsa-let-7g, and hsa-let-7i. The published studies have on IR-induced miRNA expression alterations in a variety of examined the expression of these miRNA in various cell lines cell types [12–14]. We combined our data along with other including A549, HBE135-E6E7, IM9, Jurkat, M059J, M059K, published data to compile miRNA species that are modulated TK6, and WTK1 at 0–24 h time points after exposure to in a variety of human cell types. It was important to find IR. A quick snapshot of the data assembled in Figure 1 miRNAsthatwereanalyzedindifferent studies, because a indicated that majority of these miRNA were induced in singular data point would not have allowed for the formation Jurkat, M059J, M059K, and TK6 cell lines. In contrast these of relationships between different cell types. This dataset was miRNA were downregulated in A549, HBE135-E6E7, and used for the identification of predicted miRNA target genes WTK1 cells. The expression analysis of hsa-let-7 family and the basis of the functional role of miRNAs in the cell. miRNA indicated that both A549 and WTK1 cell lines display Our goal was to establish a link between the modulation of similar signatures (Figure 1). The differences in hsa-let-7 miRNA after radiation exposure to the stress induced path- family miRNA expression between eight cell lines revealed ways in the cell. that WTK1 displayed marked down regulation in all miRNAs of interest compared to the other cell lines. This could be 3.1. Cell Lines Investigated for IR-Induced miRNA Expression due to the fact that WTK1 is characterized as an inherently Analysis. A variety of cell lines have been exploited to exam- unstable cell line [27]. Even in the absence of any radiation ine the miRNA expression profile after radiation exposure. exposure, a high percentage of WTK1 cells was reported The lymphoblast TK6 cell line has normal p53 and was used to display chromosomal aberrations such as aneuploidy, as a base to compare IR-induced miRNA from p53 negative chromatid breaks, and translocations [27]. These genomic WTK1 cell line [13]. The M059J and M059K glioma cell lines instabilities could be connected to the dysfunction in the isolated from the same tumor specimen differ in the DNA- microRNAome. Many studies have attempted to probe the dependent protein kinase activity. M059J lack this kinase role of miRNA in radiation sensitivity. The overexpression or whereas M059K express normal levels [26]. The Jurkat and inhibition of let-7g markedly influenced clonogenic survival IM9 are both of lymphoid lineages; Jurkat is of T-cell origin and cell proliferation; hsa-let-7g enhanced radiosensitivity and IM9 is of B-cell origin (ATCC; American Type Culture of these cells [15]. Overexpression of hsa-let-7g in H1299 Collection). These cells are prevalent in many research stud- cells suppressed the translation of K-RAS, and increased ies because they provide platform on which to study immu- the sensitivity to IR. Knockdown of LIN28B,anupstream nological signaling processes and the production of various regulator of hsa-let-7g, increased the level of mature hsa-let- chemokines [12, 20, 21]. The A549 and HBE135-E6E7 are 7g and the sensitivity to IR in H1299 cells [16]. The over- cancerous cell lines from lung tissue. The cytogenetic data expression of hsa-let-7a decreased the expression of K-RAS from ATCC indicate that A549 cell line displays a hypo- and radiosensitized A549 lung carcinoma cells. Inhibition of triploid chromosomal expression; a majority of cells having LIN28, a repressor of hsa-let-7, attenuated K-RAS expression 4 Comparative and Functional Genomics

Table 1: Radiation-induced miRNA expression analysis in various cell lines.

Cell line Cell type Radiation type Radiation dose (Gy) Characterized miRNA Reference A549 Basal Epithelial N/A 2.5 8 [11] HBE135-E6E7 Squamous N/A 2.5 8 [11] TK6 Lymphoblast X-radiation 0.5 21 [13] TK6 Lymphoblast X-radiation 2 21 [13] WTK1 Lymphoblast X-radiation 0.5 21 [13] WTK1 Lymphoblast X-radiation 2 21 [13] HDMEC Endothelial X-radiation 2 11 [15] AG1522 Fibroblast X-radiation 0.1 7 [17] AG1522 Fibroblast X-radiation 2 22 [17] M059J Glial X-radiation 3 19 [14] M059K Glial X-radiation 3 19 [14] IM9 B Lymphoblast Gamma 1 36 [20] IM9 B Lymphoblast Gamma 10 26 [20] IM9 B Lymphoblast Gamma 0.5 22 [21] IM9 B Lymphoblast Gamma 10 22 [21] TK6 Lymphoblast Gamma 2 20 [12] Jurkat T- cell Gamma 2 20 [12] A549 Basal epithelial Gamma 20 12 [22] A549 Basal epithelial Gamma 40 18 [22] LNCaP Epithelial X-radiation 6 15 [18] C4-2 Epithelial X-radiation 6 11 [18] HCT116 (Null) Epithelial Gamma 10 36 [19] HCT116 (WT) Epithelial Gamma 10 36 [19]

and radiosensitized A549, and pancreatic cancer ASPC1 cells on statistical learning theory. This program uses statistics [28]. and employs artificial intelligence strategy such as neural We next determined the expression status of another networks to assign a prediction score for a miRNA’s target twelve miRNA hsa-miR-142-3p, hsa-miR-142-5p, hsa-miR- gene. This process of predicting miRNA target genes is 143, hsa-miR-155, hsa-miR-15a, hsa-miR-16, hsa-miR-17- accomplished through a machine-learning algorithm, which 3p, hsa-miR-17-5p, hsa-miR-18, hsa-miR-19a, hsa-miR-19b, places a target score on a miRNA-gene association. This and hsa-miR-21 in irradiated IM9, Jurkat, M059J, M059K, miRDB database was chosen to identify miRNA target genes TK6, and WTK1 cells (Figure 2). Majority of these miRNA because for the simplicity of the database structure and were upregulated in Jurkat, M059J, and M059K cells after agreed upon thresholds for the prediction target score. The exposure to IR but were repressed in irradiated IM9 and prediction target score of 50–95 was used to filter the target WTK1 cells. Some of these IR-induced miRNAs have been genes. We constructed association networks between miR- implicated in various cancers. In chronic lymphocytic leu- NAs and the target mRNAs and visualized the target genes kemia and prostate cancer, a majority of the patients had controlled by various IR responsive miRNA with Cytoscape. deletions or downregulations of hsa-miR-15 and hsa-miR-16 The complexity associated with miRNA interactome is [29, 30]. The hsa-miR-145 and hsa-miR-143 were downreg- shown in Figure 3. As expected a singular miRNA was found ulated in colorectal neoplasia [31]. In lung cancers, miRNA to affect the regulation of hundreds of genes and a number hsa-let-7 was down regulated. The genes that code for of miRNAs were seen to work synergistically with each miRNAs have been frequently located at genomic locations other to downregulate a multitude of genes. The resulting involved in cancers and there is strong indication that miRNA : mRNA association network provided nodes and miRNAgeneactsasbothtumorsuppressorsandoncogenes connections between many miRNAs and the target mRNAs [32]. (Figure 3). This network demonstrated the overlapping mRNA targets for miRNAs in the hsa-miR-let7 family, hsa- miR-15a, hsa-miR-16, hsa-miR-18a, hsa-miR-19a, hsa-miR- 3.3. Prediction and Visualization of IR-Modulated miRNA 19b, hsa-miR-21, hsa-miR-34a, hsa-miR-34b, hsa-miR-142- Target Genes. The list of miRNAs studied among a group of 3p, hsa-miR-142-5p, hsa-miR-143, hsa-miR-145, hsa-miR- cell types was further examined to identify the target mRNAs 155, hsa-miR-197, hsa-miR-202, hsa-miR-376a, hsa-miR-575, affected by these miRNA. We utilized the miRDB for pre- and hsa-miR-609 (Figure 3). Most of the radiation-mod- dicting the miRNA target genes. The miRDB uses a support ulated miRNAs have a large number of mRNA targets. For vector machine (SVM), which is a type of algorithm based example, the number of targets genes that were identified in Comparative and Functional Genomics 5

Analysis time TK6 IM9

Analysis Jurkat M059J WTK1 (hours) M059K miRNA species time IM9 TK6 miRNA species hsa-miR-142-3p A549 WTK1 Jurkat M059J (hours) M059K hsa-miR-142-5p hsa-miR-143 HBE135-E6E7 hsa-miR-155 hsa-let-7a hsa-miR-15a hsa-let-7b 0 hsa-miR-16 hsa-miR-17-3p hsa-let-7c hsa-miR-17-5p hsa-let-7d hsa-miR-18a 0 hsa-miR-19a hsa-let-7e hsa-miR-19b hsa-let-7f hsa-miR-21 hsa-let-7g hsa-miR-142-3p hsa-miR-142-5p hsa-let-7i hsa-miR-143 hsa-let-7a hsa-miR-155 hsa-miR-15a hsa-let-7b 4 hsa-miR-16 hsa-let-7c hsa-miR-17-3p hsa-let-7d hsa-miR-17-5p 4 hsa-miR-18a hsa-let-7e hsa-miR-19a hsa-let-7f hsa-miR-19b hsa-miR-21 hsa-let-7g hsa-miR-142-3p hsa-let-7i hsa-miR-142-5p hsa-let-7a hsa-miR-143 hsa-miR-155 hsa-let-7b hsa-miR-15a hsa-miR-16 hsa-let-7c 8 hsa-miR-17-3p hsa-let-7d 8 hsa-miR-17-5p hsa-let-7e hsa-miR-18a hsa-let-7f hsa-miR-19a hsa-miR-19b hsa-let-7g hsa-miR-21 hsa-let-7i hsa-miR-142-3p hsa-miR-142-5p hsa-let-7a hsa-miR-143 hsa-let-7b hsa-miR-155 hsa-miR-15a hsa-let-7c hsa-miR-16 hsa-let-7d 12 hsa-miR-17-3p 12 hsa-let-7e hsa-miR-17-5p hsa-miR-18a hsa-let-7f hsa-miR-19a hsa-let-7g hsa-miR-19b hsa-miR-21 hsa-let-7i hsa-miR-142-3p hsa-let-7a hsa-miR-142-5p hsa-let-7b hsa-miR-143 hsa-miR-155 hsa-let-7c hsa-miR-15a hsa-let-7d hsa-miR-16 24 24 hsa-miR-17-3p hsa-let-7e hsa-miR-17-5p hsa-let-7f hsa-miR-18a hsa-let-7g hsa-miR-19a hsa-miR-19b hsa-let-7i hsa-miR-21

Figure 1: The modulation of hsa-let-7-family miRNA in various Figure 2: The modulation of various miRNAs in different cell lines. cell lines. The analysis time indicates the period of time that had The analysis time indicates the period of time that had elapsed elapsed after exposure to ionizing radiation. The green color indi- after exposure to ionizing radiation. The green color indicates cates upregulation and red color shows downregulation of a upregulation and red color shows downregulation of a miRNA. The miRNA. The yellow color signifies no change, and the gray color yellow color signifies no change, and the gray color indicates that the indicates that the miRNAs during those time points were not exam- miRNAs during those time points were not examined. The orange ined. The orange color signified some conflicting data where it was color signified some conflicting data where it was either reported a either reported a no change or downregulation in two or more no change or downregulation in two or more studies. Similarly the studies. Similarly the light green color indicates data where it was light green color indicates data where it was either reported a no either reported a no change or upregulation in two or more studies. change or upregulation in two or more studies. 6 Comparative and Functional Genomics

LHX9

VAV3 ZNF260 PCDHAC2 GABRA1 ZIC3

LCA5 INTS6 HNRNPA3 MCM8 PCDHA12 SOCS5 RBAK ZNF451 RAB11FIP2

GABRA4 ZNF845 ARID2 BACH1 PCDHA7 ZNF254 DHX40 AAK1 PCDHA8

FAR1 TWF1 DET1 XAF1 H3F3A TAPT1 IKBIP PCSK6 IGF2

ARVCF PTN DPY19L1 PCDHA11 SLC12A6 PTPN2 hsa-miR-155 CLCC1

PCDHA13 ANKFY1 ZNF761 ZNF28 PCDHA5 PCDHA3

SERPINA10 PCDHA9

CD36 PCDHA6 ZBTB41 STXBP5L IRF2BP2 MBNL3 G2E3 TAF5L

CCDC88A SLC4A8 JARID2 SEPT11 PCDHA10 HIVEP2 ZNF652 TSEN2 IL1RAP ADAT2 GLRA3 PCDHAC1 CDH6 RBMS3 HSPA12A SCG2 DCLRE1A NDUFV3 ZNF468 ZNF860 ADO PICALM PCDHA4 PCDHA1 hsa-miR-197 SLC16A12 NDFIP1 MYCN SELP ARHGEF12 PCDHA2 TDGF1 SSR3 AFAP1L2 MPHOSPH9 C3orf23 BSN UBQLN1 EVI2B HNRNPC YIPF1 PTPN9 PGPEP1 SMAP1 NID2 ETF1 BCCIP ZNF714 GPR81 ZNF799 YPEL2 ARID3B

STARD3NL PXDN hsa-miR-202

C1orf9 KIAA1609 TAB2 OR7A5 BAZ1A SNED1 CD109 RNF139 LAMA1 MYH10 WHAMM MYLK FBXO30 IP6K2 CADM2 ATF7IP DUSP1 TFAP2B ZNF443 RFX5 RSBN1 SLC1A3 PIK3IP1 IPO11 PDPK1 RLF NIPA1 RAG2 HECTD1 DMTF1 DYNC1I1 ENTPD1 KBTBD2 MOBKL3 PCDH9 WEE1 TSG101 PRKCI hsa-miR-142-3p TMC7 C12orf23 AGBL2 DICER1 hsa-miR-575 DIAPH2 MEF2A NUP43 WASL EHF CIT CD164 HELQ CCDC6 SLC24A3 LSM12 BOD1L ARID1B SGCD hsa-miR-609 MAP9 USP6NL TMEM200B KCTD9 CHSY3 KAT2B AP2A1 PKN2 DBR1 PHF19 SNRNP40 LARP6 EIF5A2 MS4A7 NBN PCDHB12 BOD1 PIGS DAZAP1 SPAG9 DCUN1D4 STARD13 SLC22A2 ITPRIPL2 MSH6 COL24A1 GRIA2 SLC11A2 PSME3 LIPF GBP5 PPP1R3A PSAT1 CCDC19 POLR3G CLCN5 STX11 VPS54 ONECUT2 SPOPL ZC3H12C PLEKHA1 RAD50 KIAA0317 RTKN2 CTAGE1 SKAP2 AFF4 ZNF658B CLK4 CSDE1 C9orf91 EXD1 FAM18B1 CPEB2 MYT1L TM9SF3 NAV1 ANGPTL5 TAF1D KLHL24 TFCP2L1 IL22RA2 ENPEP TTPAL HNRNPH3 PLXNA2 ZCCHC2 AHCYL2 B4GALT6 DNAJB4 ROBO1 ENAM TACC1 TRUB1 BNIP2 LPHN2 CBLN4 CCDC55 FIGN CAPZA2 SLC18A2 GCH1 ARSJ KIF5C PHKA1 FGD6 CDCA4 ACTN4 SEPT2 SESN1 CDC37L1 DIO2 INSR ZNF239 ARID4B CDC42SE2 USP9X ITPR1 CUL4A RPN2 PAPPA PLEKHA3 MAGI2 RCN2 ETNK1 ARHGAP5 C8orf34 ABHD10 ASH1L VHL SH3TC2 CD2AP UBE4B VAMP2 GINS3 TMF1 TPR NIN NUP50 WASF1 BTBD1 PURA TMEM55A MKL2 TBC1D4 CDK17 CNNM3 USP15 ARHGAP28 PPIL4 TBL1XR1 PLSCR4 SPICE1 HIF1A SGPP1 VAT1 CCNH UNC80 YWHAH KIF2A ACSL1 STAG1 CASK RBM12 SIAH1 ZNF705D PVRL1 PKD2L2 ABCA1 PHC3 USP31 AHR UHMK1 MKX ACTR2 CDC25A hsa-miR-142-5p TMEM85 BTRC ADIPOR2 ZNF264 PPP1R11 CARM1 ACOX1 RNASEL EDA OSGIN2 NAV3 SESTD1 TOPORS KDM5B TLK1 CPEB3 OLFML1 VPS37B SPG20 SALL1 FLOT2 MAT2B YRDC LDHA ITGAV ANO3 SACS ATP1B1 FAM134A DAP3 SOX5 COPS2 TNS1 CCNE1 XDH RC3H1 SMAD7 CBX4 PLIN4 STC1 hsa-miR-16 YAP1 COL4A4 hsa-miR-34a FAM155A ZCCHC11 MACC1 HEPHL1 PPAPDC2 RECK TGIF2 SATB2 ZNF75D CCNT2 CLCN3 DNM1L TMX3 KIAA0776 IGF2BP3 PHYHIPL RERG ADAMTS3 TMEM216 PCMT1 USP25 RELN CNOT6L SNX18 DLL1 DUSP18 RNF138 KIF23 FUT9 KITLG SLC5A3 NENF FAM70A RAP1A CCNE2 RRAS ZNF658 LRP12 FERMT2 ZNF705A STXBP3 HCN3 ZFPM2 PPM1A CCNG2 SMURF1 hsa-miR-15a RRP15 CLDN12 UBE2D1 POMP RNF217 N4BP1 TBPL1 ERGIC1 RNF128 ZFX NSL1 LGR4 MGAT4A NPAT LATS2 PPP6C UBE2Q1 SOCS6 SEMA4F PPP2R3A TRIM36 DYRK4 PIK3R1 C1QL3 GALNT7 PRPF40A SYNJ2BP RNF10 ZBTB1 ZNF680 BTLA DCLK1 PKP4 ARHGAP20 MYB PRPF4B MYO5B NOTCH1 MEGF10 TTC14 CLCN4 SLC20A2 LRP1B KDM5D MBIP MARCH6 ABRA MFN2 RASGEF1B C1D LMX1A OXCT1 DDHD1 EZH1 TMEM20 CYBASC3 SETD2 ZNF548 GFI1B FASN GGA3 TPBG NUDT13 TMEM49 KIF1B FGF2 BAI3 UBE2K RS1 CD80 ZBTB11 B4GALT1 MTMR10 ANKRD46 CD302 MARCKS HTR2A WIPI2 ODZ1 GAS7 SH2D1A WDR26 MTMR3 ZZZ3 OTUD4 PUS7 NELF ATP7A FPR3 SCOC FEM1C KLHDC8B ZNF423 MAP2 C20orf103 MRAP2 DPY19L4 SIKE1 TRIM35 CUL2 BVES BCAP29 YWHAG TPRG1 LUZP1 PPP6R3 PARP12 CLEC14A CHD9 SRPR HELZ PGM2 TSHZ3 OSTC ADAMTSL1 MAP7 SLC25A17 HPRT1 RAB40B PAPD5 TGFBI ARID4A LRP6 ELF2 INSIG1 STAM CD1D KRIT1 AK7 WBP2NL CNBP LRRC31 hsa-miR-34b ATXN2 PLAG1 MKLN1 ATXN10 CASKIN1 RP2 NUFIP2 DDX3X ATRX DNM3 SGK3 GBP4 VCL ARHGAP24 KIF21A MTMR12 PPP1R3D F2RL2 BCAT1 PDK1 ATM DSC1 PFKM PLK4 MTF2 ERI1 LONRF2 RNF149 PPP1R3B GABRB2 ANKIB1 SATB1 TRIM13 FDX1 PUS1 CDK19 ESR1 CA13 ZNF98 AKTIP DDX60L NTF3 hsa-miR-21 FAM3C NKIRAS1 LYRM7 GK5 FAM126A PITX2 NCKAP1 PELI1 CREG2 CAPZA1 HMBOX1 PBX3 C3orf17 PRKAR2B hsa-miR-18a C7 INADL ADRB2 KCNA1 FRS2 RBM44 TRPC4 SCML2 HMGA2 LIMD2 BDH2 MRS2 UBE2D3 RABGAP1 USP38 MAP7D1 IGDCC3 TSNAX GATAD2B GLRB MALT1 HSF2 TNRC6B PLAA PHF20L1 STARD9 IGF1R LIN28B GPR64 SULT1C2 SLC30A8 REEP1 SGK1 WWP1 ZNF367 FBXO11 PBRM1 LRP2 CCNJ NEDD9 STK40 BMPR1A hsa-let-7g BRCC3 LAMP2 DIP2B KLHL31 TBC1D15 GABARAPL2 TGFBR1 FIGNL2 EPT1 NETO1 hsa-let-7d YOD1 RNF216 ANGPT2 hsa-let-7e FRMD4B ST6GALNAC3 WDR43 PDCD4 AKAP12 BZW1 TMEM143 hsa-let-7f MAP4K3 PLDN hsa-let-7i GPD2 SCN3B PAIP2B SOX9 STRN3 GRB10 PHKB TRIM23 DAB2 SCN2A hsa-let-7a TRIM71 COX15 GJA3 DCLRE1B CYP4F2 ITM2B AP2B1 SLC9A6 COIL ZBTB10 SRGAP1 QKI SPSB4 RNF145 GALR1 hsa-let-7c FLI1 KCNJ2 hsa-let-7b ADCYAP1 DMD AHCTF1 ABCE1 UBR7 HCFC2 PIGA PTGR2 TPM3 PGM2L1 ZBTB33 CAMSAP1L1 SMAD2 DUSP6 ADAMTS8 GDF6 PRPF38B PPP3CA MASP1 CITED2 NRAS RDH12 PAN3 PALM2 XRN1 E2F5 DDC YTHDF2 ZNF275 LARP4 hsa-miR-145 ITLN2 LRIG3 SLMO2 KLF5 SMARCAD1 CACNA1A MYO5A GPR26 ATL2 HIC2 SULT1E1 CPB2 UBXN2B IGF2BP1 C6orf35 SNTB2 DDI2 POLH SKP1 FBXO8 GATM AKAP9 REV3L SLC5A9 DUXA PXN PRKX CAPZB RAB18 ZFYVE26 HLA-DOA GXYLT1 C1orf56 RBPMS2 LMO7 SNX24 ZIM3 ZNF546 SLC26A4 OSBPL1A PI15 MYCL1 MYO6 FAM108B1 IMPDH1 SIX4 ADD3 C7orf58 COL14A1 RPS6KA5 IGFBP5 hsa-miR-143 UBA6 PPARA PRKY MAPK6 AFTPH PRICKLE2 TAF1L EBF1 FSCN1 ATP2B2 UBE2E1 MBD6 MTX3 FNDC3A SOX6 ETV6 PAG1 UBE2E3 RINT1 SLC35F1 FOXE1 PPP3R2 HECW2 CHD2 ZNF521 RAPGEF2 MSI2 ZNF711 ARSD SYT1 THRB HIPK2 ARFGAP3 MYLK4 ATP6V1A SHISA3 FAM108C1 C10orf46 CHIC1 DEGS1 ANKRD28 TTC26 CCNL1 RNF11 NUAK1 PLCL2 RASSF2 MPZL2 HIPK3 NDFIP2 HBP1 CASP8 ZBTB44 VASH1 ABL2 TMEM27 DERL2 RIN2 VPS37A VAX1 ITGB8 FMR1 TMEM9B CPSF6 ATP10A MED13 SLC31A2 CACNA1C KRAS DLG3 FNDC3B DDX6 SEPT7 CRKL SEMA3A RBMS1 ADSS hsa-miR-19a IGF2R KCNA4 RAP2C ELL2 SLC7A11 DTNA LONRF1 PPAPDC1B KPNA6 TP53AIP1 CLTC PYROXD1 TRIM22 WDR44 SRGAP2 PSG6 TGIF1 ARMC10 G6PC2 PTPRG CBFB LIPH TRIM33 ELMOD2 hsa-miR-19b LDLR ENPP3 EIF1B ZNF230 CAST TXK PATL1 BTAF1 CRISPLD2 SULF1 DOCK10 ATP6V1B2 AFF1 SPOCK1 ARFGEF1 IFI44L CCDC47 COPS7A DSE SORBS2 ATG16L1 RGL1 SYNM CYLD RAB8B hsa-miR-376a CHST1 ELOVL5 PHTF2 SIN3B BMPR2 ACSL4 SLC7A2 DSCAML1 SLC30A9 ELK3 IMPG2 ATG14 BAMBI MTRR ATP2C1 SNIP1 SYBU CELF4 ARRDC3 TGFBR3 PMEPA1 IGSF3 KIAA0494 GRSF1 BRWD1 UBL3 ATP6V1G1 SCN9A ADNP ZNF80 ZMYND11 REEP3 PRKACB ZNF238 ZNF217 CDS1 TSPAN12 PRUNE2 RHBDL2 PDIA6 RIMKLA ATXN1 GRIK2 RHOB ABCC9 WDR1 STEAP2 ENOX2 MPPED2 S1PR1 KLHDC5 LCLAT1 ARHGAP12 SBF2 IREB2 RAP1B NCOA4 ANO1 PHF13 CGGBP1 GJA1 N6AMT1 RCOR3 MOBKL1A ENPP5 LIN9

PSG1 CAMSAP1

Figure 3: Visualization of miRNAs and their associated target genes network with Cytoscape. The interaction network shows nodes and connections between miRNAs and the target genes. The yellow nodes represent the miRNA and the pink/red nodes represent its targeted gene. The opacity of the blue edge links signiﬁes the target score between the miRNA and the target gene. The opacity of red nodes represents genes that have been identiﬁed to play roles in biological pathways.

biological pathways for hsa-miR-15a and hsa-miR-16 were 22 We also identiﬁed the genes that were controlled by (Table 2). The target genes for all these miRNA are shown multiple miRNA after exposure to IR. Figure 4 shows the list in Table 2. These interactions demonstrated that a singular of genes along with the miRNA that regulate their expression. miRNA has the capability to aﬀect the regulation of a large ThegenesthatwereregulatedbytwoormoremiRNA number of genes. It was apparent that a number of miRNAs were included in the Figure 4.ForexampleLRIG3 gene could work together to downregulate several genes. was targeted by hsa-let-7a, hsa-let-7b, hsa-let-7c, hsa-let-7d, Comparative and Functional Genomics 7

microRNA to gene association heatmap hsa-let-7a hsa-let-7b hsa-let-7c hsa-let-7d hsa-let-7e hsa-let-7f hsa-let-7g hsa-let-7i hsa-miR-142-3p hsa-miR-142-5p hsa-miR-143 hsa-miR-145 hsa-miR-155 hsa-miR-15a hsa-miR-16 hsa-miR-18a hsa-miR-197 hsa-miR-19a hsa-miR-19b hsa-miR-202 hsa-miR-21 hsa-miR-34a hsa-miR-34b hsa-miR-376a hsa-miR-575 hsa-miR-609

Gene symbol unique hits: Total Gene description LRIG3 10 Leucine-rich repeats and immunoglobulin-like domains 3 TGFBR1 10 Transforming growth factor, beta receptor 1 ADRB2 9 Adrenergic, beta-2-, receptor, surface GATM 9 Glycine amidinotransferase (L-arginine: glycine amidinotransferase) HMGA2 9 High mobility group AT-hook 2 PBX3 9 Pre-B-cell leukemia homeobox 3 USP38 9 Ubiquitin speciﬁc peptidase 38 ADAMTS8 8 ADAM metallopeptidase with thrombospondin type 1 motif, 8 CCNJ 8 Cyclin J COIL 8 Coilin DDI2 8 DNA-damage inducible 1 homolog 2 ((S. cerevisiae) E2F5 8 E2F transcription factor 5, p130-blinding FIGNL2 8 Fidgetin-like 2 FRMD4B 8 FERM domain containing 4B GDF6 8 Growth differentiation factor 6 IGF2BP1 8 Insulin-like growth factor 2 mRNA binding protein 1 KLHL31 8 Kelch-like 31 (Drosophila) LIN28B 8 Lin-28 homolog B (C. elegans) MAP4K3 8 Mitogen-activated protein kinase kinase kinase kinase 3 MRS2 8 MRS2 magnesium homeostasis factor homolog (S. cerevisiae) PRPF38B 8 PRP38 pre-mRNA processing factor 38 (yeast) domain containing B ribH 8 6, 7-dimethyl-6-ribityllumazine synthase SLC5A9 8 Solute carrier family 5 (sodium/glucose cotransporter), member 9 SMARCAD1 8 SWI/SNF-related, matrix-associated actin-dependent regulator of chromatin, subfamily a, containing DEAD/H box 1 STARD9 8 StAR-related lipid transfer (START) domain containing 9 TRIM71 8 Tripartite motif-containing 71 ZNF275 8 Zinc ﬁnger protein 275 C14orf28 7 Chromosome 14 open reading frame 28 DMD 7 Dystrophin IGDCC3 7 Immunoglobulin superfamily, DCC subclass, member 3 YOD1 7 YOD1 OTU deubiquitinating enzyme 1 homolog (S. cerevisiae) NRAS 6 Neuroblastoma RAS viral (v-ras) oncogene homolog TMEM143 6 Transmembrane protein 143 ANKIB1 5 Ankyrin repeat and IBR domain containing 1 DCLRE1B 5 DNA cross-link repair IB DDX3X 5 DEAD (Asp-Glu-Ala-Asp) box polypeptide 3, X-linked QKI 5 Quaking homolog, KH domain RNA binding (mouse) TNRC6B 5 Trinucleotide repeat containing 6B LRP2 4 Low density lipoprotein receptor-related protein 2 MOBKL3 4 MOB1, Mps One Binder kinase activator-like 3 (yeast) SGK1 4 Serum/glucocorticoid regulated kinase 1 SYT1 4 Synaptotagmin I AHCTF1 3 AT hook containing transcription factor 1 ASH1L 3 Ash1 (absent, small, or homeotic)-like (Drosophila) ATXN2 3 Ataxin 2 BMPR2 3 Bone morphogenetic protein receptor, type II (serine/threonine kinase) CASK 3 Calcium/calmodulin-dependent serine protein kinase (MAGUK family) CHIC1 3 Cysteine-rich hydrophobic domain 1 CPEB2 3 Cytoplasmic polyadenylation element binding protein 2 CPEB3 3 Cytoplasmic polyadenylation element binding protein 3 DMTF1 3 Cyclin D binding myb-like transcription factor 1 DYNC1I1 3 Dynein, cytoplasmic 1, intermediate chain 1 ENTPD1 3 Ectonucleoside triphosphote diphosphohydrolase 1 FAM108C1 3 Family with sequence similarity 108, member C1

FAM70A 3 Family with sequence similarity 70, member A FBXO30 3 F-box protein 30 HCFC2 3 Host cell factor C2 ITGB8 3 Integrin, beta 8 KBTBD2 3 Kelch repeat and BTB (POZ) domain containing 2 KCNA4 3 Potassium voltage-gated channel, shaker-related subfamily, member 4 KCNJ2 3 Potassium inwardly-rectifying channel, subfamily J, member 2 KIF21A 3 Kinesin family member 21A LOC728931 3 Adenosylmethionine decarboxylase 1 pseudogene LRP6 3 Low density lipoprotein receptor-related protein 6 MGAT4A 3 Mannosyl (alpha-1,3-)-glycoprotein beta-1,4-N-acetylglucosaminyltransferase, isozyme A PCDH9 3 Protocadherin 9 PGM2L1 3 Phosphoglucomutase 2-like 1 PLAG1 3 Pleiomorphic adenoma gene 1 RECK 3 Reversion-inducing-cysteine-rich protein with Kazal motifs RNF145 3 Ring finger protein 145 SH3TC2 3 SH3 domain and tetratricopeptide repeats 2 SLC9A6 3 Solute carrier family 9 (sodium/hydrogen exchanger), member 6 TAB2 3 TGF-beta activated kinase 1/MAP3K7 binding protein 2 TRIM23 3 Tripartite motif-containing 23 TSHZ3 3 Teashirt zinc finger homeobox 3 TTPAL 3 Tocopherol [alpha] transfer protein-like WEE1 3 WEE1 homolog (S. pombe) ABCC9 2 ATP-binding cassette, sub-family C (CFTR/MRP), member 9 ACOX1 2 Acyl-CoA oxidase 1, palmitoyl ACSL4 2 Acyl-CoA synthetase long-chain family member 4 ACTR2 2 ARP2 actin-related protein 2 homolog (yeast) ADAMTS3 2 ADAM metallopeptidase with thrombospondin type 1 motif, 3 ADNP 2 Activity-dependent neuroprotector homeobox ADSS 2 Adenylosuccinate synthase AFF1 2 AF4/FMR2 family, member 1 AFTPH 2 Aftiphilin AHCYL2 2 Adenosylhomocysteinase-like 2 ANKRD46 2 Ankyrin repeat domain 46 ANO3 2 Anoctamin 3 ARFGEF1 2 ADP-ribosylation factor guanine nucleotide-exchange factor l(brefeldin A-inhibited) ARRDC3 2 Arrestin domain containing 3 ARSD 2 Arylsulfatase D ATG14 2 ATG14 autophagy related 14 homolog (S. cerevisiae) ATG16L1 2 ATG16 autophagy related 16-like 1 (S. cerevisiae) ATP2B2 2 ATPase, Ca transporting, plasma membrane 2 ATP2C1 2 ATPase, Ca transporting, type 2C, member 1 ATP6V1B2 2 ATPase, H transporting, lysosomal 56/58 kDa, VI subunit B2 ATP7A 2 ATPase, Cu transporting, alpha polypeptide ATXN1 2 ataxin 1 BAMBI 2 BMP and activin membrane-bound inhibitor homolog (Xenopus laevis) BCAP29 2 B-cell receptor-associated protein 29 BEND4 2 BEN domain containing 4 BMPR1A 2 Bone morphogenetic protein receptor, type IA BTAF1 2 BTAF1 RNA polymerase II, B-TFIID transcription factor-associated, 170 kDa (Mot1 homolog, S. cerevisiae) BTRC 2 Beta-transducin repeat containing C10orf46 2 Chromosome 10 open reading frame 46 C11orf96 2 Chromosome 11 open reading frame 96 C1orf21 2 Chromosome 1 open reading frame 21 C1QL3 2 Complement component 1, q subcomponent-like 3 C8orf86 2 Chromosome 8 open reading frame 86 CACNA1C 2 Calcium channel, voltage-dependent, L type, alpha 1C subunit CAPZA2 2 Capping protein (actin filament] muscle Z-line, alpha 2 CARM1 2 Coactivator-associated arginine methyltransferase 1 CASP8 2 Caspase 8, apoptosis-related cysteine peptidase CAST 2 Calpastatin Figure 4: The regulation of genes with various miRNA. The genes that are controlled by two or more different miRNAs are shown. The green shade identifies the miRNA predicted to regulate that particular gene. The red shade identifies the miRNA that are not predicted to control the expression of the listed gene. 8 Comparative and Functional Genomics

Table 2: miRNA and the predicted target genes identiﬁed in biological pathways. miRNA Target genes identiﬁed in biological pathways AP2A1, PHKA1, CDC25A, SMAD7, CCNE1, SMURF1, PIK3R1, INSR, SESN1, ARHGAP5, UBE4B, miR-15a, miR-16 YWHAH, SIAH1, BTRC, RELN, UBE2Q1, FGF2, FASN, WEE1, SGK1, TGFBR1 miR-202 SMAP1, DUSP1, LAMA1 miR-155 VAV3, DET1, RAB11FIP, IGF2 miR-197 IL1RAP, ARHGEF12 DIAPH2, CUL4A, CCNH, STAG1, RAP1A, UBE2K, TPR, ACTN4, VHL, ATP1B1, ITGAV, CCNG2, miR-142-5p UBE2D1, CUL2, HIF1A miR-142-3p WASL, MYLK, TAB2 miR-575 GRIA2, TSG101, MYH10 miR-609 PRKC1, PDPK1, PPP1R3A miR-34A WASF1, PVRL1, FLOT2, VSP37B, COL4A4, DNM1L, PPM1A, CCNE2, RRAS miR-34b YWHAG, DNM3, PDK1, NCKAP1, PRKAR2E miR-18a ATM miR-21 PPP1R3B, NTF3, FRS2, VCL, PPP1R3D, PITX2, UBE2D3, WWP1 miR-let 7a, b, c, d, e, f, g, i IGF1R, ADRB2, MAP4K3, NRAS, E2F5, GDF6 miR-143 CACNA1A, PRKX, UBE2E1, KRAS, LMO7, UBE2E3, ARFGAP3 miR-376a TP53AIP1, G8PC2, PRKAC8 miR-19a, miR-19b RPS6KA5, RAPGEF2, CASP8, VPS37A, CACNA1C, CLTC, ITGB8, LDLR, BMPR2, NCOA4, RAP1B miR- 145 DAB2, AP2B1, TPM3, SMAD2, DUSP6, PPP3CA, PHHKB, SKP1ITGB8, PPP3R2, CRKL, UBA6, PXN

hsa-let-7e, hsa-let-7f, hsa-let-7g, hsa-let-7i, hsa-miR-19a,and genes were found to be associated with regulation of actin hsa-miR-19b. Similarly TGFBR1 gene was regulated by hsa- cytoskeleton, and another 20 genes belonged to endocytosis. let-7a, hsa-let-7b, hsa-let-7c, hsa-let-7d, hsa-let-7e, hsa-let- A large number of studies have documented the involvement 7f, hsa-let-7g, hsa-let-7i, hsa-miR-142-3p,andhsa-miR-145. of MAPK-signaling pathway in the response to radiation This analysis clearly indicated that individual genes were exposure [33]. The actin cytoskeleton and endocytosis path- controlled by multiple miRNA after exposure to IR. The net ways are affected in cells treated with IR [34–36]. We also effect of miRNA modulation in irradiated cells is enormous identified the participation of insulin signaling pathway in impacting many cellular functions. radiation response by mapping 19 miRNA target genes to this pathway. The involvement of insulin signaling pathway in radiation response has been reported [37]. Our analysis 3.4. Mapping of miRNA-Predicted Target Genes to Biological confirmed the participation of apoptosis, cell cycle, p53 sig- Pathways Affected by Radiation. Using the KEGG database naling, TGF-beta signaling, all known to be disturbed in cells we searched for pathways where the miRNA target genes treated with IR [38–41]. Other pathways that we identified in function in order to gain insight into the processes that could this analysis and have been documented in radiation res- be affected by miRNA modulation in irradiated cells. First ponse were adherens junction [42], focal adhesion [43], the mirDB was linked to the Gene Database from the NCBI oocyte meiosis [44], renal cell carcinoma [45], thyroid cancer (National Center for Biotechnology Information) to identify [46], and ubiquitin-mediated proteolysis [47]. the functions of the specific genes. The Gene Database from Interestingly, the interactome analysis reported in the the NCBI was retrieved for offline purposes. This database present study permitted us to discover novel pathways that provided descriptive information regarding the functions have not been previously associated with ionizing radiation of miRNA target gene. Second, the Entrez Gene ID was response. We discovered that radiation-induced miRNA linked to the KEGG pathways database to identify the genes control the expression of a number of genes that function associated with biological pathways. The miRNA target gene in aldosterone-regulated sodium reabsorption, long-term dataset was imported into DAVID, a bioinformatics tool that potentiation, and neurotrophin signaling pathways. provides functional gene-annotation. The genes associated The mineralocorticoid hormone, aldosterone is a key with various biological pathways were determined with regulator of sodium homeostasis. The aldosterone controls DAVID and the interactions were visualized with Cytoscape sodium reabsorption by regulating the cell-surface expres- (Figure 5). sion and function of the epithelial sodium channel (ENaC). The functional pathways highlighted by Cytoscape are The stimulatory effect of aldosterone on ENaC is mediated by shown in Table 3.Wewereabletomap222miRNApredicted the induction of serum- and glucocorticoid-regulated kinase targetgenesto17biologicalpathways(Table 3). 22 of these 1(SGK1)[48]. The promyelocytic leukemia zinc finger pro- genes were identified in MAPK-signaling pathway, 20 target tein (PLZF) is also upregulated by aldosterone. PLZF is Comparative and Functional Genomics 9

SESN1 TP53AIP1 CCNG2 GDF6 p53 signaling pathway BMPR2 SMAD7

TGF-beta PITX2 signaling pathway

ATM CDC25A BMPR1A STAG1 CASP8 CCNE1 E2F5 WEE1

CCNE2 CCNH Cell cycle VPS37B SMAD2 SIAH1

DAB2 AP2B1 DNM3 SKP1 DNM1L

ARFGAP3 SMAP1 YWHAH SMURF1 TSG101 YWHAG TGFBR1 VPS37A

CLTC Endocytosis AP2A1 IL1RAP STAM TAB2 LDLR MAP4K3 Oocyte meiosis ITPR1 Apoptosis RAB11FIP2 CACNA1A ADRB2 DUSP6 PPP3CA WWP1 DUSP1 BTRC CACNA1C LMO7 PPP3R2 PPM1A PVRL1 UBE2D1 PRKX Ubiquitin IGF1R mediated UBE2E3 RAPGEF2 MAPK signaling pathway Adherens junction GRIA2 PRKAR2B proteolysis UBE2E1

PRKCI UBE4B DET1 Neurotrophin Long-term UBE2D3 signaling pathway CUL4A RPS6KA5 FRS2 potentiation HIF1A VHL UBE2K UBE2Q1 NTF3 PDK1 PRKACB UBA6 Renal cell CUL2 RAP1A carcinoma

WASF1 RRAS RAP1B FGF2 WASL

ACTN4 INSR FLOT2 CRKL PIK3R1 MYH10 VCL Insulin signaling FASN DIAPH2 Regulation of pathway actin IGF2 cytoskeleton NRAS NCKAP1 PHKA1 ARHGEF12 PPP1R3B KRAS PHKB PPP1R3A

PXN G6PC2 Focal adhesion PPP1R3D VAV3 MYLK Aldosterone- ITGB8 ITGAV regulated sodium PDPK1 reabsorption LAMA1

RELN ATP1B1 ARHGAP5 Thyroid cancer SGK1 COL4A4

NCOA4 TPM3 CCDC6 TPR

Figure 5: Mapping of miRNA targets genes to biological pathways. This network depicts the miRNA target genes and the associated biological pathways. The landscape of genes to pathway interactions was visualized with Cytoscape. Pink or red nodes represent the genes and the blue nodes indicate biological pathways.

involved in cell cycle control and cell differentiation [49]. of Alzheimer’s disease [50]. The response of central ner- How the activation of aldosterone-regulated sodium reab- vous system (CNS) to the IR exposure has not been under- sorption pathway contributes to the response of irradiated stood. Perhaps signaling molecules act downstream of IGF1- cells remain to be investigated. R, and there is a checkpoint to balance excessive growth/ The neurotrophin-signaling pathway is involved in differ- “immortality” and reduced growth/“senescence” of a cell. entiation and survival of neural cells. The insulin/insulin-like Future investigations might define this connection and its growth factor 1 receptor-signaling (IGF1-R) pathway is relationship with radiation effects. linked to the neurogenic capacities of the aging brain, to The long-term potentiation pathway is associated with neurotrophin signaling, and to the molecular pathogenesis long-lasting enhancement in signal transmission between 10 Comparative and Functional Genomics

Table 3: Biological pathways controlled by IR-modulated miRNAs and their association with ionizing radiation response.

Identiﬁed pathways Number of miRNA target genes Association with IR response Reference Adherens junction 10 Yes [42] Aldosterone-regulated sodium reabsorption 7 Unknown N/A Apoptosis 9 Yes [38] Cell cycle 12 Yes [41] Endocytosis 20 Yes [35] Focal adhesion 17 Yes [43] Insulin signaling 19 Yes [37] Long-term potentiation 11 Unknown N/A MAPK signaling 22 Yes [33] Neutrotrophin signaling 12 Unknown N/A Oocyte meiosis 13 Yes [44] p53 signaling 8 Yes [39] Regulation of actin cytoskeleton 20 Yes [34] Renal cell carcinoma 9 Yes [45] TGF-β signaling 10 Yes [40] Thyroid cancer 6 Yes [46] Ubiquitin-mediated proteolysis 17 Yes [47]

two neurons. The plastic changes at synapses between neu- pathways. miRNAs may act as “hub” regulators of specific rons are partly associated with the memory. The long-term cellular responses, immediately downregulated so as to potentiation (LTP) is a major form of synaptic plasticity stimulate DNA repair mechanisms, followed by upregulation [51]. The possible implications of long-term potentiation involved in suppressing apoptosis for cell survival. Taken pathway in radiation-induced biological effects remains to be together, miRNAs may mediate signaling pathways in investigated. sequential fashion in response to radiation. Future studies Our findings suggest that the miRNA target gene inter- will be aimed to understand the effect of miRNA perturba- actome can help identify novel cellular functions that could tion on the disruption of biological pathways. Though the be altered as a result of stress induced by radiation exposure. genes that are associated with the pathways have been The ability to discover previously uncharacterized new novel determined, it is still unclear whether an activation or inhi- pathways through understanding the interactome of miRNA- bition of the pathway takes place in the cells exposed to predicted target genes and -associated pathways offers a new radiation. platform for future investigations. A deeper understanding of the miRNA expression signatures in different cell types subjected to IR exposure will not only lead to identify Conflict of Interests ff common biological pathways a ected in all cell types but will The authors declare that they have no conflict of interests. also permit to discover pathways that are only affected in certain cell types. Authors’ Contributions 4. Conclusions M. A. Chaudhry conceived the idea and planned the study. T. W. Lhakang performed the experiments and the informatics It is apparent that miRNAs are involved in controlling analysis. M. A. Chaudhry and T. W. Lhakang wrote the paper. the biological pathways associated with ionizing radiation induced stress responses. The miRNA expression alterations in irradiated cells explains the observed biological effects and Acknowledgment provides a broader perspective on understanding cellular defense mechanisms against radiation-induced insult. This M. A. Chaudhry is supported by an endowment fund, Col- investigation has provided a starting point where the role of lege of Nursing and Health Sciences, University of Vermont, miRNAs in ionizing radiation can be explored. The pathways VT, USA. affected by IR-induced miRNA provide vital information to understand the regulation of the biological processes in References cells exposed to IR. It has always been assumed that only transcriptional factors affect the gene expression and control [1] M. I. Almeida, R. M. Reis, and G. A. Calin, “MicroRNA his- biological pathways. However, the participation of miRNAs tory: discovery, recent applications, and next frontiers,” Muta- adds another set of rules dictating control of the biological tion Research, vol. 717, no. 1-2, pp. 1–8, 2011. Comparative and Functional Genomics 11

[2] P. Brodersen and O. Voinnet, “Revisiting the principles of lymphoblastic cell line,” International Journal of Oncology, vol. microRNA target recognition and mode of action,” Nature 34, no. 6, pp. 1661–1668, 2009. Reviews Molecular Cell Biology, vol. 10, no. 2, pp. 141–148, [21] H. J. Cha, K. M. Seong, S. Bae et al., “Identification of specific 2009. microRNAs responding to low and high dose γ-irradiation in [3] G. Ku and M. T. McManus, “Behind the scenes of a small RNA the human lymphoblast line IM9,” Oncology Reports, vol. 22, gene-silencing pathway,” Human Gene Therapy,vol.19,no.1, no. 4, pp. 863–868, 2009. pp. 17–26, 2008. [22] S. Shin, H. J. Cha, E. M. Lee et al., “Alteration of miRNA [4] Y. Huang, X. J. Shen, Q. Zou, S. P. Wang, S. M. Tang, and profiles by ionizing radiation in A549 human non-small cell G. Z. Zhang, “Biological functions of microRNAs: a review,” lung cancer cells,” International Journal of Oncology, vol. 35, Journal of Physiology and Biochemistry, vol. 67, no. 1, pp. 129– no. 1, pp. 81–86, 2009. 139, 2011. [23] N. L. Simone, B. P. Soule, D. Ly et al., “Ionizing radiation- [5] T. A. Farazi, J. I. Spitzer, P. Morozov, and T. Tuschl, “miRNAs induced oxidative stress alters miRNA expression,” PLoS ONE, in human cancer,” Journal of Pathology, vol. 223, no. 2, pp. vol. 4, no. 7, Article ID e6377, 2009. 102–115, 2011. [24]C.J.Marsit,K.Eddy,andK.T.Kelsey,“MicroRNAresponses [6] L. Chin, W. C. Hahn, G. Getz, and M. Meyerson, “Making to cellular stress,” Cancer Research, vol. 66, no. 22, pp. 10843– sense of cancer genomic data,” Genes and Development, vol. 10848, 2006. 25, no. 6, pp. 534–555, 2011. [25] X. Wang, “miRDB: a microRNA target prediction and func- [7] M. A. Chaudhry, “Bystander effect: biological endpoints and tional annotation database with a wiki interface,” RNA, vol. microarray analysis,” Mutation Research, vol. 597, no. 1-2, pp. 14, no. 6, pp. 1012–1017, 2008. 98–112, 2006. [26] M. J. Allalunis-Turner, G. M. Barron, R. S. Day III, K. D. [8] M. A. Chaudhry, “Biomarkers for human radiation exposure,” Dobler, and R. Mirzayans, “Isolation of two cell lines from a Journal of Biomedical Science, vol. 15, no. 5, pp. 557–563, 2008. human malignant glioma specimen differing in sensitivity to [9] M. A. Chaudhry, “Radiation-induced gene expression profile radiation and chemotherapeutic drugs,” Radiation Research, of human cells deficient in 8-hydroxy-2-deoxyguanine gly- vol. 134, no. 3, pp. 349–354, 1993. cosylase,” International Journal of Cancer, vol. 118, no. 3, pp. [27] J. L. Schwartz, R. Jordan, and H. H. Evans, “Characteristics 633–642, 2006. of chromosome instability in the human lymphoblast cell line [10] A. K. L. Leung and P. A. Sharp, “MicroRNA functions in stress WTK1,” Cancer Genetics and Cytogenetics, vol. 129, no. 2, pp. responses,” Molecular Cell, vol. 40, no. 2, pp. 205–215, 2010. 124–130, 2001. [11] I. A. Babar, F. J. Slack, and J. B. Weidhaas, “miRNA modulation [28]J.S.Oh,J.J.Kim,J.Y.Byun,andI.A.Kim,“Lin28-let7mod- of the cellular stress response,” Future Oncology, vol. 4, no. 2, ulates radiosensitivity of human cancer cells with activation pp. 289–298, 2008. of K-Ras,” International Journal of Radiation Oncology Biology [12] M. A. Chaudhry, “Real-time PCR analysis of micro-RNA Physics, vol. 76, no. 1, pp. 5–8, 2010. expression in ionizing radiation-treated cells,” Cancer Biother- [29] G. A. Calin, C. D. Dumitru, M. Shimizu et al., “Frequent apy and Radiopharmaceuticals, vol. 24, no. 1, pp. 49–56, 2009. deletions and down-regulation of micro-RNA genes miR15 [13] M. A. Chaudhry, B. Kreger, and R. A. Omaruddin, “Tran- and miR16 at 13q14 in chronic lymphocytic leukemia,” scriptional modulation of micro-RNA in human cells differing Proceedings of the National Academy of Sciences of the United in radiation sensitivity,” International Journal of Radiation States of America, vol. 99, no. 24, pp. 15524–15529, 2002. Biology, vol. 86, no. 7, pp. 569–583, 2010. [30] A. Cimmino, G. A. Calin, M. Fabbri et al., “miR-15 and miR- [14] M. A. Chaudhry, H. Sachdeva, and R. A. Omaruddin, “Rad- 16 induce apoptosis by targeting BCL2,” Proceedings of the iation-induced micro-RNA modulation in glioblastoma cells National Academy of Sciences of the United States of America, differing in DNA-repair pathways,” DNA and Cell Biology, vol. vol. 102, no. 39, pp. 13944–13949, 2005. 29, no. 9, pp. 553–561, 2010. [31] N. Yanaihara, N. Caplen, E. Bowman et al., “Unique micro- [15] M. Wagner-Ecker, C. Schwager, U. Wirkner, A. Abdollahi, and RNA molecular profiles in lung cancer diagnosis and progno- P. E. Huber, “MicroRNA expression after ionizing radiation sis,” Cancer Cell, vol. 9, no. 3, pp. 189–198, 2006. in human endothelial cells,” Radiation Oncology, vol. 5, no. 1, [32] Y. Xi, J. R. Edwards, and J. Ju, “Investigation of miRNA biology article 25, 2010. by bioinformatic tools and impact of miRNAs in colorectal [16] S. H. Jeong, H. G. Wu, and W. Y. Park, “LIN28B confers radio- cancer—regulatory relationship of c-Myc and p53 with miR- resistance through the posttranscriptional control of KRAS,” NAs,” Cancer Informatics, vol. 3, pp. 245–253, 2007. Experimental and Molecular Medicine, vol. 41, no. 12, pp. 912– [33] P. Dent, A. Yacoub, P. B. Fisher, M. P. Hagan, and S. Grant, 918, 2009. “MAPK pathways in radiation responses,” Oncogene, vol. 22, [17] O. C. Maes, J. An, H. Sarojini, H. Wu, and E. Wang, “Changes no. 37, pp. 5885–5896, 2003. in microRNA expression patterns in human fibroblasts after [34] J. Yue, Q. Wang, H. Lu, M. Brenneman, F. Fan, and Z. Shen, low-LET radiation,” Journal of Cellular Biochemistry, vol. 105, “The cytoskeleton protein filamin-A is required for an efficient no. 3, pp. 824–834, 2008. recombinational DNA double strand break repair,” Cancer [18] S. Josson, S. Y. Sung, K. Lao, L. W. K. Chung, and P. A. S. Research, vol. 69, no. 20, pp. 7978–7985, 2009. Johnstone, “Radiation modulation of microRNA in prostate [35] D. A. Gewirtz, M. L. Hilliker, and E. N. Wilson, “Promotion of cancer cell lines,” Prostate, vol. 68, no. 15, pp. 1599–1606, 2008. autophagy as a mechanism for radiation sensitization of breast [19] S. Shin, H. J. Cha, E. M. Lee et al., “MicroRNAs are signif- tumor cells,” Radiotherapy and Oncology,vol.92,no.3,pp. icantly influenced by p53 and radiation in HCT116 human 323–328, 2009. colon carcinoma cells,” International Journal of Oncology, vol. [36] C. Berndt, T. Kurz, M. Selenius, A. P. Fernandes, M. R. Edgren, 34, no. 6, pp. 1645–1652, 2009. and U. T. Brunk, “Chelation of lysosomal iron protects against [20] H. J. Cha, S. Shin, H. Yoo et al., “Identification of ioniz- ionizing radiation,” Biochemical Journal, vol. 432, no. 2, pp. ing radiation-responsive microRNAs in the IM9 human B 295–301, 2010. 12 Comparative and Functional Genomics

[37] T. Sakurai, T. Ueda, M. Kawai, H. Tobita, and J. Miyakoshi, “Protective effects of insulin-like growth factor-I on the decrease in myogenic differentiation by ionizing radiation,” International Journal of Radiation Biology,vol.85,no.2,pp. 153–158, 2009. [38] S. W. Ryter, P. K. Hong, A. Hoetzel et al., “Mechanisms of cell death in oxidative stress,” Antioxidants and Redox Signaling, vol. 9, no. 1, pp. 49–89, 2007. [39] X. P. Zhang, F. Liu, and W. Wang, “Two-phase dynamics of p53 in the DNA damage response,” Proceedings of the National Academy of Sciences of the United States of America, vol. 108, no. 22, pp. 8990–8995, 2011. [40] K. L. Andarawewa, J. Paupert, A. Pal, and M. H. Barcellos- Hoff, “New rationales for using TGFβ inhibitors in radiotherapy,” International Journal of Radiation Biology, vol. 83, no. 11- 12, pp. 803–811, 2007. [41] K. Ishikawa, H. Ishii, and T. Saito, “DNA damage-dependent cell cycle checkpoints and genomic stability,” DNA and Cell Biology, vol. 25, no. 7, pp. 406–411, 2006. [42] A. D. de Carvalho, W. de Souza, and J. A. Morgado-D´ıaz, “Morphological and molecular alterations at the junctional complex in irradiated human colon adenocarcinoma cells, Caco-2,” International Journal of Radiation Biology, vol. 82, no. 9, pp. 658–668, 2006. [43] V. Sandfort, U. Koch, and N. Cordes, “Cell adhesion-mediated radioresistance revisited,” International Journal of Radiation Biology, vol. 83, no. 11-12, pp. 727–732, 2007. [44] A. Pesty, M. Doussau, J. B. Lahaye, and B. Lefevre,` “Whole- body or isolated ovary 60Co irradiation: effects on in vivo and in vitro folliculogenesis and oocyte maturation,” Reproductive Toxicology, vol. 29, no. 1, pp. 93–98, 2010. [45] D. B. Richardson and G. Hamra, “Ionizing radiation and kidney cancer among Japanese atomic bomb survivors,” Rad- iation Research, vol. 173, no. 6, pp. 837–842, 2010. [46]P.D.Inskip,A.Ekbom,M.R.Galanti,L.Grimelius,andJ. D. Boice Jr., “Medical diagnostic x rays and thyroid cancer,” Journal of the National Cancer Institute, vol. 87, no. 21, pp. 1613–1621, 1995. [47]T.Kang,Y.Wei,Y.Honakeretal.,“GSK-3β targets Cdc25A for ubiquitin-mediated proteolysis, and GSK-3β inactivation correlates with Cdc25A overproduction in human cancers,” Cancer Cell, vol. 13, no. 1, pp. 36–47, 2008. [48] P. Fakitsas, G. Adam, D. Daidie´ et al., “Early aldosterone- induced gene product regulates the epithelial sodium channel by deubiquitylation,” Journal of the American Society of Nephrology, vol. 18, no. 4, pp. 1084–1092, 2007. [49] M. Barna, N. Hawe, L. Niswander, and P. P. Pandolfi, “Plzf regulates limb and axial skeletal patterning,” Nature Genetics, vol. 25, no. 2, pp. 166–172, 2000. [50] L. Puglielli, “Aging of the brain, neurotrophin signaling, and Alzheimer’s disease: is IGF1-R the common culprit?” Neuro- biology of Aging, vol. 29, no. 6, pp. 795–811, 2008. [51] J. A. Blundon and S. S. Zakharenko, “Dissecting the compo- nents of long-term potentiation,” Neuroscientist, vol. 14, no. 6, pp. 598–608, 2008. Copyright of Comparative & Functional Genomics is the property of Hindawi Publishing Corporation and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.