- Email: [email protected]
Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibody-based protein array
FEBS Letters xxx (2015) xxx–xxx
journal homepage: www.FEBSLetters.org
Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibody-based protein array Yazhou Cui a, Shuyang Xie b, Jing Luan a, Xiaoyan Zhou a, Jinxiang Han a,⇑ a b
Shandong Academy of Medical Sciences, Shandong Medical Biotechnological Center, Key Laboratory for Biotech Drugs of the Ministry of Health, China Department of Biochemistry and Molecular Biology, Binzhou Medical University, YanTai, Shandong, China
a r t i c l e
i n f o
Article history: Received 13 May 2015 Revised 16 June 2015 Accepted 19 June 2015 Available online xxxx Edited by Tamas Dalmay Keywords: MicroRNA Target gene Proteomic miR-206 Antibody array
a b s t r a c t This study demonstrated the feasibility and benefit of an antibody-based experimental approach to identify microRNA functional targets from hundreds of predicted genes using miR-206 as an example. Using a receptor tyrosine kinase (RTK) antibody array, we identified 7 phosphorylated RTKs that were significantly differentially regulated after miR-206-mimic transfection. We then focused on MET, the most varied RTK, and bioinformatically constructed a MET-centred network using computationally predicted miR-206 targets. Within this network, we analyzed two validated targets, PAX3 and SNX2, and one candidate target, EIF4E, may account for the inhibitory effect of miR-206 on MET phosphorylation. Luciferase and Western-blot assays indicated that EIF4E was a direct target of miR-206. This concept may also be applicable for other microRNAs and other antibody array platforms. Ó 2015 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
1. Introduction MicroRNAs (miRNAs) are important regulators of gene expression in eukaryotic cells. In humans, it is estimated that miRNAs may regulate up to 30% of the protein-coding genes. miRNAs are known to control a broad spectrum of biological processes, and their dysregulation may be associated with multiple diseases [1]. The identification of functional targets is central to understanding the molecular mechanisms by which individual miRNAs modulate cellular functions [2]. Although some computational programs based on sequence alignment have been widely applied to predict the targets of miRNAs, these bioinformatic tools still yield a high percentage of false-positives and false-negatives [3,4]. Furthermore, it is believed that a single miRNA can regulate hundreds of target mRNAs; therefore, the identification of genuine functional target genes remains a fundamental challenge. Therefore, these obstacles underscore the requirement for experimental data that determines functionally
Author contributions: JX Han and YZ Cui designed the study and wrote the manuscript. YZ Cui, SY Xie and J Luan performed the experiments and data analysis. XY Zhou participated in the data analysis. ⇑ Corresponding author at: Shandong Academy of Medical Sciences, Shandong Medical Biotechnological Center, Key Laboratory for Biotech Drugs of the Ministry of Health, 18877 Jingshi Road, Jinan 260062, China. Fax: +86 53182951586. E-mail address: [email protected] (J. Han).
important gene targets for elucidating the exact roles of miRNAs in development and disease [5]. Because miRNAs can repress the expression of target genes at the mRNA and/or protein levels, a series of experimental strategies based on gene expression microarray or proteomic tools have been proposed for microRNA target identification [6]. Because most targets may be repressed by an miRNA at the protein level without being affected at the mRNA level, proteomic methods are regarded as the most powerful experimental strategies for revealing the full spectrum of miRNA targets [7]. Currently, various quantitative proteomic methodologies have been employed in the identification of miRNA targets, such as two-dimensional difference gel electrophoresis (2D-DIGE) [8], isobaric tags for relative and absolute quantification (iTRAQ) [9], stable-isotope labelling by amino acids in cell culture (SILAC) [10], and label-free quantitative proteomics techniques [11]. These reports suggest that quantitative proteomic approaches are valuable for experimentally identifying direct and indirect functional miRNA targets. Antibody-based protein microarrays have emerged as a promising high-sensitive and high-throughput tool for quantitative proteomic research, and are superior to other mass spectrometry-based proteomic strategies in identifying low-abundance proteins, such as signalling molecules or kinases [12]. Until recently, antibody-based proteomic approaches have not been applied to miRNA target identification. Therefore, in this study, using the tumor-suppressor miRNA hsa-miR-206 as an
http://dx.doi.org/10.1016/j.febslet.2015.06.026 0014-5793/Ó 2015 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
2
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
example, we developed an antibody array-based strategy to profile miRNA-mediated receptor tyrosine kinases (RTKs) and to identify novel miRNA-related pathway and functional targets. 2. Materials and methods 2.1. Cell culture and miR-206 transfection
overlapped among at least two prediction sets were defined as candidate targets for further analysis. Furthermore, this gene set was used to search for their interaction with MET by accessing the database STRING to create a MET-centred miR-206 target gene network as described previously [14]. The links between miR-206 target genes and MET were then visualized using the Cytoscape software.
The human lung adenocarcinoma cell line A549 was obtained from the Cell Bank of the Shanghai Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China), and cultured in RPMI-1640 medium supplemented with 10% foetal bovine serum (FBS) at 37 °C in 5% CO2. A549 cells were seeded in 6-well plates to reach 75% confluence and transfected using hsa-miR-206 mimics or negative controls (Shanghai GenePharma, China) with Lipofectamine 2000 (Invitrogen). Cells were harvested after 48 h of transfection.
From each sample, a total of 30 lg of protein lysates was subjected to 12% SDS–PAGE and electro-blotted onto PVDF membranes. After blocking, the membrane was incubated overnight with anti-EIF4E antibody (1:1000, Proteintech Group) and an anti-b-actin antibody (1:2000, Keygentec, China) followed by enhanced chemiluminescence detection. The Western blots were imaged on a Gel Doc XR system (Bio-Rad).
2.2. Real-time RT-PCR analysis
2.7. Luciferase assay
Total RNA was extracted from A549 cells using TRIzol total RNA isolation reagent (Invitrogen) according to the manufacturer’s instructions. cDNA was synthesized from total RNA using human miR-206RT primer. Real-time PCR was performed with a SYBR Green PCR kit (Toyobo). U6 small nuclear RNA (snU6) was used to normalize the expression data of miR-206.
The synthetic 30 -UTR fragment of EIF4E was cloned into the pGL3-promoter vector (Promega) at the XbaI restriction site. A total of 1 105 cells were plated per well in 48-well plates prior to transfection. A total of 500 ng of the pGL3-promoter constructs and 10 ng of the Renilla luciferase plasmid pRL-TK vector (Promega) were co-transfected with the miR-206 mimic or a mimic control using Lipofectamine 2000 (Invitrogen) into A549 cells. After 48 h, the cells were collected for luciferase activity assays using the Dual-Luciferase Reporter Assay System (Promega). For each transfection, firefly luciferase activity was normalized to that of the Renilla luciferase. All assays were performed in triplicate from independent cell transfections.
2.3. RTK phosphorylation antibody microarray analysis The phosphorylation levels of 71 different human receptor tyrosine kinases (RTKs) in cell lysates were simultaneously detected using a RayBio Phosphorylation Antibody Array I Kit (RayBiotech, Inc., USA) according to the manufacturer’s instructions. Briefly, A549 cells from different groups (miR-206 mimics and the miR-negative control) were lysed and the total proteins were purified using the Cell and Tissue Protein Extraction Reagent (Kangchen, China). A BCA Protein Assay Kit (Kangchen, China) was used to determine the concentrations. A total of 50 lg of protein extract was incubated with RTK glass slide subarrays spotted with 71 different anti-RTK antibodies. Then, the antibody array chips were washed and biotinylated anti-phosphotyrosine antibodies were added to detect phosphorylated tyrosines on activated RTKs. After incubation with Alexa Fluor 555-streptavidin, the signals were visualized by a fluorescence scanner (GenePix 4000B). 2.4. Antibody array data analysis Raw data collection and the background correction of microarrays were performed using GenePix Pro 5.1. The processed miRNA intensity values were log-transformed and entered into the BRB-ArrayTools software. Analyses were performed using BRB-ArrayTools developed by Dr. Richard Simon and the BRB-ArrayTools Development Team [13]. The chips were first normalized to the positive mean value, and then quantile normalization was performed. The Class comparison between Groups of Arrays tool was used to compare the significantly differentially expressed RTKs between A549 cells with and without miR-206 transfection. A permutation test was used, and the number of permutations was 10 000. 2.5. Network analysis miR-206 targets were first predicted by PicTar 2005, miRanda v5 and TargetScan 5.1, and then a total of 106 genes that
2.6. Western blot analysis
3. Results In this study, we developed an antibody-based quantitative proteomics strategy to identify putative functional targets of an example microRNA from hundreds of predicted microRNA target genes. Our experimental schema is described in Fig. 1. To study the modulation of miR-206 on RTKs phosphorylation levels, we established a condition for ectopic overexpression of miR-206 in A549 cells. Because A549 cells have a low level of miR-206 expression, we reasoned that the magnitude of changes in the expression and modification of the proteins induced by ectopic repression would be small; therefore the inhibitory modulation of miR-206 was not performed in this study. We confirmed that the cellular level of miR-206 was 3-fold overexpressed following introduction of the miR-206 mimic compared with non-specific controls and a blank. Using an antibody-based protein array, the expression levels of a total of 71 phosphorylated RTKs were simultaneously screened in A549 cells transfected with the miR-206 mimic or an miRNA negative control. Before class comparison, the raw intensity data of the scanned protein array images were processed in 4 steps (background correction, log transformation, normalization relative to positive controls, and quantile normalization) that are typically used in oligonucleotide microarray data analysis. In this study, this procedure yielded a high-quality normalization effect (Fig. 2A). After data processing, miR-206-affected phosphorylated RTKs were identified using class comparison tools in the BRB-ArrayTools. Based on a permutation analysis of the expression data, 7 phosphorylated RTKs were significantly differentially expressed in miR-206-transfected A549 cells compared with negative controls (permutation P-value <0.05), including 2 up-regulated
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
3
Fig. 1. Experimental schema of our strategy.
Fig. 2. Processing of raw intensity data of the scanned protein array images yields high-quality normalization (A). Hierarchical clustering of identical differentially expressed RKTs between miR-206-transfected A549 cells and negative controls (B).
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
4
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
Fig. 3. Visualization of the Met-centred miR-206 target genes network.
Fig. 4. Verification of miR-206 directly targeting EIF4E using a luciferase assay. Data are shown as the mean ± S.D. *P < 0.05 using a Student’s t test (A). Western blot analysis showed that EIF4E expression was reduced in A549 cells when transfected with miR-206.
(Tie-1, NGFR) and 5 down-regulated (HGFR [MET], Tec, TYRO10, Btk, and EphA8) phosphorylated RTKs (Fig. 2B and Table S1). As miR-206 is a potential tumor-suppressor miRNA, its downregulated targets might be of more significance in promoting cancer progression. From the 5 downregulated targets of miR-206, we selected the most varied RTK, HGFR (also MET) to further explore the functional targets of miR-206. One hundred and six common target genes predicted by three computational algorithms were collected. We then captured the links of these targets with MET from the context of the STRING database, and constructed a MET-centred functional association network (Fig. 3). Thirty-one predicted miR-206 targets entered into this network included two validated genes, PAX3 and SNX2 [15,16]. Moreover, within this network, one predicted target gene, EIF4E, has been reported to be a positive MET regulator, which suggests that this gene might also be an miR-206 target functioning in MET downregulation. As indicated in Fig. 4A, the introduction of miR-206 into A549 cells with the wild-type 30 -UTR EIF4E construct, significantly inhibited luciferase activity compared to the negative control, and mutations in the binding sites of these two genes completely abolished the ability of miR-206 to regulate luciferase expression. These results demonstrate that EIF4E is a direct target of miR-206. To further confirm that miR-206 was indeed responsible for the inhibition of these two novel targets, A549 cells were transfected with the miR-206 mimic or a negative control. Western blot analysis showed while EIF4E expression was not affected by the negative-control transfection, EIF4E expression was dramatically reduced after transfection with miR-206 (Fig. 4B).
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
5
4. Discussion
Appendix A. Supplementary data
In this study, we demonstrated for the first time the feasibility of applying antibody array technology to identify functional microRNA targets from a list of putative miRNA-regulated genes. Using this strategy, a set of miR-206-repressed phosphorylated RTKs was first investigated, among which MET was the most repressed RTK. We then constructed a MET-centred network, and narrowed down a list of 106 predicted targets of miR-206 to 31 genes associated with MET expression and activation inhibition. Filtering the list by biological context drawn from the literature led to the selection of 3 genes (PAX3, EIF4E and SNX2), that have supporting experimental evidence for increasing MET expression [17–20], the suppression of these genes may at least partially account for the role of miR-206 as a tumor suppressor. Therefore, these 3 genes can be regarded as miR-206 targets with functional significance. These 3 genes include two confirmed targets, PAX3 and SNX2, and one novel target, EIF4E. We further validated that EIF4E is a direct target of miR-206 and could be repressed by miR-206 at the protein level. For miRNA target identification, this antibody array-based strategy is superior to mass spectrometry-based proteomic methods in its ability to identify low-abundance miRNA targets such as RTKs and transcription factors, which are always functionally important targets. Despite these merits, our strategy is only limited to RTK-biased targets. The target coverage of this array is far lower than other unbiased quantitative proteomic strategies. Therefore, we suggest that other protein arrays with more capacity could be used for function-oriented miRNA target identification in the future. miR-206 is downregulated in many human malignancies, and functionally, it acts as a potential tumor suppressor [21,22]. Currently, 25 miR-206 genes in 4 species have been validated; most of these are oncogene targets. Interestingly, MET itself is also a direct miR-206 target [23]. Based on our findings and other evidence, we propose that the repression of MET phosphorylation by miR-206 may be due to the following three mechanisms: first, miR-206 directly represses MET gene expression and subsequently decreases its phosphorylation level. Secondly, miR-206 targets the gene PAX3, which can directly activate MET or indirectly regulate it through MITF. Thirdly, miR-206 targets EIF4F and SNX2, thus reducing MET expression. Therefore, our results also indicate a more detailed regulatory mechanism for miR-206 in MET induction and activation, which helps better illustrate its role as a tumor suppressor. In summary, in this study, we demonstrated the utility of a protein chip- and bioinformatics-based approach for discovering functional targets of miRNAs. Here, we used miR-206 and RTK antibody arrays as a model, but this concept may also be applicable for other miRNAs and in other types of antibody arrays.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2015.06. 026.
Competing interests The authors have declared that no competing interests exist. Acknowledgments This work was supported by National Natural Science Foundation of China (81371909) and Natural Science Foundation of Shandong Province, China (ZR2010CM019).
References [1] Bartel, D.P. (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297. [2] Pasquinelli, A.E. (2012) MicroRNAs and their targets: recognition, regulation and an emerging reciprocal relationship. Nat. Rev. Genet. 13, 271–282. [3] Wang, X. (2014) Composition of seed sequence is a major determinant of microRNA targeting patterns. Bioinformatics 30, 1377–1383. [4] Saito, T. and Saetrom, P. (2010) MicroRNAs-targeting and target prediction. New Biotechnol. 27, 243–249. [5] Thomas, M., Lieberman, J. and Lal, A. (2010) Desperately seeking microRNA targets. Nat. Struct. Mol. Biol. 17, 1169–1174. [6] Thomson, D.W., Bracken, C.P. and Goodall, G.J. (2011) Experimental strategies for microRNA target identification. Nucleic Acids Res. 39, 6845–6853. [7] Li, C., Xiong, Q., Zhang, J., Ge, F. and Bi, L.J. (2012) Quantitative proteomic strategies for the identification of microRNA targets. Expert Rev. Proteomics 9, 549–559. [8] Schramedei, K., Morbt, N., Pfeifer, G., Lauter, J., Rosolowski, M., et al. (2011) MicroRNA-21 targets tumor suppressor genes ANP32A and SMARCA4. Oncogene 30, 2975–2985. [9] Hu, C.W., Tseng, C.W., Chien, C.W., Huang, H.C., Ku, W.C., Lee, S.J., Chen, Y.J. and Juan, H.F. (2013) Quantitative proteomics reveals diverse roles of miR-148a from gastric cancer progression to neurological development. J. Proteome Res. 12, 3993–4004. [10] Xiong, Q., Zhong, Q., Zhang, J., Yang, M., Li, C., et al. (2012) Identification of novel miR-21 target proteins in multiple myeloma cells by quantitative proteomics. J. Proteome Res. 11, 2078–2090. [11] Meleady, P., Gallagher, M., Clarke, C., Henry, M., Sanchez, N., et al. (2012) Impact of miR-7 over-expression on the proteome of Chinese hamster ovary cells. J. Biotechnol. 160, 251–262. [12] Lee, J.R., Magee, D.M., Gaster, R.S., LaBaer, J. and Wang, S.X. (2013) Emerging protein array technologies for proteomics. Expert Rev. Proteomics 10, 65–75. [13] Simon, R., Lam, A., Li, M.C., Ngan, M., Menenzes, S. and Zhao, Y. (2007) Analysis of gene expression data using BRB-ArrayTools. Cancer Informatics 3, 11–17. [14] Yue, X., Lan, F., Yang, W., Yang, Y., Han, L., et al. (2010) Interruption of betacatenin suppresses the EGFR pathway by blocking multiple oncogenic targets in human glioma cells. Brain Res. 1366, 27–37. [15] Goljanek-Whysall, K., Sweetman, D., Abu-Elmagd, M., Chapnik, E., Dalmay, T., Hornstein, E. and Munsterberg, A. (2011) MicroRNA regulation of the pairedbox transcription factor Pax3 confers robustness to developmental timing of myogenesis. Proc. Natl. Acad. Sci. U.S.A. 108, 11936–11941. [16] Hudson, R.S., Yi, M., Esposito, D., Watkins, S.K., Hurwitz, A.A., Yfantis, H.G., Lee, D.H., Borin, J.F., Naslund, M.J., Alexander, R.B., Dorsey, T.H., Stephens, R.M., Croce, C.M. and Ambs, S. (2012) MicroRNA-1 is a candidate tumor suppressor and prognostic marker in human prostate cancer. Nucleic Acids Res. 40, 3689– 3703. [17] Hirai, H., Verma, M., Watanabe, S., Tastad, C., Asakura, Y., et al. (2010) MyoD regulates apoptosis of myoblasts through microRNA-mediated downregulation of Pax3. J. Cell Biol. 191, 347–365. [18] Mascarenhas, J.B., Littlejohn, E.L., Wolsky, R.J., Young, K.P., Nelson, M., et al. (2010) PAX3 and SOX10 activate MET receptor expression in melanoma. Pigment Cell Melanoma Res. 23, 225–237. [19] Li, Y., Fan, S., Koo, J., Yue, P., Chen, Z.G., et al. (2012) Elevated expression of eukaryotic translation initiation factor 4E is associated with proliferation, invasion and acquired resistance to erlotinib in lung cancer. Cancer Biol. Ther. 13, 272–280. [20] Ogi, S., Fujita, H., Kashihara, M., Yamamoto, C., Sonoda, K., et al. (2013) Sorting nexin 2-mediated membrane trafficking of c-Met contributes to sensitivity of molecular-targeted drugs. Cancer Sci. 104, 573–583. [21] Zhou, J., Tian, Y., Li, J., Lu, B., Sun, M., et al. (2013) MiR-206 is down-regulated in breast cancer and inhibits cell proliferation through the up-regulation of cyclinD2. Biochem. Biophys. Res. Commun. 433, 207–212. [22] Zhang, L., Liu, X., Jin, H., Guo, X., Xia, L., et al. (2013) MiR-206 inhibits gastric cancer proliferation in part by repressing cyclinD2. Cancer Lett. 332, 94–101. [23] Yan, D., Dong Xda, E., Chen, X., Wang, L., Lu, C., et al. (2009) MicroRNA-1/206 targets c-Met and inhibits rhabdomyosarcoma development. J. Biol. Chem. 284, 29596–29604.
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
journal homepage: www.FEBSLetters.org
Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibody-based protein array Yazhou Cui a, Shuyang Xie b, Jing Luan a, Xiaoyan Zhou a, Jinxiang Han a,⇑ a b
Shandong Academy of Medical Sciences, Shandong Medical Biotechnological Center, Key Laboratory for Biotech Drugs of the Ministry of Health, China Department of Biochemistry and Molecular Biology, Binzhou Medical University, YanTai, Shandong, China
a r t i c l e
i n f o
Article history: Received 13 May 2015 Revised 16 June 2015 Accepted 19 June 2015 Available online xxxx Edited by Tamas Dalmay Keywords: MicroRNA Target gene Proteomic miR-206 Antibody array
a b s t r a c t This study demonstrated the feasibility and benefit of an antibody-based experimental approach to identify microRNA functional targets from hundreds of predicted genes using miR-206 as an example. Using a receptor tyrosine kinase (RTK) antibody array, we identified 7 phosphorylated RTKs that were significantly differentially regulated after miR-206-mimic transfection. We then focused on MET, the most varied RTK, and bioinformatically constructed a MET-centred network using computationally predicted miR-206 targets. Within this network, we analyzed two validated targets, PAX3 and SNX2, and one candidate target, EIF4E, may account for the inhibitory effect of miR-206 on MET phosphorylation. Luciferase and Western-blot assays indicated that EIF4E was a direct target of miR-206. This concept may also be applicable for other microRNAs and other antibody array platforms. Ó 2015 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
1. Introduction MicroRNAs (miRNAs) are important regulators of gene expression in eukaryotic cells. In humans, it is estimated that miRNAs may regulate up to 30% of the protein-coding genes. miRNAs are known to control a broad spectrum of biological processes, and their dysregulation may be associated with multiple diseases [1]. The identification of functional targets is central to understanding the molecular mechanisms by which individual miRNAs modulate cellular functions [2]. Although some computational programs based on sequence alignment have been widely applied to predict the targets of miRNAs, these bioinformatic tools still yield a high percentage of false-positives and false-negatives [3,4]. Furthermore, it is believed that a single miRNA can regulate hundreds of target mRNAs; therefore, the identification of genuine functional target genes remains a fundamental challenge. Therefore, these obstacles underscore the requirement for experimental data that determines functionally
Author contributions: JX Han and YZ Cui designed the study and wrote the manuscript. YZ Cui, SY Xie and J Luan performed the experiments and data analysis. XY Zhou participated in the data analysis. ⇑ Corresponding author at: Shandong Academy of Medical Sciences, Shandong Medical Biotechnological Center, Key Laboratory for Biotech Drugs of the Ministry of Health, 18877 Jingshi Road, Jinan 260062, China. Fax: +86 53182951586. E-mail address: [email protected] (J. Han).
important gene targets for elucidating the exact roles of miRNAs in development and disease [5]. Because miRNAs can repress the expression of target genes at the mRNA and/or protein levels, a series of experimental strategies based on gene expression microarray or proteomic tools have been proposed for microRNA target identification [6]. Because most targets may be repressed by an miRNA at the protein level without being affected at the mRNA level, proteomic methods are regarded as the most powerful experimental strategies for revealing the full spectrum of miRNA targets [7]. Currently, various quantitative proteomic methodologies have been employed in the identification of miRNA targets, such as two-dimensional difference gel electrophoresis (2D-DIGE) [8], isobaric tags for relative and absolute quantification (iTRAQ) [9], stable-isotope labelling by amino acids in cell culture (SILAC) [10], and label-free quantitative proteomics techniques [11]. These reports suggest that quantitative proteomic approaches are valuable for experimentally identifying direct and indirect functional miRNA targets. Antibody-based protein microarrays have emerged as a promising high-sensitive and high-throughput tool for quantitative proteomic research, and are superior to other mass spectrometry-based proteomic strategies in identifying low-abundance proteins, such as signalling molecules or kinases [12]. Until recently, antibody-based proteomic approaches have not been applied to miRNA target identification. Therefore, in this study, using the tumor-suppressor miRNA hsa-miR-206 as an
http://dx.doi.org/10.1016/j.febslet.2015.06.026 0014-5793/Ó 2015 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
2
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
example, we developed an antibody array-based strategy to profile miRNA-mediated receptor tyrosine kinases (RTKs) and to identify novel miRNA-related pathway and functional targets. 2. Materials and methods 2.1. Cell culture and miR-206 transfection
overlapped among at least two prediction sets were defined as candidate targets for further analysis. Furthermore, this gene set was used to search for their interaction with MET by accessing the database STRING to create a MET-centred miR-206 target gene network as described previously [14]. The links between miR-206 target genes and MET were then visualized using the Cytoscape software.
The human lung adenocarcinoma cell line A549 was obtained from the Cell Bank of the Shanghai Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China), and cultured in RPMI-1640 medium supplemented with 10% foetal bovine serum (FBS) at 37 °C in 5% CO2. A549 cells were seeded in 6-well plates to reach 75% confluence and transfected using hsa-miR-206 mimics or negative controls (Shanghai GenePharma, China) with Lipofectamine 2000 (Invitrogen). Cells were harvested after 48 h of transfection.
From each sample, a total of 30 lg of protein lysates was subjected to 12% SDS–PAGE and electro-blotted onto PVDF membranes. After blocking, the membrane was incubated overnight with anti-EIF4E antibody (1:1000, Proteintech Group) and an anti-b-actin antibody (1:2000, Keygentec, China) followed by enhanced chemiluminescence detection. The Western blots were imaged on a Gel Doc XR system (Bio-Rad).
2.2. Real-time RT-PCR analysis
2.7. Luciferase assay
Total RNA was extracted from A549 cells using TRIzol total RNA isolation reagent (Invitrogen) according to the manufacturer’s instructions. cDNA was synthesized from total RNA using human miR-206RT primer. Real-time PCR was performed with a SYBR Green PCR kit (Toyobo). U6 small nuclear RNA (snU6) was used to normalize the expression data of miR-206.
The synthetic 30 -UTR fragment of EIF4E was cloned into the pGL3-promoter vector (Promega) at the XbaI restriction site. A total of 1 105 cells were plated per well in 48-well plates prior to transfection. A total of 500 ng of the pGL3-promoter constructs and 10 ng of the Renilla luciferase plasmid pRL-TK vector (Promega) were co-transfected with the miR-206 mimic or a mimic control using Lipofectamine 2000 (Invitrogen) into A549 cells. After 48 h, the cells were collected for luciferase activity assays using the Dual-Luciferase Reporter Assay System (Promega). For each transfection, firefly luciferase activity was normalized to that of the Renilla luciferase. All assays were performed in triplicate from independent cell transfections.
2.3. RTK phosphorylation antibody microarray analysis The phosphorylation levels of 71 different human receptor tyrosine kinases (RTKs) in cell lysates were simultaneously detected using a RayBio Phosphorylation Antibody Array I Kit (RayBiotech, Inc., USA) according to the manufacturer’s instructions. Briefly, A549 cells from different groups (miR-206 mimics and the miR-negative control) were lysed and the total proteins were purified using the Cell and Tissue Protein Extraction Reagent (Kangchen, China). A BCA Protein Assay Kit (Kangchen, China) was used to determine the concentrations. A total of 50 lg of protein extract was incubated with RTK glass slide subarrays spotted with 71 different anti-RTK antibodies. Then, the antibody array chips were washed and biotinylated anti-phosphotyrosine antibodies were added to detect phosphorylated tyrosines on activated RTKs. After incubation with Alexa Fluor 555-streptavidin, the signals were visualized by a fluorescence scanner (GenePix 4000B). 2.4. Antibody array data analysis Raw data collection and the background correction of microarrays were performed using GenePix Pro 5.1. The processed miRNA intensity values were log-transformed and entered into the BRB-ArrayTools software. Analyses were performed using BRB-ArrayTools developed by Dr. Richard Simon and the BRB-ArrayTools Development Team [13]. The chips were first normalized to the positive mean value, and then quantile normalization was performed. The Class comparison between Groups of Arrays tool was used to compare the significantly differentially expressed RTKs between A549 cells with and without miR-206 transfection. A permutation test was used, and the number of permutations was 10 000. 2.5. Network analysis miR-206 targets were first predicted by PicTar 2005, miRanda v5 and TargetScan 5.1, and then a total of 106 genes that
2.6. Western blot analysis
3. Results In this study, we developed an antibody-based quantitative proteomics strategy to identify putative functional targets of an example microRNA from hundreds of predicted microRNA target genes. Our experimental schema is described in Fig. 1. To study the modulation of miR-206 on RTKs phosphorylation levels, we established a condition for ectopic overexpression of miR-206 in A549 cells. Because A549 cells have a low level of miR-206 expression, we reasoned that the magnitude of changes in the expression and modification of the proteins induced by ectopic repression would be small; therefore the inhibitory modulation of miR-206 was not performed in this study. We confirmed that the cellular level of miR-206 was 3-fold overexpressed following introduction of the miR-206 mimic compared with non-specific controls and a blank. Using an antibody-based protein array, the expression levels of a total of 71 phosphorylated RTKs were simultaneously screened in A549 cells transfected with the miR-206 mimic or an miRNA negative control. Before class comparison, the raw intensity data of the scanned protein array images were processed in 4 steps (background correction, log transformation, normalization relative to positive controls, and quantile normalization) that are typically used in oligonucleotide microarray data analysis. In this study, this procedure yielded a high-quality normalization effect (Fig. 2A). After data processing, miR-206-affected phosphorylated RTKs were identified using class comparison tools in the BRB-ArrayTools. Based on a permutation analysis of the expression data, 7 phosphorylated RTKs were significantly differentially expressed in miR-206-transfected A549 cells compared with negative controls (permutation P-value <0.05), including 2 up-regulated
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
3
Fig. 1. Experimental schema of our strategy.
Fig. 2. Processing of raw intensity data of the scanned protein array images yields high-quality normalization (A). Hierarchical clustering of identical differentially expressed RKTs between miR-206-transfected A549 cells and negative controls (B).
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
4
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
Fig. 3. Visualization of the Met-centred miR-206 target genes network.
Fig. 4. Verification of miR-206 directly targeting EIF4E using a luciferase assay. Data are shown as the mean ± S.D. *P < 0.05 using a Student’s t test (A). Western blot analysis showed that EIF4E expression was reduced in A549 cells when transfected with miR-206.
(Tie-1, NGFR) and 5 down-regulated (HGFR [MET], Tec, TYRO10, Btk, and EphA8) phosphorylated RTKs (Fig. 2B and Table S1). As miR-206 is a potential tumor-suppressor miRNA, its downregulated targets might be of more significance in promoting cancer progression. From the 5 downregulated targets of miR-206, we selected the most varied RTK, HGFR (also MET) to further explore the functional targets of miR-206. One hundred and six common target genes predicted by three computational algorithms were collected. We then captured the links of these targets with MET from the context of the STRING database, and constructed a MET-centred functional association network (Fig. 3). Thirty-one predicted miR-206 targets entered into this network included two validated genes, PAX3 and SNX2 [15,16]. Moreover, within this network, one predicted target gene, EIF4E, has been reported to be a positive MET regulator, which suggests that this gene might also be an miR-206 target functioning in MET downregulation. As indicated in Fig. 4A, the introduction of miR-206 into A549 cells with the wild-type 30 -UTR EIF4E construct, significantly inhibited luciferase activity compared to the negative control, and mutations in the binding sites of these two genes completely abolished the ability of miR-206 to regulate luciferase expression. These results demonstrate that EIF4E is a direct target of miR-206. To further confirm that miR-206 was indeed responsible for the inhibition of these two novel targets, A549 cells were transfected with the miR-206 mimic or a negative control. Western blot analysis showed while EIF4E expression was not affected by the negative-control transfection, EIF4E expression was dramatically reduced after transfection with miR-206 (Fig. 4B).
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026
Y. Cui et al. / FEBS Letters xxx (2015) xxx–xxx
5
4. Discussion
Appendix A. Supplementary data
In this study, we demonstrated for the first time the feasibility of applying antibody array technology to identify functional microRNA targets from a list of putative miRNA-regulated genes. Using this strategy, a set of miR-206-repressed phosphorylated RTKs was first investigated, among which MET was the most repressed RTK. We then constructed a MET-centred network, and narrowed down a list of 106 predicted targets of miR-206 to 31 genes associated with MET expression and activation inhibition. Filtering the list by biological context drawn from the literature led to the selection of 3 genes (PAX3, EIF4E and SNX2), that have supporting experimental evidence for increasing MET expression [17–20], the suppression of these genes may at least partially account for the role of miR-206 as a tumor suppressor. Therefore, these 3 genes can be regarded as miR-206 targets with functional significance. These 3 genes include two confirmed targets, PAX3 and SNX2, and one novel target, EIF4E. We further validated that EIF4E is a direct target of miR-206 and could be repressed by miR-206 at the protein level. For miRNA target identification, this antibody array-based strategy is superior to mass spectrometry-based proteomic methods in its ability to identify low-abundance miRNA targets such as RTKs and transcription factors, which are always functionally important targets. Despite these merits, our strategy is only limited to RTK-biased targets. The target coverage of this array is far lower than other unbiased quantitative proteomic strategies. Therefore, we suggest that other protein arrays with more capacity could be used for function-oriented miRNA target identification in the future. miR-206 is downregulated in many human malignancies, and functionally, it acts as a potential tumor suppressor [21,22]. Currently, 25 miR-206 genes in 4 species have been validated; most of these are oncogene targets. Interestingly, MET itself is also a direct miR-206 target [23]. Based on our findings and other evidence, we propose that the repression of MET phosphorylation by miR-206 may be due to the following three mechanisms: first, miR-206 directly represses MET gene expression and subsequently decreases its phosphorylation level. Secondly, miR-206 targets the gene PAX3, which can directly activate MET or indirectly regulate it through MITF. Thirdly, miR-206 targets EIF4F and SNX2, thus reducing MET expression. Therefore, our results also indicate a more detailed regulatory mechanism for miR-206 in MET induction and activation, which helps better illustrate its role as a tumor suppressor. In summary, in this study, we demonstrated the utility of a protein chip- and bioinformatics-based approach for discovering functional targets of miRNAs. Here, we used miR-206 and RTK antibody arrays as a model, but this concept may also be applicable for other miRNAs and in other types of antibody arrays.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2015.06. 026.
Competing interests The authors have declared that no competing interests exist. Acknowledgments This work was supported by National Natural Science Foundation of China (81371909) and Natural Science Foundation of Shandong Province, China (ZR2010CM019).
References [1] Bartel, D.P. (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297. [2] Pasquinelli, A.E. (2012) MicroRNAs and their targets: recognition, regulation and an emerging reciprocal relationship. Nat. Rev. Genet. 13, 271–282. [3] Wang, X. (2014) Composition of seed sequence is a major determinant of microRNA targeting patterns. Bioinformatics 30, 1377–1383. [4] Saito, T. and Saetrom, P. (2010) MicroRNAs-targeting and target prediction. New Biotechnol. 27, 243–249. [5] Thomas, M., Lieberman, J. and Lal, A. (2010) Desperately seeking microRNA targets. Nat. Struct. Mol. Biol. 17, 1169–1174. [6] Thomson, D.W., Bracken, C.P. and Goodall, G.J. (2011) Experimental strategies for microRNA target identification. Nucleic Acids Res. 39, 6845–6853. [7] Li, C., Xiong, Q., Zhang, J., Ge, F. and Bi, L.J. (2012) Quantitative proteomic strategies for the identification of microRNA targets. Expert Rev. Proteomics 9, 549–559. [8] Schramedei, K., Morbt, N., Pfeifer, G., Lauter, J., Rosolowski, M., et al. (2011) MicroRNA-21 targets tumor suppressor genes ANP32A and SMARCA4. Oncogene 30, 2975–2985. [9] Hu, C.W., Tseng, C.W., Chien, C.W., Huang, H.C., Ku, W.C., Lee, S.J., Chen, Y.J. and Juan, H.F. (2013) Quantitative proteomics reveals diverse roles of miR-148a from gastric cancer progression to neurological development. J. Proteome Res. 12, 3993–4004. [10] Xiong, Q., Zhong, Q., Zhang, J., Yang, M., Li, C., et al. (2012) Identification of novel miR-21 target proteins in multiple myeloma cells by quantitative proteomics. J. Proteome Res. 11, 2078–2090. [11] Meleady, P., Gallagher, M., Clarke, C., Henry, M., Sanchez, N., et al. (2012) Impact of miR-7 over-expression on the proteome of Chinese hamster ovary cells. J. Biotechnol. 160, 251–262. [12] Lee, J.R., Magee, D.M., Gaster, R.S., LaBaer, J. and Wang, S.X. (2013) Emerging protein array technologies for proteomics. Expert Rev. Proteomics 10, 65–75. [13] Simon, R., Lam, A., Li, M.C., Ngan, M., Menenzes, S. and Zhao, Y. (2007) Analysis of gene expression data using BRB-ArrayTools. Cancer Informatics 3, 11–17. [14] Yue, X., Lan, F., Yang, W., Yang, Y., Han, L., et al. (2010) Interruption of betacatenin suppresses the EGFR pathway by blocking multiple oncogenic targets in human glioma cells. Brain Res. 1366, 27–37. [15] Goljanek-Whysall, K., Sweetman, D., Abu-Elmagd, M., Chapnik, E., Dalmay, T., Hornstein, E. and Munsterberg, A. (2011) MicroRNA regulation of the pairedbox transcription factor Pax3 confers robustness to developmental timing of myogenesis. Proc. Natl. Acad. Sci. U.S.A. 108, 11936–11941. [16] Hudson, R.S., Yi, M., Esposito, D., Watkins, S.K., Hurwitz, A.A., Yfantis, H.G., Lee, D.H., Borin, J.F., Naslund, M.J., Alexander, R.B., Dorsey, T.H., Stephens, R.M., Croce, C.M. and Ambs, S. (2012) MicroRNA-1 is a candidate tumor suppressor and prognostic marker in human prostate cancer. Nucleic Acids Res. 40, 3689– 3703. [17] Hirai, H., Verma, M., Watanabe, S., Tastad, C., Asakura, Y., et al. (2010) MyoD regulates apoptosis of myoblasts through microRNA-mediated downregulation of Pax3. J. Cell Biol. 191, 347–365. [18] Mascarenhas, J.B., Littlejohn, E.L., Wolsky, R.J., Young, K.P., Nelson, M., et al. (2010) PAX3 and SOX10 activate MET receptor expression in melanoma. Pigment Cell Melanoma Res. 23, 225–237. [19] Li, Y., Fan, S., Koo, J., Yue, P., Chen, Z.G., et al. (2012) Elevated expression of eukaryotic translation initiation factor 4E is associated with proliferation, invasion and acquired resistance to erlotinib in lung cancer. Cancer Biol. Ther. 13, 272–280. [20] Ogi, S., Fujita, H., Kashihara, M., Yamamoto, C., Sonoda, K., et al. (2013) Sorting nexin 2-mediated membrane trafficking of c-Met contributes to sensitivity of molecular-targeted drugs. Cancer Sci. 104, 573–583. [21] Zhou, J., Tian, Y., Li, J., Lu, B., Sun, M., et al. (2013) MiR-206 is down-regulated in breast cancer and inhibits cell proliferation through the up-regulation of cyclinD2. Biochem. Biophys. Res. Commun. 433, 207–212. [22] Zhang, L., Liu, X., Jin, H., Guo, X., Xia, L., et al. (2013) MiR-206 inhibits gastric cancer proliferation in part by repressing cyclinD2. Cancer Lett. 332, 94–101. [23] Yan, D., Dong Xda, E., Chen, X., Wang, L., Lu, C., et al. (2009) MicroRNA-1/206 targets c-Met and inhibits rhabdomyosarcoma development. J. Biol. Chem. 284, 29596–29604.
Please cite this article in press as: Cui, Y., et al. Identification of the receptor tyrosine kinases (RTKs)-oriented functional targets of miR-206 by an antibodybased protein array. FEBS Lett. (2015), http://dx.doi.org/10.1016/j.febslet.2015.06.026