A combinative analysis of gene expression profiles and microRNA expression profiles identifies critical genes and microRNAs in oral lichen planus

Archives of Oral Biology 68 (2016) 61–65

Contents lists available at ScienceDirect

Archives of Oral Biology journal homepage: www.elsevier.com/locate/aob

A combinative analysis of gene expression profiles and microRNA expression profiles identifies critical genes and microRNAs in oral lichen planus Qing Liua,* , Xinwen Wanga , Yuan Liub , Minghui Weia , Lihua Chenc,* a State Key Laboratory of Military Stomatology, Department of Oral Medicine, School of Stomatology, Fourth Military Medical University, No. 145, Changle West Road, Xi’an, Shaanxi 710032, PR China b State Key Laboratory of Military Stomatology, Department of Oral Pathology, School of Stomatology, Fourth Military Medical University, No. 145, Changle West Road, Xi’an, Shaanxi 710032, PR China c State Key Laboratory of Military Stomatology, Department of Immunology, Fourth Military Medical University, No. 169, Changle West Road, Xi’an, Shaanxi 710032, PR China

A R T I C L E I N F O

A B S T R A C T

Article history: Received 1 July 2015 Received in revised form 1 March 2016 Accepted 29 March 2016

Objective: Oral lichen planus (OLP) is a chronic inflammatory disease but aetiology and pathogenesis has not fully elucidated. To gain insight into the mechanism of OLP, bioinformatic analysis was performed in this study. Design: GSE38616 and GSE38615 were downloaded from GEO, including 7 cases of OLP and 7 healthy controls. Differentially expressed genes (DEGs) and miRNAs (DEMs) between OLP and control were screened with package Limma of R. Potential regulatory miRNAs were screened via gene set enrichment analysis. A protein–protein interaction network was constructed for the DEGs. KEGG pathways for DEGs were revealed using Gene Set Analysis Toolkit V2. Results: After DEGs and DEMs were obtained, potential regulatory miRNAs of the DEGs were revealed and only miR-362 was differentially expressed in OLP compared with DEMs. Four targets of miR-362 were SLIT-ROBO Rho GTPase activating protein 2 (SRGAP2), vesicle-associated membrane protein 4 (VAMP4), leucine rich repeat transmembrane neuronal 4 (LRRTM4) and lysine (K)-specific demethylase5C (KDM5C). Identified DEGs were significantly enriched in olfactory transduction and ribosome pathways. Conclusion: miR-362, targeting SRGAP2 and VAMP4, may be a potential risk miRNA to regulate OLP. The findings may provide potential biomarkers for diagnosis or treatment of the disease. ã 2016 Elsevier Ltd. All rights reserved.

Keywords: Oral lichen planus Differentially expressed genes Differentially expressed microRNAs Protein–protein interaction network KEGG pathway enrichment analysis

1. Introduction Oral lichen planus (OLP) is a chronic inflammatory disease that affects the mucous membrane of the oral cavity. It is being discussed as a status with malignant potential, but controversial results are reported (Carlos & Contreras, 2004; Van der Meij, Mast, & Van der Waal, 2007). OLP affects one to two percent of adult population and is one of the most popular oral mucosal diseases.

Abbreviations: OLP, oral lichen planus; DEGs, differentially expressed genes; SRGAP2, SLITROBO Rho GTPase activating protein 2; VAMP4, vesicle-associated membrane protein 4; LRRTM4, leucine rich repeat transmembrane neuronal 4; KDM5C, lysine (K)-specific demethylase5C; DEMs, differentially expressed miRNAs; GSEA, gene set enrichment analysis; MSigDB, molecular signatures database; PPI, protein–protein interaction; DCs, dendritic cells; DCC, deleted in colorectal cancer; AP1, adaptor-related protein complex 1. * Corresponding authors. E-mail addresses: [email protected] (Q. Liu), [email protected] (L. Chen) . http://dx.doi.org/10.1016/j.archoralbio.2016.03.018 0003-9969/ ã 2016 Elsevier Ltd. All rights reserved.

To data, OLP is regarded as a T-cell-mediated chronic inflammatory mucocutaneous disease of unknown etiology (Lavanya, Jayanthi, Rao, & Ranganathan, 2011; Thongprasom, Carrozzo, Furness, & Lodi, 2011). Both antigen-specific and nonspecific mechanisms are hypothesized to well understand the pathogenesis of OLP. Antigen-specific mechanisms in OLP include antigen presentation by basal keratinocytes and antigen-specific keratinocyte killing by autocytotoxic CD8+ T cells (Zhao, Savage, Sugerman, & Walsh, 2002). Non-specific mechanism include mast cell degranulation (van der Waal, 2009; Zhao, Sugerman, Zhou, Walsh, & Savage, 2001) and matrix metalloproteinase activation in OLP lesions (Kim et al., 2006). However, the origin and pathogenesis of OLP remains obscure and not completely elucidated. MicroRNAs (miRNAs) are small non-coding RNA molecules which function in transcriptional and post-transcriptional regulation of gene expression (Chen & Rajewsky, 2007). They are involved in most biological processes. Emerging evidence has suggested a

62

Q. Liu et al. / Archives of Oral Biology 68 (2016) 61–65

direct link between miRNAs and disease (Kloosterman & Plasterk, 2006; Lu et al., 2008). Aberrant expression of miRNAs has been implicated in numerous diseases including OLP, such as miRNA146a (Arão, Guimarães, de Paula, Gomes, & Gomez, 2012), miRNA155 (Arão et al., 2012) and miRNA-27b (Zhang et al., 2012). This suggests that discovery of critical miRNAs is a new way to unveil the molecular mechanisms underlying diseases. The study by Gassling et al. (2013) focuses on the linkages between miRNAs and mRNAs, as well as their roles in the pathogenesis of OLP. A total of 11 potential miRNA-mRNA pairs were revealed (Gassling et al., 2013). To obtain more information from the high-throughput experiment, differentially expressed genes (DEGs) and differentially expressed miRNAs (DEMs) were analyzed in present study with various bioinformatic tools. Meanwhile, the potential regulatory relationship between DEGs and DEMs were also investigated, which could deepen the understandings of the pathogenesis and benefit the development of new diagnostic and therapeutic approaches. 2. Methods 2.1. Gene expression data and miRNA expression data Both gene expression data GSE38616 (Gassling et al., 2013) and miRNA expression data GSE38615 (Gassling et al., 2013) were downloaded from Gene Expression Omnibus database (GEO, http://www.ncbi.nlm.nih.gov/geo/). GSE38616 was sequenced on the platform of GPL6244 [HuGene-1_0-st] Affymetrix Human Gene 1.0 ST Array (Affymetrix Inc., Santa Clara, California, USA). And GSE38615 was based on the platform of GPL8786 [miRNA-1_0] Affymetrix miRNA Array (Affymetrix Inc., Santa Clara, California, USA). GSE38616 included 7 oral mucosa tissues from healthy controls and 7 oral mucosa tissues from OLP patients. Meanwhile, GSE38615 were obtained based on the same samples as GSE38616. According to the report by Gassling, patients fulfilled both the following clinical and histopathological criteria which based on the modified World Health Organization diagnostic criteria for OLP (Gassling et al., 2013; van der Meij, Schepman, & van der Waal, 2003). Clinically inclusion criteria were the presence of a lacelike network of slightly raised gray-white lines in a reticular pattern, the presence of bilateral, mostly symmetrical lesions, as well as atrophic, bullous, erosive, and plaque-type lesions. Histological inclusion criteria were the absence of epithelial dysplasia, the presence of well-defined bandlike zones of cellular infiltration restricted to the shallow connective tissue, and signs of “liquefactive degeneration” in the basal cell layer. Tissue samples were collected from oral mucosa at consultation hours at the Department of Cranio-Maxillofacial Surgery, University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Germany. Gassling et al. (2013) deposited GSE38616 and GSE38615, and their research was approved by the ethics board of the Christian-Albrechts-University of Kiel, Germany (Ref: D 426/08). In the study of Gassling et al. (2013), all the participants gave their written informed consent according to the Helsinki convention.

threshold to filter out DEGs, and P-value < 0.05 was used as the cutoff for screening DEMs. 2.3. Screening of potential miRNAs targeting the DEGs Based on the gene annotation information in Molecular Signatures Database (MSigDB) (Liberzon et al., 2011), the WebGestalt online tool (Wang, Duncan, Shi, & Zhang, 2013) was used to perform gene set enrichment analysis (GSEA) (Subramanian et al., 2005) for the miRNAs targeting the screened DEGs. Then, hypergeometric algorithm was used to calculate the P-values of GSEA (Kauers & Yen, 2015). And Benjamini & Hochberg method (Abbas, Kong, Liu, Jing, & Gao, 2013) was applied to adjust the P-values. The top 10 miRNAs with the samllest P-values were the top 10 most significant miRNAs, which were taken as the potetial miRNAs targeting the DEGs. 2.4. Construction of protein–protein interaction (PPI) network Protein interaction is the basis of biological processes and PPI networks are helpful in unveiling the complicated molecular mechanisms underlying OLP. We used only mammalian PPIs recorded in databases of IntAct (Kerrien et al., 2012), DIP (Xenarios et al., 2002), BIND (Bader, Betel, & Hogue, 2003) and HPRD (Peri et al., 2004), as well as in articles by Rual et al. (2005), Stelzl et al. (2005) and Ramani, Bunescu, Mooney, and Marcotte (2005). The PPIs from these databases and articles were consolidated and an integrated network was obtained. Afterwards, the DEGs involved in the integrated network were further screened to obatain their PPI network. 2.5. Pathway enrichment analysis The Kyoto Encyclopedia of Genes and Genomes (KEGG) database can be applied for systematic analysis of gene functions, which links genomic information to relevant functional information (Kanehisa & Goto, 2000; Kanehisa et al., 2014). All of the metabolic and non-metabolic pathway information were collected from KEGG database. Sbusequently, KEGG pathway enrichment analysis was performed for the DEGs based on the platform of Gene Set Analysis Toolkit V2 (Dexter Duncan, 2010; Zhang, Kirov, & Snoddy, 2005). P-value < 0.05 was set as the cutoff. 3. Results 3.1. DEGs and DEMs of OLP According to the threshold of P-value less than 0.01, a total of 725 probes were differentially expressed between OLP group and control group, corresponding to 426 DEGs. P-value < 0.05 was set as the cutoff to identify DEMs of OLP. A total of 467 DEMs were obtained, out of which 61 DEMs have been reported in previous studies. 3.2. Potential miRNAs targeting the DEGs and target DEGs of miR-362

2.2. Screening of DEGs and DEMs The mRNA and miRNA expression data were extracted from GEO and further preprocessed by GEOquery package (http:// bioconductor.org/packages/release/bioc/html/GEOquery.html) in R (Davis & Meltzer, 2007). After log2 transformation, the DEGs and DEMs between OLP group and control group were screened using Linear models for microarray data (Limma, http://bioinf. wehi.edu.au/limma) package in R (Smyth, 2004). The statistical significance of differences between OLP group and control group was evaluated by Student’s t test. P-value < 0.01 was set as the

GSEA was performed to identify regulatory miRNAs of the DEGs. The top 10 most significant miRNAs were shown in Table 1, including miR-329, miR-362 and miR-28. Compared with identified 467 DEMs, miR-362 (P = 0.0317, log fold change (FC) = 0.230497) was found to be overexpressed in OLP and regulate abnormally expressed genes. Compared regulated genes of miR-362 with identified DEGs, overlapped genes were candidate targets of miR-362 which differentially expressed in OLP. Meanwhile, target genes regulated by miR-362 were selected out: SLIT-ROBO Rho GTPase activating protein 2 (SRGAP2), vesicle-

Q. Liu et al. / Archives of Oral Biology 68 (2016) 61–65 Table 1 Top 10 potential regulatory miRNAs of the differentially expressed genes. microRNA

P-value

target

miR-329 miR-362 miR-28 miR-513 miR-516-5P miR-370 miR-30A-3P,miR-30E-3P miR-412 miR-202 miR-484

0.0027 0.0317 0.032 0.0321 0.0511 0.0793 0.0872 0.1134 0.1421 0.1495

hsa_GGTGTGT hsa_CAAGGAT hsa_AGCTCCT hsa_CCTGTGA hsa_TCCAGAT hsa_CAGCAGG hsa_ACTGAAA hsa_GGTGAAG hsa_ATAGGAA hsa_GAGCCTG

associated membrane protein 4 (VAMP4), leucine rich repeat transmembrane neuronal 4 (LRRTM4) and lysine (K)-specific demethylase5C (KDM5C). Details were listed in Table 2. Three targets of miR-362 were up-regulated and LRRTM4 was downregulated in OLP. Notably, miR-362 was identified using GSEA algorithm therefore we think that the remaining 466 miRNAs might play a negligible role in OLP compared with miR-362. 3.3. PPI network for the DEGs The PPI network was constructed for the protein products of the DEGs (Fig. 1). LRRTM4 and KDM5C were not observed in the network, since there’s no reported interactive protein for both of them. SRGAP2 and VAMP4 were two single nodes in the network, indicating there were no differentially expressed protein interacting with them. 3.4. Significant enriched pathways of DEGs Significant enriched KEGG pathways of DEGs were identified and the top 10 pathways were listed in Table 3. Olfactory transduction (P-value = 1.89E-15), ribosome (P-value = 5.87E-08) and neuroactive ligand-receptor interaction (P-value = 8.03E-08) pathways were included in the list. Especially, obvious deregulated genes in lesions were significantly involved in olfactory transduction and ribosome pathway. 4. Discussion OLP is currently discussed as a status with malignant potential, but its etiology remains unknown and mechanism is not completely elucidated. As a result, there is an urgent and great need to explore the mechanism of OLP, and develop an effective strategy for prevention. In this study, through the comparative analysis of gene expression data and miRNA expression data between patients with OLP and healthy controls, a total of 426 DEGs and 467 DEMs were revealed in OLP. Overexpressed miR-362, which targeted up-regulated SRGAP2 and VAMP4, was identified as the specific miRNA between regulatory miRNAs of DEGs and DEMs. Pathway enrichment analysis showed DEGs were significantly enriched in olfactory transduction, ribosome and neuroactive ligand-receptor interaction pathways.

63

miR-362 was revealed by GSEA as a regulator of the DEGs and it also exhibited differential expression. There are two members in miR-362 family including miR-362-3p and miR-362-5p which may have similar functions. Conflicting results have been reported for the role of miR-362 in cell proliferation. Christensen et al. (2013) have indicated that miR-362-3p induces cell cycle arrest through targeting of E2F transcription factor 1, upstream stimulating factor 2 and protein-tyrosine phosphatase non-receptor type 1 in colorectal cancer. However, Catuogno and colleagues have indicated that transfection of miR-362 may protect lung cancer cells from paclitaxel-induced apoptosis (Catuogno et al., 2013). Meanwhile, Xia et al. (2014) have found that miR-362 induces cell proliferation and apoptosis resistance in gastric cancer by activation of NF-kB signaling. In healthy people, keratinocytes forms the epithelium of the oral mucosa. However, the keratinocyte is the target cell to suffer apoptosis and attract lymphocytes and other immune cells to induce OLP (Yamamoto & Osaki, 1995; Yamamoto, Osaki, Yoneda, & Ueta, 1994). On the other hand, dendritic cells (DCs) play an important role in the immunological response as they activate T cells through antigenic stimulation (Banchereau et al., 2000). Accordingly, we infer that miR-362 might regulate cell life through inducing keratinocytes cycle arrest and proliferation of DCs in OLP. Four DEGs including SRGAP2, VAMP4, LRRTM4 and KDM5C, were the targets of miR-362. SRGAP2 is a member of the SLIT-ROBO Rho GTPase activating protein family. Previous research have demonstrated that SLIT can inhibit invasion and promote a cell cycle arrest via blocking WNT (Wingless-Type MMTV Integration Site Family) signaling or by negatively regulating CDC42 (cell division cycle 42) activity (Chedotal, Kerjan, & Moreau-Fauvarque, 2005; Saegusa, Machida, & Okayasu, 2000). Furthermore, blocking SLIT-ROBO activity decreased apoptosis in human luteal-fibroblast like cells and luteinized granulose cells from the luteinizing follicle (Dickinson, Myers, & Duncan, 2008). In the process, SLIT binding to ROBO through cytoplastic domains relieved inhibition of deleted in colorectal cancer (DCC) by netrin-1, and then this activated proapoptotic pathway through caspase-3 and caspase-9 (Stein & Tessier-Lavigne, 2001). VAMP4 is a member of the vesicleassociated membrane protein and it plays a role in trans-Golgi network-to-endosome transport (Tran, Zeng, & Hong, 2007). Previous study has indicated that VAMP4 binds to adaptor-related protein complex 1 (AP1) (Hinners et al., 2003). Interestingly, induction of AP1 resulted in activation of several genes whose products either positively or negatively regulate apoptosis to maintain the balance between the pro-apoptotic and antiapoptotic genes (Shaulian & Karin, 2002). Furthermore, apoptosis of keratinocytes could contribute to OLP. As a result, we speculate that SRGAP2 and VAMP4 may regulate OLP via inducing keratinocyte apoptosis. Notably, overexpressed miR-362 targeted three upand one down-regulated genes in OLP suggesting that these target genes were regulated by other regulatory elements, such as transcription factor. However, functional studies are necessary to indicate a causal relationship between miR-362 and the two gene expression in OLP. The KEGG pathway analysis of microarray demonstrated that obvious deregulated genes in lesions were significantly involved in olfactory transduction and ribosome pathway. A ribosome is an

Table 2 Target differentially expressed genes of miR-362. Gene symbol

Gene Full name

P-value

SRGAP2 VAMP4 LRRTM4 KDM5C

SLIT-ROBO Rho GTPase activating protein 2 vesicle-associated membrane protein 4 leucine rich repeat transmembrane neuronal 4 lysine (K)-specific demethylase5C

0.000462 0.006043 0.004794 0.003081

logFC: log (fold change).

logFC 0.487274 0.420954 0.34812 0.370548

64

Q. Liu et al. / Archives of Oral Biology 68 (2016) 61–65

Fig. 1. The protein–protein interaction network for the differentially expressed genes. Genes regulated by miR-362 are in triangles. Node color represents differential expression levels of identified genes. Node size represents significance of differentially expressed genes.

organelle as a factory to translate messenger RNA into a polypeptide chain during protein synthesis. Ribosomal proteins play important roles in DNA repair, cell proliferation, regulation of transcription and apoptosis (Chen & Ioannou, 1999; Wang et al., 2001). Consisting with this, significant DEGs involved in ribosome pathway were down-regulated in our study. Moreover, protein synthesis decreased in apoptotic keratinocytes. Therefore, we infer that ribosome pathway is an important process to induce OLP. Validation of results in other OLP samples is a major limitation in

the preliminary study. Therefore, further experimental study such as RT-PCR and western blot are needed to confirm our results. 5. Conclusions Overall, a number of DEGs and DEMs were identified in OLP. miR-362 may be a risk miRNA to induce keratinocytes apoptosis and DCs proliferation by targeting SRGAP2 and VAMP4 in OLP. GSE38616 and GSE38615 used in this study were deposited by

Table 3 Significant top 10 enriched KEGG pathways in the differentially expressed genes. KEGG ID

Pathway Name

Count

P-value

hsa_04740 hsa_03010 hsa_04080 hsa_00830 hsa_00982 hsa_00980 hsa_04260 hsa_04630 hsa_00970 hsa_00350

Olfactory transduction Ribosome Neuroactive ligand-receptor interaction Retinol metabolism Drug metabolism cytochrome P450 Metabolism of xenobiotics by cytochrome P450 Cardiac muscle contraction Jak-STAT signaling pathway Aminoacyl-tRNA biosynthesis Tyrosine metabolism

36 11 21 4 3 3 3 5 2 2

1.89E-15 5.87E-08 8.03E-08 0.0139 0.0925 0.0925 0.1797 0.1828 0.1946 0.1946

Q. Liu et al. / Archives of Oral Biology 68 (2016) 61–65

Gassling et al. and downloaded from GEO database. However, the clinical information of OLP patients and controls cannot be found in the article by Gassling et al. or in the GEO database, which may be a limitation of this study. Besides, more studies are required to confirm their roles in OLP. Further characterization of the miRNA and targets will advance our understanding of the pathogenesis of OLP and may aid in the development of novel therapies for OLP. Competing interests The authors have declared that no competing interests exist. Acknowledgement None. References Abbas, A., Kong, X. B., Liu, Z., Jing, B. Y., & Gao, X. (2013). Automatic peak selection by a Benjamini-Hochberg-based algorithm. PLoS One, 8, e53112. Arão, T. C., Guimarães, A. L. S., de Paula, A. M. B., Gomes, C. C., & Gomez, R. S. (2012). Increased miRNA-146a and miRNA-155 expressions in oral lichen planus. Archives of Dermatological Research, 304, 371–375. Bader, G. D., Betel, D., & Hogue, C. W. (2003). BIND: the biomolecular interaction network database. Nucleic Acids Research, 31, 248–250. Banchereau, J., Briere, F., Caux, C., Davoust, J., Lebecque, S., Liu, Y. J., et al. (2000). Immunobiology of dendritic cells. Annual Review of Immunology, 18, 767–811. Carlos, R., & Contreras, E. (2004). Risk of squamous cell carcinoma in 402 patients with oral lichen planus: a follow-up in an Italian population, S., Gandolfo, L., Richiardi, M., Carrozo, R., Broccoletti, M., Carbone, M., Pagano, C., Vestita, S., Rossso, F., Merleti, Oral Oncology (2004) 40: 77–83. Oral Oncology, 40, 964. Catuogno, S., Cerchia, L., Romano, G., Pognonec, P., Condorelli, G., & de Franciscis, V. (2013). miR-34c may protect lung cancer cells from paclitaxel-induced apoptosis. Oncogene, 32, 341–351. Chedotal, A., Kerjan, G., & Moreau-Fauvarque, C. (2005). The brain within the tumor: new roles for axon guidance molecules in cancers. Cell Death & Differentiation, 12, 1044–1056. Chen, F. W., & Ioannou, Y. A. (1999). Ribosomal proteins in cell proliferation and apoptosis. International Reviews of Immunology, 18, 429–448. Chen, K., & Rajewsky, N. (2007). The evolution of gene regulation by transcription factors and microRNAs. Nature Reviews Genetics, 8, 93–103. Christensen, L. L., Tobiasen, H., Holm, A., Schepeler, T., Ostenfeld, M. S., Thorsen, K., et al. (2013). MiRNA-362-3p induces cell cycle arrest through targeting of E2F1, USF2 and PTPN1 and is associated with recurrence of colorectal cancer. International Journal of Cancer, 133, 67–78. Davis, S., & Meltzer, P. S. (2007). GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics, 23, 1846–1847. Dexter Duncan, N. P. A. B. Z. (2010). WebGestalt2: an updated and expanded version of the Web-based Gene Set Analysis Toolkit. BMC Bioinformatics10. Dickinson, R. E., Myers, M., & Duncan, W. C. (2008). Novel regulated expression of the SLIT/ROBO pathway in the ovary: possible role during luteolysis in women. Endocrinology, 149, 5024–5034. Gassling, V., Hampe, J., Açil, Y., Braesen, J. H., Wiltfang, J., & Häsler, R. (2013). Diseaseassociated miRNA-mRNA networks in oral lichen planus. PLoS One, 8, e63015. Hinners, I., Wendler, F., Fei, H., Thomas, L., Thomas, G., & Tooze, S. A. (2003). AP-1 recruitment to VAMP4 is modulated by phosphorylation-dependent binding of PACS-1. EMBO Reports, 4, 1182–1189. Kanehisa, M., & Goto, S. (2000). KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Research, 28, 27–30. Kanehisa, M., Goto, S., Sato, Y., Kawashima, M., Furumichi, M., & Tanabe, M. (2014). Data, information, knowledge and principle: back to metabolism in KEGG. Nucleic Acids Research, 42, D199–205. Kauers, M., & Yen, L. (2015). On the length of integers in telescopers for proper hypergeometric terms. Journal of Symbolic Computation, 66, 21–33. Kerrien, S., Aranda, B., Breuza, L., Bridge, A., Broackes-Carter, F., Chen, C., et al. (2012). The IntAct molecular interaction database in 2012. Nucleic Acids Research, 40, D841–D846. Kim, S. G., Chae, C. H., Cho, B. O., Kim, H. N., Kim, H. J., Kim, I. S., et al. (2006). Apoptosis of oral epithelial cells in oral lichen planus caused by upregulation of BMP-4. Journal of Oral Pathology & Medicine, 35, 37–45. Kloosterman, W. P., & Plasterk, R. H. (2006). The diverse functions of microRNAs in animal development and disease. Developmental Cell, 11, 441–450.

65

Lavanya, N., Jayanthi, P., Rao, U. K., & Ranganathan, K. (2011). Oral lichen planus: an update on pathogenesis and treatment. Journal of Oral and Maxillofacial Pathology: JOMFP, 15, 127. Liberzon, A., Subramanian, A., Pinchback, R., Thorvaldsdóttir, H., Tamayo, P., & Mesirov, J. P. (2011). Molecular signatures database (MSigDB) 3.0. Bioinformatics, 27, 1739–1740. Lu, M., Zhang, Q., Deng, M., Miao, J., Guo, Y., Gao, W., et al. (2008). An analysis of human microRNA and disease associations. PLoS One, 3, e3420. Peri, S., Navarro, J. D., Kristiansen, T. Z., Amanchy, R., Surendranath, V., Muthusamy, B., et al. (2004). Human protein reference database as a discovery resource for proteomics. Nucleic Acids Research, 32, D497–501. Ramani, A. K., Bunescu, R. C., Mooney, R. J., & Marcotte, E. M. (2005). Consolidating the set of known human protein–protein interactions in preparation for large-scale mapping of the human interactome. Genome Biology, 6, R40. Rual, J. F., Venkatesan, K., Hao, T., Hirozane-Kishikawa, T., Dricot, A., Li, N., et al. (2005). Towards a proteome-scale map of the human protein–protein interaction network. Nature, 437, 1173–1178. Saegusa, M., Machida, D., & Okayasu, I. (2000). Loss of DCC gene expression during ovarian tumorigenesis: relation to tumour differentiation and progression. British Journal of Cancer, 82, 571–578. Shaulian, E., & Karin, M. (2002). AP-1 as a regulator of cell life and death. Nature Cell Biology, 4, E131–136. Smyth, G. K. (2004). Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Statistical Applications in Genetics and Molecular Biology, 3, Article3. Stein, E., & Tessier-Lavigne, M. (2001). Hierarchical organization of guidance receptors: silencing of netrin attraction by slit through a Robo/DCC receptor complex. Science, 291, 1928–1938. Stelzl, U., Worm, U., Lalowski, M., Haenig, C., Brembeck, F. H., Goehler, H., et al. (2005). A human protein–protein interaction network: a resource for annotating the proteome. Cell, 122, 957–968. Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette, M. A., et al. (2005). Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America, 102, 15545–15550. Thongprasom, K., Carrozzo, M., Furness, S., & Lodi, G. (2011). Interventions for treating oral lichen planus. Cochrane Database of Systematic Reviews, 7, . Tran, T. H. T., Zeng, Q., & Hong, W. (2007). VAMP4 cycles from the cell surface to the trans-Golgi network via sorting and recycling endosomes. Journal of Cell Science, 120, 1028–1041. van der Meij, E. H., Schepman, K.-P., & van der Waal, I. (2003). The possible premalignant character of oral lichen planus and oral lichenoid lesions: a prospective study. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology, 96, 164–171. Van der Meij, E., Mast, H., & Van der Waal, I. (2007). The possible premalignant character of oral lichen planus and oral lichenoid lesions: a prospective five-year follow-up study of 192 patients. Oral Oncology, 43, 742–748. van der Waal, I. (2009). Oral lichen planus and oral lichenoid lesions; a critical appraisal with emphasis on the diagnostic aspects. Medicina Oral Patologia Oral Y Cirugia Bucal, 14, e310–e314. Wang, Q., Yang, C., Zhou, J., Wang, X., Wu, M., & Liu, Z. (2001). Cloning and characterization of full-length human ribosomal protein L15 cDNA which was overexpressed in esophageal cancer. Gene, 263, 205–209. Wang, J., Duncan, D., Shi, Z., & Zhang, B. (2013). Web-based gene set analysis toolkit (WebGestalt): update 2013. Nucleic Acids Research, 41, W77–W83. Xenarios, I., Salwinski, L., Duan, X. J., Higney, P., Kim, S. M., & Eisenberg, D. (2002). DIP, the Database of Interacting Proteins: a research tool for studying cellular networks of protein interactions. Nucleic Acids Research, 30, 303–305. Xia, J. T., Chen, L. Z., Jian, W. H., Wang, K. B., Yang, Y. Z., He, W. L., et al. (2014). MicroRNA-362 induces cell proliferation and apoptosis resistance in gastric cancer by activation of NF-kappaB signaling. Journal of Translational Medicine, 12, 33. Yamamoto, T., & Osaki, T. (1995). Characteristic cytokines generated by keratinocytes and mononuclear infiltrates in oral lichen planus. Journal of Investigative Dermatology, 104, 784–788. Yamamoto, T., Osaki, T., Yoneda, K., & Ueta, E. (1994). Cytokine production by keratinocytes and mononuclear infiltrates in oral lichen planus. Journal of Oral Pathology and Medicine, 23, 309–315. Zhang, B., Kirov, S., & Snoddy, J. (2005). WebGestalt: an integrated system for exploring gene sets in various biological contexts. Nucleic Acids Research, 33, W741–748. Zhang, W. Y., Liu, W., Zhou, Y. M., Shen, X. M., Wang, Y. F., & Tang, G. Y. (2012). Altered microRNA expression profile with miR-27b down-regulation correlated with disease activity of oral lichen planus. Oral Diseases, 18, 265–270. Zhao, Z., Sugerman, P., Zhou, X., Walsh, L., & Savage, N. (2001). Mast cell degranulation and the role of T cell RANTES in oral lichen planus. Oral Diseases, 7, 246–251. Zhao, Z., Savage, N., Sugerman, P., & Walsh, L. (2002). Mast cell/T cell interactions in oral lichen planus. Journal of Oral Pathology & Medicine, 31, 189–195.