br Lastly given the scarcity of studies
Lastly, given the scarcity of studies about regulation by miRNAs shared between comprehensive sets of tumor suppressor genes and oncogenes acting on different cell signaling pathways, we investigated whether validated and predicted data sources could identify miRNAs/ miRNA families with potential tumor suppressor and oncogenic func-tions. The mir-192 family was significantly overrepresented as reg-ulator of tumor suppressor genes, having experimentally validated in-teractions with genes implicated in essential DNA repair pathways, including double-strand break repair by homologous recombination (BRCA1, BRCA2, and RAD51), mismatch repair (MSH6), and nucleotide excision repair induced by ultraviolet light (XPA). An upregulation of this miRNA family was reported in multiple tumor types including gastric cancer, hepatocellular carcinoma, neuroblastoma, pancreatic ductal adenocarcinoma, and esophageal squamous cell carcinoma (Feinberg-Gorenshtein et al., 2013; Zhao et al., 2013; Chen et al., 2014; Tan et al., 2014; Li et al., 2015). Preliminary evidences suggest that this expression pattern could enhance cell proliferation and migration, re-duce apoptosis and promote Okadaic acid progression (Feinberg-Gorenshtein et al., 2013; Zhao et al., 2013; Li et al., 2015). Of note, mir-192 was previously identified as an oncomiR in gastric tumorigenesis (Jin et al., 2011; Zhang et al., 2018), while its biological effects in other cancers have been only partially elucidated. These findings support our hypothesis that miRNAs within this family may have a broader onco-genic function. However, mir-192 was found downregulated in colon, colorectal, and lung tumors (Feng et al., 2011; Karaayvaz et al., 2011; Chiang et al., 2012). We also identified some miRNA families that might act as effectors in tumor suppression, including mir-128, mir-1471, mir-483, mir-3170 and mir-218. Recent reports corroborate the assigned role to some of these miRNA families: mir-128 exerts pro-
Fig. 2. Regulatory networks and Venn diagrams obtained by combining validated and predicted interactions between miRNAs and tumor suppressor genes/oncogenes. miRNAs (A) and miRNA families (B) significantly overrepresented (P < 0.01) as regulators (diamond nodes) of expression between the groups of tumor suppressor genes (green circle nodes) and oncogenes studied (red circle nodes) are shown in these interactomes. Red-outlined diamonds indicate miRNAs with potential tumor suppressor function, whereas the green-outlined ones represent potentially oncogenic miRNAs. Solid lines denote experimentally validated interactions and dashed lines are high con-fidence computationally predicted interactions. Interactome (A) was built considering only miRNAs with at least three target genes among tumor suppressor genes and oncogenes. Intersections among the sets of validated and predicted miRNAs found to modulate expression of tumor suppressor genes (C) and oncogenes (D) are indicated in Venn diagrams. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
apoptotic effects and it was found as a tumor suppressor in certain tumor types (Adlakha and Saini, 2013; Hauser et al., 2015; Wu et al., 2015; Shan et al., 2016); mir-483 suppresses the proliferation of glioma and squamous cell carcinoma cells (Wang et al., 2012; Bertero et al., 2013); and emerging tumor-suppressing roles have been described for mir-218 in prostate, breast and lung cancers (Liu et al., 2016; Song et al., 2016; Guan et al., 2017). Up to now, little is known about mir-1471 and mir-3170 roles' in cancer and, if experimentally proven, they might represent novel tumor-suppressive miRNA families. Moreover, mir-34a is a known tumor-suppressive miRNA (Raver-Shapira et al., 2007) and it was found regulating several oncogenes included in the present study, validating our in silico analysis. Importantly, in a feed-forward loop fashion, mir-34a has a positive effect on p53 transcrip-tional activity and protein stability (encoded by tumor suppressor gene TP53), by targeting multiple p53 inhibitor genes (e.g., the MDM4 on-cogene) (Mandke et al., 2012; Okada et al., 2014). In turn, p53 upre-gulates this miRNA, contributing to apoptosis and senescence (Raver-Shapira et al., 2007; Navarro and Lieberman, 2015). In fact, TP53 and MDM4 genes appeared as direct targets of mir-34a in our analysis. Overall, our results using this computational analysis are in agreement with already available literature data, suggesting that it is a suitable strategy to identify potentially tumor-suppressive and oncogenic miRNAs.