The purpose of systems biology is to combine comprehensive biological data from varied experimental approaches to realize complex interactions at the molecular stage [11]. One of the important protein components of the human immune system is CD25 expressed on cell surface of Treg cells [1]. CD4+CD25+ Tregs cells inside TME known as TI-Tregs possess the essential function in cancer immune escape [12]. Albeit the molecular mechanisms of CD25-mediated signaling pathways has been revealed, some inquiries are still unanswered. Using systems biology approach, the molecular interactors including lncRNAs, miRNAs and proteins for significant protein of CD25 will be unraveled. Revealing the interactions between CD25 and other molecules in TME may be promised to overcome the cancer immune escape and gain the cancer successful treatment [13].
With this aim, we found that the CD25 protein interactors were included IL2, IL2RB, IL2RG, STAT5B, FOXP3, STAT5A, CSF2, JAK3, STAT3, IL5, JAK1, LCK, IL3, CD8A, CD3E, STAT1, AKT1, JAK2, CSF2RA and SOCS1. Also, the enrichment analysis showed that the top five significant signaling pathways were included JAK-STAT signaling pathway, Th17 cell differentiation, Th1 and Th2 cell differentiation, Measles and pathways in cancer. Indeed, JAK-STAT signaling pathway and pathways in cancer were related with high score based on the tree illustration in ShinyGO v0.60 web server (Fig. 2d). Among the evaluated miRNAs targeting CD25 mRNA, starBase v3.0 revealed that hsa-miR-30a-5p, hsa-miR-30b-5p, hsa-miR-30c-5p, hsa-miR-30d-5p, hsa-miR-30e-5p, hsa-miR-140-5p, hsa-miR-143-3p, hsa-miR-211-5p and hsa-miR-324-3p were engaged in lncRNA interaction with OIP5-AS1. SFPEL-LPI demonstrated that the lncRNA had the interaction with the following proteins: ELAVL1, IGF2BP3, IGF2BP2, IGF2BP1, RNA-binding protein FUS (FUS), TARDBP, AGO2, TIA1, PTBP1, AGO3, AGO4, SRSF1, Putative helicase MOV-10 (MOV10), AGO1, TNRC6A, RBFOX2, Transcriptional repressor protein YY1 (YY1), Transcription factor Sp1 (SP1), PTEN, Polycomb protein SUZ12 (SUZ12), SF1 and REST (Table 3, Fig. 3). Generally, SFPEL-LPI utilizes the sequence features of lncRNAs and proteins. Also, this web server calculates multiple similarities of protein-protein and lncRNA-lncRNA using protein and lncRNA sequences and recognized lncRNA-protein interactions. Next, SFPEL-LPI merges multiple features and similarities with a feature projection ensemble learning frame [10].
This lncRNA is an important non-coding RNA involved in many cellular processes. In fact, the lncRNA with NONCODE ID: NONHSAT041930 or OIP5-AS1 abbreviated from OIP5 antisense RNA 1, is a mammalian lncRNA functioning in the cytoplasm [14]. The OIP5-AS1 has been focused for its role in the development of brain and eye [15]. Kim et al. (2016) reported that the lncRNA could inhibit HuR binding to target mRNAs. Therefore, it repressed the HuR-elicited proliferative phenotypes. In fact, they reported as the study of HeLa cells that OIP5-AS1 sponges ELAVL1 [16]. Also, Kim et al. (2017) illustrated that the lncRNA had the interaction with GAK mRNA, advancing GAK mRNA decay and sodecreasing GAK protein levels and reducing cell proliferation [17]. Also, Zhang et al. (2019) concluded OIP5-AS1 played as a ceRNA to make proliferation, migration and invasion of primary HemECs via regulating miR-195-5p/NOB1 axis. Indeed, ceRNAs perform as the molecular sponges for a particular miRNA via their miRNA binding sites [18]. Because of this function, also known as MREs, they de-repressed all target genes from the respective miRNA family [19].
In another point of view, the present co-expression analysis showed that OIP5-AS1 was co-expressed in normal vs. tumor bladder cancer based on the RNA-seq data. Clearly, this lncRNA was co-expressed in aminoacyl-tRNA biosynthesis pathway with p-value= 3.33E-15 and Bonferroni correction= 6.02E-13. Also, this cancer had the co-expressed OIP5-AS1 in other four top predicted pathways such as DNA replication, RNA degradation, cell cycle and spliceosome (Table 4). This analysis in breast cancer showed that pathways including pathways in cancer, neurotrophin signaling pathway, spliceosome, purine metabolism and aminoacyl-tRNA biosynthesis possessed the co-expressed OIP5-AS1 in the normal vs. tumor tissue (Table S4). This characteristic in head and neck cancer was engaged in these cellular processes including endocytosis, T cell receptor signaling pathway, neurotrophin signaling pathway, MAPK signaling pathway and lysosome (Table S5). In case of kidney cancer, purine metabolism, prostate cancer pathways, insulin signaling, endocytosis and RNA degradation were involved by p-value of 1.32E-12, 5.94E-12, 1.89E-11, 3.66E-10 and 3.87E-10, respectively (Table S6). The liver cancer showed that the regulation of actin cytoskeleton, neurotrophin signaling pathway, focal adhesion, pancreatic cancer pathway and ribosome were engaged in OIP5-AS1 co-expression (Table S7). For lung cancer, MAPK signaling pathway, purine metabolism, lysosome, huntingtons disease and pyrimidine metabolism demonstrated the co-expressed OIP5-AS1 (Table S8). For prostate cancer, focal adhesion, MAPK signaling pathway, chemokine signaling pathway, regulation of actin cytoskeleton apoptosis and apoptosis were shared the co-expression of OIP5-AS1 (Table S9). Notably, the thyrioid cancer illustrated that neurotrophin signaling pathway, RNA degradation, lysine degradation, aminoacyl-tRNA biosynthesis and renal cell carcinoma pathways had the shared feature from the point of view of OIP5-AS1 co-expression by p-value of 5.55E-16, 3.75E-14, 2.82E-11, 1.06E-10 and 1.48E-10, respectively (Table S10).
As a recent report, OIP5-AS1 lncRNA could adjust cell proliferation and apoptosis by miR-410 and its target KLF10/PTEN/AKT [20]. In addition, the researchers recognized a putative ceRNA network for lncRNAs of AC008124.1, OPI5-AS1 and NEAT1 in breast tumors [21]. Also, it was studied in two human osteosarcoma cell lines, MG63 and SaOS2, that OIP5-AS1 led cisplatin resistance via provoking the LPAATβ/PI3K/AKT/mTOR signaling pathway as the sponge for miR-340-5p [22]. In another study in osteosarcoma tissues and cells, the silencing of OIP5-AS1 inhibited the proliferation and also speeded up the apoptosis, and G0/G1 cycle arrest. Indeed, OIP5-AS1/miR-223/CDK14 performed the modulation on the tumorigenesis of osteosarcoma [23]. It was resulted that the over-expression of miR-367-3p, piR-30188 and PIWIL3 or knockdown of OIP5-AS1 effected on the suppression of glioma progression [24]. In undifferentiated oral tumors, the over-expression of OIP5-AS1 could be proposed for the poor clinical result and elevated cancer stemness [25]. Also, OIP5-AS expression was significantly reduced in non-small cell lung cancer tissues against adjacent non-cancerous tissues in whole samples and in male patients [26].
The obtained results pertained to the interaction of OIP5-AS1 with CD25 and its interactors-targeting miRNAs revealed that OIP5-AS1 acts possibly as reported function of molecular sponge in the regulation of CD25 expression and its protein interactors by the relative miRNAs. OIP5-AS1 showed the interaction with CD25-targeting miRNAs. In fact, hsa-miR-152-3p (with medium score of mirDIP), hsa-miR-137 (with medium score of mirDIP), hsa-miR-148a-3p (with medium score of mirDIP), hsa-miR-143-3p (with very high score of mirDIP), hsa-miR-92a-3p (with medium score of mirDIP), hsa-miR-4659a-3p (with medium score of mirDIP), hsa-miR-4659b-3p (with medium score of mirDIP), hsa-miR-1305 (with high score of mirDIP), hsa-miR-92b-3p (with medium score of mirDIP), hsa-miR-3606-3p (with medium score of mirDIP), hsa-miR-1277-5p (with high score of mirDIP) and hsa-miR-32-5p (with medium score of mirDIP), ranging from 1 to 0.996 score in LncBase Predicted v.2, targeted CD25 mRNA and interacted with OIP5-AS1, possibly. Also, from the view point of miRNAs targeting protein interactors of CD25 including hsa-miR-152-3p (with very high score of mirDIP targeting JAK1), hsa-miR-137 (with high score of mirDIP targeting STAT3), hsa-miR-148a-3p (with high score of mirDIP targeting STAT1), hsa-miR-143-3p (with high score of mirDIP targeting AKT1), hsa-miR-92a-3p (with medium score of mirDIP targeting JAK3), hsa-miR-4659a-3p (with medium score of mirDIP targeting CSF2), hsa-miR-4659b-3p (with medium score of mirDIP targeting CSF2), hsa-miR-1305 (with high score of mirDIP targeting IL2RB), hsa-miR-92b-3p (with medium score of mirDIP targeting JAK3), hsa-miR-3606-3p (with medium score of mirDIP targeting JAK2), hsa-miR-1277-5p (with high score of mirDIP targeting CSF2) and hsa-miR-32-5p (with medium score of mirDIP targeting JAK2), ranging from 1 to 0.996 score in LncBase Predicted v.2, had probably the interaction with OIP5-AS1. Consequently, OIP5-AS1 can possibly control the signaling pathways in which CD25 were engaged in as the mentioned top five signaling pathways for CD25 function including JAK-STAT signaling pathway, Th17 cell differentiation, Th1 and Th2 cell differentiation, Measles and pathways in cancer resulted from the enrichment and network analyses. Also, the above considered cancers including bladder, breast, head and neck, kidney, liver, lung, prostate and thyroid possessed probably the clue of the involvement of OIP5-AS1 in CD25-pertained signaling pathways which were analyzed as the co-expressed lncRNA in this research.