Abstract
Background
The highly conservative miR-15/107 family (also named as miR-15/107 gene group) including ten miRNA members is currently recognized strongly implicated in multiple human disorders. Some studies focus on the entire family rather than individual miRNA for a bigger picture, while there is also certain signature dysregulation for some of the individual miRNA implicated even in the same disorder.
Methods
Faced with the exponential growth of experimental evidence, our study tries to analyze their function and target interactions using various bioinformatics tools.
Results
Firstly, the evolutionary conservative “AGCAGC” sequence and possible clustered transcriptional pattern were described. Secondly, both the experimentally validated and bioinformatically predicted miRNA-target gene relationship of the entire family was analyzed to understand the mechanism of underlying collective effects for target regulation from the miR-15/107 family. Moreover, pathway analysis among miR-15/107 family was performed and displayed in detail, while its impact on cell proliferation is experimentally validated. Eventually, the dysregulation of miR-15/107 in diseases was discussed.
Conclusions
In summary, our study proposes that the collective functions and implication of miR-15/107 family in various human diseases are achieved relying on the massive overlap** target genes. While the minor differences within target gene interaction among family members could also explain the signature behavior for some of the individual miRNA in aspects such as its disease-specific dysregulation and various participation in pathways.
Similar content being viewed by others
Background
General background of evolutionarily conservative miR-15/107 family
The miR-15/107 family (also called as miR-15/107 gene group) contains multiple highly conservative miRNA members, including miR-15a-5p, miR-15b-5p, miR-16-5p, miR-103a-3p, miR-107, miR-195-5p, miR-424-5p, miR-497-5p, miR-503-5p and miR-646 [1]. Mature miRNA form of this entire family was found highly expressed in eleven human tissues including the cerebral cortex, frontal cortex, primary visual cortex, thalamus, heart, lung, liver, kidney, spleen, stomach and skeletal muscle [2]. These miRNAs share a common “AGCAGC” sequence within the crucial No.2-No.7 seed binding region for their targets [1], which leads to the potential functional overlap among this family.
Family behavior vs. individual signature
Since this miRNA family was recognized [3], the researchers started to capture the impact of the entire family rather than focus exclusively on a single one of such miRNA, which helped us better understand their biological role [
The overlap** target genes were analyzed and a cluster dendrogram of miR-15/107 family was plotted accordingly (Fig. 2). Concerning the profile of their target genes, it was shown that there are three pairs of miRNAs including miR-15a-5p/miR-15b-5p, miR-103a-3p/miR-107 and miR-424/miR-497 closely resembling each other, while three of them including the miR-6838-5p, miR-503-5p and miR-646 vary independently.
Collective effects of target regulation from the miR-15/107 family
Since there are massive overlap** target genes among the miR-15/107 family, we further calculated and emphasized the collective effects of target regulation from multi members. According to the regulatory network calculated by miRTargetLink database, we can see that dozens of target genes (located in the central part of this network) could be affected by this family collectively, especially for genes such as CCNE1, CCND1, VEGFA and so on which were simultaneously subject to regulation from more than 5 members of this family (Fig. 3).
Pathway analysis of the miR-15/107 family
miR-15/107 regulated pathways were calculated using mirPath v3.0 web-server, and the results showed that 39 KEGG pathways (see Additional file 3) were significantly (P < 0.05 in Fisher’s Exact Test) regulated by the miR-15/107 family (Fig. 4). Pathway calculation was based on merging and meta-analysis algorithms according to experimentally validated miRNA target interactions, hence we could find that cluster dendrogram based on miRNA regulated pathways was consistent with that of miRNA target gene. The pathways regulated by the miR-15a-5p/miR-15b-5p, miR-103a-3p/miR-107 and miR-424/miR-497 closely resembled each other, while the miR-6838-5p, miR-503-5p and miR-646 vary independently. Besides, the cluster dendrogram of these pathways further showed that the most significantly regulated pathways included the fatty acid metabolism/biosynthesis/degradation/elongation, various signaling during carcinogenesis, and some crucial pathways for cell survival such as cell cycle, meiosis, adherent junction et al.
Cell cycle as a representative pathway regulated by the miR-15/107 family
Here, we take the cell cycle pathway as an example, to uncover the implication of miR-15/107 family in determining cell fate. miR-15/107 targeted genes within the cell cycle pathway was highlighted in Fig. 5 and listed in Additional file 4. The results showed that 47 of 65 cell cycle related genes (up to 72%) were regulated by miR-15/107 family. Particularly, some of these target genes play multi-functional roles as molecular contacts within the pathway, such as CDKN2A, RBL1, E2F5, TFDP1, RBX1, SKP2, and MCM3. Besides, cell cycle related complex machineries including ORC (origin recognition complex) and MCM (mini-chromosome maintenance complex) were also regulated by miR-15/107. More importantly, there were up to 15 genes subject to regulation from more than 5 members from the miR-15/107 family simultaneously. These bioinformatics evidence suggested a strong implication of miR-15/107 family in cell cycle regulation, while such prediction was also widely supported by experimental evidence (see Additional file 5). Taken together, miR-15/107 family was strongly implicated in multiple pathways including the cell cycle regulation.
Members of the miR-15/107 gene family have an inhibitory effect on cell proliferation
Based on our previous signaling pathway prediction, this family may play an important role in the cell cycle pathway. We validated their role during cell proliferation in human synovial fibroblast cell line SW982 after gain of miR-15/107 function for 48 h. The results showed that the miRNA mimic for miR-16, miR-497, miR-503, miR-646 and miR-6838 displayed a significant inhibition on cell proliferation (Fig. 6). The results demonstrated that this gene family indeed has impact on cell cycle signaling as we predicted.
Dysregulation of miR-15/107 in diseases and prospect of therapeutics
According to the archives in miR2Disease, dysregulated miR-15/107 family was reported in various human diseases, especially in cardiac hypertrophy [24], chronic lymphocytic leukemia (CLL) [25] and prostate cancer [26] (Table 3). More than 5 of such members were simultaneously found abnormally expressed in those diseases, while there are also some of the members individually participated in various diseases such as miR-103 in cerebellar neurodegeneration [27].
These dysregulated miRNAs offered a prospect of therapeutics. Some lncRNAs were discovered to serve as miRNA sponge and since might alter disease progression. For example, miR-107 is found upregulated in glioma cell lines and binds to LncRNA nuclear paraspeckle assembly transcript 1 (NEAT1). NEAT1 silencing inhibits glioma progression, and NEAT1 induces glioma progression by regulating miR-107 as its endogenous sponge [28]. Besides, lncRNA RP11-79H23.3 might suppress the pathogenesis and development of bladder cancer by acting as a sponge for miR-107 to increase PTEN expression [29].
Discussion
In this study, function and target interactions of miR-15/107 family was analyzed using various bioinformatics tools. Firstly, the evolutionary conservative “AGCAGC” sequence, possible clustered transcriptional pattern and tissue specific expression profile were described. Secondly, both the experimentally validated and bioinformatically predicted miRNA-target gene relationship of the entire family was fully interpreted to understand the mechanism of underlying collective effects. Moreover, pathway analysis among miR-15/107 family was performed and displayed in detail. Eventually, the dysregulation of miR-15/107 in diseases was discussed.
According to the previous reports, the implication of miR-15/107 family in various diseases has aroused much attention from multiple fields. In chronic lymphocytic lymphoma (CLL), downregulated miR-15a/miR-16-1 was widely studied [30,31,32]. Their CLL related target genes includes BCL2/MCL1/CCND1/WNT3A [30], BAZ2A/RNF41/RASSF5/MKK3/LRIG1 [31] et al. The miR-15/107 was also strongly implicated in Alzheimer’s disease. Alzheimer is firstly found to be associated with miR-107, it’s levels decreased significantly even in patients with the earliest stages of pathology [33]. The role of miR-15/107 family during AD pathogenesis was also found related with the target suppressing on CDK5R1/p35. Our study focuses on explaining how this miR-15/107 family displayed a collective function and similar implication in various human diseases. We agree that the major contributor should be the massive overlap** miRNA: mRNA target interaction shared crossing the entire family. For this part, we have displayed plenty of convincing evidence achieved from both the computationally prediction and the previous experimentally validation.
As above-mentioned, the common “AGCAGC” sequence within the 5′ seed binding region is the crucial characteristic of miR-15/107 family. Hence, plenty of shared target genes could be widely expected. Besides the important 5′ portion similarity, 3’portion also contribute to target binding. For example, 3′ portion of miR-103/107 appears to play a role in causing miRNAs to bind preferentially with CDS of target mRNAs [35, 36]. Moreover, we considered that the coordinately regulating network derived from those clustered miRNAs might contribute to the majority of the present complicated regulation pattern to lots of target genes, and eventually lead to the multi-miRNA/target interaction network as we described.
Conclusions
Our study proposes that the collective functions and implication of miR-15/107 family in various human diseases are achieved relying on the massive overlap** target genes. While the minor differences within target gene interaction among family members could also explain the signature behavior for some of the individual miRNA in aspects such as its disease-specific dysregulation and participation in pathways.