miRe2e: a full end-to-end deep model based on transformers for prediction of pre-miRNAs.

MOTIVATION: MicroRNAs (miRNAs) are small RNA sequences with key roles in the regulation of gene expression at post-transcriptional level in different species. Accurate prediction of novel miRNAs is needed due to their importance in many biological processes and their associations with complicated diseases in humans. Many machine learning approaches were proposed in the last decade for this purpose, but requiring handcrafted features extraction to identify possible de novo miRNAs. More recently, the emergence of deep learning (DL) has allowed the automatic feature extraction, learning relevant representations by themselves. However, the state-of-art deep models require complex pre-processing of the input sequences and prediction of their secondary structure to reach an acceptable performance. RESULTS: In this work, we present miRe2e, the first full end-to-end DL model for pre-miRNA prediction. This model is based on Transformers, a neural architecture that uses attention mechanisms to infer global dependencies between inputs and outputs. It is capable of receiving the raw genome-wide data as input, without any pre-processing nor feature engineering. After a training stage with known pre-miRNAs, hairpin and non-harpin sequences, it can identify all the pre-miRNA sequences within a genome. The model has been validated through several experimental setups using the human genome, and it was compared with state-of-the-art algorithms obtaining 10 times better performance. AVAILABILITY AND IMPLEMENTATION: Webdemo available at https://sinc.unl.edu.ar/web-demo/miRe2e/ and source code available for download at https://github.com/sinc-lab/miRe2e. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

EsxA mainly contributes to the miR-155 overexpression in human monocyte-derived macrophages and potentially affect the immune mechanism of macrophages through miRNA dysregulation.

BACKGROUND/PURPOSE: Mycobacterium tuberculosis is a successful intracellular pathogen that uses multiple proteins to survive within macrophages, one of the most remarkable is the virulence factor EsxA. In this study, we evaluate the participation of EsxA in the miRNAs expression profile of human monocyte-derived macrophages (hMDM), to mapping out the contribution of this virulence factor in the miRNA profile and how these changes can influence and alter immune-related processes and pathways. METHODS: The cytotoxic effect of rEsxA on hMDM was evaluated by the neutral red assay. The evaluation of miRNA expression profile in infected and rEsxA-stimulated hMDM was done using TaqMan Low Density Assays, and in silico analyses was carried on to construct Protein-Protein Interaction network of miRNAs targets. RESULTS: miR-155 was the only miRNA upregulated consistently in hMDM infected with M. tuberculosis H37Rv or stimulated with rEsxA. In hMDM stimulated with rEsxA, we found 25 miRNA's dysregulated (8 up-regulated and 17 down-regulated). The most significant were the miR-155 and miR-622 that has been observed in the analysis carried out with two different endogenous controls (U6 snRNA and RNU44) for the normalization of expression analysis. This result suggests that rEsxA induces the deregulation of miRNAs that potentially target genes in key pathways for the infection control, like the MAPK signaling pathway, cytokines, and chemokine signaling pathways, and several connected pathways involved in mycobacterial uptake, vesicular traffic, and endosome maturation. CONCLUSION: Higher expression levels of miR-155 suggest potential roles of these miRNA in EsxA-dependent immune subversion.

Structural insights of the pre-let-7 interaction with LIN28B.

The Let-7:LIN28 regulatory loop is a paradigm in miRNA regulation. LIN28 harbors two RNA binding domains, which interact with well-conserved sequences in pre-let-7 RNAs, the GNGAY and the GGAG motifs. Here, the differential binding between LIN28B and pre-let-7 members was associated with the structural characteristics of the pre-let-7 family mapped by SHAPE, uncovering diverse structural patterns within pre-let-7 members. Pre-let-7 mutants supported a relevant role of the GGAG motif location and the preE-stem stability for the interaction with LIN28B. Based on these results, we propose a core RNA structure for LIN28B interaction.

Identification of key sequence features required for microRNA biogenesis in plants.

MicroRNAs (miRNAs) are endogenous small RNAs of ∼21 nt that regulate multiple biological pathways in multicellular organisms. They derive from longer transcripts that harbor an imperfect stem-loop structure. In plants, the ribonuclease type III DICER-LIKE1 assisted by accessory proteins cleaves the precursor to release the mature miRNA. Numerous studies highlight the role of the precursor secondary structure during plant miRNA biogenesis; however, little is known about the relevance of the precursor sequence. Here, we analyzed the sequence composition of plant miRNA primary transcripts and found specifically located sequence biases. We show that changes in the identity of specific nucleotides can increase or abolish miRNA biogenesis. Most conspicuously, our analysis revealed that the identity of the nucleotides at unpaired positions of the precursor plays a crucial role during miRNA biogenesis in Arabidopsis.

Characterization and discovery of miRNA and miRNA targets from apomictic and sexual genotypes of Eragrostis curvula.

BACKGROUND: Weeping lovegrass (Eragrostis curvula [Shrad.] Nees) is a perennial grass found in semi-arid regions that is well adapted for growth in sandy soils and drought conditions. E. curvula constitutes a polymorphic complex that includes cytotypes with different ploidy levels (from 2x to 8x), where most polyploids are facultative apomicts, although both sexual reproduction and full apomixis have been reported in this species. Apomixis is thought to be associated with silencing of the sexual pathway, which would involve epigenetic mechanisms. However, a correlation between small RNAs and apomixis has not yet been conclusively established. RESULTS: Aiming to contribute to the elucidation of their role in the expression of apomixis, we constructed small RNA libraries from sexual and apomictic E. curvula genotypes via Illumina technology, characterized the small RNA populations, and conducted differential expression analysis by comparing these small RNAs with the E. curvula reference transcriptome. We found that the expression of two genes is repressed in the sexual genotype, which is associated with specific microRNA expression. CONCLUSION: Our results support the hypothesis that in E. curvula the expression of apomixis leads to sexual repression.

Predicting novel microRNA: a comprehensive comparison of machine learning approaches.

MOTIVATION: The importance of microRNAs (miRNAs) is widely recognized in the community nowadays because these short segments of RNA can play several roles in almost all biological processes. The computational prediction of novel miRNAs involves training a classifier for identifying sequences having the highest chance of being precursors of miRNAs (pre-miRNAs). The big issue with this task is that well-known pre-miRNAs are usually few in comparison with the hundreds of thousands of candidate sequences in a genome, which results in high class imbalance. This imbalance has a strong influence on most standard classifiers, and if not properly addressed in the model and the experiments, not only performance reported can be completely unrealistic but also the classifier will not be able to work properly for pre-miRNA prediction. Besides, another important issue is that for most of the machine learning (ML) approaches already used (supervised methods), it is necessary to have both positive and negative examples. The selection of positive examples is straightforward (well-known pre-miRNAs). However, it is difficult to build a representative set of negative examples because they should be sequences with hairpin structure that do not contain a pre-miRNA. RESULTS: This review provides a comprehensive study and comparative assessment of methods from these two ML approaches for dealing with the prediction of novel pre-miRNAs: supervised and unsupervised training. We present and analyze the ML proposals that have appeared during the past 10 years in literature. They have been compared in several prediction tasks involving two model genomes and increasing imbalance levels. This work provides a review of existing ML approaches for pre-miRNA prediction and fair comparisons of the classifiers with same features and data sets, instead of just a revision of published software tools. The results and the discussion can help the community to select the most adequate bioinformatics approach according to the prediction task at hand. The comparative results obtained suggest that from low to mid-imbalance levels between classes, supervised methods can be the best. However, at very high imbalance levels, closer to real case scenarios, models including unsupervised and deep learning can provide better performance.

Regulatory roles of miR-155 and let-7b on the expression of inflammation-related genes in THP-1 cells: effects of fatty acids.

The main aim of this investigation was to study the regulatory roles of let-7b and miR-155-3p on the expression of inflammation-associated genes in monocytes, macrophages, and lipopolysaccharide (LPS)-activated macrophages (AcM). A second goal was to analyze the potential modulatory roles of different fatty acids, including oleic, palmitic, eicosapentaenoic (EPA), and docosahexaenoic (DHA), on the expression of these miRNAs in the three cell types. This hypothesis was tested in human acute monocytic leukemia cells (THP-1), which were differentiated into macrophages with 2-O-tetradecanoylphorbol-13-acetate (TPA) and further activated with LPS for 24 h. Monocytes, macrophages, and AcM were transfected with a negative control, or mimics for miR-155-3p and miR-let-7b-5p. The expression of both miRNAs and some proinflammatory genes was analyzed by qRT-PCR. Interestingly, let-7b mimic reduced the expression of IL6 and TNF in monocytes, and SERPINE1 expression in LPS-activated macrophages. However, IL6, TNF, and SERPINE1 were upregulated in macrophages by let-7b mimic. IL6 expression was higher in the three types of cells after transfecting with miR-155-3p mimic. Similarly, expression of SERPINE1 was increased by miR-155-3p mimic in monocytes and macrophages. However, TLR4 was downregulated by miR-155-3p in monocytes and macrophages. Regarding the effects of the different fatty acids, oleic acid increased the expression of let-7b in macrophages and AcM and also increased the expression of miR-155 in monocytes when compared with DHA but not when compared with non-treated cells. Overall, these results suggest anti- and proinflammatory roles of let-7b and miR-155-3p in THP-1 cells, respectively, although these outcomes are strongly dependent on the cell type. Noteworthy, oleic acid might exert beneficial anti-inflammatory effects in immune cells (i.e., non-activated and LPS-activated macrophages) by upregulating the expression of let-7b.

Conformational sampling of the intrinsically disordered dsRBD-1 domain from Arabidopsis thaliana DCL1.

DCL1 is the ribonuclease that carries out miRNA biogenesis in plants. Substrate pri-miRNA recognition by DCL1 requires two double stranded RNA binding domains located at the C-terminus of the protein. We have previously shown that the first of these domains, DCL1-A, is intrinsically disordered and folds upon binding pri-miRNA. Integrating NMR and SAXS data, we study here the conformational landscape of free DCL1-A through an ensemble description. Our results reveal that secondary structure elements, corresponding to the folded form of the protein, are transiently populated in the unbound state. The conformation of one of the dsRNA binding regions in the free protein shows that, at a local level, RNA recognition proceeds through a conformational selection mechanism. We further explored the stability of the preformed structural elements via temperature and urea destabilization. The C-terminal helix is halfway on the folding pathway in free DCL1-A, constituting a potential nucleation site for the final folding of the protein. In contrast, the N-terminal helix adopts stable non-native structures that could hinder the correct folding of the protein in the absence of RNA. This description of the unfolded form allows us to understand details of the mechanism of binding-induced folding of the protein.

Genome-wide pre-miRNA discovery from few labeled examples.

Motivation: Although many machine learning techniques have been proposed for distinguishing miRNA hairpins from other stem-loop sequences, most of the current methods use supervised learning, which requires a very good set of positive and negative examples. Those methods have important practical limitations when they have to be applied to a real prediction task. First, there is the challenge of dealing with a scarce number of positive (well-known) pre-miRNA examples. Secondly, it is very difficult to build a good set of negative examples for representing the full spectrum of non-miRNA sequences. Thirdly, in any genome, there is a huge class imbalance (1: 10 000) that is well-known for particularly affecting supervised classifiers. Results: To enable efficient and speedy genome-wide predictions of novel miRNAs, we present miRNAss, which is a novel method based on semi-supervised learning. It takes advantage of the information provided by the unlabeled stem-loops, thereby improving the prediction rates, even when the number of labeled examples is low and not representative of the classes. An automatic method for searching negative examples to initialize the algorithm is also proposed so as to spare the user this difficult task. MiRNAss obtained better prediction rates and shorter execution times than state-of-the-art supervised methods. It was validated with genome-wide data from three model species, with more than one million of hairpin sequences each, thereby demonstrating its applicability to a real prediction task. Availability and implementation: An R package can be downloaded from https://cran.r-project.org/package=miRNAss. In addition, a web-demo for testing the algorithm is available at http://fich.unl.edu.ar/sinc/web-demo/mirnass. All the datasets that were used in this study and the sets of predicted pre-miRNA are available on http://sourceforge.net/projects/sourcesinc/files/mirnass. Contact: cyones@sinc.unl.edu.ar. Supplementary information: Supplementary data are available at Bioinformatics online.

The regulatory role of IL-6R in hepatitis B-associated fibrosis and cirrhosis.

This study investigated the expression and regulation of IL-6R in hepatitis B-associated moderate hepatic fibrosis and cirrhosis. Liver tissues, peripheral blood monocytes (PBMs) and serum were collected from 26 hepatitis B patients with liver fibrosis and 35 hepatitis B patients with liver cirrhosis. The levels of Il-6r mRNA expression in these samples were examined by quantitative real-time PCR and IL-6R protein levels were analyzed by western blot and ELISA. MiRNAs that regulate IL-6R expression were predicted by bioinformatics analysis, and validated by dual luciferase reporter assay. Compared with the hepatic fibrosis group, IL-6R was significantly upregulated at both mRNA and protein levels in liver tissues, PBMs and serum samples from the hepatic cirrhosis group (P<0.05). The 3'UTR of Il-6r mRNA was predicted to contain a miR-30b binding site and IL-6R was identified as a possible target of miR-30b. MiR-30b expression was significantly downregulated in samples from hepatic cirrhosis patients compared with hepatic fibrosis patients (P<0.05). In conclusion, IL-6R was upregulated while miR-30b was decreased in patients with liver cirrhosis. The miR-30 can directly regulate the expression of IL-6R.

Graph-based Clustering of miRNA Sequences.

AIMS: The many miRNAs discovered so far have been divided into biologically representative families, aiming at organizing and systematizing their study so to promote, mainly, a better understanding of their functionalities. Clustering miRNA sequences can corroborate the family-based organizations as well as helping to explore sequences belonging to the same cluster as potentially having similar biological functions. OBSERVATIONS: Considering that members of the same miRNA family tend to biologically function in similar ways, a well-structured family can help detecting miRNA functions which have not been associated yet with any existing family. METHODS: The work described in this paper empirically investigates the suitability of organizing miRNAs as families, using a clustering algorithm based on a particular type of graph i.e., minimal spanning tree (MST), for clustering miRNA sequences. Seven miRNA families stored in the online miRBase database have been used as input to the MST-based clustering algorithm and clustering results have been compared to assess the suitability of identirying them. RESULTS: The motivations for the experiments were to identify refinements in miRNA family organizations that could be pursued and, also, investigate how the chosen graph-based clustering algorithm would perform in miRNA related domains. CONCLUSION: Interesting and useful results, particularly related to detecting information based on the visualization of the final induced graphs, and their induced connected components (clusters), are presented and discussed. Particularly, experiments results suggested the possibility of refining some miRNA families by grouping some of their miRNAs as sub-families.

Cestode parasites release extracellular vesicles with microRNAs and immunodiagnostic protein cargo.

Intercellular communication is crucial in multiple aspects of cell biology. This interaction can be mediated by several mechanisms including extracellular vesicle (EV) transfer. EV secretion by parasites has been reported in protozoans, trematodes and nematodes. Here we report that this mechanism is present in three different species of cestodes, Taenia crassiceps, Mesocestoides corti and Echinococcus multilocularis. To confirm this we determined, in vitro, the presence of EVs in culture supernatants by transmission electron microscopy. Interestingly, while T. crassiceps and M. corti metacestodes secrete membranous structures into the culture media, similar vesicles were observed in the interface of the germinal and laminated layers of E. multilocularis metacestodes and were hardly detected in culture supernatants. We then determined the protein cargo in the EV-enriched secreted fractions of T. crassiceps and M. corti conditioned media by LC-MS/MS. Among the identified proteins, eukaryotic vesicle-enriched proteins were identified as expected, but also proteins used for cestode disease diagnosis, proteins related to neurotransmission, lipid binding proteins as well as host immunoglobulins and complement factors. Finally, we confirmed by capillary electrophoresis the presence of intravesicular RNA for both parasites and detected microRNAs by reverse transcription-PCR. This is the first report of EV secretion in cestode parasites and of an RNA secretion mechanism. These findings will provide valuable data not only for basic cestode biology but also for the rational search for new diagnostic targets.

Use of Mature miRNA Strand Selection in miRNAs Families in Cervical Cancer Development.

Aberrant miRNA expression is well recognized as a cancer hallmark, nevertheless miRNA function and expression does not always correlate in patients tissues and cell lines studies. In addition to this issue, miRNA strand usage conduces to increased cell signaling pathways modulation diversifying cellular processes regulation. In cervical cancer, 20 miRNA families are involved in carcinogenesis induction and development to this moment. These families have 5p and 3p strands with different nucleotide (nt) chain sizes. In general, mature 5p strands are larger: two miRNAs of 24 nt, 24 miRNAs of 23 nt, 35 miRNAs of 22 nt and three miRNAs of 21 nt. On the other hand, the 3p strands lengths observed are: seven miRNAs of 23 nt, 50 miRNAs of 22 nt, six miRNAs of 21 nt and four miRNAs of 20 nt. Based on the analysis of the 20 miRNA families associated with cervical cancer, 67 3p strands and 65 5p strands are selected suggesting selectivity and specificity mechanisms regulating cell processes like proliferation, apoptosis, migration, invasion, metabolism and Warburg effect. The insight reviewed here could be used in the miRNA based therapy, diagnosis and prognosis approaches.

MicroRNA-34 directly targets pair-rule genes and cytoskeleton component in the honey bee.

MicroRNAs (miRNAs) are key regulators of developmental processes, such as cell fate determination and differentiation. Previous studies showed Dicer knockdown in honeybee embryos disrupt the processing of functional mature miRNAs and impairs embryo patterning. Here we investigated the expression profiles of miRNAs in honeybee embryogenesis and the role of the highly conserved miR-34-5p in the regulation of genes involved in insect segmentation. A total of 221 miRNAs were expressed in honey bee embryogenesis among which 97 mature miRNA sequences have not been observed before. Interestingly, we observed a switch in dominance between the 5-prime and 3-prime arm of some miRNAs in different embryonic stages; however, most miRNAs present one dominant arm across all stages of embryogenesis. Our genome-wide analysis of putative miRNA-target networks and functional pathways indicates miR-34-5p is one of the most conserved and connected miRNAs associated with the regulation of genes involved in embryonic patterning and development. In addition, we experimentally validated that miR-34-5p directly interacts to regulatory elements in the 3'-untranslated regions of pair-rule (even-skipped, hairy, fushi-tarazu transcription factor 1) and cytoskeleton (actin5C) genes. Our study suggests that miR-34-5p may regulate the expression of pair-rule and cytoskeleton genes during early development and control insect segmentation.

Expression profiles of eNOS, iNOS and microRNA-27b in the corpus cavernosum of rats submitted to chronic alcoholism and Diabetes mellitus

Purpose: To evaluate the expression of endothelial and inducible NOS in addition to the miRNA-27b in the corpus cavernosum and peripheral blood of healthy rats, diabetic rats, alcoholic rats and rats with both pathologies. Methods: Forty eight Wistar rats were divided into four groups: control (C), alcoholic (A), diabetic (D) and alcoholic-diabetic (AD). Samples of the corpus cavernosum were prepared to study protein expressions of eNOS and iNOS by immunohistochemistry and expression of miRNA-27b in the corpus cavernosum and peripheral blood. Results: Immunohistochemistry for eNOS and iNOS showed an increase in cavernosal smooth muscle cells in the alcoholic, diabetic and alcoholic-diabetic groups when compared with the control group. Similarly, the mRNA levels for eNOS were increased in cavernosal smooth muscle (CSM) in the alcoholic, diabetic and alcoholic-diabetic groups and miRNA-27b were decreased in CSM in the alcoholic, diabetic and alcoholic-diabetic groups. Conclusion: The major new finding of our study was an impairment of relaxation of cavernosal smooth muscle in alcoholic, diabetic, and alcoholic-diabetic rats that involved a decrease in the nitric oxide pathway by endothelium-dependent mechanisms accompanied by a change in the corpus cavernosum contractile sensitivity.(AU)

The regulatory role of IL-6R in hepatitis B-associated fibrosis and cirrhosis

This study investigated the expression and regulation of IL-6R in hepatitis B-associated moderate hepatic fibrosis and cirrhosis. Liver tissues, peripheral blood monocytes (PBMs) and serum were collected from 26 hepatitis B patients with liver fibrosis and 35 hepatitis B patients with liver cirrhosis. The levels of Il-6r mRNA expression in these samples were examined by quantitative real-time PCR and IL-6R protein levels were analyzed by western blot and ELISA. MiRNAs that regulate IL-6R expression were predicted by bioinformatics analysis, and validated by dual luciferase reporter assay. Compared with the hepatic fibrosis group, IL-6R was significantly upregulated at both mRNA and protein levels in liver tissues, PBMs and serum samples from the hepatic cirrhosis group (P<0.05). The 3′UTR of Il-6r mRNA was predicted to contain a miR-30b binding site and IL-6R was identified as a possible target of miR-30b. MiR-30b expression was significantly downregulated in samples from hepatic cirrhosis patients compared with hepatic fibrosis patients (P<0.05). In conclusion, IL-6R was upregulated while miR-30b was decreased in patients with liver cirrhosis. The miR-30 can directly regulate the expression of IL-6R.

Automatic learning of pre-miRNAs from different species.

BACKGROUND: Discovery of microRNAs (miRNAs) relies on predictive models for characteristic features from miRNA precursors (pre-miRNAs). The short length of miRNA genes and the lack of pronounced sequence features complicate this task. To accommodate the peculiarities of plant and animal miRNAs systems, tools for both systems have evolved differently. However, these tools are biased towards the species for which they were primarily developed and, consequently, their predictive performance on data sets from other species of the same kingdom might be lower. While these biases are intrinsic to the species, their characterization can lead to computational approaches capable of diminishing their negative effect on the accuracy of pre-miRNAs predictive models. We investigate in this study how 45 predictive models induced for data sets from 45 species, distributed in eight subphyla/classes, perform when applied to a species different from the species used in its induction. RESULTS: Our computational experiments show that the separability of pre-miRNAs and pseudo pre-miRNAs instances is species-dependent and no feature set performs well for all species, even within the same subphylum/class. Mitigating this species dependency, we show that an ensemble of classifiers reduced the classification errors for all 45 species. As the ensemble members were obtained using meaningful, and yet computationally viable feature sets, the ensembles also have a lower computational cost than individual classifiers that rely on energy stability parameters, which are of prohibitive computational cost in large scale applications. CONCLUSION: In this study, the combination of multiple pre-miRNAs feature sets and multiple learning biases enhanced the predictive accuracy of pre-miRNAs classifiers of 45 species. This is certainly a promising approach to be incorporated in miRNA discovery tools towards more accurate and less species-dependent tools. The material to reproduce the results from this paper can be downloaded from http://dx.doi.org/10.5281/zenodo.49754 .

Mirnacle: machine learning with SMOTE and random forest for improving selectivity in pre-miRNA ab initio prediction.

BACKGROUND: MicroRNAs (miRNAs) are key gene expression regulators in plants and animals. Therefore, miRNAs are involved in several biological processes, making the study of these molecules one of the most relevant topics of molecular biology nowadays. However, characterizing miRNAs in vivo is still a complex task. As a consequence, in silico methods have been developed to predict miRNA loci. A common ab initio strategy to find miRNAs in genomic data is to search for sequences that can fold into the typical hairpin structure of miRNA precursors (pre-miRNAs). The current ab initio approaches, however, have selectivity issues, i.e., a high number of false positives is reported, which can lead to laborious and costly attempts to provide biological validation. This study presents an extension of the ab initio method miRNAFold, with the aim of improving selectivity through machine learning techniques, namely, random forest combined with the SMOTE procedure that copes with imbalance datasets. RESULTS: By comparing our method, termed Mirnacle, with other important approaches in the literature, we demonstrate that Mirnacle substantially improves selectivity without compromising sensitivity. For the three datasets used in our experiments, our method achieved at least 97% of sensitivity and could deliver a two-fold, 20-fold, and 6-fold increase in selectivity, respectively, compared with the best results of current computational tools. CONCLUSIONS: The extension of miRNAFold by the introduction of machine learning techniques, significantly increases selectivity in pre-miRNA ab initio prediction, which optimally contributes to advanced studies on miRNAs, as the need of biological validations is diminished. Hopefully, new research, such as studies of severe diseases caused by miRNA malfunction, will benefit from the proposed computational tool.

MiR-577 inhibits pancreatic ß-cell function and survival by targeting fibroblast growth factor 21 (FGF-21) in pediatric diabetes.

Pancreatic ß-cell dysfunction is a central component of the pathogenesis of pediatric diabetes. MicroRNA (miRNA) have become one of the most encouraging and fruitful fields in biological research, and have been implicated as new players in the pathogenesis of diabetes and diabetes-associated complications. The role of miRNA in diabetes begins with the development of pancreatic islets. Fibroblast growth factor (FGF)-21 enhances glucose uptake in adipocytes, protecting transgenic animals from diet-induced obesity when overexpressed, and lowers blood glucose and triglyceride levels in diabetic animals (when administered); therefore, it is a good way to treat diabetes. However, the mechanism of miRNA in regulation of FGF21 is not known. In this study, FGF-21 was predicted to be the target of miR-577. Therefore, we investigated the effects of miR-577 on ß-cell function and survival by targeting FGF-21. We demonstrated that, although FGF-21 does not acutely stimulate insulin secretion in isolated islets from normal rats, it increases insulin secretion and insulin content in diabetic islets and protects ß-cells from apoptosis via the activation of extracellular signal-regulated kinase 1/2 and Akt signaling pathways.