Micro-RNA Feedback Loops Modulating the Calcineurin/NFAT Signaling Pathway

Nuclear factor of activated T cells (NFAT) is a family of transcription factors important for innate and adaptive immune responses. NFAT activation is tightly regulated through the calcineurin/NFAT signaling pathway. There is increasing evidence on non-coding RNAs such as miRNAs playing a crucial role in regulating transcription factors and signaling pathways. However, not much is known about microRNAs (miRNAs) targeting the calcineurin/NFAT signaling pathway involved in immune response in human. In this study, a comprehensive pathway level analysis has been carried out to identify miRNAs regulating the calcineurin/NFAT signaling pathway. Firstly, by incorporating experimental data and computational predictions, 191 unique miRNAs were identified to be targeting the calcineurin/NFAT signaling pathway in humans. Secondly, combining miRNA expression data from activated T cells and computational predictions, 32 miRNAs were observed to be induced by NFAT transcription factors. Finally, 11 miRNAs were identified to be involved in a feedback loop to modulate the calcineurin/NFAT signaling pathway activity. This data demonstrate the potential role of miRNAs as regulators of the calcineurin/NFAT signaling pathway. The present study thus emphasizes the importance of pathway level analysis to identify miRNAs and understands their role in modulating signaling pathways and transcription factor activity.


Introduction
Nuclear factor of activated T cells (NFAT) is a family of transcription factors that was first described as a part of a protein complex, which altered transcription of the interleukin-2 (IL-2) gene. The NFAT family consists of five members: NFATc1 (NFAT2), NFATc2 (NFAT1), NFATc3 (NFAT4), NFATc4 (NFAT3) and NFAT5, in which the first four NFATs are regulated by calcium signaling [1,2]. NFAT transcription factor activities are tightly regulated through the calcineurin/NFAT signaling pathway. This pathway is an essential player in immune responses including cytokines' production of TNF-α and IL-10 [1]. When T cells are activated, the intracellular calcium level increases and binds to a protein called calmodulin. The calcium-bound calmodulin interacts with calcineurin, which in turn dephosphorylates NFATs into its active form. Dephosphorylated NFATs translocate from the cytoplasm to the nucleus and induce NFAT-dependent transcription [2,3]. The calcineurin/NFAT signaling pathway is thus crucial for the adaptive immune response and T cell activation [4]. Together with its role in adaptive immunity, NFATs are also involved in innate immune responses against fungal and bacterial infection [5,6].
MicroRNAs (miRNAs) are short (18)(19)(20)(21)(22) nucleotides in length), endogenous, non-coding single-stranded RNAs that regulate expression of target mRNAs. miRNAs have emerged as one of the key regulators of gene expression [7]. Nearly one-third of the genes in the human genome are estimated to be regulated by the~2000 mature miRNAs so far identified [8]. miRNAs can activate or inhibit the function of a signaling pathway by translational inhibition of the component genes [9,10]. TarBase 7 [45,46], DIANA [47,48] and miRNAMap [49,50], was used to identify miRNAs targeting the selected genes regulating NFAT activation. Most of these algorithms incorporate experimental validation of miRNA expression data. Since these algorithms are based on different prediction methods, sequences and/or structural interaction between miRNA and mRNA, the prediction overlap of these algorithms remains very low. However, combined, prediction algorithms are shown to give better a prediction of the targets [51]. These algorithms incorporate features that include sequence composition, secondary structure and species conservation to predict miRNA targets. The miRNA targets predicted by all six algorithms were used for further analysis. The computational analysis predicted 4679 combinations of mRNA-miRNA interactions involving 1961 unique miRNAs targeting the calcineurin/NFAT signaling pathway (Supplementary data). Table 1. Members of the calcineurin/NFAT signaling pathway analyzed in this study. Twenty-three key members of the calcineurin/NFAT signaling pathway were selected based on literature survey.

NFATc1
Family of transcription factors that are expressed in the vertebrates. These transcription factors induce expression of cytokine genes important for immune response.
Calcium-dependent, calmodulin-stimulated protein phosphatase. It dephosphorylates the NFAT transcription factors and facilitates nuclear transport of NFAT.

HOMER2
Is a negative regulator of T cell activation and binds with NFAT to compete with calcineurin binding. [34] ITPR1 Intracellular channel that mediates calcium release from the endoplasmic reticulum. [38,56] MAP3K7 Selectively induces calcineurin-NFAT signaling through direct phosphorylation of RCAN1. [35] TAB1 Phosphorylated RCAN1 and inhibits RCAN1 from binding to calcineurin.
Works as a negative inhibitor of the NFAT by phosphorylation and mediates its cytoplasmic translocation. [57] LRRK2 Negative regulator of the transcription factor NFAT and is a component of a complex that includes the large non-coding RNA NRON (an NFAT repressor). [58]

CABIN1
Binds specifically to the activated form of calcineurin and inhibits calcineurin-mediated signal transduction and NFAT nuclear translocation. [36] AKAP5 Inhibits calcineurin-dependent dephosphorylation of NFAT [59] CALM1 Calmodulin is a calcium binding protein that activates the calcineurin. [37] STIM1 Ca2+ sensor proteins on the endoplasmic reticulum, which is an indispensable part in the activation of store-operated Ca2+ channels (SOC). [60] ORAI1 Calcium channel subunit that is activated by the calcium sensor STIM1 when calcium stores are depleted. [61] The miRNA targeting the calcineurin/NFAT signaling pathway was further refined by incorporating experimental data on miRNA target validation. Experimental validation data was collected from the miRTarBase database [28,62]. A subset of 230 miRNAs with experimentally validated data on targeting the calcineurin/NFAT signaling pathway was selected for subsequent analysis (Supplementary data). This miRNA subset was further refined by incorporating miRNA-expressed data from human immune cells (Supplementary data). Since the calcineurin/NFAT signaling pathway is known to be important for regulating the T cell development and immune response [4], miRNAs expressed in immune cells were collected from miRmineHuman miRNA expression (http://guanlab.ccmb.med.umich.edu/mirmine/index.html) database [63]. By incorporating this expression data, 191 miRNAs targeting the calcineurin/NFAT signaling pathway were identified (Supplementary data). The negative regulators of this pathway, Glycogen synthase kinase 3 beta (GSK3B), Dyrk1A, CSNK1A1 and RCAN1, were among the highly targeted genes, with 26, 14, 14 and 11 miRNA targets, respectively. The positive regulators of this pathway, PPP3R1, MAP3K7, TAB2, CALM1 and ITPR1, were also observed to be targeted by 19, 14, 14, 15 and 12 miRNAs, respectively (Supplementary data).

miRNAs Induced by NFAT Transcription Factor Family
miRNAs are known to be transcribed mainly in two different ways: (a) as part of an mRNA or (b) individually with a separate promoter [64]. The majority of the miRNAs are thought to form an independent transcription unit and the rest are transcribed as part of the annotated genes [65]. However, it is not clearly understood which miRNAs are transcribed as part of an mRNA or have independent transcription. Not many studies are done on transcription factors that transcribe these miRNAs. However, a number of computational prediction methods have been developed to predict transcription factors that can bind to the miRNA promoter region and potentially transcribe it [66]. There are ChIP-seq experimental data available on transcription factors inducing miRNA expression in ChIPBase (http://deepbase.sysu.edu.cn/chipbase/index.php) [67,68]. However, miRNAs induced by the NFAT family of transcription factors are not available from the above study.
In the current study, miRNAs induced by the NFAT transcription factors were predicted by combining computational predictions and published experimental data. To conduct this analysis, miRNA sequence coordinates were collected from the miRBase database [8]. Promoter regions of the miRNAs were obtained through the UCSC genome browser [69,70] and promoter sequences for 1443 human miRNAs were collected from UCSC. The NFAT binding sites on these promoter regions were predicted using transcription factor prediction algorithms TFSEARCH (http://diyhpl.us/~bryan/irc/protocol-online/protocol-cache/TFSEARCH.htmls) [71] and PROMO (http://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_8.3) [66,72]. Prediction results from these algorithms were combined to identify 536 miRNAs with an NFAT transcription factor binding site on the promoter region (Supplementary data).
This computational prediction of the miRNAs induced by NFAT transcription factors was further refined by incorporating published miRNA expression data in activated and non-activated T cells in humans [32]. In an activated T cell, NFAT translocate into the nucleus and starts transcribing genes under its control. Thus, it was hypothesized that the miRNAs induced by NFATs would get transcribed in activated T cells and would get upregulated in activated T cells compared to non-activated T cells. Based on this hypothesis, miRNA expression data for human T cells were further analyzed. Ninety-five miRNAs upregulated by 1.5-fold or more in active T cells were selected (Supplementary data). A subset of these miRNAs was observed to have an NFAT transcription factor binding site on their promoter region. Thirty-two miRNAs from this subset that were identified to have an NFAT binding region and to be upregulated in activated T cells were selected, because they are potentially induced by NFAT transcription factor ( Figure 1).  Figure 1. List of 32 miRNAs potentially induced by NFAT transcription factor with the fold change of expression in activated T cells. These miRNAs were observed to be highly overexpressed in activated T cells compared to non-activated T cells.

miRNAs in Negative or Positive Feedback Loop to Modulate the Calcineurin/NFAT Signaling Pathway Activity
Signaling pathways such as the calcineurin/NFAT pathway are tightly regulated through positive or negative feedback loops to reduce fluctuations in gene expression [61][62][63]. Previous studies have shown the role of miRNAs in such feedback loops [26]. While miRNAs operate through a repressive mechanism, their function in signaling pathways can result in activation or repression of the pathways. Not much is known about the miRNAs involved in a feedback loop to regulate the calcineurin/NFAT signaling pathway. An miRNA feedback loop can be of two types: a negative feedback loop, where miRNAs transcribed by NFAT transcription factor inhibit NFAT activation (translocation to nucleus) by repressing NFAT or NFAT activators (type 1), and a positive feedback loop, where miRNAs transcribed by NFAT transcription factors repress the expression of the NFAT inhibitor, which results in the activation of NFAT (type 2) ( Figure 2). In this study, by combining miRNAs targeting the calcineurin/NFAT signaling pathway and miRNAs induced by NFAT transcription factor, miRNAs involved in feedback loops to modulate the calcineurin/NFAT signaling pathway were identified. Computational predictions combined with experimentally validated data have identified 191 miRNAs targeting the calcineurin/NFAT signaling pathway and 32 miRNAs induced by NFAT transcription factors. By combining these two sets of miRNAs, a subset of 11 miRNAs (Table 2) were observed to be potentially involved in a feedback loop to regulate the calcineurin/NFAT signaling pathway. Out of the 11 miRNAs, six (hsa-miR-21-3p, hsa-let-7b-5p, hsa-miR-17-5p, hsa-miR-19a-3p, hsa-miR-92b-3p and hsa-miR-17-3p) are involved in a negative feedback loops that modulates NFAT activity down and five miRNAs (hsa-miR-21-5p,

miRNAs in Negative or Positive Feedback Loop to Modulate the Calcineurin/NFAT Signaling Pathway Activity
Signaling pathways such as the calcineurin/NFAT pathway are tightly regulated through positive or negative feedback loops to reduce fluctuations in gene expression [73][74][75]. Previous studies have shown the role of miRNAs in such feedback loops [26]. While miRNAs operate through a repressive mechanism, their function in signaling pathways can result in activation or repression of the pathways. Not much is known about the miRNAs involved in a feedback loop to regulate the calcineurin/NFAT signaling pathway. An miRNA feedback loop can be of two types: a negative feedback loop, where miRNAs transcribed by NFAT transcription factor inhibit NFAT activation (translocation to nucleus) by repressing NFAT or NFAT activators (type 1), and a positive feedback loop, where miRNAs transcribed by NFAT transcription factors repress the expression of the NFAT inhibitor, which results in the activation of NFAT (type 2) ( Figure 2).

miRNAs in Negative or Positive Feedback Loop to Modulate the Calcineurin/NFAT Signaling Pathway Activity
Signaling pathways such as the calcineurin/NFAT pathway are tightly regulated through positive or negative feedback loops to reduce fluctuations in gene expression [61][62][63]. Previous studies have shown the role of miRNAs in such feedback loops [26]. While miRNAs operate through a repressive mechanism, their function in signaling pathways can result in activation or repression of the pathways. Not much is known about the miRNAs involved in a feedback loop to regulate the calcineurin/NFAT signaling pathway. An miRNA feedback loop can be of two types: a negative feedback loop, where miRNAs transcribed by NFAT transcription factor inhibit NFAT activation (translocation to nucleus) by repressing NFAT or NFAT activators (type 1), and a positive feedback loop, where miRNAs transcribed by NFAT transcription factors repress the expression of the NFAT inhibitor, which results in the activation of NFAT (type 2) (Figure 2). In this study, by combining miRNAs targeting the calcineurin/NFAT signaling pathway and miRNAs induced by NFAT transcription factor, miRNAs involved in feedback loops to modulate the calcineurin/NFAT signaling pathway were identified. Computational predictions combined with experimentally validated data have identified 191 miRNAs targeting the calcineurin/NFAT signaling pathway and 32 miRNAs induced by NFAT transcription factors. By combining these two sets of miRNAs, a subset of 11 miRNAs (Table 2) were observed to be potentially involved in a feedback loop to regulate the calcineurin/NFAT signaling pathway. Out of the 11 miRNAs, six (hsa-miR-21-3p, hsa-let-7b-5p, hsa-miR-17-5p, hsa-miR-19a-3p, hsa-miR-92b-3p and hsa-miR-17-3p) are involved in a negative feedback loops that modulates NFAT activity down and five miRNAs (hsa-miR-21-5p, In this study, by combining miRNAs targeting the calcineurin/NFAT signaling pathway and miRNAs induced by NFAT transcription factor, miRNAs involved in feedback loops to modulate the calcineurin/NFAT signaling pathway were identified. Computational predictions combined with experimentally validated data have identified 191 miRNAs targeting the calcineurin/NFAT signaling pathway and 32 miRNAs induced by NFAT transcription factors. By combining these two sets of miRNAs, a subset of 11 miRNAs (Table 2) were observed to be potentially involved in a feedback loop to regulate the calcineurin/NFAT signaling pathway. Out of the 11 miRNAs, six (hsa-miR-21-3p, hsa-let-7b-5p, hsa-miR-17-5p, hsa-miR-19a-3p, hsa-miR-92b-3p and hsa-miR-17-3p) are involved in a negative feedback loops that modulates NFAT activity down and five miRNAs (hsa-miR-21-5p, hsa-miR-181c-5p, hsa-let-7c-5p, hsa-let-7b-3p and hsa-miR-155-5p) are involved in a positive feedback loops facilitating NFAT activation.

Discussion
The calcineurin/NFAT signaling pathway is an important immune regulatory pathway in vertebrates. This pathway regulates the activation of a family of NFAT transcription factors, which upon activation regulate the gene expression of important immune regulatory cytokines [1]. Over the past decade, the importance of miRNAs in the regulation of signaling pathways, such as the calcineurin/NFAT signaling pathway, has become more and more evident [76]. miRNAs through their repressive function can modulate diverse signal transduction pathways [9]. However, apart from a few miRNAs, not much is known about the miRNAs involved in regulating the calcineurin/NFAT signaling pathway [11,15,17]. In the present study, computational predictions, experimental validation and miRNA expression data were combined to identify miRNA feedback loops regulating the calcineurin/NFAT signaling pathway. The concept of feedback regulation of pathways has been known for over 100 years [77]. A variety of studies in the literature have identified positive or negative feedback loops in the calcineurin/NFAT signaling pathway [1,25,35,53]. However, miRNA feedback loops are much less studied, and currently no information is available on miRNA feedback loops regulating the calcineurin/NFAT signaling pathway. The present study describes 11 miRNAs (Table 2) potentially involved in feedback loops to regulate the calcineurin/NFAT signaling pathway. These miRNAs were systematically identified based on their involvement in targeting the calcineurin/NFAT pathway and on computational predictions to be a potential target of NFAT transcription factor. miRNAs observed to be involved in negative feedback loops, hsa-miR-21-3p, hsa-let-7b-5p, hsa-miR-17-5p, hsa-miR-19a-3p, hsa-miR-92b-3p and hsa-miR-17-3p, are targeting the NFAT activators, thus modulating down the NFAT activity. miRNAs observed to be part of the positive feedback loops, hsa-miR-21-5p, hsa-miR-181c-5p, hsa-let-7c-5p, hsa-let-7b-3p and hsa-miR-155-5p, target the negative regulators of the NFAT, thus facilitating NFAT activation (Figure 3). The let-7 miRNAs, identified in this study to be involved in both negative and positive feedback loops, are part of the highly conserved miRNA families across different species. The let-7 family of proteins is described to be expressed in immune cells (miRmine database) and is overexpressed in T cells upon activation [32]. Various experimental validation studies have shown that the let-7 family of miRNAs targets members of the calcineurin/NFAT signaling pathway (hsa-let-7b-3p, hsa-let-7c-5p targeting GSK3 and hsa-let-7b-5p targeting NFATc1) [66,67]. Let-7f, one member of the let-7 miRNA family, has previously been shown to positively modulate NF-κB signaling by targeting an inhibitor of the pathway, A20 [64]. Thus, upregulation of let-7f is shown to be beneficial for the immune cells to control infection [64]. The current study demonstrates the potential role of hsa-let-7b-3p and hsalet-7c-5p in a positive feedback loop by targeting the negative regulator of NFAT, GSK3. This data suggest, similar to the NF-κB signaling, the potential role of let-7 in activating the NFAT signaling pathway by targeting GSK3, which would be beneficial for immune response.
The next set of miRNAs observed to be involved in feedback loops, hsa-miR-17-3p, hsa-miR-17-5p, hsa-miR-19a-3p and hsa-miR-92b-3p, are part of the miR-17~92 cluster. In humans, the miR-17~92 cluster is composed of six members from four different seed families-miR-17, miR-18a, miR-19a, miR-20a, miR-19b-1, and miR-92-1. In mice, overexpression of the miR-17~92 cluster has been shown to be associated with an increased CD4+ T cell population and autoimmune disease [68]; however, similar data is not available for humans. Transcription factors c-Myc and E2F are shown to be associated with activation of the miR-17~92 cluster miRNAs. In addition, the miR-17~92 cluster is involved in a feedback loop regulating E2F family members [68]. The four members of the miR-17~92 cluster, hsa-miR-17-3p, hsa-miR-17-5p, hsa-miR-19a-3p and hsa-miR-92b-3p, from this study have been observed to be involved in a negative feedback loop to target the calcineurin/NFAT signaling pathway. Three of them are observed to be targeting the genes associated with calcium release (hsa-miR-17-3p targeting STIM1, hsa-miR-19a-3p and hsa-miR-92b-3p targeting ITPR1). This data suggests that members of the miR-17~92 cluster are potentially involved in negatively regulating the calcineurin/NFAT signaling pathway, potentially to reduce over-activation of the immune response. The let-7 miRNAs, identified in this study to be involved in both negative and positive feedback loops, are part of the highly conserved miRNA families across different species. The let-7 family of proteins is described to be expressed in immune cells (miRmine database) and is overexpressed in T cells upon activation [32]. Various experimental validation studies have shown that the let-7 family of miRNAs targets members of the calcineurin/NFAT signaling pathway (hsa-let-7b-3p, hsa-let-7c-5p targeting GSK3 and hsa-let-7b-5p targeting NFATc1) [78,79]. Let-7f, one member of the let-7 miRNA family, has previously been shown to positively modulate NF-κB signaling by targeting an inhibitor of the pathway, A20 [76]. Thus, upregulation of let-7f is shown to be beneficial for the immune cells to control infection [76]. The current study demonstrates the potential role of hsa-let-7b-3p and hsa-let-7c-5p in a positive feedback loop by targeting the negative regulator of NFAT, GSK3. This data suggest, similar to the NF-κB signaling, the potential role of let-7 in activating the NFAT signaling pathway by targeting GSK3, which would be beneficial for immune response.
The next set of miRNAs observed to be involved in feedback loops, hsa-miR-17-3p, hsa-miR-17-5p, hsa-miR-19a-3p and hsa-miR-92b-3p, are part of the miR-17~92 cluster. In humans, the miR-17~92 cluster is composed of six members from four different seed families-miR-17, miR-18a, miR-19a, miR-20a, miR-19b-1, and miR-92-1. In mice, overexpression of the miR-17~92 cluster has been shown to be associated with an increased CD4+ T cell population and autoimmune disease [80]; however, similar data is not available for humans. Transcription factors c-Myc and E2F are shown to be associated with activation of the miR-17~92 cluster miRNAs. In addition, the miR-17~92 cluster is involved in a feedback loop regulating E2F family members [80]. The four members of the miR-17~92 cluster, hsa-miR-17-3p, hsa-miR-17-5p, hsa-miR-19a-3p and hsa-miR-92b-3p, from this study have been observed to be involved in a negative feedback loop to target the calcineurin/NFAT signaling pathway. Three of them are observed to be targeting the genes associated with calcium release (hsa-miR-17-3p targeting STIM1, hsa-miR-19a-3p and hsa-miR-92b-3p targeting ITPR1). This data suggests that members of the miR-17~92 cluster are potentially involved in negatively regulating the calcineurin/NFAT signaling pathway, potentially to reduce over-activation of the immune response.
Bazzoni et al. have demonstrated the miRNA-mRNA interaction between hsa-miR-155-5p and GSK3, and explored the potential involvement of hsa-miR-155 in the regulation of innate immune response [81]. There are several studies exploring the role of hsa-miR-155 in regulating immune responses, and in mice miR-155 has been shown to be crucial for T cell differentiation and proliferation [80]. In humans, miR-155 expression was shown to be associated with transcription factor NF-κB; however, in mice, miR-155 was shown to be activated via the NFAT pathway [81,82]. However, no study has yet explored the role of NFAT transcription factor as an inducer of hsa-miR-155 expression in human immune cells. In this study, the transcription factor binding region analysis identified a potential NFAT binding site on the hsa-mir-155 promoter region and also demonstrated the role of hsa-miR-155 in positive feedback loops. Thus, similar to let-7, by targeting GSK3 hsa-miR-155 potentially activates the calcineurin/NFAT pathway, which is important for T cell proliferation.
Similar to miR-155, hsa-miR-21 has also been shown to be a major regulator of Th1 versus Th2 T cell response [83]. In this study, hsa-miR-21-5p has been shown to target CK1 (CSNK1A1) and to be involved in a positive feedback loop activating the calcineurin/NFAT signaling pathway. A microarray study done by Gabriely et al. has validated the miRNA-mRNA interaction between hsa-miR-21-5p and CK1 (CSNK1A1) [84]. Being involved in a positive feedback loop and by targeting CK1, miR-21 might be contributing to T cell proliferation through activating the NFAT transcription factor. Interestingly, hsa-miR-21-3p has been observed to be targeting the CALM1 gene and to be involved in a negative feedback loop. It is known that miRNAs from the same pre-miRNA could have different functional properties depending on the differences in the seed sequence or nucleotides at the 5 1 and 3 1 ends [85]. The second miRNA target of CSNK1A1, hsa-miR-181c, however, has not been studied in detail for its function in T cell activation and immune responses. hsa-miR-181c is part of the miR-181 family, consisting of miR-181a, miR-181b and miR-181c. Kishore et al., using the cross-linking and immunoprecipitation (CLIP) method, have validated the interaction between CSNK1A1 and hsa-miR-181c [86]. In the present study, hsa-miR-181c has been described to be potentially involved in a positive feedback loop to activate the calcineurin/NFAT pathway by silencing the negative of the pathway, CSNK1A1. These data thus suggest a potential role of hsa-miR-181c in silencing CSNK1A1, which would activate the NFAT pathway and immune response.
Taken together, feedback loops involving miRNAs demonstrate the crucial role of miRNAs in modulating the calcineurin/NFAT signaling pathway. Additional comprehensive pathway level studies including experimental validations need to be carried out to identify these miRNAs and understand their role in signaling pathways.  [46] were used for computational prediction of miRNAs targeting 23 members of the calcineurin/NFAT pathway. The algorithms were used using their default parameters and miRNAs predicted by at least one of the prediction algorithms were selected for subsequent analysis. The computational prediction identified 4679 combinations of mRNA-miRNA interactions with 1961 unique miRNAs targeting the calcineurin/NFAT signaling pathway (Supplementary data).

Experimental Validation Data on miRNA Targets
miRNA target validation data was collected from the database miRTarBase (http://mirtarbase. mbc.nctu.edu.tw/index.php) [62]. List of all miRNAs targeting the 23 members of the calcineurin/NFAT pathway were collected. Experimental validation data on 840 miRNA-mRNA interactions were collected from miRTarBase (Supplementary data).

Upregulated miRNA Selection from Activated Human T Cells
Expression profile of miRNAs from activated or non-activated T cells for human was collected from previously published datasets [32]. miRNAs were sorted based on their expression fold change in activated T cells compared to the non-activated cells. Seventy-three miRNAs with fold change >1.5 were selected for further analysis (Supplementary data).

Conclusions
A combination of computational prediction algorithms and experimental data available on miRNA expression and target predictions was used to identify possible miRNA feedback loops modulating the calcineurin/NFAT signaling pathway in humans. The feedback loop mechanism described in this study highlights the central role of miRNAs in regulating NFAT activity. Further studies need to be undertaken, especially experimental validation, which will lead to novel insights into the nature of miRNA modulation of the calcineurin/NFAT signaling pathway. Overall, this study demonstrates the importance of a comprehensive systems-level analysis in identifying the indirect role of miRNAs in activating or silencing pathway response, such as that of the calcineurin/NFAT signaling pathway.

Acknowledgments:
The author sincerely thanks Alberto Polleri, Katalin Takacs and Ella Palmer for their ideas, comments and support, and MRC Clinical Science Centre for the financial support.
Author Contributions: S.K. conceived and designed the study, analyzed the data and wrote the paper.

Conflicts of Interest:
The author declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

NFAT
Nuclear factor of activated T cells PPP3CA Calmodulin