InteractiveFly: GeneBrief

mir-9a stem loop: Biological Overview | References

BIOLOGICAL OVERVIEW

Transcription factors and microRNAs (miRNAs) are two important classes of trans-regulators in differential gene expression. Transcription factors occupy cis-regulatory motifs in DNA to activate or repress gene transcription, whereas miRNAs specifically pair with seed sites in target mRNAs to trigger mRNA decay or inhibit translation. Dynamic spatiotemporal expression patterns of transcription factors and miRNAs during development point to their stage- and tissue-specific functions. Recent studies have focused on miRNA functions during development; however, much remains to explore regarding how the expression of miRNAs is initiated and how dynamic miRNA expression patterns are achieved by transcriptional regulatory networks at different developmental stages. This study has focused on the identification, regulation and function of miRNAs during the earliest stage of Drosophila development, when the maternal-to-zygotic transition (MZT) takes place. Eleven miRNA clusters comprise the first set of miRNAs activated in the blastoderm embryo. The transcriptional activator Zelda is required for their proper activation and regulation, and Zelda binding observed in genome-wide binding profiles is predictive of enhancer activity. In addition, other blastoderm transcription factors, comprising both activators and repressors, the activities of which are potentiated and coordinated by Zelda, contribute to the accurate temporal and spatial expression of these miRNAs, which are known to function in diverse developmental processes. Although previous genetic studies showed no early phenotypes upon loss of individual miRNAs, this analysis of the mir-1; miR-9a double mutant revealed defects in gastrulation, demonstrating the importance of co-activation of miRNAs by Zelda during the MZT (Fu, 2014).

Similar to protein-coding genes, miRNA genes are regulated by sophisticated spatial and temporal signals to ensure their proper production in specific cell types. The muscle-specific transcription factors Twist (Twi) and Mef2 are key activators of mir-1 in Drosophila. Genomic studies have also identified regulators of miRNAs, such as Dorsal (Dl), c-Myc (Diminutive -- FlyBase) and Ecdysone. However, for many miRNAs, particularly those differentially expressed across developmental stages, the regulatory networks that control their transcription remain unknown. This study examined the gene network that regulates miRNA functions during the maternal-to-zygotic transition (MZT), a time when developmental control is transferred from maternal products preloaded into the egg to the embryo's own genome, which in Drosophila is activated ~1 hour after fertilization. During the MZT, thousands of maternal RNAs are degraded and hundreds of newly synthesized RNAs appear; thus, the MZT represents a major reprogramming event of the early transcriptome. Previous studies have reported that the zinc-finger transcription factor Zelda (Vielfaltig -- FlyBase) plays a key role during the MZT in Drosophila, collectively activating batteries of genes involved in early developmental processes, such as sex determination, cellularization and axis patterning. Interestingly, Zelda also activates the miR-309 cluster of eight miRNAs, which is involved in the clearance of many maternally loaded mRNAs (Bushati, 2008). Since Zelda plays such an extensive role in zygotic genome activation, possibly as a pioneer factor to prime genes for transcriptional activation, this study investigated the possibility that Zelda activates the miRNAs expressed during the MZT (Fu, 2014).

This study identified a group of miRNAs (11 clusters) that are zygotically expressed in cellular blastoderm embryos; Zelda was shown to regulate all 11. The enhancers of several miRNAs were localized by virtue of Zelda ChIP binding; Zelda binding sites, also known as CAGGTAG sites or TAGteam sites, in these enhancers were shown to be essential for proper activation. It was further shown that anteroposterior (AP) and dorsoventral (DV) patterning factors work together with Zelda to ensure timely and robust transcriptional activation of these miRNAs, contributing to their accurate spatial expression patterns. The reduced and disrupted miRNA expression seen in Zelda mutants affects their downstream functions in maternal mRNA degradation, cell death gene repression and Hox gene regulation. Ventral midline defects were observed during gastrulation in mir-1; miR-9a double mutants, that were not seen in either single mutant, suggesting that the coordinated activation of miRNAs by Zelda is crucial for their combinatorial function. This analysis offers a systems-level view and understanding of the early gene network. Zelda sits as a major hub in the network, globally activating both protein-coding and non-coding genes, thereby orchestrating the early developmental processes (Fu, 2014).

This study identified the set of miRNAs expressed in 2- to 3-h Drosophila embryos, a time when the MZT is well underway and the fate map of the embryo is being established. These early expressed miRNAs are regulated globally by Zelda, both directly via binding to cis-regulatory enhancers and indirectly by affecting the expression of additional transcriptional regulators. Together with previously published data on specific miRNAs, it was possible to integrate the early miRNAs, their upstream regulators and downstream targets into the early gene network (Fu, 2014).

Using expression profiling data from 2- to 3-h wild-type and zelda mutant embryos, blastoderm-specific pri-miRNA transcription units, which included seven intergenic and four intronic miRNAs (clusters) were identified. Since the expression levels of all 11 miRNAs were affected in Zelda mutants, it was possible to better distinguish blastoderm-specific isoforms, particularly in the case of intronic miRNAs, such asmir-11, which resides in an intron of E2f . Moreover, maternal E2f expression could be differentiated from zygotic expression by observing the intronic signal, which was clearly downregulated in zelda mutants (Fu, 2014).

The early miRNAs exhibit strikingly different expression patterns, and it is noteworthy that, similar to the protein-coding targets of Zelda, two different strategies are used to regulate these miRNAs. Some miRNAs, such as miR-9a, were completely abolished in Zelda mutants, indicating that Zelda is their sole activator, whereas others, such as mir-1, were affected temporally and/or spatially, indicating that Zelda works together with other factors to establish robust and precise domains of expression. For example, mir-1 downregulation in Zelda mutants is likely to be due to the cumulative effect of loss of direct inputs from Zelda and the delayed expression of twi that occurs in Zelda mutants. Thus, the effect onmir-1 is the result of a breakdown in the Zelda-Twi-mir-1 feedforward loop (Fu, 2014).

Cis-regulatory modules/enhancers of miRNAs have been predicted based on the presence of transcription factor binding or specific chromatin marks, and verified in only some cases. For example, two regions upstream of mir-1 that bind Twi/Dl were shown to drive a mir-1-like expression pattern. It was reasoned that it could be possible to locate enhancers of all early miRNAs by simply looking for Zelda-bound regions upstream of the pri-miRNA transcription units, especially since Zelda is a global activator during the MZT. This approach worked well; eight of nine enhancers recapitulated endogenous-like expression. Mutation of Zelda binding sites in enhancers further demonstrated direct Zelda input. As proof of principle, enhancers of two genes, miR-9a and mir-1, were analyzed and it was shown that mutation of the Zelda binding sites had the same effect as eliminating Zelda in trans. These results indicate that Zelda directly regulates the early expressed miRNAs, often in conjunction with other transcription factors, many of which are also regulated by Zelda. Zelda is a major hub in the early network, establishing multiple feedforward loops and closely linking the transcription factors and miRNAs expressed in this stage (Fu, 2014).

The MZT is a key event during the development of an organism, whereby the transcriptome is reprogrammed in the first few hours of development. This requires the clearance of previous information (maternal mRNA degradation) and the initiation of a new program (zygotic genome activation). The maternal mRNA degradation machinery comprises both maternally derived and zygotically derived pathways. In Drosophila, Smaug (Smg), a maternally loaded RNA-binding protein, is central to the mRNA clearance pathway. By recruiting the CCR4-NOT deadenylation complex, Smg destabilizes two-thirds of the maternal mRNAs that undergo degradation (i.e. that are unstable) upon egg activation (Tadros, 2007). By contrast, miR-309 is a key component of the zygotically derived pathway to clear mRNAs (Bushati, 2008). When analyzing the maternal RNAs upregulated in zelda mutants, it was noted in this study that 81% of them (434) depend on zygotic degradation pathways; 125 of the 434 genes are also upregulated in miR-309 mutants (Bushati, 2008), indicating that Zelda, by activating miR-309, is involved in maternal RNA degradation. Therefore, Zelda plays important roles in both of the hallmark events of the MZT. Interestingly, the miR-309 targets account for only ~30% of the unstable maternal RNAs upregulated in Zelda mutants, and another 14% are putative targets of the other early miRNAs, indicating that Zelda might activate additional zygotic pathways to mediate maternal mRNA degradation (Fu, 2014).

Several miRNAs, in addition to miR-309, have been shown to target specific mRNAs in the early embryo; for example, miR-iab-4 and miR-iab-4as target Hox genes. However, although each of the miRNAs is predicted to target hundreds of genes, in many cases the individual miRNA loss-of-function phenotypes are relatively mild. There are several explanations for this phenomenon: (1) the miRNA does not function at the time that it is expressed, but might function later; (2) miRNAs 'fine-tune' the expression levels of their target genes, which might not be reflected in obvious phenotypes when they are mutated; and (3) miRNA functions are redundant, such that knockdown of one miRNA may be compensated by another (Fu, 2014).

To better address the functions of miRNAs, investigators have used several genetic approaches: gain-of-function assays, using sensitized genetic backgrounds, and assaying double mutants. For example, the miR-6; mir-11 double mutant exhibits increased apoptosis, leading to lower survival rates compared with either of the single miRNA mutants. Using a similar approach, this study observed fully penetrant gastrulation defects in mir-1; miR-9a double mutants. Neither single mutant is embryonic lethal, nor shows any sign of ventral furrow defects; however, mir-1 mutants are larval lethal and display muscle defects, while miR-9a mutants show wing margin defects in adulthood. Importantly, the double-mutant phenotype is the earliest phenotype seen for any known miRNA, or combination of miRNAs, thus far tested. These results support the idea that co-activation of miRNAs by Zelda is required for normal development (Fu, 2014).

The mir-1; miR-9a double-mutant phenotype resembles, to some extent, the ventral furrow defects observed in RhoGEF2 loss-of-function mutant. Rho signaling is involved in the cell shape changes associated with ventral furrow invagination, and loss of Rho signaling results in very disorganized invagination. Curiously, mir-1 and miR-9a are both predicted to target RhoGAP68F, a negative regulator of Rho signaling. RhoGAP68F is maternally loaded and cleared during the MZT. It is possible that the gastrulation phenotype of the mir-1; miR-9a double mutant is caused in part by excess activity of RhoGAP68F. Although no obvious upregulation of RhoGAP68F transcripts were seen in mir-1; miR-9a mutant embryos by in situ hybridization, it is possible that subtle upregulation of RhoGAP68F, combined with effects on other predicted targets, all contribute to the gastrulation defects observed in the double mutant (Fu, 2014).

The co-activation of groups of miRNAs by master regulators such as Zelda may be crucial for miRNA activity during development, as revealed by the severe gastrulation phenotype of the mir-1; miR-9a double mutant. Such coordinated activation of miRNAs might also occur at later stages in development in tissues in which Zelda is expressed, such as the central nervous system. In the future, various combinations of mutations in miRNA genes that are co-regulated by Zelda, or other key factors, might unveil additional functions of miRNAs across development stages (Fu, 2014).

Drosophila miR-9a targets the ECM receptor Dystroglycan to canalize myotendinous junction formation

Establishment of intercellular interactions between various cell types of different origin is vital for organism development and tissue maintenance. Therefore, precise timing, expression pattern, and amounts of extracellular matrix (ECM) proteins must be tightly regulated. Particularly, the ECM is important for the development and function of myotendinous junctions (MTJs). This study finds that precise levels of the ECM receptor Dystroglycan (Dg) are required for MTJ formation in Drosophila and that Dg levels in this process are controlled by miR-9a. In the embryo, Dg is enriched at the termini of the growing muscles facing the tendon matrix and absent from miR-9a-expressing tendons. This gradient of Dg expression is crucial for proper muscle-tendon attachments and is adjusted by miR-9a. In addition to Dg, miR-9a regulates the expression of several other critical muscle genes, and it is therefore proposed that during embryogenesis, miR-9a specifically controls the expression of mesodermal genes to canalize MTJ morphogenesis (Yatsenko, 2014).

To achieve successful results, despite the extreme fluctuation of internal cues, genetic background, and external conditions, embryonic development must be stabilized. Coordinated transcription factor networks are prominent regulatory features of cell fate establishment during embryonic development and adult life. It is now becoming evident that, in conjunction with transcription factors, at least three epigenetic elements help to form a reciprocal regulatory circuit to maintain cell identity and differentiation: chromatin structure, DNA methylation, and microRNAs (miRNAs). miRNAs, based on their paradoxical properties, e.g., being highly evolutionarily conserved, but not essential, have been proposed to play a role in generating biological robustness as canalization factors to buffer gene expression against perturbation or variability. As canalization factors, miRNAs have previously been shown to liquidate transcripts resulting from aberrant gene expression or leaky splicing The previously described in vivo cases of miRNA-based regulation mostly are examples of simple pairs, in which one miRNA is targeting one gene. However, increasing evidence suggests that functionally related genes are clustered at the level of DNA sequence, histone modifications, chromatin loops, or chromosome territories and are under similar transcriptional control. Taking into account that, first, the gene expression in general is a noisy process that incidentally allows leaky expression of 'neighboring' genes and, second, that one miRNA can regulate multiple genes, it is logical to propose that as a canalization factor one miRNA should be capable of regulation of multiple genes that are involved in the same signaling network. Therefore, this study investigated whether this type of miRNA-based regulation, employed to confer robustness of embryonic development, actually takes place (Yatsenko, 2014).

Assembly of muscle tissue requires communication between mesoderm-derived myotubes and ectoderm-originated epidermal muscle-attachment cells or tendons. Since tendon cells invaginate into mesoderm, some mechanism that reassures the robustness of their identity must exist. Initially, the pretendon cells send signals to the myotubes and direct myotube attraction and adhesion to their target cells; subsequently, the muscle cells communicate a reciprocal signal to the epidermal muscle attachment cells, initiating their terminal differentiation into tendon-like cells. This suggests the necessity of a microenvironment that will allow for both a rapid and precise signal transduction between these ectodermally and mesodermally derived cell types (Yatsenko, 2014).

Importantly, the process of muscle guidance and attachment in Drosophila is remarkably similar to that of vertebrates, as both are greatly dependent on the extracellular matrix (ECM) gradient that is established through differential recruitment and clustering of transmembrane receptors by extracellular-presented signaling molecules. During Drosophila embryonic development, the initial determination of myoblast fate is controlled by high expression of the basic-helix-loop-helix protein Twist; after the myoblast division and fusion, multinucleated myotubes are formed. At stage 12-14, myotubes undergo a substantial transformation: not only do they continue to grow through cell fusion, but they also change their shape and form elongated filopodia at the leading edge that help to find their proper tendon cells in the epidermis. At the same time, the tendon cells also undergo a series of cell shape rearrangements, including apical constriction and apical-basal elongation, which results in the formation of epidermal furrows. When myotubes reach their targets, the surface of the myotube facing the tendon cells loses filopodia and multiple adhesion complex molecules accumulate at the muscle attachment site toward the tendon cell in order to form a stable adhesion complex. While the signaling crosstalk between these cell types has been extensively studied, it is not clear whether a genetic program exists that would aid cells that are subjected to similar spatiotemporal signaling to undergo distinct developmental programs. The role of miRNAs in this process, vital for muscle physiology, has not yet been analyzed; however, vigorous cell rearrangements and cell fate specifications that take place during establishment of the muscle attachment suggest a need for a mechanism that enhances robustness of the process by attenuating leaky transcripts (Yatsenko, 2014).

This study found that Drosophila miR-9a is involved in canalization of myotendinous junction (MTJ) assembly. Deficiency of miR-9a affects embryonic survival, a phenotype that can be rescued by specific expression of this miRNA in tendon cells. The survival of miR-9a mutants depends on the speed of embryonic development that reciprocally correlates with transcriptional noise. miR-9a is expressed in epidermally derived tendon cells, while many miR-9a predicted targets are essential muscle genes that are misregulated due to miR-9a loss and gain of function. Moreover, exogenous expression of miR-9a in mesoderm completely abolishes muscle formation. Therefore, it was hypothesized that miR-9a adjusts tendon cell differentiation by preventing misexpression of muscle genes resulting from stress or aberrant transcription. To prove this hypothesis, putative miR-9a targets were expressed in tendon cells, and ectopic heartless (htl), wishful thinking (wit), and Dystroglycan (Dg) in tendons was found to cause muscle attachment and embryonic lethality phenotypes similar to those found in miR-9a mutants. In particular, it was found that the muscular-dystrophy-associated ECM receptor, Dg, is regulated posttranscriptionally via the miRNA, miR-9a. During the early embryonic stages, Dg is present in all epidermal cells; however, for proper assembly of muscle attachment sites it is essential that Dg is eliminated from epidermally derived tendon cells, with miR-9a modulating the precision of this expression. Dg establishes a specific ECM gradient that influences muscle-tendon signaling; therefore, its differential localization is crucial for proper muscle-tendon attachments and is adjusted by miR-9a. When Dg is misexpressed in tendon cells, the composition of the tendon matrix is affected, resulting in aberrant muscle attachments and embryonic death (Yatsenko, 2014).

This study shows that the muscular-dystrophy-associated ECM receptor Dg can be posttranscriptionally regulated by miR-9a. During embryogenesis, both miR-9a and Dg have dynamic expression patterns that become mutually exclusive in the regions of muscle-tendon connections. Dg protein is present in all ectodermal cells, except for the ones that are differentiating into epithelial tendon cells and are also expressing miR-9a. The data show that the elimination of Dg from tendon precursor cells is required for accurate muscle-tendon matrix assembly. The miR-9a ensures that Dg is not misexpressed in tendon precursors due to leaky transcription, as these epidermal cells invaginate into and reside within the mesoderm (Yatsenko, 2014).

Embryonic development is an extremely dynamic process in which rapid cell specifications and rearrangements take place, features indicative of the need for stabilization. miRNAs have been implicated in stabilization of biological robustness in different animal systems. miRNAs are involved in the stabilization of the process of muscle-tendon attachment in the developing Drosophila embryo. In particular, the data imply that miR-9a acts as a backup mechanism in tendons to diminish the effects of leaky expression of a group of muscle genes. When two adjacent cells have different cell fates, evolutionarily it would make a lot of sense for a canalization factor in one cell type to regulate multiple genes critical for the differentiation of the other cell type. Apparently, many essential muscle differentiation genes are miR-9a predicted targets, and their ectopic expression in ectodermal tendon cells causes embryonic lethality and abnormal MTJs. Moreover, exogenous expression of miR-9a in the mesoderm completely abolishes muscle formation. With this in mind, it was hypothesized that miR-9a specifically acts as a guardian to prevent aberrant muscle gene expression in the epidermal tendon precursor cells (Yatsenko, 2014).

It has already been shown that mir-124 and miR-9a (Bejarano, 2010, Biryukova, 2009, Li, 2006 and Li, 2013) act to canalize nonneuronal versus neuronal fates. Genes expressed in the nervous system are highly enriched for miR-9a binding sites , and the current data show that multiple muscle genes also have miR-9a target sites. Since miR-9a is ectoderm specific and genes expressed in ectodermal tissues avoid miR-9a sites, the findings insinuate that miR-9a can act as the key ectodermal canalization factor that protects ectodermal cell fate by repressing genes of the sibling tissues (such as muscle and nervous). This should reinforce the robustness of ectodermal cell differentiation. It would be interesting to find in the future if miRNAs that canalize mesodermal or endodermal fate exist and to address the question of whether there are more examples of the miRNAs that target multiple genes from the same functional network (Yatsenko, 2014).

One of the muscle genes that this study showed is a bona fide miR-9a target is the ECM receptor, Dg. The transmembrane protein Dg has a distinct expression pattern at the MTJ. It is present at the membrane of the developing muscle and is enriched at the myotube ends; however, it is absent from the tendon cell membranes. Similarly, a restricted expression pattern of Dg is required for neuromuscular junction (NMJ) establishment in vertebrates (Xenopus), with Dg being present at the entire muscle membrane and showing enrichment at the NMJ site, where it acts as a sink for the ECM component agrin, preventing its binding to muscle-specific kinase (MuSK). At the site of nerve contact, in the absence of Dg, agrin can bind to MuSK, allowing acetylcholine receptor aggregation and synaptic development Thus, due to the distinct expression patterns in cells that form connections via the ECM, Dg is able to establish the ECM gradient, which is also essential for proper formation of MTJs in the developing Drosophila embryo. It would be interesting to investigate if there is a regulatory molecule that is differentially distributed between muscle and tendon due to specific binding to Dg at the MTJ (Yatsenko, 2014).

The extracellular environment of the cell is a complex organization of ECM receptors, matrix proteins, and the regulatory molecules that reside in it. Also, it continuously changes during development and allows rapid communication between different cells to coordinate tissue formation. Therefore, changes in the composition of the ECM can have a profound effect on an organism's development. This study shows that miR-9a-based regulation of Dg is needed to adjust the ECM composition at the MTJ. Regulation of the affinity of the transmembrane adhesion receptor integrins has a key role during development as it generates strong adhesion of cells to the insoluble ECM. This study shows that, at the MTJ, Dg also acts as a receptor regulating ECM gradient at the tendon matrix, since Dg levels affect the amount of the ECM protein Lan. In addition, Dg can modulate expression of a key ECM receptor, βPS Integrin. These data are consistent with previous findings revealing a regulatory pathway between the DGC and integrin receptors and lends the idea that Dg is involved in selective regulation of integrin gene expression Moreover, integrin overexpression alleviates the development of muscular dystrophy phenotypes in mdx mice, supporting the possibility that Dg and integrin compensate for each other in mediating cell-ECM adhesion. Additionally, this study showed that this regulation can be cell nonautonomous, since abnormal Dg levels, through modification of Lan amounts, affect integrin expression in the neighboring cells. In intestinal epithelial cells, the DGC coprecipitates with β1-integrin, suggesting a possible direct interaction among these proteins where the strength of this interaction depends on the Lan type. Moreover, it has been shown that increased Lan expression can ameliorate muscular dystrophy in mice. These results, on the one hand, support the findings that alterations in Lan levels influence the expression of the ECM receptors but, on the other hand, pose an interesting question of why, depending on the animal's genetic background (dystrophic or not), the increased levels of Lan have positive or negative effects on MTJs and muscles. The beneficial influence of Lan was seen so far only when it was upregulated in dystrophic animals: muscular dystrophy mdx mouse (dystrophin deficient), dy^w-/- merosin-deficient congenital muscular dystrophy mouse model (Lan deficient), and muscular dystrophy zebrafish model (Dg reduced). All above muscular-dystrophy-related components per se are required for accurate Lan localization and distribution, suggesting that restoration of Lan levels in the ECM has favorable effects on dystrophic muscle maintenance. Previous studies did not address what would happen if Lan were overexpressed in the otherwise normal background or they were differentially upregulated in muscle or tendons. Since elegant studies on the role of Laminin-111 (Lan-111) in muscle development and maintenance (Van Ry, 2014) propose Lan protein therapy as a treatment option for muscular dystrophy patients, it would be important in the future to study, using different models, the effects of differential Lan expression on muscle-tendon attachments during development and to determine the levels that can be tolerated without induction of deleterious effects on muscle maintenance and tendon attachments during adulthood (Yatsenko, 2014).

Taken together, previous studies and the current findings show that the amounts and types of the ECM receptors affect ECM constitution and govern its remodeling, and then via 'dynamic reciprocity' the ECM readjusts intracellular signaling, gene expression, and morphology of the cells and tissues. The crosstalk between tendons and muscles depends on differentially expressed ECM receptor Dg that, together with integrins, helps to establish the ECM gradient. The information about the tendon matrix composition is communicated back to the muscles and tendon cells that readjust their ECM receptor expression profiles in order to reinforce and stabilize the MTJ. Providing a link between the ECM and cytoskeleton, Dg acts as a vital signal-transducing element that allows for communication between the cell's outer environment and inner milieu. In vertebrates, Dg is implicated in multiple biological processes: for example, formation of spatiotemporally regulated microenvironments necessary for muscle fiber morphogenesis at the MTJ. In Drosophila, in addition to its function in muscle maintenance, Dg is involved in control of neuron behavior; modulation of the concentration of postsynaptic and synaptic proteins, in particular the ECM component Lan at the neuromuscular junction, and regulation of miRNA expression profiles. The currently increasing amount of research on the diverse roles of Dg during development demonstrates its critical role in multiple developmental circuits, suggesting that there is a necessity for precise and dynamic regulation of Dg levels. Despite the vast data about posttranslational regulation of Dg activity, Dg posttranscriptional regulation has not been studied. The current data show that Dg can be regulated by miRNAs and this regulation has an important functional role at the MTJ. Since the human homolog of Drosophila Dg (Dag1) also contains multiple predicted miRNA biding sites, it would be important to study if miRNAs also play a role in regulation of Dg in mammals. Even though there are numerous studies in vertebrate models indicating that MTJ assembly affects muscle development, the role of aberrant MTJ in muscle maintenance and function in muscular dystrophy patients is greatly underappreciated; therefore, understanding of the miRNA-based mechanisms controlling the ECM assembly at the MTJ may suggest new directions for muscular dystrophy research (Yatsenko, 2014).

miR-9a minimizes the phenotypic impact of genomic diversity by buffering a transcription factor

Gene expression has to withstand stochastic, environmental, and genomic perturbations. For example, in the latter case, 0.5%-1% of the human genome is typically variable between any two unrelated individuals. Such diversity might create problematic variability in the activity of gene regulatory networks and, ultimately, in cell behaviors. Using multigenerational selection experiments, this study finds that for the Drosophila proneural network, the effect of genomic diversity is dampened by miR-9a-mediated regulation of senseless expression. Reducing miR-9a regulation of the Senseless transcription factor frees the genomic landscape to exert greater phenotypic influence. Whole-genome sequencing identified genomic loci that potentially exert such effects. A larger set of sequence variants, including variants within proneural network genes, exhibits these characteristics when miR-9a concentration is reduced. These findings reveal that microRNA-target interactions may be a key mechanism by which the impact of genomic diversity on cell behavior is dampened (Cassidy, 2013).

An outstanding question is whether cell behaviors are actively made insensitive to the immense genomic diversity that exists between individuals. This study presents one mechanism by which this may be achieved, and this mechanism was experimentally validated in the context of a natural biological system. Regulation of the transcription factor Sens by miR-9a renders cell fate less sensitive to the varied genomic landscapes of individuals. This effect can be explained by the ability of miR-9a to create a threshold response. Because miR-9a creates a threshold concentration of sens transcript that must be crossed before cell fate is switched, genomic variants that perturb sens transcription are rendered ineffective if the threshold is not passed. If miR-9a is less effective at attenuating sens, then these same variants would be more likely to trigger the cell fate switch. Several variants were found to affect genes that regulate sens transcription, and these variants correlate with a greater impact on the cell fate switch. The variants that affect these genes are not likely sufficient to perturb the network because a constellation of other loci also showed signs of influence. Each of these might have weak and probabilistic effects on sens transcription, which in combination can potentially perturb the network. The impact of such probabilistic fluctuations would be suppressed by the miR-9a-engineered threshold (Cassidy, 2013).

This study is the first demonstration of a molecular mechanism that buffers genomic diversity via gene expression. Transcription factors have previously been invoked as buffering agents of genome variation, although how they achieve this is not known. Their impact on buffering can be significant, as witnessed in classic selection experiments with the sc¹ mutation. Impaired Ac/Sc activity generates increased h², a result that is recapitulated here. Genome buffering can also occur by posttranslational means that involve protein chaperones (Cassidy, 2013).

Although miRNAs have features that theoretically make them suitable for buffering, clearly, this function is not generalized to all miRNAs in all cells. miR-9a shows this function, whereas miR-7 does not. Yet, both miRNAs are highly conserved from fly to human, and the tissue specificity of their expression is also highly conserved. It is proposed that the difference can be found in each miRNA's target within the fate switch network. miR-7 indirectly activates ac/sc transcription, and in turn, its transcription is activated by Ac/Sc. Thus, genome-induced fluctuation of Ac/Sc activity would be amplified and not dampened by miR-7. It could be that miR-7 functions in a nonhomeostatic way in this particular network (Cassidy, 2013).

It is rare for genome variants to have deterministic effects on outcome, and yet the personal genome is being heralded as a new instrument for predicting disease risk and prognosis. Genome variation that correlates with certain outcomes could become a powerful predictor for prevention and treatment. However, counteracting this relationship will be buffering mechanisms that resemble the one is described in this study. The strength and efficacy of such mechanisms will affect the probability that certain variants are disease causative. Buffering mechanisms can also affect the evolution of diseases, in particular cancer. Cancers evolve by clonal expansion, genome instability, and clonal selection It is suggested that tumor heterogeneity is not only manifested by clones of cells with distinct genomes but by the variable buffering of these genomes within a tumor. In support, it has been found that individual tumor cells from a common genetic lineage are phenotypically heterogeneous with respect to growth and responsiveness to therapy. This phenotypic heterogeneity then enables selection for heritable features that promote cell survival and growth in the evolving environment of the tumor (Cassidy, 2013).

The subtle change in miR-9a copy number had large effects on buffering genomic diversity. Hence, the fluid variation in gene copy number, commonly seen in tumor cells, might have an impact in ways previously unforeseen. Moreover, epigenetic variation of gene expression in tumor cells could have similar consequences. In this regard, hypermethylation of the human miR-9 promoter is frequently observed in various carcinomas; this leads to reduced expression of the miRNA. For renal carcinoma, the epigenetic modification is associated with increased risk for recurrence. The current results suggest that reduced miRNA gene expression might affect the ability of cancer cells to evolve under natural and therapeutic conditions (Cassidy, 2013).

In conclusion, this study has identified a miRNA that inhibits the potential for genomic diversity to express itself at the level of a cell phenotype. Because selection experiments are applicable to many genes and cell types, this integrated approach should aid in understanding how genomic diversity is buffered in other organisms for traits that include disease risk and prognosis (Cassidy, 2013).

Repressive gene regulation synchronizes development with cellular metabolism

Metabolic conditions affect the developmental tempo of animals. Developmental gene regulatory networks (GRNs) must therefore synchronize their dynamics with a variable timescale. This study found that layered repression of genes couples GRN output with variable metabolism. When repressors of transcription or mRNA and protein stability are lost, fewer errors in Drosophila development occur when metabolism is lowered. This study demonstrates the universality of this phenomenon by eliminating the entire microRNA family of repressors. Development to maturity can be largely rescued when metabolism is reduced. Using a mathematical model that replicates GRN dynamics, lowering metabolism was found to suppress the emergence of developmental errors by curtailing the influence of auxiliary repressors on GRN output. This study experimentally shows that gene expression dynamics are less affected by loss of repressors when metabolism is reduced. Thus, layered repression provides robustness through error suppression and may provide an evolutionary route to a shorter reproductive cycle (Cassidy, 2019).

Developmental patterns arise from directed dynamics of cell-cell signaling and gene regulation. The sensory organs of Drosophila are a classic system with which to study these phenomena. A broad collection of gene mutations has specific effects on the formation of various sensory organs. The affected genes encode transcription factors, microRNAs, signaling factors, and other gene regulators. This study used such mutations to readdress the relationship between reduced metabolism and phenotype suppression. This was done by scoring Drosophila sensory mutant phenotypes under conditions of reduced energy metabolism (Cassidy, 2019).

To reduce metabolism, animals were generated that had genetic ablation of their insulin-producing cells (IPCs) in the brain. IPC ablation limits synthesis and release of insulin-like peptides (ILPs) and reduces the amount of glucose cells consume. It also reduces the abundance of the mitochondrial ATP synthase enzyme complex in cells. Moreover, inhibition of ILP production reduces the whole-body metabolic rate of Drosophila, as measured by calorimetry. IPC ablation results in 70% slower development and small but normally proportioned adults. Therefore, reduced ILP production by IPC ablation broadly decreases cellular energy metabolism and slows development (Cassidy, 2019).

Yan is a transcription factor that maintains cells of the developing compound eye in a progenitor-like state. The protein is transiently expressed in cells and is cleared from differentiating photoreceptor (R) cells by multiple repressors acting on its transcription, mRNA stability, and protein stability. The microRNA miR-7 represses post-transcriptional expression of Yan in the developing eye. When the mir-7 gene was specifically ablated in the compound eye of an otherwise wild-type animal, it resulted in small malformed adult eyes caused by errors in R cell differentiation. This phenotype was highly penetrant in genetically mosaic animals. However, when energy metabolism was slowed by IPC ablation, loss of mir-7 was much less important for the formation of correctly patterned eyes. Mutations affecting post-translational modification of Yan were examined. The epidermal growth factor (EGF) and Sevenless (Sev) receptor tyrosine kinases activate MAP kinase in the progenitors of R7 photoreceptors. MAP kinase phosphorylates Yan protein, leading to its ubiquitin-mediated degradation. This clearance of Yan protein enables differentiation of R7 cells, and when sev is mutated, cells completely fail to differentiate as R7 photoreceptors. However, slowing metabolism allowed a small but significant number of sev mutant cells to become R7 photoreceptors. Importantly, because the sev mutant makes no protein products, rescue of the mutant phenotype was not simply due to more functional Sev protein molecules being present in slowly metabolizing cells (Cassidy, 2019).

The transcription factor Hairy directly represses transcription of the achaete and scute genes during selection of cells for bristle fates. Mutation of hairy causes some individuals to develop ectopic large bristles. However, this effect of hairy mutation was strongly suppressed when energy metabolism was slowed. A similar effect was seen on a cis-regulatory module (CRM) that represses transcription of the wingless (wg) gene. The Sternopleural (Sp-1) mutation is present in a CRM located on the 3' flank of wg, causing wg misexpression and development of ectopic bristles. However, the ectopic bristle phenotype of the wgSp-1 mutant was almost totally reversed under conditions of slowed energy metabolism (Cassidy, 2019).

In conclusion, IPC ablation suppressed developmental phenotypes caused by mutations in genes that repress other genes at the transcriptional, post-transcriptional, and post-translational levels (Cassidy, 2019).

This study has shown that multi-layered weak repression within GRNs plays an unexpected function in synchronizing gene expression dynamics with the variable pace of the developmental program. Multiple repressors are required for accelerated development when metabolism is high, and they become functionally redundant when metabolism is low. Multiple repressors therefore allow reliable development across a broader range of metabolic conditions than tolerated otherwise (Cassidy, 2019).

This model explains long-standing observations linking nutrient limitation to suppression of mutant phenotypes. Presumably, such mutations cripple regulatory genes acting on developmental GRNs. This model might also offer an explanation for why animals that undergo above-normal growth exhibit compromised development. Wild-type GRNs might function across a limited range of metabolism, with functionality breaking down when metabolism exceeds that range (Cassidy, 2019).

Another mechanism to explain phenotype suppression relies on a steady-state and not dynamic perspective of gene expression. Genome-wide gene expression patterns could conceivably change with organismal growth rate. This is the case for chemostat-grown yeast cells, where the expression of 27% of all genes correlates with growth rate. Most genes associated with stress response are overexpressed when cells grow at a slow rate. Such stress-responsive expression could modulate global processes such as protein folding and turnover, among others, and attenuate phenotypes when metabolism is slowed. Indeed, molecular chaperones have been found to affect the penetrance of diverse gene mutations in C. elegans and Drosophila. However, this steady-state model does not explain why gene expression dynamics are conditionally dependent on the availability of repressors. This study found that repression of Yan and Sens expression by microRNAs becomes more redundant when metabolic rates are slowed. Nevertheless, phenotype suppression might be due to a combination of mechanisms, including steady-state stress response and gene expression dynamics (Cassidy, 2019).

Varied analyses carried out in this study suggest that the relationship between metabolism and gene expression dynamics is widespread. The entire family of 466 microRNAs in Drosophila melanogaster was found to become much less essential for development when energy metabolism is slowed. The extensive literature on microRNA function in Drosophila implicates them in practically all facets of the fruit fly's life. Various explanations have been provided for why this family of weak repressors has flourished in the animal kingdom, chief among them the idea that they act as buffers for gene expression. This study now posits that microRNAs also provide a robust means for developmental processes to accommodate fluctuations in metabolism (Cassidy, 2019).

Raising animals at lower temperatures can suppress the phenotypes of mutations that are not classical ts alleles. Indeed, loss of sens repression by miR-9a has less impact on bristle development when temperature is lowered. Because metabolic rate varies with temperature, it is possible that temperature-dependent phenotype suppression may also be attributed to a relaxed requirement for coupling gene expression dynamics to a metabolism-dependent timescale. This notion was explored using a modeling framework, and the results are inconclusive. It is anticipated that error suppression will be weak when temperature-modulated expression dynamics are the sole cause. This is because temperature should affect the rates of both anabolic and catabolic processes involved in gene expression. In contrast, limiting ATP availability or protein translation reduces the rates of anabolic reactions but not all catabolic reactions. This asymmetric effect on different steps in gene expression is a major reason why gene repression becomes less important when ATP availability or protein translation is limited (Cassidy, 2019).

Metabolic conditions drive variation of the intrinsic developmental tempo of each species. This study has shown that layered weak repression within GRNs enables these fluctuations to occur without causing developmental errors. Metabolic conditions change in both space and time. Perhaps the selective advantage of a reliable developmental outcome amidst variable environmental conditions is a driving force in the evolution of gene regulatory networks (Cassidy, 2019).

The FTD/ALS-associated RNA-binding protein TDP-43 regulates the robustness of neuronal specification through microRNA-9a in Drosophila

TDP-43 is an evolutionarily conserved RNA-binding protein currently under intense investigation for its involvement in the molecular pathogenesis of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). TDP-43 is normally localized in the nucleus, but translocated to the cytoplasm in diseased neurons. The endogenous functions of TDP-43 in the nervous system remain poorly understood. This study shows that the loss of Drosophila TDP-43 (dTDP-43) results in an increased production of sensory bristles and sensory organ precursor (SOP) cells on the notum of some but not all flies. The location of ectopic SOPs varies among mutant flies. The penetrance of this novel phenotype is dependent on the gender and sensitive to environmental influences. A similar SOP phenotype was also observed on the wing and in the embryos. Overexpression of dTDP-43 causes both loss and ectopic production of SOPs. Ectopic expression of ALS-associated mutant human TDP-43 (hTDP-43(M337V) and hTDP-43(Q331K)) produces a less severe SOP phenotype than hTDP-43(WT), indicating a partial loss of function of mutant hTDP-43. In dTDP-43 mutants, miR-9a expression is significantly reduced. Genetic interaction studies further support the notion that dTDP-43 acts through miR-9a to control the precision of SOP specification. These findings reveal a novel role for endogenous TDP-43 in neuronal specification and suggest that the FTD/ALS-associated RNA-binding protein TDP-43 functions to ensure the robustness of genetic control programs (Li, 2013).

This study identified a novel function for the evolutionarily conserved RNA-binding protein TDP-43 in neural specification in Drosophila. Unlike several well-studied transcription factors in the Notch-Delta lateral inhibition pathway, TDP-43 seems to function as a 'robustness' factor in a manner similar to that of miR-9a. In the absence of TDP-43 activity, ectopic bristles appear at one of several locations and SOP specification becomes more sensitive to environmental influences. Thus, TDP-43 seems to serve as a 'gatekeeper' to ensure the reproducible execution of genetic control programs and to canalize developmental phenotypes (Li, 2013).

Since mRNA synthesis can fluctuate, regulation of mRNA metabolism plays a central role in gene expression. miRNAs are a key class of regulatory molecules that ensure the robustness of developmental programs and may play a key role in canalization, which is partly due to the modest effect of each miRNA on expression levels of multiple mRNA targets. Hundreds of RNA-binding proteins are encoded by the Drosophila genome or in other species, and many simultaneously regulate multiple mRNAs. It is likely some other RNA-binding proteins also serve as robustness factors through direct interactions with target mRNAs (Li, 2013).

The current findings show that the 'robustness' function of TDP-43 is mediated in part by miR-9a in one specific neural developmental process. The molecular functions of TDP-43 in development as revealed in this study may provide a new perspective on its endogenous role in neurodegeneration. Although the loss of TDP-43 leads to an early lethal phenotype in mouse embryos, loss of nuclear TDP-43 does not necessarily result in rapid neuronal cell death. Gene expression in individual cells is tightly regulated and also varies significantly, in part due to stochastic biochemical events. Chronic loss of nuclear TDP-43 or compromise of its buffering function by genetic mutations may cause an imbalance in protein homeostasis in human neurons before eventual neurodegeneration, during which miRNAs may play an essential role downstream of TDP-43. Both TDP-43 and miR-9 are highly conserved through evolution and their interaction may occur in mammalian cells as well. It is interesting to note that miR-9 is significantly downregulated in Huntington's disease and a mouse model of spinal motor neuron disease. Thus, downregulation of miR-9 and possibly other miRNAs as well in stressed neurons in which TDP-43 has been depleted from the nucleus may be a common contributing factor in different neurodegenerative disorders (Li, 2013).

miR-9a prevents apoptosis during wing development by repressing Drosophila LIM-only

Loss of Drosophila mir-9a induces a subtle increase in sensory bristles, but a substantial loss of wing tissue. This study established that the latter phenotype is largely due to ectopic apoptosis in the dorsal wing primordium, and the wing development could be rescued in the absence of this microRNA by dorsal-specific inhibition of apoptosis. Such apoptosis was a consequence of de-repressing Drosophila LIM-only (dLMO), which encodes a transcriptional regulator of wing and neural development. Cell-autonomous elevation of endogenous dLMO and a GFP-dLMO 3'UTR sensor was observed in mir-9a mutant wing clones, and heterozygosity for dLMO rescued the apoptosis and wing defects of mir-9a mutants. Evidence is provided that dLMO, in addition to senseless, contributes to the bristle defects of the mir-9a mutant. Unexpectedly, the upregulation of dLMO, loss of Cut, and adult wing margin defects seen with mir-9a mutant clones were not recapitulated by clonal loss of the miRNA biogenesis factors Dicer-1 or Pasha, even though these mutant conditions similarly de-repressed miR-9a and dLMO sensor transgenes. Therefore, the failure to observe a phenotype upon conditional knockout of a miRNA processing factor does not reliably indicate the lack of critical roles of miRNAs in a given setting (Bejarano, 2010).

Because of their relative ease of detection, dominant alleles and X-linked mutants constituted a high proportion of the classical spontaneous mutants isolated by Morgan and colleagues. Bridges isolated Bx[1] in 1923, and genetic tests by Green in the early 1950s established that Bx was due to overactivity of the locus. In fact, the recessive allele Bx[r] was associated with a duplication of the region, indicating that as little as a two-fold increase in Bx activity could interfere with wing development. In 1979, Lifschytz and Green further proposed that Bx might be due to a mutation in a cis-acting repressor site in the heldup locus. Indeed, the cloning of Bx by the Cohen, Jan, and Segal labs in 1998 finally revealed that Bx and heldup were gain- and loss-of-function alleles of the dLMO gene, respectively. Moreover, most spontaneous Bx alleles proved to be transposable element insertions in the dLMO 3' UTR, and new Bx mutants were easily obtained by imprecise excisions of a downstream transposable element, so as to delete dLMO 3' UTR sequence. Collectively, these 85 years of research indicated that 3'UTR-mediated post-transcriptional repression of dLMO is critical for normal development (Bejarano, 2010).

A small number of other gain-of-function mutants in Drosophila and C. elegans result from the loss of 3' UTR regulatory elements, and many of these are now appreciated to be key genetic switch targets of miRNAs. The current studies, together with concurrent work from Heitzler and colleagues (Biryukova, 2009), establish dLMO as one of a handful of genes whose loss of miRNA-mediated repression leads to a severe morphological defect. This study found that the lack of mir-9a results in upregulation of dLMO, aberrant apoptosis in the wing pouch, and failure to completely specify and develop the wing margin. Importantly, dorsal-specific expression of miR-9a, dorsal-specific inhibition of cell death (using p35 or Diap1), or heterozygosity for dLmo, all strongly reduced ectopic apoptosis and restored adult wing development in mir-9a null animals (Bejarano, 2010).

Ectopic apoptosis in the wing pouch has previously been reported to result in loss of wing margin. However in other cases, the disc is able to compensate for cell loss in the face of ectopic apoptosis, so that no loss of margin is observed in the adult wing. In the mir-9a mutant, this study shows that excess apoptosis is coupled with a margin specification defect. Despite an ability to rescue the mutant by inhibiting apoptosis, it cannot be ruled out that ectopic apoptosis by itself might be insufficient to induce adult margin loss; perhaps it requires the sensitized background evidenced by the demonstrable failure to fully activate Cut in the third instar. It is also noted that Bx was recently reported not to be suppressed by inhibiting apoptosis (Bejarano, 2008), which might be at odds with the current conclusions. However, that study examined Bx/Y hemizygotes, which are substantially stronger in phenotype than Bx/X heterozygotes. It is clear that elevation of dLMO yields a variety of patterning defects that are not seen in mir-9a mutants. It is inferred that loss of miR-9a results in apoptosis and wing margin defects that are attributable to de-repression of dLMO, but that elevation of dLMO can clearly generate developmental phenotypes that are not simply due to excess apoptosis (Bejarano, 2010).

A minor, but quantifiable, consequence of lacking mir-9a is the development of a small number of ectopic sensory organs. This is demonstrably due to the de-repression of the proneural factors Sens and dLMO. Therefore, even though computational approaches provide evidence for hundreds of conserved miR-9a targets, including compelling 'anti-target' relationships with a large number of neural genes, the bulk of its morphologically evident phenotypes can be accounted for by the failure to repress only two target genes, sens and dLmo. In addition to miR-9a, Drosophilid species encode miR-9b and miR-9c, as well as the ancestrally related miR-79. The function of these miRNAs remains to be studied, but conventional knowledge of miRNA targeting suggests that they may have overlapping target capacity since they have similar seeds. It is conceivable that the analysis of double or triple mir-9 mutants may reveal additional targets that mediate compelling phenotypes. Nevertheless, it is clear that miR-9a serves a function to repress dLmo and sens that cannot be substantially compensated by the remaining miR-9-related genes (Bejarano, 2010).

dLMO and miR-9 are both highly conserved between invertebrates and vertebrates. However, vanishingly few miRNA:target interactions have been preserved over this evolutionary distance, indicating that these post-transcriptional target networks are much more plastic than the genes themselves. Therefore, the existence of a key miR-9a:dLmo regulatory connection in flies does not necessary imply that human miR-9 regulates LMO genes, and human LMO genes lack conserved canonical miR-9 seed sites. Heitzler and colleagues proposed that mammalian LMO2 is a conserved target of miR-9 (Biryukova, 2009). However, the candidate site contains a G:U seedpair, a feature that is detrimental to, although not necessarily incompatible with miRNA targeting. Directed studies are needed to assess whether this site alone confers repression by miR-9 g (Bejarano, 2010).

On the other hand, the necessity of restricting LMO activity might well prove to be a conserved feature of invertebrate and vertebrate biology. As in Drosophila, vertebrate LMO proteins can dominantly interfere with LDB:Islet complexes, indicating that its overactivity is especially 'dangerous'. Indeed, elevation of LMO proteins has myriad consequences for downstream transcriptional networks, and LMO2 is in fact a T-cell oncogene. Intriguingly, LMO2, which normally regulates hematopoetic development, has a highly conserved 8mer seed for miR-223. Recent studies demonstrated that mir-223 mutant mice exhibit hematopoetic defects, and that mir-223 deletion has consequences for the neutrophil transcriptome and proteome. While the depth of peptide sampling was insufficient to report on LMO2 status, the microarray data demonstrated LMO2 to be the 60th most-upregulated mRNA across the transcriptome of mir-223 knockout cells. Indeed, it has been recently reported that suppression of LMO2 by miR-223 regulates erythropoiesis. It is notef that its paralog LMO1 contains a highly conserved canonical site for miR-181, another miRNA with a demonstrated function in the hematopoietic system. These observations suggest that the regulation of vertebrate LMO genes by hematopoietic miRNAs, and its potential relevance to cancer, deserves further study (Bejarano, 2010).

To date, relatively few Drosophila or vertebrate miRNA genes have been analyzed using bona fide mutant alleles. As an approximation, many researchers have taken to analyzing the effects of conditional knockout of miRNA biogenesis factors, such as Dicer. This manipulation is presumed to break all miRNA regulatory links, thereby serving as a plausibility test of whether miRNAs might be required in a given setting. Acknowledged drawbacks of this approach include uncertainty as to whether one or many miRNAs might contribute to a given phenotype, and whether phenotypes are a direct or indirect cause of miRNA loss. However, a caveat that is little considered is the potentially canceling effects of removing 'all' miRNAs, so that loss of one miRNA might be compensated for by the concomitant loss of another miRNA(s). While such an outcome might seem to require highly unlikely coincidences, it may be plausible if it is considered that most biological processes are under both positive and negative control, and that most genes are themselves miRNA targets (Bejarano, 2010).

During development of the Drosophila wing primordium, this study has shown that clones lacking mir-9a upregulate dLMO and induce wing notching, whereas dcr-1 and pasha-mutant clones do not. Additionally, mir-9a mutant clones exhibit a more severe phenotype than dcr-1 mutant clones with respect to loss of wing margin, both in the third instar wing pouch and in the adult wing. Although the cells analyzed were homozygous mutant for substantial periods of time (72-96 hours), perdurance of miRNAs on account of Dcr-1 or Pasha proteins inherited by mutant cells conceivably contributes to the phenotypic disparity. For example, perdurance may explain the incomplete phenocopy of bantam mutant discs by dcr-1 or pasha 'whole disc' mutants. However, potential perdurance is not reconciled with the comparable upregulation of miR-9a and dLMO sensor activity in dcr-1 and mir-9a homozygous mutant cells, which report on similar loss of miR-9a activity in these clones. Together, these data suggest that mir-9a mutant cells exhibit phenotypes that are intrinsically different from those of dcr-1 or pasha mutant cells (Bejarano, 2010).

In summary, the failure to observe a phenotype in cells or tissues that are mutant for a general miRNA biogenesis factor cannot reliably be taken as evidence that miRNAs lack substantial roles in the setting of interest. Reciprocally, the observation that loss of miRNA-mediated regulation from a single target gene (e.g. failure to repress dLmo in mir-9a mutant wings) can be of greater phenotypic impact than loss of 'all' miRNA-mediated regulation (e.g, in dcr-1, pasha double mutant wing clones) highlights the disproportionate consequence of releasing particular miRNA targets from amidst a regulatory web that is inferred to encompass most animal transcripts (Bejarano, 2010).

Drosophila mir-9a regulates wing development via fine-tuning expression of the LIM only factor, dLMO

MicroRNAs are short non-coding endogenous RNAs that are implicated in regulating various aspects of plants and animal development, however their functions in organogenesis are largely unknown. This study reports that mir-9a belonging to the mir-9 family, regulates Drosophila wing development through a functional target site in the 3' untranslated region of the Drosophila LIM only protein, dLMO. dLMO is a transcription cofactor, that directly inhibits the activity of Apterous, the LIM-HD factor required for the proper dorsal identity of the wings. Deletions of the 3' untranslated region, including the mir-9a site, generate gain-of-function dLMO mutants (Beadex) associated with high levels of dLMO mRNA and protein. Beadex mutants lack wing margins, a phenotype also observed in null mir-9a mutants. mir-9a and dLMO are co-expressed in wing discs and interact genetically for controlling wing development. Lack of mir-9a results in overexpression of dLMO, while gain-of-function mir-9a mutant suppresses dLMO expression. These data indicate that a function of mir-9a is to ensure the appropriate stoichiometry of dLMO during Drosophila wing development. The mir-9a binding site is conserved in the human counterpart LMO2, the T-cell acute leukemia oncogene, suggesting that mir-9 might apply a similar strategy to maintain LMO2 expression under a detrimental threshold (Biryukova, 2009).

The dorsoventral (DV) axis of wings is specified, at the second-instar larval stage, by the activity of Apterous (Ap) in the presumptive dorsal cells of the wing primordium. Ap induces Fringe and Serrate within the whole dorsal territory. Fringe, a crucial effector of the Notch receptor, leads to a differential affinity between the ligands Delta and Serrate. Thus, Notch activity is restricted to the DV boundary, where intense signaling by Notch leads to expression of Wingless (Wg). Levels of Ap and the subsequent Ap-dependent expression of Serrate and Fringe are regulated by Chip (the Drosophila Ldb factor) and dLMO during second and early third-instar stages. Ap activity depends strictly on its cofactor Chip and requires the formation of an Ap-Chip-Chip-Ap tetramer. Ap induces expression of its repressor dLMO, which competes with Ap for binding Chip. As a consequence, within the dorsal domain, dLMO downregulates Ap and limits levels of Serrate and Fringe, that are critical for wing cell survival. In the analysed Bx mutants, Ap expression was observed to be strongly misregulated. In addition, no proper DV boundary formation was found in the wing pouch. Wg is expressed along the DV boundary in wing pouch, where it acts locally to induce nearby cells to adopt a wing margin identity. Wg expression was found to be severely affected in the analysed Bx mutants. The dLMO-dependent repression of Serrate and Fringe in early third-instar, is necessary to allow secondary Delta and Serrate expression domains in an Ap-independent mode of regulation. This Ap-independent induction of Serrate maintains Wg and Cut expression in a Notch-dependent positive feed back loop. However, since Ap activity continues to be required for dorsal cell fate specification, it is important to limit Ap downregulation by dLMO. It is suggested that mir-9a could have a critical role in this process (Biryukova, 2009).

The stoichiometry of dLMO is relevant for proper wing development. This study analyzed series of new Bx, the gain-of-function alleles of dLMO that are associated with lack of miRNA sites in the 3' UTR. Among several putative candidates, the investigation focussed on the function of a very ancient miRNA mir-9a, that is conserved from Drosophila to human. dLMO was shown to be a mir-9a target gene that is involved in the control of the Drosophila wing development. Both loss and gain-of-function mir-9a mutants exhibit wing margin defects. Analysis of dLMO expression in loss-of-function mir-9a mutants revealed increased levels of dLMO in the wing discs. Reciprocally, in mir-9a gain-of-function mutants, a reduced expression of dLMO was detected. Strong genetic interactions were observed between dLMO and mir-9a during wing development. Loss of wing margin of mir-9a null flies is fully rescued by removing its target gene, dLMO. Moreover, the wing margin defects in gain-of-function Bx mutants can be rescued by overexpression of mir-9a. These data provide evidence that dLMO is the main target gene of mir-9a for wing development. Loss of wing margin phenotype of mir-9a null flies has been previously attributed to an excess of another mir-9a target gene, senseless (sens). However, this study showed that the wing shape is less sensitive to the dosage of Sens than to the dosage of dLMO. Flies carrying two extra copies of sens⁺ do not have a modified wing shape, whereas a fly carrying only one extra copy of sens⁺, shows a gap in the posterior wing margin, a phenotype observed in weak Bx mutants. According to current estimation, a medium to strong Bx mutant, overexpresses 2- to 4-fold the normal level of dLMO transcripts. Loss of wing margin in mir-9a null mutants is more likely a consequence of the overexpression of dLMO or a synergism between overexpressed dLMO and Sens. Besides, no genetic interaction was found between dLMO and Sens during wing development. In this context, it is suggested that mir-9a might maintain the required threshold of dLMO protein levels that are critical for wing cell survival and fine tune the appropriate level of Ap (Biryukova, 2009).

mir-9a was recently described to restrict the generation of sensory organs by downregulating Sens expression within the proneural fields and assuming the robustness of the Notch lateral signaling. Interestingly, dLMO behaves as an early proneural activator in the thorax (Asmar, 2008). dLMO is expressed within the proneural clusters, and later after the SOP selection, it is excluded from the mature neurons, like mir-9a. Thus, when dLMO is downregulated by mir-9a, its proneural activity no longer interferes with the Notch lateral signaling that represses proneural fate within the same cells. It is suggested that regulation of both dLMO and Sens by mir-9a represent further level of complexity for gene regulation that contributes to sensory organ development (Biryukova, 2009).

dLMO plays an important role in multiple biological processes in Drosophila. The gene encodes two functional isoforms dLMO-PA and dLMO-PB that are differentially expressed during development. It has been demonstrated that the major isoform, dLMO-PA participates with Chip in the assembly of a proneural transcriptional complex that includes the GATA factor Pannier and the bHLHs Ac/Sc and Daughterless (Da). Therefore, dLMO acts as an early coactivator of the proneural genes ac/sc. dLMO mutants have consistent pleiotropic phenotypes, for instance, amorphic dLMO mutants show a hypoplasia of neurons. Loss-of-function dLMO mutants exhibit strong locomotion defects and changes in cocaine responsiveness, hence dLMO function is required for circadian pacemaker neurons in the brain. For these aspects of neurogenesis and neurological responsiveness, the gain-of-function Bx mutants show opposite phenotypes with regards to the loss-of-function ones. Since Bx alleles encode more stable truncated mRNAs, the stoichiometry of dLMO might be relevant for diverse dLMO functions. dLMO 3' UTR contains multiple motifs involved in negative post-transcriptional regulation, including Brd-boxes, AU-rich elements (AREs) and miRNA sites. Interestingly, the non-canonical Brd boxes in the dLMO 3' UTR carry the same wobble at position 6 in the seed, that might decrease a tuning efficiency by Brd box related miRNAs, mir-79 and mir-4. Both ARE-binding proteins and specific miRNAs can bind the AREs sequences, modulating the translational regulation of the transcripts. The diversity of miRNA sites might reflect specificities, redundancies and cooperations with the mir-9a site for adjusting dLMO transcripts under a detrimental level in regulating both neurogenesis and wing development, or controlling other functions of dLMO. For instance, dLMO was identified recently as a potential mir-14 target gene. Although mir-14 is expressed during larval stages, loss of mir-14 did not show any overt wing phenotype nor did it enhance the wing margin defects of Bx¹/+ wings. Therefore, mir-14 might directly regulate the dLMO protein levels that are required in other developmental contexts in which dLMO activity is involved (Biryukova, 2009).

The mammalian homolog of dLMO, LMO2, is expressed in the mesenchyme of the developing mouse limb bud, suggesting a conserved function of LMO2 between insects and mammals. Lmx-1, a LIM-HD protein like Ap, is expressed in the mesenchyme of the dorsal limb bud during development. Loss of Lmx-1b function causes a biventral phenotype, implicating Lmx-1b as a primary dorsalizing activity in the mouse limb. Like dLMO, LMO2 is expressed in both the dorsal and ventral compartments during limb patterning. Lhx-2, the mammalian ortholog of Ap, is able to rescue ap mutant phenotypes as efficiently as the fly Ap protein. LMO2 or other LMO proteins could interact with Lmx-1 or Lhx-2, in a manner similar to the proposed interaction between Ap and dLMO in Drosophila wings. LMO2 is also expressed at the somite boundaries. Many genes required for formation of the D-V boundary in the developing limb, such as members of the fringe, Wnt, and Notch gene families, also play an important role during somitogenesis. The presence of LMO2 RNA at the somite boundaries might indicate a conserved role of LMO gene family members in the context of boundary formation, including the limb bud, somite, and insect wing disc (Biryukova, 2009).

LMO2 is known as a master regulator of haematopoiesis in mouse and human. Its stoichiometry is critical for proper T-cell differentiation during early haematopoiesis. LMO2 is activated via chromosomal translocation in T-cell acute leukemia (T-ALL). In Drosophila neurogenesis, dLMO is a member of a transcriptional complex similar to the one that controls Tal-1 expression during human haematopoiesis. Interestingly, the human LMO2 3' UTR contains multiple miRNA target sites, including hsa-mir-9 and hsa-mir-9*, that are orthologs of the Drosophila dme-mir-9a and dme-mir-79, respectively. Furthermore, hsa-mir-9 is expressed within the mammalian haematopoietic system. In human, the enforced expression of LMO2 in a significant fraction of T-ALL results from loss of the upstream transcriptional mechanisms that normally downregulate the expression of this oncogene during T-cell development. Therefore, it is suggested that a similar mechanism of LMO2 regulation by mir-9 may operate during normal human haematopoiesis and can be disrupted in pathological conditions, like T-cell acute leukemias and large B-cell non-Hodgkin's lymphomas (Biryukova, 2009).

Post-transcriptional regulation; miR-9a targets senseless

MicroRNAs (miRNAs) have been implicated in regulating various aspects of animal development, but their functions in neurogenesis are largely unknown. Loss of miR-9a function in the Drosophila peripheral nervous system leads to ectopic production of sensory organ precursors (SOPs), whereas overexpression of miR-9a results in a severe loss of SOPs. A strong genetic interaction occurs between miR-9a and senseless (sens) in controlling the formation of SOPs in the adult wing imaginal disc. Moreover, miR-9a suppresses Sens expression through its 3' untranslated region. miR-9a is expressed in epithelial cells, including those adjacent to SOPs within proneural clusters, suggesting that miR-9a normally inhibits neuronal fate in non-SOP cells by down-regulating Sens expression. These results indicate that miR-9a ensures the generation of the precise number of neuronal precursor cells during development (Li, 2006).

miR-9a is one of the miRNAs that are highly expressed in the mammalian brain and 100% conserved at the nucleotide sequence from flies to humans, suggesting an important role in brain development and/or function. miR-9a loss-of-function alleles were generated; homozygous mutant flies developed into adulthood at the expected Mendelian ratio. Adult mutant flies are grossly normal and fertile, indicating that miR-9a is not required for viability or fertility. This finding is different from the reported severe dorsal closure defects and embryonic lethal phenotype generated by antisense 2 O-methyl oligoribonucleotide-mediated depletion of miR-9a. Interestingly, Drosophila miR-9a is not expressed in mature neurons, but is expressed in epithelial cells, including the proneural clusters that give rise to SOPs. Detailed analysis of embryonic PNS development revealed an unexpected finding that miR-9a mutants have an increased number of sensory neurons that elaborate extensive dendritic arbors underneath the epithelial cell layer, such as ddaE and ddaF neurons. The duplicated neurons occupy the same dendritic field and appear to have similar dendritic branching patterns. Indeed, the average numbers of dendritic ends of ddaE and ddaF neurons in abdominal segments 3-5 were similar in wild-type and miR-9a mutant larvae and MARCM clones, indicating that loss of miR-9a activity affected the number of these sensory neurons only but had no cell-autonomous effect on their dendritic branching patterns (Li, 2006).

The effect of miR-9a on the number of embryonic sensory neurons has two major features: (1) the ectopic ddaE or ddaF neurons were generated as a result of ectopic SOPs and not cell fate transformation within a cell lineage, suggesting miR-9a affects an early step in neurogenesis, consistent with its embryonic expression pattern; (2) both the expressivity and penetrance of this defect were relatively low. This finding supports the idea that miRNAs, at least in this particular case, are not developmental switches, but instead function as a fine-tuning mechanism to ensure the accuracy of a particular developmental process. In this study, focus was placed on the formation of SOPs in adult flies. Like embryos, only 14% of miR-9a mutant flies exhibited ectopic SOPs on the notum, again indicating a fine-tuning role for miR-9a in controlling SOP formation. However, analysis of the miR-9a mutant phenotype in adult flies also indicates that miRNAs can have dramatic effects on some other developmental processes. For instance, miR-9a is widely expressed in the wing disc, and 100% of miR-9a mutant flies exhibited a severe posterior wing margin defect, suggesting that cell proliferation and/or survival are much more sensitive to changes in the expression levels of the proteins regulated by miR-9a (Li, 2006).

How does miR-9a exert its effect on SOP formation? Sens is a zinc finger transcription factor required to maintain high-level expression of proneural gene in SOPs and to suppress their expression in non-SOP cells. Several findings in this study demonstrate that Sens is a key target of miR-9a regulation and is essential for mediating miR-9a function in SOP formation. (1) The wing margin defects in miR-9a mutant flies were remarkably similar to that caused by overexpression of Sens by the UAS-Gal4 system or in Lyra1 mutants. (2) miR-9a was expressed at a much lower level in SOPs than in adjacent epithelial cells, correlating with the high level of Sens expression in SOPs and the low level of Sens in non-SOP cells in proneural clusters. The inability to use immunostaining to detect subtle changes of Sens expression level in non-SOP cells due to miR-9a loss of function could be attributed to the following reasons: Sens expression is primarily down-regulated at the transcriptional level in the non-SOP proneural cells, and miR-9a's function is limited to preventing translation of the leaky/residual sens mRNA. The alteration in Sens level, due to loss of miR-9a function in the non- SOP cells, is sufficient to initiate the production of ectopic SOPs, but it may not be dramatic enough to be detected by immunostaining. (3) The sens 3' UTR contains three miR-9a-binding sites and is the best predicted target of miR-9a. (4) Wild-type but not mutant miR-9a precursors down-regulated reporter gene expression through the sens 3' UTR in transfected cells. (5) Overexpression of miR-9a in vivo inhibited Sens expression. It was observed that Sens expression along the wing margin in the dorsal compartment is lower than in the ventral compartment in some wing discs when miR-9a is expressed by ap-Gal4. The failure to completely suppress Sens expression in the dorsal compartment is probably due to the fact that, at this developmental stage, Sens expression in the wing margins is controlled by proneural genes, unlike SOPs in the notum region where Sens expression is maintained by itself. (6) miR-9a and sens showed strong genetic interactions in controlling SOP formation (Li, 2006).

These findings provide an experimental example to support the notion that miRNAs and their mRNA targets are often expressed in cells adjacent to each other. The differential expression of Sens in SOPs and adjacent neuroepithelial cells is essential for the production of a precise number of SOPs during development. A model is proposed in which miR-9a functions in non-SOPs cells to further suppress Sens expression at the translational level, as a complementary mechanism to the transcriptional inhibition of Sens expression by E(spl). Loss of miR-9a function increases Sens protein level, not so dramatically but just enough to convert Sens in some neuroepithelial cells from a transcription repressor into an activator of proneural genes, therefore resulting in the formation of a small number of ectopic SOPs. However, unlike many other genes essential for neurogenesis, such as Notch and Delta, miR-9a does not function as an absolute switch. Instead, it only ensures the accurate differential Sens expression and fine-tunes this developmental. Overexpression of miR-9a in the wing imaginal disc could dramatically inhibit the formation of sensory organs on the notum, suggesting that misregulation of miR-9a expression itself could potentially have severe developmental consequences. Since both miR-9a and E(spl) have similar functions in non-SOP cells, it is possible that both genes may be regulated by a similar transcriptional mechanism. Indeed, binding sites for the Achaete-Scute complex and Su(H) are present in the regulatory region of miR-9a. Taken together, these studies presented here have uncovered another layer of gene regulation during early neurogenesis in the Drosophila PNS. miR-9a is 100% conserved at the nucleotide level from flies to humans. Moreover, the human miR-9a is highly expressed in fetal but not in adult brains. Therefore, a similar mechanism of miR-9a function may operate during mammalian neurogenesis as well (Li, 2006).

Search PubMed for articles about Drosophila Mir-9a

Bejarano, F., Smibert, P. and Lai, E. C. (2010). miR-9a prevents apoptosis during wing development by repressing Drosophila LIM-only. Dev Biol 338: 63-73. PubMed ID: 19944676

Biryukova, I., Asmar, J., Abdesselem, H. and Heitzler, P. (2009). Drosophila mir-9a regulates wing development via fine-tuning expression of the LIM only factor, dLMO. Dev. Biol. 327: 487-496. PubMed ID: 19162004

Bushati, N., Stark, A., Brennecke, J. and Cohen, S. M. (2008). Temporal reciprocity of miRNAs and their targets during the maternal-to-zygotic transition in Drosophila. Curr Biol 18: 501-506. PubMed ID: 18394895

Cassidy, J. J., Jha, A. R., Posadas, D. M., Giri, R., Venken, K. J., Ji, J., Jiang, H., Bellen, H. J., White, K. P. and Carthew, R. W. (2013). miR-9a minimizes the phenotypic impact of genomic diversity by buffering a transcription factor. Cell 155: 1556-1567. PubMed ID: 24360277

Cassidy, J. J., Bernasek, S. M., Bakker, R., Giri, R., Pelaez, N., Eder, B., Bobrowska, A., Bagheri, N., Nunes Amaral, L. A. and Carthew, R. W. (2019). Repressive gene regulation synchronizes development with cellular metabolism. Cell 178(4):980-992. PubMed ID: 31353220

Fu, S., Nien, C. Y., Liang, H. L. and Rushlow, C. (2014). Co-activation of microRNAs by Zelda is essential for early Drosophila development. Development 141: 2108-2118. PubMed ID: 24764079

Li, Y., Wang, F., Lee, J.A. and Gao, F.-B. (2006). MicroRNA-9a ensures the precise specification of sensory organ precursors in Drosophila. Genes Dev 20: 2793-2805. Medline abstract: 17015424

Li, Z., Lu, Y., Xu, X. L. and Gao, F. B. (2013). The FTD/ALS-associated RNA-binding protein TDP-43 regulates the robustness of neuronal specification through microRNA-9a in Drosophila. Hum Mol Genet 22: 218-225. PubMed ID: 23042786

Tadros, W., Goldman, A. L., Babak, T., Menzies, F., Vardy, L., Orr-Weaver, T., Hughes, T. R., Westwood, J. T., Smibert, C. A. and Lipshitz, H. D. (2007). SMAUG is a major regulator of maternal mRNA destabilization in Drosophila and its translation is activated by the PAN GU kinase. Dev Cell 12: 143-155. PubMed ID: 17199047

Van Ry, P. M., Minogue, P., Hodges, B. L. and Burkin, D. J. (2014). Laminin-111 improves muscle repair in a mouse model of merosin-deficient congenital muscular dystrophy. Hum Mol Genet 23: 383-396. PubMed ID: 24009313

Yatsenko, A. S. and Shcherbata, H. R. (2014). Drosophila miR-9a targets the ECM receptor Dystroglycan to canalize myotendinous junction formation. Dev Cell 28: 335-348. PubMed ID: 24525189

date revised: 11 June 2022

Home page: The Interactive Fly © 2011 Thomas Brody, Ph.D.