Transcription factor redundancy and tissue-specific regulation: evidence from functional and physical network connectivity

HLH-1 Modulates Muscle Proteostasis During Caenorhabditis elegans Larval Development

Muscle proteostasis is shaped by the myogenic transcription factor MyoD which regulates the expression of chaperones during muscle differentiation. Whether MyoD can also modulate chaperone expression in terminally differentiated muscle cells remains open. Here we utilized a temperature-sensitive (ts) conditional knockdown nonsense mutation in MyoD ortholog in C. elegans, HLH-1, to ask whether MyoD plays a role in maintaining muscle proteostasis post myogenesis. We showed that hlh-1 is expressed during larval development and that hlh-1 knockdown at the first, second, or third larval stages resulted in severe defects in motility and muscle organization. Motility defects and myofilament organization were rescued when the clearance of hlh-1(ts) mRNA was inhibited, and hlh-1 mRNA levels were restored. Moreover, hlh-1 knockdown modulated the expression of chaperones with putative HLH-1 binding sites in their promoters, supporting HLH-1 role in muscle maintenance during larval development. Finally, mild disruption of hlh-1 expression during development resulted in earlier dysregulation of muscle maintenance and function during adulthood. We propose that the differentiation transcription factor, HLH-1, contributes to muscle maintenance and regulates cell-specific chaperone expression post differentiation. HLH-1 may thus impact muscle proteostasis and potentially the onset and manifestation of sarcopenia.

Isoform-specific functions of PPARγ in gene regulation and metabolism

In this study, Hu et al. investigated the specific functions of the two main PPARγ isoforms by generating mouse lines in which endogenous PPARγ1 and PPARγ2 were epitope-tagged to interrogate isoform-specific genomic binding, and mice deficient in either PPARγ1 or PPARγ2 to assess isoform-specific gene regulation. They show that PPARγ isoforms have specific and separable metabolic functions that may be targeted to improve therapy for insulin resistance and diabetes.

Peroxisome proliferator-activated receptor γ (PPARγ) is a nuclear receptor that is a vital regulator of adipogenesis, insulin sensitivity, and lipid metabolism. Activation of PPARγ by antidiabetic thiazolidinediones (TZD) reverses insulin resistance but also leads to weight gain that limits the use of these drugs. There are two main PPARγ isoforms, but the specific functions of each are not established. Here we generated mouse lines in which endogenous PPARγ1 and PPARγ2 were epitope-tagged to interrogate isoform-specific genomic binding, and mice deficient in either PPARγ1 or PPARγ2 to assess isoform-specific gene regulation. Strikingly, although PPARγ1 and PPARγ2 contain identical DNA binding domains, we uncovered isoform-specific genomic binding sites in addition to shared sites. Moreover, PPARγ1 and PPARγ2 regulated a different set of genes in adipose tissue depots, suggesting distinct roles in adipocyte biology. Indeed, mice with selective deficiency of PPARγ1 maintained body temperature better than wild-type or PPARγ2-deficient mice. Most remarkably, although TZD treatment improved glucose tolerance in mice lacking either PPARγ1 or PPARγ2, the PPARγ1-deficient mice were protected from TZD-induced body weight gain compared with PPARγ2-deficient mice. Thus, PPARγ isoforms have specific and separable metabolic functions that may be targeted to improve therapy for insulin resistance and diabetes.

SAResNet: self-attention residual network for predicting DNA-protein binding

Knowledge of the specificity of DNA-protein binding is crucial for understanding the mechanisms of gene expression, regulation and gene therapy. In recent years, deep-learning-based methods for predicting DNA-protein binding from sequence data have achieved significant success. Nevertheless, the current state-of-the-art computational methods have some drawbacks associated with the use of limited datasets with insufficient experimental data. To address this, we propose a novel transfer learning-based method, termed SAResNet, which combines the self-attention mechanism and residual network structure. More specifically, the attention-driven module captures the position information of the sequence, while the residual network structure guarantees that the high-level features of the binding site can be extracted. Meanwhile, the pre-training strategy used by SAResNet improves the learning ability of the network and accelerates the convergence speed of the network during transfer learning. The performance of SAResNet is extensively tested on 690 datasets from the ChIP-seq experiments with an average AUC of 92.0%, which is 4.4% higher than that of the best state-of-the-art method currently available. When tested on smaller datasets, the predictive performance is more clearly improved. Overall, we demonstrate that the superior performance of DNA-protein binding prediction on DNA sequences can be achieved by combining the attention mechanism and residual structure, and a novel pipeline is accordingly developed. The proposed methodology is generally applicable and can be used to address any other sequence classification problems.

FHOD formin and SRF promote post-embryonic striated muscle growth through separate pathways in C. elegans

Previous work with cultured cells has shown transcription of muscle genes by serum response factor (SRF) can be stimulated by actin polymerization driven by proteins of the formin family. However, it is not clear if endogenous formins similarly promote SRF-dependent transcription during muscle development in vivo. We tested whether formin activity promotes SRF-dependent transcription in striated muscle in the simple animal model, Caenorhabditis elegans. Our lab has shown FHOD-1 is the only formin that directly promotes sarcomere formation in the worm's striated muscle. We show here FHOD-1 and SRF homolog UNC-120 both support muscle growth and also muscle myosin II heavy chain A expression. However, while a hypomorphic unc-120 allele blunts expression of a set of striated muscle genes, these genes are largely upregulated or unchanged by absence of FHOD-1. Instead, pharmacological inhibition of the proteasome restores myosin protein levels in worms lacking FHOD-1, suggesting elevated proteolysis accounts for their myosin deficit. Interestingly, proteasome inhibition does not restore normal muscle growth to fhod-1(Δ) mutants, suggesting formin contributes to muscle growth by some alternative mechanism. Overall, we find SRF does not depend on formin to promote muscle gene transcription in a simple in vivo system.

Plant PHR Transcription Factors: Put on A Map

The phosphate starvation response (PHR) protein family exhibits the MYB and coiled-coil domains. In plants, within the either 5' untranslated regions (UTRs) or promoter regions of phosphate starvation-induced (PSI) genes are characteristic cis-regulatory elements, namely PHR1 binding sequence (P1BS). The most widely studied PHR protein family members, such as AtPHR1 in Arabidopsis thaliana (L.) and OsPHR2 in Oryza sativa (L.), may activate the gene expression of a broad range of PSI genes by binding to such elements in a phosphate (Pi) dependent manner. In Pi signaling, PHR transcription factors (TFs) can be selectively activated or deactivated by other proteins to execute the final step of signal transduction. Several new proteins have been associated with the AtPHR1/OsPHR2 signaling cascade in the last few years. While the PHR TF transcriptional role has been studied intensively, here we highlight the recent findings of upstream molecular components and other signaling pathways that may interfere with the PHR final mode of action in plants. Detailed information about transcriptional regulation of the AtPHR1 gene itself and its upstream molecular events has been reviewed.

The potential application of genome editing by using CRISPR/Cas9, and its engineered and ortholog variants for studying the transcription factors involved in the maintenance of phosphate homeostasis in model plants

Phosphorus (P), an essential macronutrient, is pivotal for growth and development of plants. Availability of phosphate (Pi), the only assimilable P, is often suboptimal in rhizospheres. Pi deficiency triggers an array of spatiotemporal adaptive responses including the differential regulation of several transcription factors (TFs). Studies on MYB TF PHR1 in Arabidopsis thaliana (Arabidopsis) and its orthologs OsPHRs in Oryza sativa (rice) have provided empirical evidence of their significant roles in the maintenance of Pi homeostasis. Since the functional characterization of PHR1 in 2001, several other TFs have now been identified in these model plants. This raised a pertinent question whether there are any likely interactions across these TFs. Clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9) system has provided an attractive paradigm for editing genome in plants. Here, we review the applications and challenges of this technique for genome editing of the TFs for deciphering the function and plausible interactions across them. This technology could thus provide a much-needed fillip towards engineering TFs for generating Pi use efficient plants for sustainable agriculture. Furthermore, we contemplate whether this technology could be a viable alternative to the controversial genetically modified (GM) rice or it may also eventually embroil into a limbo.

UNC-120/SRF independently controls muscle aging and lifespan in Caenorhabditis elegans

Aging is commonly defined as the loss of global homeostasis, which results from progressive alteration of all organs function. This model is currently challenged by recent data showing that interventions that extend lifespan do not always increase the overall fitness of the organism. These data suggest the existence of tissue-specific factors that regulate the pace of aging in a cell-autonomous manner. Here, we investigated aging of Caenorhabditis elegans striated muscles at the subcellular and the physiological level. Our data show that muscle aging is characterized by a dramatic decrease in the expression of genes encoding proteins required for muscle contraction, followed by a change in mitochondria morphology, and an increase in autophagosome number. Myofilaments, however, remain unaffected during aging. We demonstrated that the conserved transcription factor UNC-120/SRF regulates muscle aging biomarkers. Interestingly, the role of UNC-120/SRF in the control of muscle aging can be dissociated from its broader effect on lifespan. In daf-2/insulin/IGF1 receptor mutants, which exhibit a delayed appearance of muscle aging biomarkers and are long-lived, disruption of unc-120 accelerates muscle aging but does not suppress the lifespan phenotype of daf-2 mutant. Conversely, unc-120 overexpression delays muscle aging but does not increase lifespan. Overall, we demonstrate that UNC-120/SRF controls the pace of muscle aging in a cell-autonomous manner downstream of the insulin/IGF1 receptor.

A network diffusion approach to inferring sample-specific function reveals functional changes associated with breast cancer

Guilt-by-association codifies the empirical observation that a gene's function is informed by its neighborhood in a biological network. This would imply that when a gene's network context is altered, for instance in disease condition, so could be the gene's function. Although context-specific changes in biological networks have been explored, the potential changes they may induce on the functional roles of genes are yet to be characterized. Here we analyze, for the first time, the network-induced potential functional changes in breast cancer. Using transcriptomic samples for 1047 breast tumors and 110 healthy breast tissues from TCGA, we derive sample-specific protein interaction networks and assign sample-specific functions to genes via a diffusion strategy. Testing for significant changes in the inferred functions between normal and cancer samples, we find several functions to have significantly gained or lost genes in cancer, not due to differential expression of genes known to perform the function, but rather due to changes in the network topology. Our predicted functional changes are supported by mutational and copy number profiles in breast cancers. Our diffusion-based functional assignment provides a novel characterization of a tumor that is complementary to the standard approach based on functional annotation alone. Importantly, this characterization is effective in predicting patient survival, as well as in predicting several known histopathological subtypes of breast cancer.

A Differentiation Transcription Factor Establishes Muscle-Specific Proteostasis in Caenorhabditis elegans

Safeguarding the proteome is central to the health of the cell. In multi-cellular organisms, the composition of the proteome, and by extension, protein-folding requirements, varies between cells. In agreement, chaperone network composition differs between tissues. Here, we ask how chaperone expression is regulated in a cell type-specific manner and whether cellular differentiation affects chaperone expression. Our bioinformatics analyses show that the myogenic transcription factor HLH-1 (MyoD) can bind to the promoters of chaperone genes expressed or required for the folding of muscle proteins. To test this experimentally, we employed HLH-1 myogenic potential to genetically modulate cellular differentiation of Caenorhabditis elegans embryonic cells by ectopically expressing HLH-1 in all cells of the embryo and monitoring chaperone expression. We found that HLH-1-dependent myogenic conversion specifically induced the expression of putative HLH-1-regulated chaperones in differentiating muscle cells. Moreover, disrupting the putative HLH-1-binding sites on ubiquitously expressed daf-21(Hsp90) and muscle-enriched hsp-12.2(sHsp) promoters abolished their myogenic-dependent expression. Disrupting HLH-1 function in muscle cells reduced the expression of putative HLH-1-regulated chaperones and compromised muscle proteostasis during and after embryogenesis. In turn, we found that modulating the expression of muscle chaperones disrupted the folding and assembly of muscle proteins and thus, myogenesis. Moreover, muscle-specific over-expression of the DNAJB6 homolog DNJ-24, a limb-girdle muscular dystrophy-associated chaperone, disrupted the muscle chaperone network and exposed synthetic motility defects. We propose that cellular differentiation could establish a proteostasis network dedicated to the folding and maintenance of the muscle proteome. Such cell-specific proteostasis networks can explain the selective vulnerability that many diseases of protein misfolding exhibit even when the misfolded protein is ubiquitously expressed.

Molecular chaperones protect proteins from misfolding and aggregation. In multi-cellular organisms, the composition and expression levels of chaperones vary between tissues. However, little is known of how such differential expression is regulated. We hypothesized that the cellular differentiation that regulates the cell-type specific expression program may be involved in establishing a cell-type specific chaperone network. To test this possibility, we addressed the myogenic commitment transcription factor HLH-1 (CeMyoD) that converts embryonic cells to muscle cells in Caenorhabditis elegans. We demonstrated that HLH-1 regulates the expression of muscle chaperones during muscle differentiation. Moreover, we showed that HLH-1-dependent expression of chaperones is required for embryonic development and muscle function. We propose that cellular differentiation results in cell-specific differences in the chaperone network that may be detrimental in terms of the susceptibility of neurons and muscle cells to protein misfolding diseases.

Overlapping cell population expression profiling and regulatory inference in C. elegans

Background

Understanding gene expression across the diverse metazoan cell types during development is critical to understanding their function and regulation. However, most cell types have not been assayed for expression genome-wide.

Results

We applied a novel approach we term “Profiling of Overlapping Populations of cells (POP-Seq)” to assay differential expression across all embryonic cells in the nematode Caenorhabditis elegans. In this approach, we use RNA-seq to define the transcriptome of diverse partially overlapping FACS-sorted cell populations. This identified thousands of transcripts differentially expressed across embryonic cells. Hierarchical clustering analysis identified over 100 sets of coexpressed genes corresponding to distinct patterns of cell type specific expression. We identified thousands of candidate regulators of these clusters based on enrichment of transcription factor motifs and experimentally determined binding sites.

Conclusions

Our analysis provides new insight into embryonic gene regulation, and provides a resource for improving our knowledge of tissue-specific expression and its regulation throughout C. elegans development.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-016-2482-z) contains supplementary material, which is available to authorized users.

Collaborative regulation of development but independent control of metabolism by two epidermis-specific transcription factors in Caenorhabditis elegans

Cell fate specification is typically initiated by a master regulator, which is relayed by tissue-specific regulatory proteins (usually transcription factors) for further enforcement of cell identities, but how the factors are coordinated among each other to "finish up" the specification remains poorly understood. Caenorhabditis elegans epidermis specification is initiated by a master regulator, ELT-1, that activates its targets, NHR-25 and ELT-3, two epidermis-specific transcription factors that are important for development but not for initial specification of epidermis, thus providing a unique paradigm for illustrating how the tissue-specific regulatory proteins work together to enforce cell fate specification. Here we addressed the question through contrasting genome-wide in vivo binding targets between NHR-25 and ELT-3. We demonstrate that the two factors bind discrete but conserved DNA motifs, most of which remain in proximity, suggesting formation of a complex between the two. In agreement with this, gene ontology analysis of putative target genes suggested differential regulation of metabolism but coordinated control of epidermal development between the two factors, which is supported by quantitative analysis of expression of their specific or common targets in the presence or absence of either protein. Functional validation of a subset of the target genes showed both activating and inhibitory roles of NHR-25 and ELT-3 in regulating their targets. We further demonstrated differential control of specification of AB and C lineage-derived epidermis. The results allow us to assemble a comprehensive gene network underlying C. elegans epidermis development that is likely to be widely used across species and provides insights into how tissue-specific transcription factors coordinate with one another to enforce cell fate specification initiated by its master regulator.

Integrative analysis of C. elegans modENCODE ChIP-seq data sets to infer gene regulatory interactions

The C. elegans modENCODE Consortium has defined in vivo binding sites for a large array of transcription factors by ChIP-seq. In this article, we present examples that illustrate how this compendium of ChIP-seq data can drive biological insights not possible with analysis of individual factors. First, we analyze the number of independent factors bound to the same locus, termed transcription factor complexity, and find that low-complexity sites are more likely to respond to altered expression of a single bound transcription factor. Next, we show that comparison of binding sites for the same factor across developmental stages can reveal insight into the regulatory network of that factor, as we find that the transcription factor UNC-62 has distinct binding profiles at different stages due to distinct cofactor co-association as well as tissue-specific alternative splicing. Finally, we describe an approach to infer potential regulators of gene expression changes found in profiling experiments (such as DNA microarrays) by screening these altered genes to identify significant enrichment for targets of a transcription factor identified in ChIP-seq data sets. After confirming that this approach can correctly identify the upstream regulator on expression data sets for which the regulator was previously known, we applied this approach to identify novel candidate regulators of transcriptional changes with age. The analysis revealed nine candidate aging regulators, of which three were previously known to have a role in longevity. We experimentally showed that two of the new candidate aging regulators can extend lifespan when overexpressed, indicating that this approach can identify novel functional regulators of complex processes.