Learning Processes in Hierarchical Pairs Regulate Entire Gene Expression in Cells

Expression of numerous genes is precisely controlled in a cell in various contexts. While genetic and epigenetic mechanisms contribute to this regulation, how each mechanism cooperates to ensure the proper expression patterns of whole gene remains unclear. Here, I theoretically show that the repetition of simple biological processes makes appropriate whole-gene expression only if the appropriateness of current pattern is roughly detectable. A learning pair model is developed, in which two factors autonomously approach the target ratio by repeating two stochastic processes; competitive amplification with a small addition term and decay depending on the difference between the current and target ratios. Furthermore, thousands of factors are self-regulated in a hierarchical-pair architecture, in which the activation degrees competitively amplify, while transducing the activation signal, and decay at four different probabilities. Changes in whole-gene expression during human early embryogenesis and hematopoiesis are reproduced in simulation using this epigenetic learning process in a single genetically-determined hierarchical-pair architecture of gene regulatory cascades. On the background of this learning process, I propose the law of biological inertia which means that a living cell basically maintains the expression pattern while renewing the contents.

Gene Regulation Analysis Reveals Perturbations of Autism Spectrum Disorder during Neural System Development

Autism spectrum disorder (ASD) is a neurodevelopmental disorder that impedes patients’ cognition, social, speech and communication skills. ASD is highly heterogeneous with a variety of etiologies and clinical manifestations. The prevalence rate of ASD increased steadily in recent years. Presently, molecular mechanisms underlying ASD occurrence and development remain to be elucidated. Here, we integrated multi-layer genomics data to investigate the transcriptome and pathway dysregulations in ASD development. The RNA sequencing (RNA-seq) expression profiles of induced pluripotent stem cells (iPSCs), neural progenitor cells (NPCs) and neuron cells from ASD and normal samples were compared in our study. We found that substantially more genes were differentially expressed in the NPCs than the iPSCs. Consistently, gene set variation analysis revealed that the activity of the known ASD pathways in NPCs and neural cells were significantly different from the iPSCs, suggesting that ASD occurred at the early stage of neural system development. We further constructed comprehensive brain- and neural-specific regulatory networks by incorporating transcription factor (TF) and gene interactions with long 5 non-coding RNA(lncRNA) and protein interactions. We then overlaid the transcriptomes of different cell types on the regulatory networks to infer the regulatory cascades. The variations of the regulatory cascades between ASD and normal samples uncovered a set of novel disease-associated genes and gene interactions, particularly highlighting the functional roles of ELF3 and the interaction between STAT1 and lncRNA ELF3-AS 1 in the disease development. These new findings extend our understanding of ASD and offer putative new therapeutic targets for further studies.

Kinetic networks identify key regulatory nodes and transcription factor functions in early adipogenesis

Sequence-specific transcription factors (TFs) bind DNA, modulate chromatin structure, regulate gene expression, and orchestrate transcription cascades. Activation and repression of TFs drive tightly controlled regulatory programs that lead to cellular processes such as differentiation. We measured chromatin accessibility and nascent transcription at seven time points over the first four hours of induced adipogenesis of 3T3-L1 mouse preadipocytes to construct dynamic gene regulatory networks. Regulatory networks describe successive waves of TF binding and dissociation followed by direct regulation of proximal genes. We identified 14 families of TFs that coordinate with and antagonize each other to regulate early adipogenesis. We developed a compartment model to quantify individual TF contributions to RNA polymerase initiation and pause release rates. Network analysis showed that the glucocorticoid receptor and AP1 drive immediate gene activation, including induction of Twist2. Twist2 is a highly interconnected node within the network and its expression leads to repression of target genes. Although Twist2's role in adipogenesis has not been previously appreciated, both Twist2 knockout mice and Setleis syndrome (Twist2-/-) patients lack subcutaneous and brown adipose tissue. We found that kinetic networks integrating chromatin structure and nascent transcription dynamics identify key genes, TF functions, and coordinate interactions within regulatory cascades.

Differential nuclear import sets the timing of protein access to the embryonic genome

The development of a fertilized egg to an embryo requires the proper temporal control of gene expression. During cell differentiation, timing is often controlled via cascades of transcription factors (TFs). However, in early development, transcription is often inactive, and many TF levels are constant, suggesting that unknown mechanisms govern the observed rapid and ordered onset of gene expression. Here, we find that in early embryonic development, access of maternally deposited nuclear proteins to the genome is temporally ordered via importin affinities, thereby timing the expression of downstream targets. We quantify changes in the nuclear proteome during early development and find that nuclear proteins, such as TFs and RNA polymerases, enter nuclei sequentially. Moreover, we find that the timing of the access of nuclear proteins to the genome corresponds to the timing of downstream gene activation. We show that the affinity of proteins to importin is a major determinant in the timing of protein entry into embryonic nuclei. Thus, we propose a mechanism by which embryos encode the timing of gene expression in early development via biochemical affinities. This process could be critical for embryos to organize themselves before deploying the regulatory cascades that control cell identities.

Proton Pumping and Non-Pumping Terminal Respiratory Oxidases: Active Sites Intermediates of These Molecular Machines and Their Derivatives

Terminal respiratory oxidases are highly efficient molecular machines. These most important bioenergetic membrane enzymes transform the energy of chemical bonds released during the transfer of electrons along the respiratory chains of eukaryotes and prokaryotes from cytochromes or quinols to molecular oxygen into a transmembrane proton gradient. They participate in regulatory cascades and physiological anti-stress reactions in multicellular organisms. They also allow microorganisms to adapt to low-oxygen conditions, survive in chemically aggressive environments and acquire antibiotic resistance. To date, three-dimensional structures with atomic resolution of members of all major groups of terminal respiratory oxidases, heme-copper oxidases, and bd-type cytochromes, have been obtained. These groups of enzymes have different origins and a wide range of functional significance in cells. At the same time, all of them are united by a catalytic reaction of four-electron reduction in oxygen into water which proceeds without the formation and release of potentially dangerous ROS from active sites. The review analyzes recent structural and functional studies of oxygen reduction intermediates in the active sites of terminal respiratory oxidases, the features of catalytic cycles, and the properties of the active sites of these enzymes.

Ratiometric RNA labeling allows dynamic multiplexed analysis of gene circuits in single cells

ABSTRACTBiological processes are highly dynamic and are regulated by genes that connect with one and another, forming regulatory circuits and networks. Understanding how gene regulatory circuits operate dynamically requires monitoring the expression of multiple genes in the same cell. However, it is limited by the relatively few distinguishable fluorescent proteins. Here, we developed a multiplexed real-time transcriptional imaging method based on two RNA stem-loop binding proteins, and employed it to analyze the temporal dynamics of synthetic gene circuits. By incorporating different ratios of MS2 and PP7 stem-loops, we were able to monitor the real-time nascent transcriptional activities of up to five genes in the same cell using only two fluorescent proteins. Applying this multiplexing capability to synthetic linear or branched gene regulatory cascades revealed that propagation of transcriptional dynamics is enhanced by non-stationary dynamics and is dictated by the slowest regulatory branch in the presence of combinatorial regulation. Mathematical modeling provided further insight into temporal multi-gene interactions and helped to understand potential challenges in regulatory inference using snapshot single-cell data. Ratiometric multiplexing should scale exponentially with additional labelling channels, providing a way to track the dynamics of larger circuits.

Limited Association between Schizophrenia Genetic Risk Factors and Transcriptomic Features

Schizophrenia is a polygenic disorder with many genomic regions contributing to schizophrenia risk. The majority of genetic variants associated with schizophrenia lie in the non-coding genome and are thought to contribute to transcriptional regulation. Extensive transcriptomic dysregulation has been detected from postmortem brain samples of schizophrenia-affected individuals. However, the relationship between schizophrenia genetic risk factors and transcriptomic features has yet to be explored. Herein, we examined whether varying gene expression features, including differentially expressed genes (DEGs), co-expression networks, and central hubness of genes, contribute to the heritability of schizophrenia. We leveraged quantitative trait loci and chromatin interaction profiles to identify schizophrenia risk variants assigned to the genes that represent different transcriptomic features. We then performed stratified linkage disequilibrium score regression analysis on these variants to estimate schizophrenia heritability enrichment for different gene expression features. Notably, DEGs and co-expression networks showed nominal heritability enrichment. This nominal association can be partly explained by cellular heterogeneity, as DEGs were associated with the genetic risk of schizophrenia in a cell type-specific manner. Moreover, DEGs were enriched for target genes of schizophrenia-associated transcription factors, suggesting that the transcriptomic signatures of schizophrenia are the result of transcriptional regulatory cascades elicited by genetic risk factors.

smartPARE: An R Package for Efficient Identification of True mRNA Cleavage Sites

Degradome sequencing is commonly used to generate high-throughput information on mRNA cleavage sites mediated by small RNAs (sRNA). In our datasets of potato (Solanum tuberosum, St) and Phytophthora infestans (Pi), initial predictions generated high numbers of cleavage site predictions, which highlighted the need of improved analytic tools. Here, we present an R package based on a deep learning convolutional neural network (CNN) in a machine learning environment to optimize discrimination of false from true cleavage sites. When applying smartPARE to our datasets on potato during the infection process by the late blight pathogen, 7.3% of all cleavage windows represented true cleavages distributed on 214 sites in P. infestans and 444 sites in potato. The sRNA landscape of the two organisms is complex with uneven sRNA production and cleavage regions widespread in the two genomes. Multiple targets and several cases of complex regulatory cascades, particularly in potato, was revealed. We conclude that our new analytic approach is useful for anyone working on complex biological systems and with the interest of identifying cleavage sites particularly inferred by sRNA classes beyond miRNAs.

Identification of developmentally important genes in Silene latifolia through chemical genetics and transcriptome profiling

Dioecious plants possess diverse sex determination systems and unique mechanisms of reproductive organ development; however, little is known about how sex-linked genes shape the expression of regulatory cascades that lead to developmental differences between sexes. In Silene latifolia, a dioecious plant with stable dimorphism in floral traits, early experiments suggested that female-regulator genes act on the factors that determine the boundaries of the flower whorls. To identify these regulators, we sequenced the transcriptome of male flowers with fully developed gynoecia induced by rapid demethylation in the parental generation. As the hermaphrodite flower trait is holandric (transmitted only from male to male, inherited on the Y chromosome), we screened for genes that are differentially expressed between male, female, and hermaphrodite flowers. Dozens of candidate genes that are upregulated in hermaphrodite flowers compared to male and female flowers were detected and found to have putative roles in floral organization, affecting the expression of floral MADS-box and other genes. Amongst these genes, eight candidates were found to promote gynoecium formation in female and hermaphrodite flowers, affecting organ size, whorl boundary, and the expression of mainly B class flower genes. To complement our transcriptome analysis, we closely examined the floral organs in their native state using a field emission environmental scanning electron microscope. Our results reveal the principal regulatory pathways involved in sex-specific flower development in the classical model of dioecy, S. latifolia.