Molecular Systems Biology最新文献

Synthetic biology approaches such as whole-cell biosensing and 'sense-and-respond' therapeutics aim to enlist the vast sensing repertoire of gut microbes to drive cutting-edge clinical and research applications. However, well-characterised circuit components that sense health- and disease-relevant conditions within the gut remain limited. Here, we extend the flexibility and power of a biosensor screening platform using bacterial memory circuits. We construct libraries of sensory components sourced from diverse gut bacteria using a bespoke two-component system identification and cloning pipeline. Tagging unique strains using a hypervariable DNA barcode enables parallel tracking of thousands of unique clones, corresponding to ~150 putative biosensors, in a single experiment. Evaluating sensor activity and performance heterogeneity across various in vitro and in vivo conditions using mouse models, we identify several biosensors of interest. Validated hits include biosensors with relevance for autonomous control of synthetic functions within the mammalian gut and for non-invasive monitoring of inflammatory disease using faecal sampling. This approach will promote rapid biosensor engineering to advance the development of synthetic biology tools for deployment within complex environments.

Precision oncology led to the establishment and widespread application of molecular tumor boards (MTBs)-multidisciplinary units combining molecular and clinical assessment of individual cancer cases for swift selection of personalized treatments. Whole-exome or gene panel sequencing, combined with transcriptomic, immunohistochemical, and other molecular analyses, often permits dissection of molecular drivers of a tumor and identification of its potential targetable vulnerabilities, instructing clinical oncologists on sometimes unconventional treatment options. However, cancer drivers are often unleashed mutation-independently, especially in breast and gynecological cancers, and deleterious mutations are not always pathogenic. To complement the MTB arsenal, we chart here the molecular toolset we call Signalomics that permits fast and robust assessment of a panel of oncogenic signaling pathways in fresh tumor samples. Using transcriptional reporters introduced in primary tumor cells, this approach identifies the pathways overactivated in a given tumor and validates their sensitivity to targeted therapies, providing actionable insights for personalized treatment strategies. Integration of Signalomics into MTB workflows bridges the gap between molecular profiling and functional pathway analysis, refining clinical treatment decisions and advancing precision oncology.

Gene gains and losses are a major driver of genome evolution; their precise characterization can provide insights into the origin and diversification of major lineages. Here, we examined gene family evolution of 1154 genomes from nearly all known species in the medically and technologically important yeast subphylum Saccharomycotina. We found that yeast gene family evolution differs from that of plants, animals, and filamentous ascomycetes, and is characterized by smaller overall gene numbers yet larger gene family sizes for a given gene number. Faster-evolving lineages (FELs) in yeasts experienced significantly higher rates of gene losses-commensurate with a narrowing of metabolic niche breadth-but higher speciation rates than their slower-evolving sister lineages (SELs). Gene families most often lost are those involved in mRNA splicing, carbohydrate metabolism, and cell division and are likely associated with intron loss, metabolic breadth, and non-canonical cell cycle processes. Our results highlight the significant role of gene family contractions in the evolution of yeast metabolism, genome function, and speciation, and suggest that gene family evolutionary trajectories have differed markedly across major eukaryotic lineages.

Large-scale metabolomic analyses of pan-cancer cell line panels have provided significant insights into the relationships between metabolism and cancer cell biology. Here, we took a pathway-centric approach by transforming targeted metabolomic data into ratios to study associations between reactant and product metabolites in a panel of cancer and non-cancer cell lines. We identified five clusters of cells from various tissue origins. Of these, cells in Cluster 4 had high ratios of TCA cycle metabolites relative to pyruvate, produced more lactate yet consumed less glucose and glutamine, and greater OXPHOS activity compared to Cluster 3 cells with low TCA cycle metabolite ratios. This was due to more glutamine cataplerotic efflux and not glycolysis in cells of Cluster 4. In silico analyses of loss-of-function and drug sensitivity screens showed that Cluster 4 cells were more susceptible to gene deletion and drug targeting of glutamine metabolism and OXPHOS than cells in Cluster 3. Our results highlight the potential of pathway-centric approaches to reveal new aspects of cellular metabolism from metabolomic data.

In response to infection or vaccination, lymph nodes must select antigen-reactive B-cells while eliminating auto-reactive B-cells. B-cells are instructed via B-cell receptor (BCR), which binds antigen, and CD40 receptor by antigen-recognizing T-cells. How BCR and CD40 signaling are integrated quantitatively to jointly determine B-cell fate decisions remains unclear. Here, we developed a differential-equations-based model of BCR and CD40 signaling networks activating NFκB. The model recapitulates NFκB dynamics upon BCR and CD40 stimulation, and when linked to established cell decision models of cell cycle and survival control, the resulting cell population dynamics. However, upon costimulation, NFκB dynamics were correctly predicted but the predicted potentiated population expansion was not observed experimentally. We found that this discrepancy was due to BCR-induced caspase activity that may trigger apoptosis in founder cells, unless timely NFκB-induced survival gene expression protects them. Iterative model predictions and sequential co-stimulation experiments revealed how complex non-monotonic integration of BCR and CD40 signals controls positive and negative selection of B-cells. Our work suggests a temporal proof-reading mechanism for regulating the stringency of B-cell selection during antibody responses.

Chemical and genetic perturbations, such as those induced by small molecules and CRISPR, effect complex changes in the molecular states of cells. Despite advances in high-throughput single-cell perturbation screening technology, the space of possible perturbations is far too large to measure exhaustively. Here, we introduce PerturbNet, a flexible deep generative model designed to predict the distribution of cell states induced by unseen chemical or genetic perturbations. PerturbNet accurately predicts gene expression changes in response to unseen small molecules based on their chemical structures while also accounting for key covariates such as dosage and cell type. Moreover, PerturbNet accurately predicts the distribution of single-cell gene expression states following CRISPR activation or CRISPR interference by leveraging gene functional annotations. Our approach significantly outperforms previous methods, particularly for predicting the effects of perturbing completely unseen genes. Finally, we demonstrate for the first time that amino acid sequence embeddings can be used to predict gene expression changes induced by missense mutations. We use PerturbNet to predict the effects of all point mutations in GATA1 and nominate variants that significantly impact the cell state distribution of human hematopoietic stem cells. Using a crystal structure of GATA1 bound to DNA, we validate that these large-effect variants occur in the core DNA-contact region of GATA1 and tend to involve large changes in amino acid side-chain volume.

The gut is exposed to a wide range of proteins, including ingested proteins and those produced by the resident microbiota. While ingested prion-like proteins can propagate across species, their implications for disease development remain largely unknown. Here, we apply a multidisciplinary approach to examine the relationship between the biophysical properties of exogenous prion-like proteins and the phenotypic consequences of ingesting them. Through computational analysis of gut bacterial proteins, we identified an enrichment of prion-like sequences in Helicobacter pylori. Based on these findings, we rationally designed a set of synthetic prion-like sequences that form amyloid fibrils, interfere with amyloid-beta-peptide aggregation, and trigger prion propagation when introduced in the yeast Sup35 model. When C. elegans were fed bacteria expressing these prion-like proteins, they lost associative memory and exhibited increased lipid oxidation. These data suggest a link between memory impairment, the conformational state of aggregates, and oxidative stress. Overall, this work supports gut microbiota as a reservoir of exogenous prion-like sequences, especially H. pylori, and the gut as an entry point for molecules capable of triggering cognitive dysfunction.

Proteins operate within dense interconnected networks, with interactions necessary both for stabilising proteins and enabling them to execute their molecular functions. Remarkably, protein-protein interaction networks operating within tumour cells continue to function despite widespread genetic perturbations. Previous work has demonstrated that tumour cells tolerate perturbations of paralogs better than perturbations of singleton genes, but the underlying mechanisms remain poorly understood. Here, we systematically profile the proteomic response of tumours and cell lines to gene loss. We find many examples of proteomic compensation, where loss of one gene causes increased abundance of a paralog, and collateral loss, where gene loss causes reduced paralog abundance. Compensation is enriched among paralog pairs that are central in the protein-protein interaction network and whose interaction partners perform essential functions. Compensation is also significantly more likely to be observed between synthetic lethal pairs. Our results support a model whereby loss of one gene results in increased protein abundance of its paralog, stabilising the protein-protein interaction network. Consequently, tumour cells may become dependent on the paralog for survival, creating potentially targetable vulnerabilities.

Kidney fibrosis, characterized by excessive extracellular matrix deposition, is a progressive disease that, despite affecting 10% of the population, lacks specific treatments and suitable biomarkers. This study presents a comprehensive, time-resolved multi-omics analysis of kidney fibrosis using an in vitro model system based on human kidney PDGFRβ⁺ mesenchymal cells aimed at unraveling disease mechanisms. Using transcriptomics, proteomics, phosphoproteomics, and secretomics, we quantified over 14,000 biomolecules across seven time points following TGF-β stimulation. This revealed distinct temporal patterns in the expression and activity of known and potential kidney fibrosis markers and modulators. Data integration resulted in time-resolved multi-omic network models which allowed us to propose mechanisms related to fibrosis progression through early transcriptional reprogramming. Using siRNA knockdowns and phenotypic assays, we validated predictions and regulatory mechanisms underlying kidney fibrosis. In particular, we show that several early-activated transcription factors, including FLI1 and E2F1, act as negative regulators of collagen deposition and propose underlying molecular mechanisms. This work advances our understanding of the pathogenesis of kidney fibrosis and provides a resource to be further leveraged by the community.