bioRxiv : the preprint server for biology最新文献

Aberrant expression of the BAP1 tumor suppressor gene is a prominent risk factor for several tumor types and is important in tumor evolution and progression. Here we performed integrated multi-omic analyses using data from The Cancer Genome Atlas (TCGA) for 33 cancer types and over 10,000 individuals to identify alterations leading to BAP1 disruption. We combined existing variant calls and new calls derived from a de novo local realignment pipeline across multiple independent variant callers, increasing somatic variant detection by 41% from 182 to 257, including 11 indels ≥40bp. The expanded detection of mutations highlights the power of new tools to uncover longer indels and impactful mutations. We developed an expression-based BAP1 activity score and identified a transcriptional profile associated with BAP1 disruption in cancer. BAP1 has been proposed to play a critical role in controlling tumor plasticity and normal cell fate. Leveraging human and mouse liver datasets, BAP1 loss in normal cells resulted in lower BAP1 activity scores and lower scores were associated with a less-differentiated phenotype in embryonic cells. Together, our expanded BAP1 mutant samples revealed a transcriptional signature in cancer cells, supporting BAP1's influences on cellular plasticity and cell identity maintenance.

Following prolonged activity blockade, amplitudes of miniature excitatory postsynaptic currents (mEPSCs) increase, a form of plasticity termed "homeostatic synaptic plasticity." We previously showed that a presynaptic protein, the small GTPase Rab3A, is required for full expression of the increase in miniature endplate current amplitudes following prolonged blockade of action potential activity at the mouse neuromuscular junction in vivo (Wang et al., 2011), but it is unknown whether this form of Rab3A-dependent homeostatic plasticity shares any characteristics with central synapses. We show here that homeostatic synaptic plasticity of mEPSCs is impaired in mouse cortical neuron cultures prepared from Rab3A^-/- and mutant mice expressing a single point mutation of Rab3A, Rab3A Earlybird mice. To determine if Rab3A is involved in the well-established homeostatic increase in postsynaptic AMPA-type receptors (AMPARs), we performed a series of experiments in which electrophysiological recordings of mEPSCs and confocal imaging of synaptic AMPAR immunofluorescence were assessed within the same cultures. We found that Rab3A was required for the increase in synaptic AMPARs following prolonged activity blockade, but the increase in mEPSC amplitudes was not always accompanied by an increase in postsynaptic AMPAR levels, suggesting other factors may contribute. Finally, we demonstrate that Rab3A is acting in neurons because only selective loss of Rab3A in neurons, not glia, disrupted the homeostatic increase in mEPSC amplitudes. This is the first demonstration that neuronal Rab3A is required for homeostatic synaptic plasticity and that it does so partially through regulation of the surface expression of AMPA receptors.

Huntington's disease (HD) and spinocerebellar ataxia type 3 (SCA3) are the two most prevalent polyglutamine (polyQ) neurodegenerative diseases, caused by CAG (encoding glutamine) repeat expansion in the coding region of the huntingtin (HTT) and ataxin-3 (ATXN3) proteins, respectively. We have earlier reported that the activity, but not the protein level, of an essential DNA repair enzyme, polynucleotide kinase 3'-phosphatase (PNKP), is severely abrogated in both HD and SCA3 resulting in accumulation of double-strand breaks in patients' brain genome. While investigating the mechanistic basis for the loss of PNKP activity and accumulation of DNA double-strand breaks leading to neuronal death, we observed that PNKP interacts with the nuclear isoform of 6-phosphofructo-2-kinase fructose-2,6-bisphosphatase 3 (PFKFB3). Depletion of PFKFB3 markedly abrogates PNKP activity without changing its protein level. Notably, the levels of both PFKFB3 and its product fructose-2,6 bisphosphate (F2,6BP), an allosteric modulator of glycolysis, are significantly lower in the nuclear extracts of post-mortem brain tissues of HD and SCA3 patients. Supplementation of F2,6BP restored PNKP activity in the nuclear extracts of patients' brain. Moreover, intracellular delivery of F2,6BP restored both the activity of PNKP and the integrity of transcribed genome in neuronal cells derived from striatum of HD mouse. Importantly, supplementing F2,6BP rescued the HD phenotype in Drosophila, suggesting F2,6BP to serve in vivo as a cofactor for the proper functionality of PNKP and thereby, of brain health. Our results thus provide a compelling rationale for exploring the therapeutic use of F2,6BP and structurally related compounds for treating polyQ diseases.

The size of subcellular structures must be tightly controlled to maintain normal cell function. Despite its importance, few studies have determined how the size of organelles or other structures is maintained during development, when cells are growing, dividing, and rearranging. The developing egg chamber is a powerful model in which to study the relative growth rates of subcellular structures. The egg chamber contains a cluster of sixteen germline cells, which are connected through intercellular bridges called ring canals. As the egg chamber grows, the germline cells and the ring canals that connect them increase in size. Here, we demonstrate that ring canal size scaling is related to lineage; the largest, "first born" ring canals increase in size at a relatively slower rate than ring canals derived from subsequent mitotic divisions. This lineage-based scaling relationship is maintained even if directed transport is reduced, ring canal size is altered, or in egg chambers with twice as many germline cells. Analysis of lines that produce larger or smaller mature eggs reveals different strategies could be used to alter final egg size.

Summary statement: Using the fruit fly egg chamber as a model, this study demonstrates that the size and scaling of germline intercellular bridges vary based on lineage.

The prevalence of multidrug resistant (MDR) bacterial infections continues to rise as the development of antibiotics needed to combat these infections remains stagnant. MDR enterococci are a major contributor to this crisis. A potential therapeutic approach for combating MDR enterococci is bacteriophage (phage) therapy, which uses lytic viruses to infect and kill pathogenic bacteria. While phages that lyse some strains of MDR enterococci have been identified, other strains display high levels of resistance and the mechanisms underlying this resistance are poorly defined. Here, we use a CRISPR interference (CRISPRi) screen to identify a genetic locus found on a mobilizable plasmid from Enterococcus faecalis involved in phage resistance. This locus encodes a putative serine recombinase followed by a Type IV restriction enzyme (TIV-RE) that we show restricts the replication of phage phi47 in E. faecalis. We further find that phi47 evolves to overcome restriction by acquiring a missense mutation in a TIV-RE inhibitor protein. We show that this inhibitor, termed type IV restriction inhibiting factor A (tifA), binds and inactivates diverse TIV-REs. Overall, our findings advance our understanding of phage defense in drug-resistant E. faecalis and provide mechanistic insight into how phages evolve to overcome antiphage defense systems.

α- and β-tubulin form heterodimers, with GTPase activity, that assemble into microtubules. Like other GTPases, the nucleotide-bound state of tubulin heterodimers controls whether the molecules are in a biologically active or inactive state. While α-tubulin in the heterodimer is constitutively bound to GTP, β-tubulin can be bound to either GDP (GDP-tubulin) or GTP (GTP-tubulin). GTP-tubulin hydrolyzes its GTP to GDP following assembly into a microtubule and, upon disassembly, must exchange its bound GDP for GTP to participate in subsequent microtubule polymerization. Tubulin dimers have been shown to exhibit rapid intrinsic nucleotide exchange in vitro, leading to a commonly accepted belief that a tubulin guanine nucleotide exchange factor (GEF) may be unnecessary in cells. Here, we use quantitative binding assays to show that BuGZ, a spindle assembly factor, binds tightly to GDP-tubulin, less tightly to GTP-tubulin, and weakly to microtubules. We further show that BuGZ promotes the incorporation of GTP into tubulin using a nucleotide exchange assay. The discovery of a tubulin GEF suggests a mechanism that may aid rapid microtubule assembly dynamics in cells.

Membrane potential is a property of all living cells¹. Nevertheless, its physiological role in non-excitable cells is poorly understood. Resting membrane potential is typically considered fixed and under tight homeostatic control². Contrary to this paradigm, we find that membrane potential is a dynamic property that directly reflects mechanical forces acting on the cell and that cells use membrane potential to assess their biomechanical state. We show that several important mechano-sensitive signal transduction pathways, like MAPK and Hippo^3-9, are directly controlled by membrane potential and this signaling is mediated by upstream membrane-bound receptors, including FAT1. We further show that mechano-transduction via membrane potential plays a critical role in the homeostasis of epithelial tissues, setting cellular biomass density and cell number density by controlling proliferation and cell elimination. In epithelial scratch wound assays, as well as Xenopus tadpole tail regeneration, we observe a wave of depolarization caused by a drop in cellular biomass density due to mechanical stretch and we show that this depolarization wave is critical for wound closure. Together, these data are explained by a first-principles biophysical model, which demonstrates that membrane potential is physically coupled to mechanical pressure and cellular biomass density. Membrane potential thereby provides a quasi-instantaneous, global readout of the biophysical state of the cell and in turn regulates cell growth, resulting in homeostatic feedback control of biomass density and cell number density in tissues. This interplay may be an ancient mechanism for growth control in multi-cellular organisms and its misregulation may play an important role in tumorigenesis.

Understanding protein function and developing molecular therapies require deciphering the cell types in which proteins act as well as the interactions between proteins. However, modeling protein interactions across biological contexts remains challenging for existing algorithms. Here, we introduce Pinnacle, a geometric deep learning approach that generates context-aware protein representations. Leveraging a multi-organ single-cell atlas, Pinnacle learns on contextualized protein interaction networks to produce 394,760 protein representations from 156 cell type contexts across 24 tissues. Pinnacle's embedding space reflects cellular and tissue organization, enabling zero-shot retrieval of the tissue hierarchy. Pretrained protein representations can be adapted for downstream tasks: enhancing 3D structure-based representations for resolving immuno-oncological protein interactions, and investigating drugs' effects across cell types. Pinnacle outperforms state-of-the-art models in nominating therapeutic targets for rheumatoid arthritis and inflammatory bowel diseases, and pinpoints cell type contexts with higher predictive capability than context-free models. Pinnacle's ability to adjust its outputs based on the context in which it operates paves way for large-scale context-specific predictions in biology.

As bacterial symbionts transition from a motile free-living state to a sessile biofilm state, they must coordinate behavior changes suitable to each lifestyle. Cyclic diguanylate (c-di-GMP) is an intracellular signaling molecule that can regulate this transition, and it is synthesized by diguanylate cyclase (DGC) enzymes and degraded by phosphodiesterase (PDE) enzymes. Generally, c-di-GMP inhibits motility and promotes biofilm formation. While c-di-GMP and the enzymes that contribute to its metabolism have been well-studied in pathogens, considerably less focus has been placed on c-di-GMP regulation in beneficial symbionts. Vibrio fischeri is the sole beneficial symbiont of the Hawaiian bobtail squid (Euprymna scolopes) light organ, and the bacterium requires both motility and biofilm formation to efficiently colonize. C-di-GMP regulates swimming motility and cellulose exopolysaccharide production in V. fischeri. The genome encodes 50 DGCs and PDEs, and while a few of these proteins have been characterized, the majority have not undergone comprehensive characterization. In this study, we use protein overexpression to systematically characterize the functional potential of all 50 V. fischeri proteins. All 28 predicted DGCs and 14 predicted PDEs displayed at least one phenotype consistent with their predicted function, and a majority of each displayed multiple phenotypes. Finally, active site mutant analysis of proteins with the potential for both DGC and PDE activities revealed potential activities for these proteins. This work presents a systems-level functional analysis of a family of signaling proteins in a tractable animal symbiont and will inform future efforts to characterize the roles of individual proteins during lifestyle transitions.

The fusion of traditional chemical descriptors with Graph Neural Networks (GNNs) offers a compelling strategy for enhancing ligand-based virtual screening methodologies. A comprehensive evaluation revealed that the benefits derived from this integrative strategy vary significantly among different GNNs. Specifically, while GCN and SchNet demonstrate pronounced improvements by incorporating descriptors, SphereNet exhibits only marginal enhancement. Intriguingly, despite SphereNet's modest gain, all three models-GCN, SchNet, and SphereNet-achieve comparable performance levels when leveraging this combination strategy. This observation underscores a pivotal insight: sophisticated GNN architectures may be substituted with simpler counterparts without sacrificing efficacy, provided that they are augmented with descriptors. Furthermore, our analysis reveals a set of expert-crafted descriptors' robustness in scaffold-split scenarios, frequently outperforming the combined GNN-descriptor models. Given the critical importance of scaffold splitting in accurately mimicking real-world drug discovery contexts, this finding accentuates an imperative for GNN researchers to innovate models that can adeptly navigate and predict within such frameworks. Our work not only validates the potential of integrating descriptors with GNNs in advancing ligand-based virtual screening but also illuminates pathways for future enhancements in model development and application. Our implementation can be found at https://github.com/meilerlab/gnn-descriptor.