Biology subjects

Zhang, J.

Publications and source records attributed to Zhang, J..

At least 325 records · Page 18Linked to original sources

bioRxiv2026

Calcium directs actin assembly via allosteric activation of formin INF2

Calcium signals spatiotemporally orchestrate cytoskeletal dynamics, typically through Rho GTPase-mediated signaling cascades, yet the direct molecular transducers that convert local calcium transients into spatially controlled actin assembly have remained elusive. Here, we uncover a structure-based mechanism by which calcium-bound calmodulin (Ca2+-CaM) directly activates formin INF2 to drive actin assembly. We demonstrate that Ca2+-CaM binds to the diaphanous inhibitory domain (DID) of INF2 with nanomolar affinity, inducing allosteric conformational changes that sterically disrupt INF2 autoinhibition. The unexpected bipartite Ca2+-CaM binding interface on INF2 enables ultrasensitive decoding of local calcium microdomains, such as ER-mitochondrial contacts, where activated INF2 promotes actin polymerization to facilitate mitochondrial fission. We further show that the Charcot-Marie-Tooth disease-associated INF2 R91G mutation enhances Ca2+-CaM binding via optimized interfacial dynamics, suggesting a gain-of-function disease mechanism. Our work establishes the Ca2+-CaM-INF2 axis as a direct molecular transducer linking spatial calcium signals to actin-dependent organelle dynamics, defining a Rho-GTPase-independent paradigm for calcium-cytoskeleton communication, with broad implications for INF2-linked pathologies.

biochemistry↗

bioRxiv2026

A general and extensible algorithmic framework to biological sequence alignment across scales and applications

Sequence alignment underpins nearly every facet of modern genomics, from genetic testing and cancer profiling to functional genome annotation. Yet, despite decades of algorithmic innovation, most existing aligners remain narrowly optimized for specific tasks, fragmenting analytical workflows and limiting reproducibility. Here we introduce the Versatile Alignment Toolkit (VAT), a unified algorithmic framework that generalizes existing seeding and genome-indexing strategies within a single, transparent architecture. VAT employs a novel multi-view indexing scheme that integrates multiple seeding strategies and supports run-time seed-length parameterization without reindexing. A radix clustering algorithm based on hardware-efficient in-register bitonic sort accelerates multi-view table construction, ensuring scalability across large datasets. VAT delivers consistently high performance across diverse alignment tasks, including short- and long-read mapping, homology search, and whole-genome alignment, while maintaining algorithmic simplicity and adaptability. By bridging previously isolated alignment paradigms, VAT substantially reduces workflow complexity, enhances computational efficiency, and establishes an extensible foundation for future sequencing technologies. We anticipate this unification will set a new standard for flexible and reproducible sequence alignment in biomedical research and clinical genomics.

bioinformatics↗

bioRxiv2026

Mechanical Stiffening Promotes Growth, Invasion, and Mitogen-activated protein kinase kinase (MEK) Inhibitor Resistance in 3D Plexiform Neurofibroma Cultures

Plexiform neurofibromas with neurofibromatosis type I (pNF1s) are benign peripheral nerve sheath tumors caused by NF1 loss, leading to dysregulated RAS/mitogen-activated protein kinase (MAPK) signaling. While mitogen-activated protein kinase kinase (MEK) inhibitors, selumetinib and mirdametinib, can reduce tumor volume, surgical resection remains the primary treatment for immediate debulking and symptom relief. Complete removal is often limited by tumor infiltration along nerve plexuses, and residual tumors may undergo postsurgical tissue remodeling, producing localized regions of stiffened extracellular matrix (ECM). The impact of ECM stiffness on pNF1 progression and drug response is unknown. Using patient-derived pNF1 tumor cells cultured in 3D hydrogels with defined stiffness (1.5 kPa soft, 7 kPa stiff), we found that stiff ECM promoted spread morphology, increased growth, and progressive intracellular softening. Stiff ECM reduced lysyl oxidase (LOX) expression, reflecting mechanoadaptive ECM remodeling, and upregulated P-glycoprotein, leading to decreased sensitivity to selumetinib. These results provide the first evidence that ECM stiffening, such as that arising from postsurgical remodeling, directly drives pNF1 progression and therapeutic resistance. Our findings highlight mechanobiology as a key regulator of tumor behavior and support targeting ECM mechanics to improve clinical outcomes in NF1 patients.

cancer biology↗

bioRxiv2026

MDA5 multimerization on LINE RNA drives pathogenic extracellular immune complexes in autoimmunity

Autoantibodies are hallmarks of many autoimmune diseases, but their potential pathogenic roles, particularly for those targeting intracellular proteins, remain unclear. Anti-MDA5-positive dermatomyositis (anti-MDA5 DM) is characterized by autoantibodies against the intracellular protein MDA51,2, a conserved innate immune receptor that recognizes viral dsRNA by forming filaments3. Here, using four patient-derived monoclonal autoantibodies (mAbs), we reconstitute and define the molecular architecture, biogenesis, and immunological activity of pathogenic MDA5 immune complexes. Our cryo-EM analysis revealed that these mAbs bind dsRNA-scaffolded MDA5 filaments in at least two distinct binding modes, each exhibiting striking epitope convergence using germline-encoded residues. Extracellular immune complexes formed between mAbs and filamentous, but not monomeric, MDA5 potently activate multiple innate immune pathways, with the magnitude of activation determined by antibody binding mode and immune complex stoichiometry. Antibody bivalency further crosslinks MDA5 filaments into higher-order aggregates with heightened immunostimulatory activity, demonstrating an active role of autoantibodies in shaping immune complex architecture. Analysis of patient plasma reveals elevated levels of extracellular MDA5 filaments and identifies LINE retroelement-derived dsRNA as a structural scaffold. Notably, MDA5 immune complexes induce endogenous LINE dsRNA expression, likely promoting additional MDA5 filament formation and extracellular release through inflammatory cell death. These data thus support a self-amplifying inflammatory cycle as a pathogenic mechanism for anti-MDA5 DM. Collectively, our study defines a broadly applicable architectural principle, in which higher-order organization and binding modes of autoantibodies--beyond antibody affinity or nucleic acid presence alone--govern innate immune activation.

biochemistry↗

bioRxiv2026

Hidden causal inference delineates dynamic lncRNA regulation in autism spectrum disorder

Long non-coding RNAs (lncRNAs) are increasingly implicated in autism spectrum disorder (ASD), yet their causal regulatory mechanisms remain poorly characterized. Conventional static causal inference methods fail to capture the dynamic interplay of lncRNAs across diverse brain biological contexts. Here, we develop a hidden causality-based method Clean to infer dynamic lncRNA causal regulation in ASD. Our analysis reveals that dark causality, representing a hybrid form of positive and negative interdependencies, dominates lncRNA-mediated regulation in ASD. Moreover, dynamic lncRNA causal regulation emerges in diverse brain biological contexts, and 14 ASD risk lncRNAs significantly contribute to disease pathogenesis through disrupted gene regulatory programs. We also illustrate that lncRNA expression signatures enable robust classification of sex and cell type, offering potential diagnostic lncRNA biomarkers. Notably, 20 immune-related lncRNAs are prominently involved in the immune, suggesting a potential role for neuroimmune regulation. Additionally, cell-cell interaction networks diverge substantially between ASD and Normal cohorts. This study establishes a paradigm for delineating the dynamic causal regulation of lncRNAs, underscoring their diagnostic and therapeutic potential in ASD.

bioinformatics↗

bioRxiv2026

A brain dynamic model based on graph neural network reflect the inter-region interaction of cortical areas

A central objective in neuroscience is to elucidate how the brain generates complex dynamic activity through the interactions of brain areas. In this study, we utilized Interaction Network, a graph neural network model, to develop a computational framework for predicting whole-brain cortical blood oxygenation level dependent (BOLD) signals. We derived an Inter-Regional Interaction (IRI) metric to quantify information exchange among brain areas probing the underlying dynamical mechanisms. In addition, the total IRI emitted from each brain region was calculated and defined as the IRI sent by region (RS-IRI). Our model predicted the following 10 time points BOLD activity from initial BOLD signals, and achieved a mean absolute error of 0.04. The predicted functional connectivity (FC) achieves a correlation coefficient of 0.97 compared to the empirical FC. The fluctuation amplitude of the IRI increases with the length of the connection and the largest RS-IRI oscillation amplitude is observed in visual areas. The RS-IRI demonstrates a hierarchical organization, characterized by more concentrated distributions in association regions and larger fluctuation amplitudes in unimodal regions. Applying our approach to Alzheimers disease (AD), we demonstrate that the frequency-specific amplitudes of IRI oscillations discriminate AD patients from healthy controls and correlate with Mini-Mental State Examination scores. Together, this work presents a deep learning-based framework for modeling brain dynamics as well a quantitative index of inter-areal interactions, and offers a new perspective for disease characterization. Author SummaryThe human brain comprises distinct regions that interact through complex fiber tracts, forming the functional dynamics for diverse cognitive processes. We employed fMRI to assess functional activity and DTI to reconstruct fiber tract connectivity. To elucidate how brain function emerges from these inter-regional interactions, we developed a novel computational framework based on Graph Neural Network (GNN) to model the brains interactive dynamics for its capacity to uncover hidden and intricate patterns within data. From this model, we derived a quantitative metric termed Inter-Regional Interaction (IRI), which characterized the fine-grained, dynamic fluctuations in communication between brain areas. Our results suggest that this GNN-based model can accurately simulate brain functional activity and provide a quantitative description of neural interaction patterns. Applying this model to a cohort of Alzheimers disease patients, we demonstrated that the IRI metric not only effectively distinguished patients from healthy controls but also significantly correlated with clinical cognitive performance (MMSE scores). This approach advances our understanding of the fundamental principles of brain function and offers a promising tool for identifying the underlying mechanisms of neurological disorders.

neuroscience↗

bioRxiv2026

Sonogenetic control of cardiomyocytes and cardiac pacing using exogenous Transient Receptor Potential A1 channels

Electronic cardiac pacemakers are the standard of care for treating arrhythmias but, even for wireless pacemakers, they require an intracardiac implant procedure and are invasive. Lead-related complications, including life-threatening infections, and limited suitability in pediatric patients restrict pacemaker use. Biological pacing has emerged as a hardware-free alternative, leveraging cell reprogramming approaches to generate pacemaker-like cardiomyocytes. Despite progress, biological strategies face challenges related to complex signaling pathways, intercellular coupling, durability of effect, efficacy, and safety. In addition, once ion channels or cells are implanted, biological pacemakers offer limited real-time programmability, whereas electronic pacemakers enable continuous adjustment of pacing output based on cardiac performance. Here we propose a hybrid approach that combines genetic sensitization with external control by exploring the feasibility of sonogenetic cardiac pacing. We express the ultrasound-sensitive ion channel hsTRPA1 in cardiomyocytes and use noninvasive ultrasound to modulate channel activity and cardiac function. Using calcium imaging, we show that hsTRPA1 potentiates ultrasound-evoked responses in cardiomyocytes. Furthermore, cardiac expression of hsTRPA1 in mice increases heart rate in response to ultrasound delivered noninvasively through the intact chest, without evidence of a thermal mechanism under the conditions tested. Together, these results establish a proof of concept for sonogenetic cardiac pacing and define an initial acoustic and genetic parameter space compatible with preserving baseline cardiac function. More broadly, this work positions sonogenetics as a minimally invasive, wireless strategy for cardiac rhythm control and motivates future optimization and safety studies toward translation.

bioengineering↗

bioRxiv2026

Dysregulated RNA G-quadruplex binding proteins reveal shifted stress responses in Alzheimer's Disease

Single-cell sequencing has reshaped the research paradigm of Alzheimers Disease (AD) by revealing the heterogeneous transcriptional states of brain cells. Large single-cell cohort datasets, such as ROSMAP and SEA-AD, provide valuable atlases but primarily focus on RNA expression and chromatin accessibility, often overlooking the roles of RNA secondary structures. Although RNA G-quadruplexes (rG4s) are increasingly recognized as regulators of neurodegeneration, their interacting RNA-binding proteins (RBPs) remain poorly understood. We analyzed multiple independent scRNA-seq datasets to investigate rG4-associated biological functions across AD-relevant cell types and transcriptional states. We discovered that rG4-binding RBPs (RG4BPs) directly regulate essential glial cell functions, which are frequently impaired in AD. At the transcriptional state level, distinct sets of dysregulated RG4BPs correspond to the impaired state-specific biological features of astrocytes and microglia during AD progression, including a shift from an acute protective state to a chronic stress-associated state. This progressive state transition is accompanied by glial exhaustion and accumulated rG4s, which ultimately compromise glial support functions. We report several RG4BPs that are known regulators, including CIRBP, HSP90AA1, VIM, and PICALM, whose altered expression is associated with either pathological activation or severe functional impairment. These dysregulated RG4BPs provide a lens to examine the mechanism of rG4 accumulation in AD and position RG4BPs as novel targets for understanding and potentially intervening in AD progression.

neuroscience↗

bioRxiv2026

Molecular atlas of key food odorants reveals structured aroma organization and enables generative aroma design

Food aromas arise from complex combinations of odorants, yet how these combinations are organized across foods to define aroma identity remains unclear. Decoding this "aroma code" could help bridge the sensory gap between traditional foods and sustainable alternatives, which often struggle with off-notes or to replicate consumer-expected flavor profiles. Here we present KFO-Atlas, a molecular atlas of 896 key food odorants curated from 2,282 food aroma formulations. Analysis shows that food aromas are built from sparse and structured sets of odorants. Plant-derived foods span a broad and diverse aroma space, whereas animal-derived foods tend to exhibit more similar odorant sets. In specific cases, distinct plant- and animal-based foods converge on similar odorant compositions through shared reaction pathways. Building on these insights, we develop a generative AI model that produces category-targeted aroma formulations and validate its outputs by blinded human sensory evaluation. As a proof of principle, the model reconstructs meat-like aromas using exclusively plant-derived odorants, demonstrating a data-driven route to address sensory bottlenecks in sustainable food products.

systems biology↗

bioRxiv2026

Computational neural dynamics of goal-directed visual attention in macaques

Goal-directed visual attention requires the dynamic integration of task goals with perceptual and mnemonic processes across distributed cortical networks. Using large-scale recordings from V4, IT, OFC, and LPFC, we identified distinct neural populations selective for attention and category. Population dynamics robustly represented visual categories during cue presentation, sustained cue information across delays, and differentiated categories and attentional states during search. Cue-related activity predicted subsequent search efficiency, linking pre-search processing to behavioral performance. The orthogonal subspace provided a crucial latent representational structure for encoding and maintaining task-relevant information across search stages. Foveal attention enhanced peripheral representations by both increasing pattern separation and reshaping representational geometry in a non-linear, context-dependent manner. Search dynamics further reflected fixation history and target detection, which modulated both response strength and representational structure. Finally, V4 and IT encoded the spatial geometry of the search array, preserving its layout. Together, these findings highlight population-level dynamics as critical mechanisms supporting goal-directed visual search.

neuroscience↗

bioRxiv2026

Molecular Imaging of Collagen Turnover in Myocardial Infarction

Cardiac fibrosis is a key contributor to cardiomyopathy after myocardial infarction (MI). Existing imaging techniques can detect established fibrotic changes; however, they lack sensitivity for ongoing collagen turnover--a dynamic process involving the denaturation of collagen triple helix. Molecular imaging of this process could enhance risk assessment and aid in the development of anti-fibrotic treatments. This study aimed to evaluate 99mTc-(HE)-(GPO), a radiotracer designed to target denatured collagen, as a biomarker of collagen turnover after MI. Methods99mTc-(HE)-(GPO) incorporates glycine-proline-hydroxyproline (GPO) repeats and can hybridize with denatured single- or double-stranded collagen. MI was induced in mice by ligation of the left anterior descending artery; sham-operated animals served as controls. At 2 weeks post-MI, animals underwent myocardial perfusion imaging or contrast-enhanced CT to detect the infarct zone, followed by SPECT/CT imaging using 99mTc-(HE)-(GPO) or a control scrambled tracer. Tracer uptake was quantified in vivo and ex vivo with gamma counting and autoradiography. Different aspects of fibrosis were examined by tissue analysis, along with autoradiography with a matrix metalloproteinase-targeted radiotracer, 99mTc-RYM1. Tracer binding was also assessed in human cardiac tissue through ex vivo autoradiography. Results99mTc-(HE)-(GPO) SPECT/CT revealed significantly higher tracer uptake in the infarct zone of MI mice compared to the remote zone and sham controls (P < 0.0001 for both). Tracer uptake was confirmed by autoradiography, which showed a strong correlation between SPECT and autoradiography (R = 0.81, P < 0.01). The scrambled tracer exhibited minimal cardiac uptake, demonstrating the specificity of 99mTc-(HE)-(GPO) signal. Denatured collagen staining and 99mTc-RYM1 autoradiography showed similar patterns as ex vivo 99mTc-(HE)-(GPO) autoradiography, while the ratio of denatured collagen to procollagen in the infarct zone significantly increased from day 3 to 2 weeks post-MI. Finally, 99mTc-(HE)-(GPO) bound to human fibrotic (but not normal) cardiac tissue. Conclusion99mTc-(HE)-(GPO) enables non-invasive detection of denatured collagen as a marker of collagen remodeling in vivo, offering a promising tool for assessing fibrotic remodeling after MI. Collagen, procollagen, and denatured collagen, along with MMP activation, exhibit distinct patterns, and their combined imaging may provide a comprehensive molecular fingerprint of cardiac fibrosis, advancing personalized management of cardiomyopathy.

bioengineering↗

bioRxiv2026

Selective Immune Silencing by Targeted TGF-β Agonists

Depletion of pathogenic T and B cells is a pillar of therapies for autoimmune, inflammatory, and transplantation-related immunological diseases. However, adverse events, safety concerns in immunocompromised patients, and disease relapse limit clinical utility. Here, we exploit the immunosuppressive properties of a transforming growth factor beta (TGF-{beta}) mimic repurposed from helminths for cell type-specific therapeutic silencing, as a new approach to complement existing therapies. Mouse CD4 and CD8 T cell-targeted TGF-{beta} agonists selectively and potently silence antigen-specific T cell responses in OVA-immunized mice by suppressing pro-inflammatory effector, cytotoxic, and T follicular helper programs, while skewing cells toward a quiescent state biased toward regulatory and type 17 T cell phenotypes. Similarly, human CD4 and CD8 T cell-targeted TGF-{beta} agonists precisely and effectively suppress live-attenuated influenza vaccine (LAIV)-induced T cell activation and expansion in human spleen organoids. Correspondingly, CD4 T cell-targeted TGF-{beta} agonist effectively ameliorated disease activity and promoted disease remission in CD4 T cell-driven models of autoimmune neuroinflammation and allergic airway inflammation, demonstrating efficacy in both prophylactic and established inflammatory settings. Moreover, both CD4 and CD8 T cell-targeted TGF-{beta} agonists ameliorated disease activity in graft-versus-host disease. Additionally, a human CD19 B cell-targeted TGF-{beta} agonist robustly inhibits germinal center B cell-to-plasmablast maturation and antibody responses in LAIV-stimulated human spleen organoids. These early-stage results suggest that cell-selective TGF-{beta} agonism merits further investigation as a versatile therapeutic approach for the precise silencing of pathogenic adaptive immune responses.

immunology↗

bioRxiv2026

Joint Segmental Duplication Co-option Drives Human-specific Transcriptional Readthrough and Expression Fine-tuning of NPEPPS-TBC1D3

Segmental duplication (SD) is a major driver of functional changes in evolution and disease. Many genes embedded within SDs, such as NPEPPS and TBC1D3, display substantial copy number variation (CNV) across individuals. Yet, the precise identification of the functional copies and their transcriptional outputs remains largely unstudied. Focusing on NPEPPS and TBC1D3, we illustrate human-specific expression fine-tuning mechanisms associated with readthrough transcripts. We identified a human-specific NPEPPS-TBC1D3 digenic genomic structure that originated from a joint SD pair and became fixed across populations. Experiments demonstrate that this structure generates NPEPPS-TBC1D3 readthrough transcripts, which are the predominant isoforms of TBC1D3 expression in various cell types, fine-tuning its protein level. Furthermore, a human-specific hypomethylation signal within an upstream CpG island of NPEPPS precisely pinpoints the expressed TBC1D3 paralog. Moreover, we reveal transcriptional readthrough events are [~]3-fold enriched for joint-SD-associated transcriptional readthrough (JSDTR) and identify 109 JSDTR gene pairs, including neurodevelopmentally important pairs and clinically interesting SERF1A/B-SMN1/2. Taken together, our findings comprehensively describe an example of how a joint SD event shaped evolution and suggest that JSDTR is a broad mechanism for the emergence of new functions.

genetics↗

bioRxiv2026

Plasma proteomics reveal heterogeneous subtypes of depression linked to inflammation and aging

The clinical heterogeneity of depression has defied biological classification, limiting personalized treatment. Previous neuroimaging- or symptom-based subtyping of depression failed to clarify the underlying pathoetiology, while plasma proteins which integrates signals from multiple organ systems, offers a promising way to define biologically grounded subtypes. Using plasma proteomics from 2,127 incident depression cases in a cohort of 53000 individuals, we identified three biologically distinct subtypes differing in inflammation, aging, and metabolic profiles. The most prevalent subtype ( inflammation/ageing) was characterized by aging-related inflammation, poorest prognosis with hippocampal atrophy and highest suicide risk, mediated by age-related amygdala atrophy; this subtype had highest anhedonia burden. A distinct inflammation/energy dysregulation group had metabolic pathway enrichment with high inflammation and lifestyle risk factors (smoking) but no ageing trend, predominantly physical/psychomotor symptoms and decreased thalamic volume. In contrast, the inflammation-resilient group had the lowest inflammatory proteomic loading, lowest depression severity, more resilient lifestyle and increased hippocampal volume. These proteomic signatures, detectable years before symptom onset, enable risk stratification and suggest subtype-specific targeted physical and lifestyle interventions.

neuroscience↗

bioRxiv2026

Integrated Multi-omics Analysis Identifies Molecular Signatures of Regular Physical Activity in Human Peripheral Blood

Regular physical activity is well established to protect against metabolic disorders and bolster immunity; yet, the underlying molecular and cellular mechanisms remain incompletely understood. We integrated plasma metabolomics and lipidomics with single-cell transcriptomic and chromatin accessibility profiles to decode the systemic impact of physical activity on human immunity and metabolism. Our data reveal that regular physical activity is linked to a coordinated metabolic signature marked by enhanced fatty acid oxidation and antioxidant defense. In circulating immune cells, regularly active individuals exhibited synchronous enhancement at both the chromatin accessibility and transcriptional levels for antigen presentation-related genes in antigen-presenting cells (APCs), particularly in classical monocytes, naive B cells, and switched memory B cells. Meanwhile, cytotoxic programs in CD8+ cytotoxic T and mature NK cells showed epigenetic pre-activation of effector function regulators. Intercellular communication analysis further revealed that regular exercise enhanced MHC-I/II signaling between APCs and T cells and suppressed inflammatory signaling networks. Together, these findings elucidate molecular mechanisms underlying the health benefits of regular exercise and offer a theoretical basis for enhancing public health and preventing chronic diseases.

bioinformatics↗

bioRxiv2026

Deep learning-guided design of cell type-specific AAV promoters

Precise cell type targeting is critical for both clinical and experimental applications of adeno-associated viral (AAV) vectors, yet engineering vectors with cell type-specific activity remains a challenge. Here, we compared three strategies leveraging single-cell chromatin accessibility data to design cell type-specific AAV promoters, including a deep learning-based method to generate de novo regulatory sequences. When applied to target retinal ganglion cells or horizontal cells in mouse retina, deep learning-guided design consistently outperformed rational approaches, yielding synthetic promoters with stronger and more specific expression in vivo. Synthetic AAV promoters supported diverse transgenes, enabling the recording and ablation of targeted cells. Promoter activity was also maintained in human retinal organoids, suggesting that deep learning-designed sequences may be suitable for translation. Our findings highlight the potential of deep learning to synthesize cell type-specific AAV promoters and establish a versatile platform for cell type targeting with broad implications for gene therapy and basic research.

synthetic biology↗

bioRxiv2026

A Bear-Specific Coding Exon in TMEM41B Is Associated with Cold Adaptation

AbstractBears exhibit an unusual form of hibernation in which core body temperature remains relatively high despite profound suppression of whole-body metabolism. How such a warm yet hypometabolic state is achieved remains poorly understood. Here, through integrative analysis of 238 RNA-seq datasets spanning seven species, we identify transmembrane protein 41B (TMEM41B) as an evolutionarily conserved factor associated with thermogenic capacity. Loss of Tmem41b in mice and Drosophila results in severe cold intolerance, whereas its overexpression enhances thermogenic performance, supporting a conserved role in heat production. Comparative genomic analyses reveal a bear-specific 51-amino acid N-terminal extension in TMEM41B, arising from a 5' untranslated region mutation that converts a premature stop codon into coding sequence. Interactome profiling indicates that this bear-specific isoform differs from the canonical TMEM41B in its protein interaction landscape, including reduced association with components of mitochondria-associated endoplasmic reticulum membranes such as voltage-dependent anion-selective channel protein 1 (VDAC1). Consistent with this altered interaction profile, the extended isoform is associated with reduced mitochondrial oxidative activity. Functional analyses further suggest that the 51- amino acid extension modulates multiple cellular programs, including suppression of oxidative phosphorylation and myogenic differentiation alongside activation of osteogenic pathways, collectively biasing cells toward a hypometabolic state. Notably, expression of the polar bear TMEM41B isoform in mice induces a torpor-like phenotype under fasting conditions. Together, these findings identify TMEM41B as a conserved regulator of thermogenic capacity and suggest that evolutionary modification of its N terminus may contribute to metabolic suppression and energy conservation during bear hibernation.

evolutionary biology↗

bioRxiv2026

Single-Cell Multi-Omics Insights into BMI-Mediated Immune-Related Disease Risk

This study is based on the Chinese Immune Multi-Omics Atlas (CIMA) cohort and performs an integrated single-cell multi-omics analysis of peripheral blood mononuclear cells (PBMCs) from 210 individuals aged 20-29, comprising 3,311,699 scRNA-seq cells and 1,839,860 scATAC-seq cells, to decipher immune characteristics across different body mass index (BMI) categories. By integrating gene regulatory network (GRN) and cell type-level molecular quantitative trait loci (xQTL) data, and conducting summary-data-based Mendelian randomization (SMR) analysis in conjunction with genome-wide association studies (GWAS) of BMI-mediated immune-related diseases, we identified key disease-associated cell types and molecular features. Using rheumatoid arthritis (RA) as an example, multi-omics evidence revealed that in MAIT-SLC4A10 cells of underweight individuals, increased chromatin accessibility at the CCR6 locus (chr6:167119627-167120128) and enhanced transcriptional activity of RORC/RORA collectively drive elevated CCR6 expression, which may represent a potential molecular mechanism underlying the higher risk of RA in underweight populations. Our study leverages single-cell multi-omics data to systematically dissect the molecular mechanisms by which different BMI categories mediate susceptibility to immune-related diseases through coordinated regulation of immune cell chromatin accessibility, transcription factor activity, and downstream gene expression.

bioinformatics↗