Biology subjects

Tan, N. S.

Publications and source records attributed to Tan, N. S..

8 recordsLinked to original sources

bioRxiv2026

Gut Roseburia hominis-derived 10-hydroxystearic acid alleviates metabolic disorder by restoring hepatic fatty acid metabolism

Metabolic disorders and health-promoting dietary interventions represent complementary contexts of metabolic impairment and improvement, providing an opportunity to identify gut microbial taxa associated with both disease status and dietary response. Here, we integrated five independent human cohorts spanning metabolic disorders and dietary interventions and identified Roseburia hominis as a gut microbial taxon that consistently decreased in metabolic disorders and increased after dietary interventions. Beyond species abundance, strain-level genetic variation in R. hominis provided additional information for distinguishing metabolic disorder status and stratifying cardiometabolic severity. Integration of phenotype-associated variants with circulating metabolomics prioritized ohyA, encoding an oleate hydratase, as a candidate functional determinant linking R. hominis genetic variation to host metabolic heterogeneity. An ohyA-based genetic burden score was associated with adverse cardiometabolic traits and changed longitudinally during dietary intervention, while circulating metabolites statistically accounted for part of the associations between ohyA variants and host phenotypes. Functional experiments demonstrated that R. hominis OhyA protein converts oleic acid to 10-hydroxystearic acid (10-HSA), whereas an OhyA carrying five prioritized missense substitutions showed markedly impaired enzymatic activity. In high-fat diet-fed mice, 10-HSA attenuated body-weight gain, improved glucose homeostasis and insulin sensitivity, and reduced dyslipidemia and hepatic steatosis. Multi-omics profiling revealed limited remodeling of the gut microbiota but pronounced hepatic transcriptional and metabolic responses, characterized by activation of PPAR-associated lipid-metabolic pathways, altered fatty-acid oxidation and desaturation, and remodeling of hepatic lipid composition. Together, these findings establish a diet-responsive R. hominis-OhyA-10-HSA axis linking microbial variation to host metabolic dysfunction and show that 10-HSA can improve metabolic health in mice.

microbiology↗

bioRxiv2026

Gut microbial phenylalanine metabolism contributes to MASLD in a dietary protein-dependent manner

Metabolic dysfunction-associated steatotic liver disease (MASLD) affects nearly a third of adults worldwide, yet how diet shapes the microbiota's contribution remains unclear. In a diet-induced mouse model varying protein content independently of caloric intake, microbiota depletion ameliorated hepatic steatosis and liver injury under protein- sufficient but not protein-restricted conditions. Caecal and plasma metabolomics implicated microbial phenylalanine catabolism and its host conjugates phenylacetylglutamine (PAGln) and phenylacetylglycine (PAGly). PAGln supplementation exacerbated steatosis and liver injury without altering body weight, adiposity or glucose homeostasis. Individual-antibiotic perturbation with metagenomics linked bacterial phenylalanine catabolic capacity to disease severity, and in gnotobiotic mice, genetic disruption of bacterial phenylalanine-to-phenylacetic acid conversion lowered circulating conjugates and attenuated steatosis and fibrosis. In a human cohort, circulating PAGln was selectively elevated in cardiometabolic MASLD. These findings identify dietary protein availability as a modifier of the contribution of microbial phenylalanine metabolism to MASLD and highlight PAGln as a microbiota-derived metabolite that exacerbates hepatic disease.

microbiology↗

bioRxiv2025

Bioengineered human tonsil organoids as an immuno-engineering platform for evaluating immune functions

Although animal models remain the preclinical gold standard, they are constrained by interspecies differences in adaptive immunity, including antibody class-switching, affinity maturation, and multicellular interactions. To address this gap, we established a more physiologically relevant platform, a bioengineered tonsil organoid (BTO) system that combines autologous tonsil-derived stromal cells with immune cells to generate highly uniform organoids, preserving multicellular complexity and sustained stromal-immune interactions. BTO cultures mounted robust, antigen-specific recall responses to influenza, tetanus, and COVID-19 antigens. Notably, stimulation with ovalbumin and SARS-CoV-1 spike protein expanded antigen-recognizing B cells, indicating capacity for naive responses. Digital spatial transcriptomics and proteomics analysis confirmed immune-stromal crosstalk underpinning both humoral and cellular responses of BTO after challenge. These data highlight stromal cells as active scaffolds that not only maintain 3D architecture but also facilitate immune activation, memory recall, and recognition of naive antigens. Together, BTO offers a human-relevant model for mechanistic immunology and for pre-clinical evaluation on immunotherapies and vaccines not achievable with 2D cultures or animal models.

bioengineering↗

bioRxiv2025

Leveraging Large Language Models for Redundancy-Aware Pathway Analysis and Deep Biological Interpretation

Extracting coherent, biologically meaningful insights from vast, complex multi-omics data remains challenging. Currently, pathway enrichment analysis serves as a cornerstone for the functional interpretation of such data. However, conventional approaches often suffer from extensive functional redundancy caused by shared molecular components and overlapping pathway definitions across databases. This redundancy can obscure key biological signals and compromise the interpretability of pathway enrichment results. Here, we present MAPA (Functional Module Identification and Annotation for Pathway Analysis Results Using Large Language Models [LLM]), an open-source computational framework that resolves redundancy and enhances pathway analysis result interpretation. MAPA computes functional similarity between pathways using LLM-based text embeddings, enabling comparison across different databases. It constructs pathway similarity networks and identifies functional modules via community detection algorithms. Crucially, MAPA employs LLMs for automated functional annotation, integrating Retrieval-Augmented Generation (RAG) to generate comprehensive and real-time biological summaries and reduce hallucinations. Benchmarking demonstrated MAPAs superior performance: the biotext embedding similarity showed a large effect size (Cliffs {delta} = 0.96) compared with the Jaccard index ({delta} = 0.73), and module identification achieved high accuracy (Adjusted Rand Index [ARI] = 0.95) versus existing methods (ARI = 0.23-0.33). Human expert evaluation confirmed that MAPAs annotations match expert-quality interpretations. Finally, a multi-omics aging case study illustrates that MAPA uncovers coherent functional modules and generates insights extending beyond conventional pathway analyses. Collectively, MAPA represents a significant advance in redundancy-aware pathway analysis, transforming pathway enrichment results from fragmented lists into biologically coherent narratives. By leveraging the capabilities of LLMs, MAPA offers researchers a robust, scalable tool for deriving deep mechanistic insights from complex and vast multi-omics datasets, marking a new direction for AI-driven bioinformatics.

bioinformatics↗

bioRxiv2024

MegaMASLD: An interactive platform for exploring stratified transcriptomic signatures in MASLD progression

Liver transcriptomic data from patients with metabolic dysfunction-associated steatotic liver disease (MASLD) offers valuable resource for deciphering pathogenic molecular drivers. Here, we performed a Mega-analysis of MASLD Liver Transcriptomes (MegaMASLD) which reanalysed raw RNAseq data of over 800 livers in a standardized and integrative manner, aiming to unravel druggable molecular events in MASLD. Our analysis revealed a progressive transcriptomic shift predominantly associated with immunopathologies during MASLD progression. The differential transcriptomes produced a MASLD gene signature useful for quantitative assessment of MASLD severity but failed to faithfully recapitulate the exact histological staging. Instead, a histologic-independent unsupervised clustering analysis predicted a high-risk group prone to develop metabolic dysfunction-associated steatohepatitis (MASH), characterized by aberrant changes in humoral immune response and antibody repertoires. These findings were supported by another histologic-independent pseudotime analysis, which also identified several potentially targetable molecular switches, including FGFR, PDGFR, PAK, PRKG1 and CAMK kinase families, activated at various transitory phases of MASLD. The robust analysis has enabled risk stratification and deepened our understanding of the dynamic molecular events driving MASLD, thereby offering new options to enhance precision medicine of MASLD. An online web tool featuring MegaMASLD is available at https://bioanalytics-hs.shinyapps.io/MegaMASLD/.

molecular biology↗

bioRxiv2022

PAX4 loss of function alters human endocrine cell development and influences diabetes risk

Diabetes is a major chronic disease with an excessive healthcare burden on society1. A coding variant (p.Arg192His) in the transcription factor PAX4 is uniquely and reproducibly associated with an altered risk for type 2 diabetes (T2D) in East Asian populations2-7, whilst rare PAX4 alleles have been proposed to cause monogenic diabetes8. In mice, Pax4 is essential for beta cell formation but neither the role of diabetes-associated variants in PAX4 nor PAX4 itself on human beta cell development and/or function are known. Here, we demonstrate that non-diabetic carriers of either the PAX4 p.Arg192His or a newly identified p.Tyr186X allele exhibit decreased pancreatic beta cell function. In the human beta cell model, EndoC-{beta}H1, PAX4 knockdown led to impaired insulin secretion, reduced total insulin content, and altered hormone gene expression. Deletion of PAX4 in isogenic human induced pluripotent stem cell (hiPSC)-derived beta-like cells resulted in derepression of alpha cell gene expression whilst in vitro differentiation of hiPSCs from carriers of PAX4 p.His192 and p.X186 alleles exhibited increased polyhormonal endocrine cell formation and reduced insulin content. In silico and in vitro studies showed that these PAX4 alleles cause either reduced PAX4 expression or function. Correction of the diabetes-associated PAX4 alleles reversed these phenotypic changes. Together, we demonstrate the role of PAX4 in human endocrine cell development, beta cell function, and its contribution to T2D-risk.

developmental biology↗

bioRxiv2021

Endothelial-immune crosstalk contributes to vasculopathy in non-alcoholic fatty liver disease

The top cause of mortality in patients with non-alcoholic fatty liver disease (NAFLD) is cardiovascular complications. However, the mechanisms of NAFLD-associated vasculopathy remain understudied. We developed blood outgrowth endothelial cell (BOEC) models from NAFLD and healthy subjects. NAFLD BOECs exhibited global transcriptional upregulation of chemokine hallmarks and human leukocyte antigens. In mouse models of diet-induced NAFLD, we further confirmed enhanced endothelial expressions of CXCL12 in the aortas and liver vasculatures. To elucidate endothelial-immune crosstalk, we performed immunoprofiling by single-cell analysis, uncovering T cell intensification and potentially T-helper type 1 inflammation in NAFLD patients. Functionally, interference of the CXCL12-CXCR4 axis by small molecule AMD3100 selectively modulated the chemotaxis of patient-derived CD4+ T cells and natural killer cells towards NAFLD BOECs, restoring endothelial barrier integrity. Clinically, we detected three folds more circulating damaged endothelial cells in NAFLD patients than healthy controls. Our work provides insights for modulation of interactions with effector immune subsets to mitigate endothelial injury in NAFLD.

cell biology↗

bioRxiv2021

Modulated TRPC1 expression predicts sensitivity of breast cancer to doxorubicin and magnetic field therapy: segue towards a precision medicine approach.

BackgroundChemotherapy is the mainstream treatment modality for invasive breast cancer. Nonetheless, chemotherapy-associated adverse events can result in a patient terminating treatment. We show that transient receptor potential channel 1 (TRPC1) expression level predicts breast cancer sensitivity to doxorubicin (DOX) and pulsed electromagnetic field (PEMF) therapies. MethodsThe effects of PEMFs were examined with respect to: 1) the growth of MCF-7 cells in vitro; 2) MCF-7 tumors implanted into a chicken chorioallantoic membrane (CAM) model and; 3) patient-derived and MCF-7 breast cancer xenografts in mice. Potential synergisms between DOX and PEMF therapies were examined in these model systems and under conditions of TRPC1 overexpression or silencing in vitro. ResultsPEMF exposure impaired the survival of MCF-7 cells, but not that of nonmalignant MCF10A breast cells. The effects of PEMF- and DOX-therapies synergized in vitro at compromising MCF-7 cell growth. Synergism could be corroborated in vivo with patient-derived xenograft mouse models, wherein PEMF exposure alone or in combination with DOX reduced tumor size. Stable overexpression of TRPC1 enhanced the vulnerability of MCF-7 cells to both DOX and PEMF exposure and promoted proliferation, whereas chronic DOX exposure reduced TRPC1 expression, induced chemoresistance, precluded response to PEMF exposure and mitigated proliferation. Markers of metastasis including SLUG, SNAIL, VIMENTIN, and E-CADHERIN as well as invasiveness were also positively correlated with TRPC1 channel expression. ConclusionThe presented data supports a potential role of PEMF-therapy as an effective companion therapy to DOX-based chemotherapy for the treatment of breast cancers characterized by elevated TRPC1 expression levels.

cancer biology↗