Biology subjects

Liu, H.

Publications and source records attributed to Liu, H..

At least 325 records · Page 18Linked to original sources

bioRxiv2024

A single high-zinc activation enhancer can control two genes oriented head-to-head in C. elegans

Enhancers play critical roles in gene expression, but a full understanding of their complex functions has yet to be defined. The cellular response to excess zinc levels in C. elegans requires the HIZR-1 transcription factor, which binds the high-zinc activation (HZA) enhancer in the promoters of multiple target genes. Cadmium hijacks the excess zinc response by binding and activating HIZR-1. By analyzing the genome-wide transcriptional response to excess zinc and cadmium, we identified two positions in the genome where head-to-head oriented genes are both induced by metals. In both examples, a single predicted HZA enhancer is positioned between the two translational start sites. We hypothesized that a single enhancer can control both head-to-head genes, an arrangement that has not been extensively characterized. To test this hypothesis, we used CRISPR genome editing to precisely delete the HZAmT enhancer positioned between mtl-2 and T08G5.1; in this mutant, both head-to-head genes display severely reduced zinc-activated transcription, whereas zinc-activated transcription of more distant genes was not strongly affected. Deleting the HZAcF enhancer positioned between cdr-1 and F35E8.10 caused both head-to-head genes to display reduced cadmium-activated transcription, whereas cadmium-activated transcription of more distant genes was not strongly affected. These studies rigorously document that a single HZA enhancer can control two head-to-head genes, advancing our understanding of the diverse functions of enhancers. Article SummaryEnhancers are critical for gene expression, but a full understanding their functions has yet to be elucidated. We discovered two positions in the C. elegans genome where a pair of head-to-head oriented genes are both transcriptionally activated by excess zinc and/or cadmium; in both cases, one high-zinc activation (HZA) enhancer is positioned between the translation start sites. When these HZA enhancers were deleted by genome engineering, both head-to-head genes lost metal-activated transcription. These results demonstrate that a single HZA enhancer can control two head-to-head genes, advancing the understanding of enhancer function.

genetics↗

bioRxiv2024

Beta-DIA: Integrating learning-based and function-based feature scores to optimize the proteome profiling of single-shot diaPASEF mass spectrometry data

We present a freely available diaPASEF data analysis software, Beta-DIA, that utilizes deep learning methods to score coelution consistency in retention time-ion mobility dimensions and spectrum similarity. Beta-DIA integrates these learning-based scores with traditional function-based scores, enhancing the qualitative analysis performance. In some low detection datasets, Beta-DIA identifies twice as many protein groups as DIA-NN. The success of Beta-DIA has paved another way for the application of deep learning in fundamental proteome profiling.

bioinformatics↗

bioRxiv2024

CellBinDB: A Large-Scale Multimodal Annotated Dataset for Cell Segmentation with Benchmarking of Universal Models

In recent years, cell segmentation techniques have played a critical role in the analysis of biological images, especially for quantitative studies. Deep learning-based cell segmentation models have demonstrated remarkable performance in segmenting cell and nucleus boundaries, however, they are typically tailored to specific modalities or require manual tuning of hyperparameters, limiting their generalizability to unseen data. Comprehensive datasets that support both the training of universal models and the evaluation of various segmentation techniques are essential for overcoming these limitations and promoting the development of more versatile cell segmentation solutions. Here, we present CellBinDB, a large-scale multimodal annotated dataset established for these purposes. CellBinDB contains more than 1,000 annotated images, each labeled to identify the boundaries of cells or nuclei, including 4,6-Diamidino-2-Phenylindole (DAPI), Single-stranded DNA (ssDNA), Hematoxylin and Eosin (H&E), and Multiplex Immunofluorescence (mIF) staining, covering over 30 normal and diseased tissue types from human and mouse samples. Based on CellBinDB, we benchmarked seven state-of-the-art and widely used cell segmentation technologies/methods, and further analyzed the effects of four cell morphology indicators and image gradient on the segmentation results.

cell biology↗

bioRxiv2024

The impact of transplant location on the gut microbiome and resistome in patients undergoing hematopoietic stem cell transplantation at home versus in the hospital

ObjectivesHome-based hematopoietic stem cell transplantation (HCT) is a novel approach that has the potential to improve outcomes, however, the impact of transplant location on the gut microbiome remains uncharacterized. We hypothesized that patients randomized to undergo home HCT would have higher gut taxonomic diversity and lower antimicrobial resistance (AMR) gene abundance compared to those undergoing standard hospital HCT. MethodsWe identified 28 patients enrolled in Phase II randomized trials of home (n=16) v. hospital (n=12) HCT at Duke and performed shotgun metagenomic sequencing of stools to compare taxonomic and AMR gene composition between groups. We performed a secondary analysis of patients from each group transplanted at an outpatient infusion clinic with those who underwent standard inpatient HCT ("outpatient" v. "inpatient"). ResultsNo significant differences in duration of hospitalization were found in those randomized to home v. hospital HCT. Taxonomic and AMR gene - and {beta}-diversity were comparable. In contrast, secondary analyses demonstrated that patients from both home and hospital groups transplanted at an outpatient infusion clinic spent significantly less time in the hospital and demonstrated higher taxonomic -diversity and differential {beta}-diversity compared to standard inpatient HCT, although AMR gene -diversity did not differ, and comparisons were confounded by both differences in transplant type and use of antibiotics. ConclusionsRandomization by transplant location did not impact the gut microbiota to the same extent as the duration of hospitalization, although secondary analyses were heavily confounded. Even when taxonomic differences were observed, AMR genes were similar between groups.

microbiology↗

bioRxiv2024

Restoration of PKM1 improves functional maturation of human stem-cell derived-β cell by regulating PEP metabolism

Human stem cell-derived {beta} (SC-{beta}) cells still exhibit limited glucose response required for insulin secretion due to glycolytic bottlenecks, yet how these metabolic abnormalities impact glucose response and functional maturation of SC-{beta} cells remains unclear. In this study, we identified a metabolic checkpoint located at PEP accumulation that impeded the functional maturation, which was rescued by restoration of pyruvate kinase 1 (PKM1). Glucose-tracing metabolomics in human stem cell-derived islets revealed abnormal glycolytic PEP accumulation at resting condition, resulting in impaired calcium response and insulin secretion upon high glucose or glycolytic metabolite stimulation. Mechanistically, elevated PEP significantly raised intracellular basal calcium levels, leading to downregulated expression of genes involved in TCA cycle elucidated by single cell transcriptomics. Furthermore, the activity of pyruvate kinase, which metabolizes PEP, was notably reduced due to low PKM1 expression. By overexpressing PKM1, the impairment of TCA-related genes caused by PEP accumulation was reversed via modulating PEP metabolism, resulting in enhanced calcium responses and insulin secretion upon high glucose stimulation. Together, we discovered a novel role of PKM1-regulated PEP metabolism in mediating the functional maturation of human SC-{beta} cells. This study highlights the importance of metabolic reprogramming in human SC-{beta} cell maturation, advancing cell therapy approaches for diabetes treatment.

cell biology↗

bioRxiv2024

Pan-Cancer Genetic Profiles of Mitotic DNA Integrity Checkpoint Protein Kinases

BACKGROUNDThe mitotic DNA integrity checkpoint signaling pathway is potentially involved in cancers that regulate genomic stability where protein kinases play a pivotal role. 16 total protein kinase genes are involved in this pathway: ATM, BRSK1, CDK1, CDK2, CHEK1, CHEK2, MAP3K20, NEK11, PLK1, PLK2, PLK3, PRKDC, STK33, TAOK1, TAOK2, and TAOK3. This study aims to provide pan-cancer profiles of the protein kinases in mitotic DNA integrity checkpoint signaling gene set for potential prognostic and diagnostic purposes, as well as future potential therapeutic targets for cancer in a clinical setting. METHODSMulti-omic data was acquired for the 16 genes; over 9000 samples of 33 types of cancer were analyzed to create pan-cancer profiles of SNV, CNV, methylation, mRNA expression, pathway crosstalk, and microRNA regulation networks. RESULTSThe SNV profile showed that most of these genes have a high SNV mutation frequency across some cancer types, such as UCEC and SKCM. The CNVs of some of these genes are associated with the survival of UCEC, KIRP, and LGG. BRCA, KIRC, LUAD, and STAD might be affected by the mRNA expression of these genes which might involve regulation of copy number, methylation, and miRNA. In addition, these genes also cross-talk with some known cancer pathways. CONCLUSIONThe protein kinases in mitotic DNA integrity checkpoint signaling may play a role in cancer development and, with adequate research, could potentially be developed as biomarkers for cancer diagnosis and prognosis. However, further efforts are necessary to validate their clinical value for diagnosis and prognosis and to develop practical applications in clinical settings. Nevertheless, these pan-cancer profiles offer a better overall understanding as well as useful information for future reference regarding mitotic DNA integrity checkpoint signaling in cancer.

genetics↗

bioRxiv2024

Interpretable PROTAC degradation prediction with structure-informed deep ternary attention framework

Proteolysis Targeting Chimeras (PROTACs) are heterobifunctional ligands that form ternary complexes with Protein Of Interests (POIs) and E3 ligases, exploiting the ubiquitin-proteasome system to degrade disease-associated proteins, promising to drug the undruggable. While PROTAC research primarily relies on costly and time-consuming wet experimental approaches, deep learning offers a promising avenue to accelerate development and reduce expenses. However, existing deep learning methods for PROTAC degradation prediction often overlook the significance of hierarchical molecular representation and protein structural information, hindering effective data modeling. Moreover, their black-box nature limits the interpretability of computational outcomes, failing to provide intuitive insights into substructure interactions within the PROTAC system. This study introduces PROTAC-STAN, a structure-informed deep ternary attention network (STAN) framework for interpretable PROTAC degradation prediction. PROTAC-STAN represents PROTAC molecules across atom, molecule, and property hierarchies and incorporates structure information for POIs and E3 ligases using a protein language model infused with structural data. Furthermore, it simulates interactions among three entities at the atom and amino acid levels via a novel ternary attention network tailored for the PROTAC system, providing unprecedented insights into the degradation mechanism. By integrating hierarchical PROTAC molecule representation, structural embedding of POI and E3 ligase, and ternary attention network modeling interactions, our approach substantially improves prediction accuracy by 10.95% while enabling significant model interpretability via atomic and residue level visualization of molecule and complex. Experiments on the refined public PROTAC dataset demonstrate that PROTAC-STAN outperforms state-of-the-art baselines in overall performance. The excellent performance of PROTAC-STAN is anticipated to establish it as a foundational tool in future research on PROTAC-related drugs, thereby accelerating the development of this field.

bioinformatics↗

bioRxiv2024

Neuronal Detection Triggers Systemic Digestive Shutdown in Response to Adverse Food Sources

The ability to sense and adapt to adverse food conditions is essential for survival across species, but the detailed mechanisms of neuron-digestive crosstalk in food sensing and adaptation remain poorly understood. This study identifies a novel mechanism by which animals detect unfavorable food sources through neurons and initiate a systemic response to shut down digestion, thus safeguarding against potential harm. Specifically, we demonstrate that NSY-1, expressed in AWC neurons, detects Staphylococcus saprophyticus (SS) as an unfavorable food source, prompting the animal to avoid and halt digestion of SS. Upon detection, the animals activate the AWCOFF neural circuit, leading to a systemic digestive shutdown, which is mediated by NSY-1-dependent STR-130. Additionally, NSY-1 mutation triggers the production of insulin peptides, including INS-23, which interact with the DAF-2 receptor to modulate SS digestion and affects the expression of intestinal BCF-1. These findings uncover a crucial survival strategy through neuron-digestive crosstalk, where the NSY-1 pathway in AWC neurons orchestrates food evaluation and initiates digestive shutdown to adapt effectively to harmful food sources.

physiology↗

bioRxiv2024

Compressive forces induce epigenetic activation of aged human dermal fibroblasts through ERK signalling pathway

Age-related changes in human dermal fibroblasts (HDFs) contribute to impaired wound healing and skin aging. While these changes result in altered mechanotransduction, the epigenetic basis of rejuvenating aging cells remains a significant challenge. This study investigates the effects of compressive forces on nuclear mechanotransduction and epigenetic rejuvenation in aged HDFs. Using a compressive force application model, the activation of HDFs through alpha-smooth muscle actin ([a]-SMA) is demonstrated. Sustained compressive forces induce significant epigenetic modifications, including chromatin remodeling and altered histone methylation patterns. These epigenetic changes correlate with enhanced cellular migration and rejuvenation. Small-scale drug screening identifies the extracellular signal-regulated kinase (ERK) signaling pathway as a key mediator of compression-induced epigenetic activation. Furthermore, implanting aged cell spheroids to an aged skin model and subjecting the tissue with compressing forces resulted in increased collagen I protein levels. Collectively, these findings demonstrate that applying compressive force to aged fibroblasts activates global epigenetic changes through the ERK signaling pathway, ultimately rejuvenating cellular functions with potential applications for wound healing and skin tissue regeneration. Significance StatementPartial rejuvenation of aging cells is desirable but is still a major challenge. In this paper, we demonstrate that aged human dermal fibroblasts, embedded in a 3D collagen hydrogel matrix as spheroids, subjected to external static compressive force exhibit partial rejuvenation. Through immunofluorescence, small-scale inhibitor screen and gene expression analysis, we identify some of the critical mechanotransduction pathways in this process. Collectively, our results provide compelling evidence that tissue compression results in the activation of potential rejuvenation pathways in aging cells.

cell biology↗

bioRxiv2024

MARVEL: Microenvironment Annotation by Supervised Graph Contrastive Learning

Recent advancements in in situ molecular profiling technologies, including spatial proteomics and transcriptomics, have enabled detailed characterization of the microenvironment at cellular and subcellular levels. While these techniques provide rich information about individual cells spatial coordinates and expression profiles, extracting biologically meaningful spatial structures from the data remains a significant challenge. Current methodologies often rely on unsupervised clustering followed by cell type annotation based on differentially expressed genes within each cluster and most of the time will require other information as the reference (e.g., HE-stained images). This is labor-intensive and demands extensive domain knowledge. To address these challenges, we propose a supervised graph contrastive learning framework, MARVEL. MARVEL is a supervised graph contrastive learning method that can effectively embed local microenvironments represented by cell neighbor graphs into a continuous representation space, facilitating various downstream microenvironment annotation scenarios. By leveraging partially annotated examples as strong positives, our approach mitigates the common issues of false positives encountered in conventional graph contrastive learning. Using real-world annotated data, we demonstrate that MARVEL outperforms existing methods in three key microenvironment-related tasks: transductive microenvironment annotation, inductive microenvironment querying, and the identification of novel microenvironments across different slices.

bioinformatics↗

bioRxiv2024

MethylGPT: a foundation model for the DNA methylome

DNA methylation serves as a powerful biomarker for disease diagnosis and biological age assessment. However, current analytical approaches often rely on linear models that cannot capture the complex, context-dependent nature of methylation regulation. Here we present MethylGPT, a transformer-based foundation model trained on 226,555 (154,063 after QC and deduplication) human methylation profiles spanning diverse tissue types from 5,281 datasets, curated 49,156 CpG sites, and 7.6 billion training tokens. MethylGPT learns biologically meaningful representations of CpG sites, capturing both local genomic context and higher-order chromosomal features without external supervision. The model demonstrates robust methylation value prediction (Pearson R=0.929) and maintains stable performance in downstream tasks with up to 70% missing data. Applied to age prediction across multiple tissue types, MethylGPT achieves superior accuracy compared to existing methods. Analysis of the models attention patterns reveals distinct methylation signatures between young and old samples, with differential enrichment of developmental and aging-associated pathways. When finetuned to mortality and disease prediction across 60 major conditions using 18,859 samples from Generation Scotland, MethylGPT achieves robust predictive performance and enables systematic evaluation of intervention effects on disease risks, demonstrating potential for clinical applications. Our results demonstrate that transformer architectures can effectively model DNA methylation patterns while preserving biological interpretability, suggesting broad utility for epigenetic analysis and clinical applications.

systems biology↗

bioRxiv2024

Novel environment exposure drives temporally defined and region-specific chromatin accessibility and gene expression changes in the hippocampus

The interaction of mammals with a novel environment (NE) results in structural and functional changes in multiple brain areas, including the hippocampus. This experience-dependent circuit reorganization is driven in part by changes in gene expression however, the dynamic sensory experience-driven chromatin states and the diverse cell type specific gene expression programs that are regulated by novel experiences are not well described. We employed single- nucleus multiomics (snRNA- and ATAC-seq) and bulk RNA-seq of the hippocampal DG, CA3, and CA1 regions to characterize the temporal evolution of cell-type-specific chromatin accessibility and gene expression changes that occur in 14 different cell types of the hippocampus upon exposure of mice to a novel environment. We observe strong hippocampal regional specificity in excitatory neuron chromatin accessibility and gene expression as well as great diversity in the inhibitory neuron and non-neuronal transcriptional responses. The novel environment-regulated genes in each cell type were enriched for genes that encode secreted factors, and cell-type-specific expression of their cognate receptors identified promising candidates for the modulation of learning and memory processes. Our characterization of the effect of novel experience on chromatin revealed thousands of cell-type-specific changes in chromatin accessibility. Coordinated analysis of chromatin accessibility and gene expression changes within individual cell types identified Fos/AP-1 as a key driver of novel experience-induced changes in chromatin accessibility and cell-type-specific gene expression. Together, these data provide a rich resource of hippocampal chromatin accessibility and gene expression profiles across diverse cell types in response to novel experience, a physiological stimulus that affects learning and memory.

neuroscience↗

bioRxiv2024

Prenatal stress increases corticosterone levels in offspring by impairing placental glucocorticoid barrier function

To investigate the association between prenatal stress (PS) and corticosterone, and its influence on DNA methylation of genes related to the placental glucocorticoid (GC) barrier, including 11{beta}-HSD2, P-gp, NR3C1, and FKBP5. The PS model was established through chronic unpredictable mild stress (CUMS). DNA methylation of GC-related genes was analyzed using reduced representation bisulfite sequencing (RRBS), and the results were confirmed using MethylTarget sequencing. The mRNA and protein expression levels of these genes were detected through qRT-PCR and Western blotting, respectively. Plasma corticosterone levels are elevated in pregnant female rats exposed to PS conditions and their offspring. Compared to the offspring of the prenatal control (OPC) group, the offspring of the prenatal stress (OPS) group showed down-regulation in both mRNA and protein expression of DNMT 3A and DNMT 3B, while up-regulation was observed in the expression of DNMT1. RRBS analyses identified P-gp and FKBP5 as hypermethylated genes, including a total of 43 differentially methylated sites (DMS) and 2 differentially methylated regions (DMR). MethylTarget sequencing revealed that both genes had 15 differentially methylated CpG sites. This study provides preliminary evidence that PS disrupts the placental GC barrier through abnormal gene expression caused by hypermethylation of GC-related genes, resulting in elevated corticosterone levels in offspring and affecting their growth and development.

developmental biology↗

bioRxiv2024

Comparative spatiotemporal single cell transcriptomes reveal rewiring of pre-existing regulations during emergence of Kranz anatomy in C4 grasses

Many of the worlds most productive food and bioenergy crops use C4 photosynthesis, which have high photosynthetic efficiency due to a Kranz anatomy-based CO2 concentrating mechanism. Here, we took a comparative transcriptomics approach using single cell spatial transcriptomes of leaf primordia for maize (Zea mays) and single cell RNA-seq (scRNA-seq) atlases of corresponding leaf tissues for three C4 species (Zea mays, Sorghum bicolor, Setaria viridis) and one C3 rice (Oryza sativa) to study regulatory networks involved in the development and evolution of Kranz anatomy. We show that the formation of Kranz anatomy involves extensive recruitment and modification of pre-existing regulatory modules, especially the SHR-SCR module and the auxin signaling pathway. Members of the INDETERMINATE DOMAIN (IDD) family transcription factors, including IDD7 and IDDP1, contribute to the modification of the involved modules. These extensive recruitment and modification of existing genetic regulatory programs underlie the repeated emergence of C4 photosynthesis during evolution.

plant biology↗

bioRxiv2024

Directing cellular transitions on gene graph-enhanced cell state manifold

A select few genes act as pivotal drivers in the process of cell state transitions. However, finding key genes involved in different transitions is challenging. To address this problem, we present CellNavi, a deep learning-based framework designed to predict genes that drive cell state transitions. CellNavi builds a driver gene predictor upon a cell state manifold, which captures the intrinsic features of cells by learning from large-scale, high-dimensional transcriptomics data and integrating gene graphs with directional connections. Our analysis shows that CellNavi can accurately predict driver genes for transitions induced by genetic, chemical, and cytokine perturbations across diverse cell types, conditions, and studies. By leveraging a biologically meaningful cell state manifold, it is proficient in tasks involving critical transitions such as cellular differentiation, disease progression, and drug response. CellNavi represents a substantial advancement in driver gene prediction and cell state manipulation, opening new avenues in disease biology and therapeutic discovery.

cell biology↗

bioRxiv2024

AMPK activation promotes transcriptional activation of TFEB through its dephosphorylation

Transcription Factor EB (TFEB) is a critical regulator of lysosomal biogenesis, autophagy and energy homeostasis through controlling expression of genes belonging to the coordinated lysosomal expression and regulation network. AMP-activated protein kinase (AMPK) has been reported to phosphorylate TFEB at three conserved C-terminal serine residues (S466, S467, S469) and these phosphorylation events were essential for transcriptional activation of TFEB. In sharp contrast to this proposition, here we demonstrate that AMPK activation leads to dephosphorylation of the C-terminal sites, and that AMPK is dispensable for mTORC1-mediated/torin1-sensitive TFEB activation. We show that a synthetic peptide encompassing C-terminal serine residues of TFEB is a poor substrate of AMPK. Treatment of cells with AMPK activator (MK-8722) or mTOR inhibitor (torin1) robustly dephosphorylated TFEB not only at mTORC1-targeted N-terminal serine sites, but also at the C-terminal sites. Loss of function of AMPK abrogated MK-8722-but not torin1-induced dephosphorylation and induction of the vast majority of TFEB target genes.

biochemistry↗

bioRxiv2024

PolSpec: polarisation-based detection for spectral classification of optical signals

Spectrally resolved imaging is typically realised using bandpass filters, which are inefficient when they reject "out-of-band" photons or using angularly dispersive devices with at least one image dimension requiring scanning to acquire a full hyperspectral dataset and therefore sequential data acquisition, unless cascades of dichroic beamsplitters are employed, for which cost and experimental complexity scale with the number of spectral channels. Here we present a new approach, "PolSpec", to realise rapid and flexible widefield hyperspectral imaging with lower cost and complexity using polarisation optics instead of dielectric coatings or dispersive devices. PolSpec utilises Lyot filters that provide continuously varying transmission across the desired spectral range to generate orthogonal "spectral modulation vectors" that can represent specific spectral signatures with significantly lower data volumes than full spectral profiles. We demonstrate single-shot widefield hyperspectral imaging using a polarisation-resolving camera and rapid, electronically reconfigurable, more photon efficient, hyperspectral imaging using a liquid crystal variable retarder.

biophysics↗

bioRxiv2024

α-Synuclein Strain Dynamics Correlate with Cognitive Shifts in Parkinson's Disease

-Synuclein (-syn) strains can serve as discriminators between Parkinsons disease (PD) and related -synucleinopathies. The relationship between -syn strain dynamics and clinical performance as patients transition from normal cognition (NC) to cognitive impairment (CI) is not known. Here, we show that the biophysical properties and neurotoxicity of -syn strains change as PD cognitive status transitions from NC to mild cognitive impairment (PD-MCI) and dementia (PD-D). Both cross-sectional and longitudinal analyses reveal distinct -syn strains in PD patients correlating to their level of cognitive impairment. Machine learning (ML) was employed to achieve high classification accuracy. The combination of thioflavin T (ThT) maximal fluorescence intensity (mfi), max slope of rise curve (forming rate), lag time (tlag), 20% time (t20), and half-time (t50), dynamic light scattering (DLS) (peak number, [1/2] peak size, [1/2] peak intensity) and neurotoxicity together with demographic variables for model training yielded superior performance (89[~]99% accuracy in the 4- and 2- classification schema) compared to individual features alone in classifying cognitive status. For the longitudinal study, DLS peak number emerged as the strongest predictor of cognitive transition (HR = 0.12, P = 0.002), with the optimal predictive model combining DLS peak number, sex, education, DLS peak 1 size, and DLS peak 2 polydispersity achieving high accuracy (C-index of [~]93%). This study presents evidence that individuals with PD have different -syn strains correlating to their cognitive status and highlights the potential of -syn strain dynamics to guide future diagnosis, management, and stratification of PD patients. One Sentence SummaryDistinct features of -syn strains change with cognitive decline in Parkinsons disease and AI-based analysis incorporating these combined characteristics serves as a powerful tool for PD clinical stratification.

neuroscience↗