Biology subjects

Zheng, S.

Publications and source records attributed to Zheng, S..

At least 145 records · Page 8Linked to original sources

bioRxiv2022

Probing lung function at high spatiotemporal resolution using a novel crystal ribcage

Understanding the dynamic pathogenesis and treatment response in pulmonary diseases requires probing the lung at cellular resolution in real-time. Despite recent progress in intravital imaging, optical imaging of the lung during active respiration and circulation has remained challenging. Here, we introduce the crystal ribcage: a transparent ribcage that (i) allows truly multiscale optical imaging of the lung in health and disease from whole-organ to single cell, (ii) enables the modulation of lung biophysics and immunity through intravascular, intrapulmonary, intraparenchymal, and optogenetic interventions, and (iii) preserves the 3-D architecture, air-liquid interface, cellular diversity, and respiratory-circulatory functions of the lung. Utilizing these unprecedented capabilities on murine models of primary and metastatic lung tumors, respiratory infection, pulmonary fibrosis, emphysema, and acute lung injury we probed how disease progression remodels the respiratory-circulatory functions at the single alveolus and capillary levels. In cancer, we identified the earliest stage of tumorigenesis that compromises alveolar and capillary functions, a key state with consequences on tumor progression and treatment response. In pneumonia, we mapped mutual links between the recruited immune cells and the alveolar-capillary functions. We found that neutrophil migration is strongly and reversibly responsive to vascular pressure with implications for understanding of how lung physiology, altered by disease and anatomical location, affects immune cell activities. The crystal ribcage and its broad applications presented here will facilitate further studies of real-time remodeling of the alveoli and capillaries during pathogenesis of nearly any pulmonary disease, leading to the identification of new targets for treatment strategies.

bioengineering↗

bioRxiv2022

Chromosome-level genome assembly of Torreya grandis provides insights into the origin and evolution of gymnosperm-specific sciadonic acid biosynthesis

Species in genus Torreya are nut trees that produce dry fruits with a wide assortment of functions. Here, we report the 19-Gb chromosome-level genome assembly of T. grandis. The genome is shaped by an ancient whole genome duplication and recurrent LTR retrotransposon bursts. Comparative genomic analyses reveal key genes involved in reproductive organ development, cell wall biosynthesis and seed storage. Two genes encoding a C18 {Delta}9-elongase and a C20 {Delta}5-desaturase are identified in T. grandis to be responsible for sciadonic acid biosynthesis and both are present in diverse plant lineages except angiosperms. We demonstrate that the histidine-rich boxes of the {Delta}5-desaturase are crucial for its catalytic activity. Methylome analysis reveals that methylation valleys of the T. grandis seed genome harbor genes associated with important seed activities, including cell wall and lipid biosynthesis. Moreover, seed development is accompanied by DNA methylation changes that possibly fuel energy production. This study provides important genomic resource for gymnosperms and unravels key enzymes for biosynthesis of sciadonic acid as a hallmark metabolite of gymnosperms.

plant biology↗

bioRxiv2022

Long-term, super-resolution HIDE imaging of the inner mitochondrial membrane in live cells with a cell-permeant lipid probe

The densely packed inner mitochondrial membrane (IMM) is vital for bioenergy generation and its dynamics control mitochondrial health and cellular homeostasis. IMM structure is complex, however, and imaging its dynamics with high temporal and spatial resolution is complicated by the photosensitivity of IMM-resident enzymes. Here we describe the cell-permeant, lipid-like acridine orange derivative MAO-N3 and use it to assemble high-density, environmentally sensitive (HIDE) probes that selectively label and image the IMM in live cells. MAO-N3 pairs with multiple SPAAC-reactive fluorophores to support HIDE imaging via confocal, Structured Illumination, Single Molecule Localization, and Stimulated Emission Depletion microscopy, all with significantly improved resistance against photobleaching. The HIDE probes generated using MAO-N3 require no genetic manipulations, are non-toxic in model cell lines and primary cardiomyocytes, even under conditions that amplify the effects of mitochondrial toxins, and visualize the IMM for up to 12.5 hours with unprecedented spatial and temporal resolution.

cell biology↗

bioRxiv2022

DepLink: an R Shiny app to systematically link genetic and pharmacologic dependencies of cancer

Large-scale genetic and pharmacologic dependency maps are generated to reveal genetic vulnerabilities and drug sensitivities of cancer. However, user-friendly software is needed to systematically link such maps. Here we present DepLink, an R Shiny server to identify genetic and pharmacologic perturbations that induce similar effects on cell viability or molecular changes. DepLink integrates heterogeneous datasets of genome-wide CRISPR loss-of-function screens, high-throughput pharmacologic screens, and perturbation expression signatures. The datasets are systematically connected by four complementary modules tailored for different query scenarios. In summary, DepLink enables easy navigation, visualization, and linkage of rapidly evolving cancer dependency maps.

bioinformatics↗

bioRxiv2022

OpenPBTA: An Open Pediatric Brain Tumor Atlas

SummaryPediatric brain and spinal cancer are the leading disease-related cause of death in children, thus we urgently need curative therapeutic strategies for these tumors. To accelerate such discoveries, the Childrens Brain Tumor Network and Pacific Pediatric Neuro-Oncology Consortium created a systematic process for tumor biobanking, model generation, and sequencing with immediate access to harmonized data. We leverage these data to create OpenPBTA, an open collaborative project which establishes over 40 scalable analysis modules to genomically characterize 1,074 pediatric brain tumors. Transcriptomic classification reveals that TP53 loss is a significant marker for poor overall survival in ependymomas and H3 K28-altered diffuse midline gliomas and further identifies universal TP53 dysregulation in mismatch repair-deficient hypermutant high-grade gliomas. OpenPBTA is a foundational analysis platform actively being applied to other pediatric cancers and inform molecular tumor board decision-making, making it an invaluable resource to the pediatric oncology community. In BriefThe OpenPBTA is a global, collaborative open-science initiative which brought together researchers and clinicians to genomically characterize 1,074 pediatric brain tumors and 22 patient-derived cell lines. Shapiro, et. al create over 40 open-source, scalable modules to perform cancer genomics analyses and provide a richly-annotated somatic dataset across 58 brain tumor histologies. The OpenPBTA framework can be used as a model for large-scale data integration to inform basic research, therapeutic target identification, and clinical translation. HighlightsOpenPBTA collaborative analyses establish resource for 1,074 pediatric brain tumors NGS-based WHO-aligned integrated diagnoses generated for 641 of 1,074 tumors RNA-Seq analysis infers medulloblastoma subtypes, TP53 status, and telomerase activity OpenPBTA will accelerate therapeutic translation of genomic insights

genomics↗

bioRxiv2022

SE(3) Equivalent Graph Attention Network as an Energy-Based Model for Protein Side Chain Conformation

Protein design energy functions have been developed over decades by leveraging physical forces approximation and knowledge-derived features. However, manual feature engineering and parameter tuning might suffer from knowledge bias. Learning potential energy functions fully from crystal structure data is promising to automatically discover unknown or high-order features that contribute to the proteins energy. Here we proposed a graph attention network as an energy-based model for protein conformation, namely GraphEBM. GraphEBM is equivariant to the SE(3) group transformation, which is the important principle of modern machine learning for molecules-related tasks. GraphEBM was benchmarked on the rotamer recovery task and outperformed both Rosetta and the state-of-the-art deep learning based methods. Furthermore, GraphEBM also yielded promising results on combinatorial side chain optimization, improving 22.2%{chi} 1 rotamer recovery to the PULCHRA method on average.

bioinformatics↗

bioRxiv2022

Endothelial TIE1 restricts angiogenic sprouting to coordinate vein assembly in synergy with its homologue TIE2

ObjectiveVascular growth followed by vessel specification is crucial for the establishment of a hierarchical blood vascular network. We have here investigated mechanisms underlying venogenesis, particularly the molecular control over venous fate acquisition during vascular development. Approach and ResultsWe analyzed the function of TIE1 as well as its synergy with TIE2 in the regulation of vein formation by employing genetic mouse models targeting Tie1 and Tek. Cardinal vein growth appeared normal in TIE1 deficient mice, whereas TIE2 deficiency altered the identity of cardinal vein endothelial cells with the aberrant expression of DLL4. Interestingly, the parallel growth of murine cutaneous veins along with arteries, which was initiated at approximately embryonic day 13.5, was retarded in mice lack of TIE1. Tie1 deletion disrupted also venous integrity, displaying increased sprouting angiogenesis and vascular bleeding. Abnormal venous sprouts with defective arteriovenous alignment were also observed in the mesenteries of Tie1 deleted mice. Mechanistically, TIE1 deficiency resulted in the decreased expression of venous regulators including TIE2 while angiogenic regulators were upregulated. The alteration of TIE2 level by TIE1 insufficiency was further confirmed by the siRNA-mediated knockdown of Tie1 in cultured endothelial cells. Additionally, combining the endothelial deletion of Tie1 with one null allele of Tek resulted in a progressive increase of vein-associated angiogenesis leading to the formation of vascular tufts in retinas, whereas the loss of Tie1 alone produced only a relatively mild venous defect. ConclusionsFindings from this study imply that TIE1 and TIE2 act in a synergistic manner to restrict sprouting angiogenesis during vein formation.

developmental biology↗

bioRxiv2022

Tangled gene-for-gene interactions mediate co-evolution of the rice NLR immune receptor Pik and blast fungus effector proteins

Studies focused solely on single organisms can fail to identify the networks underlying host-pathogen gene-for-gene interactions. Here, we integrate genetic analyses of rice (Oryza sativa, host) and rice blast fungus (Magnaporthe oryzae, pathogen) and uncover a new pathogen recognition specificity of the rice nucleotide-binding domain and leucine-rich repeat protein (NLR) immune receptor Pik, which mediates resistance to M. oryzae expressing the avirulence effector gene AVR-Pik. Rice Piks-1, encoded by an allele of Pik-1, recognizes a previously unidentified effector encoded by the M. oryzae avirulence gene AVR-Mgk1, which is found on a mini-chromosome. AVR-Mgk1 has no sequence similarity to known AVR-Pik effectors, and is prone to deletion from the mini-chromosome mediated by repeated Inago2 retrotransposon sequences. AVR-Mgk1 is detected by Piks-1 and by other Pik-1 alleles known to recognize AVR-Pik effectors; recognition is mediated by AVR-Mgk1 binding to the integrated heavy metal-associated (HMA) domain of Piks-1 and other Pik-1 alleles. Our findings highlight how complex gene-for-gene interaction networks can be disentangled by applying forward genetics approaches simultaneously to the host and pathogen. We demonstrate dynamic co-evolution between an NLR integrated domain and multiple families of effector proteins.

plant biology↗

bioRxiv2022

Tracing back primed resistance in cancer via sister cells

Exploring non-genetic evolution of cell states during cancer treatments has become attainable by recent advances in lineage-tracing methods. However, transcriptional changes that drive cells into resistant fates may be subtle, necessitating high resolution analysis. We developed ReSisTrace that uses shared transcriptomic features of synchronised sister cells to predict the states that prime treatment resistance. We applied ReSisTrace in ovarian cancer cells perturbed with olaparib, carboplatin or natural killer (NK) cells. The pre-resistant phenotypes were defined by cell cycle and proteostatic features, reflecting the traits enriched in the upcoming subclonal selection. Furthermore, DNA repair deficiency rendered cells susceptible to both DNA damaging agents and NK killing in a context-dependent manner. Finally, we leveraged the pre-resistance profiles to predict and validate small molecules driving cells to sensitive states prior to treatment. In summary, ReSisTrace resolves pre-existing transcriptional features of treatment vulnerability, facilitating both molecular patient stratification and discovery of synergistic pre-sensitizing therapies.

cancer biology↗

bioRxiv2022

Trackable and scalable LC-MS metabolomics data processing using asari

Significant challenges still exist in the computational processing of data from LC-MS metabolomic experiments into metabolite features. In this study, we examine the issues of provenance and reproducibility in the current software tools. The inconsistency among these tools is attributed to the deficiencies of mass alignment and controls of feature quality. To address these issues, we have developed a new open-source software tool, asari, for LC-MS metabolomics data processing. Asari is designed with a set of new algorithmic framework and data structures, and all steps are explicitly trackable. Asari compares favorably to other tools in feature detection and quantification. It offers substantial improvement of computational performance over current tools, and is highly scalable.

biochemistry↗

bioRxiv2022

Cross-modal Graph Contrastive Learning with Cellular Images

Constructing discriminative representations of molecules lies at the core of a number of domains such as drug discovery, material science, and chemistry. State-of-the-art methods employ graph neural networks (GNNs) and self-supervised learning (SSL) to learn the structural representations from unlabeled data, which can then be fine-tuned for downstream tasks. Albeit powerful, these methods that are pre-trained solely on molecular structures cannot generalize well to the tasks involved in intricate biological processes. To cope with this challenge, we propose using high-content cell microscopy images to assist in learning molecular representation. The fundamental rationale of our method is to leverage the correspondence between molecular topological structures and the caused perturbations at the phenotypic level. By including cross-modal pre-training with different types of contrastive loss functions in a unified framework, our model can efficiently learn generic and informative representations from cellular images, which are complementary to molecular structures. Empirical experiments demonstrated that the model transfers non-trivially to a variety of downstream tasks and is often competitive with the existing SSL baselines, e.g., a 15.4% absolute Hit@10 gains in graph-image retrieval task and a 4.0% absolute AUC improvements in clinical outcome predictions. Further zero-shot case studies show the potential of the approach to be applied to real-world drug discovery.

bioinformatics↗

bioRxiv2022

TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction

Illuminating interactions between proteins and small drug molecules is a longstanding challenge in the field of drug discovery. Despite the importance of understanding these interactions, most previous works are limited by hand-designed scoring functions and insufficient conformation sampling. The recently-proposed graph neural network-based methods provides alternatives to predict protein-ligand complex conformation in a one-shot manner. However, these methods neglect the geometric constraints of the complex structure and weaken the role of local functional regions. As a result, they might produce unreasonable conformations for challenging targets and generalize poorly to novel proteins. In this paper, we propose Trigonometry-Aware Neural networKs for binding structure prediction, TANKBind, that builds trigonometry constraint as a vigorous inductive bias into the model and explicitly attends to all possible binding sites for each protein by segmenting the whole protein into functional blocks. We construct novel contrastive losses with local region negative sampling to jointly optimize the binding interaction and affinity. Extensive experiments show substantial performance gains in comparison to state-of-the-art physics-based and deep learning-based methods on commonly-used benchmark datasets for both binding structure and affinity predictions with variant settings.

biophysics↗

bioRxiv2022

The SARS-CoV-2 accessory factor ORF7a downregulates MHC class I surface expression

The pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in over 500 million infections and more than six million deaths worldwide. Although the viral genomes of SARS-CoV-1 and SARS-CoV-2 share high sequence homology, the clinical and pathological features of COVID-19 differ profoundly from those of SARS. It is apparent that changes in viral genes contribute to the increased transmissibility of SARS-CoV-2 and pathology of COVID-19. Cytotoxic T lymphocytes play a key role in the elimination of virus-infected cells, mediated by recognition of virus-derived peptides that are presented on MHC class I molecules. Here, we show that SARS-CoV-2 can interfere with antigen presentation thereby evading immune surveillance. SARS-CoV-2 infection of monkey and human cell lines resulted in reduced cell-surface expression of MHC class I molecules. We identified a single viral gene product, the accessory factor open reading frame 7a (ORF7a), that mediates this effect. ORF7a interacts with HLA class I molecules in the ER, resulting in ER retention or impaired HLA heavy chain (HC) trafficking to the Golgi. Ultimately, these actions result in reduced HLA class I surface expression on infected cells. Whereas ORF7a from SARS-CoV-2 reduces surface HLA class I levels, the homologous ORF7a from the 2002 pandemic SARS-CoV-1 did not, suggesting that SARS-CoV-2 ORF7a acquired the ability to downregulate HLA-I during evolution of the virus. We identified a single amino acid in the SARS-CoV-1 ORF7a luminal domain that, upon mutating to the corresponding SARS-CoV-2 ORF7a sequence, induced a gain-of-function in HLA surface downregulation. By abrogating HLA class I antigen presentation via ORF7a, SARS-CoV-2 may evade host immune responses by inhibiting anti-viral cytotoxic T cell activity, thereby contributing to the pathology of COVID-19.

microbiology↗

bioRxiv2022

The LEPIS-HuR-TMOD4 axis regulates hepatic cholesterol homeostasis and accelerates atherosclerosis

Long noncoding RNAs (lncRNAs) play an important role in the entire progression of atherosclerosis. In this study, we identified an uncharacterized lncRNA, Liver Expressions by PSRC1 Induce Specifically (LEPIS). The expression of LEPIS and its potential target tropomodulin 4 (TMOD4) in the liver of ApoE-/- mice fed a high-fat diet was increased. An ApoE-/- mouse model with the overexpression of LEPIS or TMOD4 in liver was established, and we found that both LEPIS and TMOD4 increased the burden of atherosclerosis and reduced hepatic cholesterol levels. Further study revealed that LEPIS and TMOD4 affect the expression of genes related to hepatic cholesterol homeostasis, including proprotein convertase subtilisin/kexin type9 (PCSK9) and low-density lipoprotein receptor (LDLR), which are closely related to hypercholesterolemia. Mechanistically, human antigen R (HuR), an RNA-binding protein, was shown to be critical for the regulation of TMOD4 by LEPIS. Further, we found that overexpression of LEPIS promoted the shuttling of HuR from the nucleus to the cytoplasm, enhanced the stability of TMOD4 mRNA, and in turn promoted the expression of TMOD4. In addition, TMOD4 was found to affect intracellular cholesterol levels through PCSK9. These results suggest that the LEPIS-HuR-TMOD4 axis is a potential intervention target for hepatic cholesterol homeostasis and atherosclerosis. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=183 SRC="FIGDIR/small/490400v1_ufig1.gif" ALT="Figure 1"> View larger version (59K): org.highwire.dtl.DTLVardef@c4af60org.highwire.dtl.DTLVardef@d5e7d1org.highwire.dtl.DTLVardef@1667d08org.highwire.dtl.DTLVardef@19abddd_HPS_FORMAT_FIGEXP M_FIG C_FIG

molecular biology↗

bioRxiv2022

Predicting A/B compartments from histone modifications using deep learning

Genomes fold into organizational units in the 3D space that can influence critical biological functions. In particular, the organization of chromatin into A and B compartments segregates its active regions from inactive regions. Compartments, evident in Hi-C contact matrices, have been used to describe cell-type specific changes in the A/B organization. However, obtaining Hi-C data for all cell and tissue types of interest is prohibitively expensive, which has limited the widespread consideration of compartment status. We present a prediction tool called Compartment prediction using Recurrent Neural Network (CoRNN) that models the relationship between the compartmental organization of the genome and histone modification enrichment. Our model predicts A/B compartments, in a cross-cell type setting, with an average area under the ROC curve of 90.9%. Our cell type-specific compartment predictions show high overlap with known functional elements. We investigate our predictions by systematically removing combinations of histone marks and find that H3K27ac and H3K36me3 are the most predictive marks. We then perform a detailed analysis of loci where compartment status cannot be accurately predicted from these marks. These regions represent chromatin with ambiguous compartmental status, likely due to variations in status within the population of cells. These ambiguous loci also show highly variable compartmental status between biological replicates in the same GM12878 cell type. Finally, we demonstrate the generalizability of our model by predicting compartments in independent tissue samples. Our software and trained model are publicly available at https://github.com/rsinghlab/CoRNN.

bioinformatics↗

bioRxiv2022

Cadmium is detrimental to Caenorhabditis elegans via anetwork involving circRNA, lncRNA and phosphorylatedprotein

Cadmium (Cd) as a heavy metal causes serious environmental pollution and multiple organ and system damage in human. However, little is known about the specific molecular mechanisms of the associated regulatory networks. In this study, we selected Caenorhabditis elegans (C. elegans) to investigate the effects of Cd exposure as it acts as an acknowledged and established genetic model organism. A total of 26 differentially-expressed circular RNA (DEcircRNAs), 143 lncRNAs (DElncRNAs), 69 microRNAs (DEmiRNAs) and 6209 mRNAs (DEmRNAs) were found and identified, which might influence reproductive function, aging processes and nervous system functions through regulating the levels of circRNAs and lncRNAs and the controlling of regulatory networks of circRNA/lncRNA-miRNA-mRNA. Based on quantitative PCR, four DEcircRNAs and three DElncRNAs were confirmed to have different expression levels between the Cd-treated and control group. Further, 5 protein-coding genes might be regulated by DElnRNAs through cis-acting and 114 by trans-acting elements. Additionally, 42 differentially regulative phosphopeptides were detected and 4 novel pairs of transcription factors (TFs)-kinase-substrate that might be influenced by Cd exposure were constructed by phosphoproteomics. Our findings suggest that Cd might influence multi-functions and the aging process of C. elegans and may inhibit the expression of TFs to reduce phosphorylated levels of the corresponding protein. SynopsisCadmium exists widely in soil, water and air. This study manifested the regulatory network involving circRNA, lncRNA and phosphorylated protein in C.elegans after Cd exposure, which revealing the potential molecular mechanism underlying the toxic effect caused by Cd. Graphic Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=82 SRC="FIGDIR/small/486470v1_ufig1.gif" ALT="Figure 1"> View larger version (26K): org.highwire.dtl.DTLVardef@1a012bborg.highwire.dtl.DTLVardef@2dc128org.highwire.dtl.DTLVardef@1e43ebaorg.highwire.dtl.DTLVardef@1d236b8_HPS_FORMAT_FIGEXP M_FIG C_FIG

molecular biology↗

bioRxiv2022

Co-expression of recombinant human collagen α1(III) chain with viral prolyl 4-Hydroxylase in P. pastoris GS115

Prolyl 4-hydroxylase (P4H) is essential to maintain the stable triple-helix structure and function of human collagen 1([SHcy]) chain (COL3A1). To obtain hydroxylated human COL3A1, the human COL3A1 and the viral P4H A085R were co-expressed in P. pastoris GS115. The sequence of human COL3A1 without N-terminal and C-terminal was selected for expression. Colony PCR analysis and sequencing after transfection showed that the target gene had inserted successfully. Real-time quantitative PCR (RT-qPCR) indicated that human COL3A1 and P4H were expressed at the mRNA levels. SDS-PAGE and Western blotting analysis of supernatant from the recombinant methylotrophic yest culture showed that recombinant human COL3A1 (rhCOL3A1) was secreted into the culture medium with an apparent molecular mass of approximately 130 kDa. It was noted that the rhCOL3A1 expession quantity was higest at 120 h of induction. Furthermore, mass spectrometry analysis demonstrated that the rhCOL3A1 was expressed successfully. His-tagged rhCOL3A1 protein was purified by Ni-affinity column.

biochemistry↗

bioRxiv2022

Secreted dengue virus NS1 is predominantly dimeric and in complex with high-density lipoprotein

Severe dengue infections are characterized by endothelial dysfunction shown to be associated with the secreted nonstructural protein 1 (sNS1), making it an attractive vaccine antigen and biotherapeutic target. To uncover the biologically relevant structure of sNS1, we obtained infection-derived sNS1 (isNS1) from DENV-infected Vero cells through immunoffinity purification instead of recombinant sNS1 (rsNS1) overexpressed in insect or mammalian cell lines. We found that isNS1 appeared as an approximately 250 kDa complex of NS1 and ApoA1 and further determined the cryoEM structures of isNS1 and its complex with a monoclonal antibody/Fab. Indeed, we found that the major species of isNS1 is a complex of the NS1 dimer partially embedded in a High-Density Lipoprotein (HDL) particle. Cross-linking mass spectrometry (XL-MS) studies confirmed that the isNS1 interacts with the major HDL component ApoA1 through interactions that map to the NS1 wing and hydrophobic domains. Furthermore, our studies demonstrated that the sNS1 in sera from DENV-infected mice and a human patient form a similar complex as isNS1. Our results report the molecular architecture of a biological form of sNS1 which may have implications for the molecular pathogenesis of dengue. SummaryCryoEM structures of secreted dengue virus NS1 protein reveal dimers in complex with high-density lipoprotein.

microbiology↗