Biology subjects

Xiong, X.

Publications and source records attributed to Xiong, X..

At least 55 records · Page 3Linked to original sources

bioRxiv2024

In-silico modeling and interpretation of RBP binding disentangle m6A-RBP interaction

RNA binding protein (RBP) binding and N6-methyladenosine (m6A) are both essential post-transcriptional regulatory layers for RNA fate decisions. However, the intricate mechanism underlying the interaction between m6A and RBP binding remains underexplored. Here, we develop TransRBP, an interpretable deep learning framework, to model the base-resolution binding of RBPs from RNA sequences and to subsequently investigate the interaction between m6A and RBPs. TransRBP achieves a median accuracy of 0.59 across 32 m6A-related RBPs, representing a 28% increase over the state-of-the-art model. Using gradient-based interpretation, we demonstrate that the binding motifs of the m6A-related RBPs strongly enrich for splicing consensus, laying a foundation for studying the RBP-dependent crosstalk between m6A and splicing. Moreover, we develop an in-silico mutagenesis assay to assess the impact of m6A on RBPs, and utilize the self-attention mechanism to elucidate the interplay between RBP binding and m6A. We further uncover 1,806 variant-RBP combinations with the in-silico mutagenesis, revealing variants that strongly alter RBP binding for genetic diseases including Parkinsons disease, autism, and cardiomyopathy. In particular, we identify m6A cis-acting variants that alter RBP binding in an m6A-proximal manner, including the binding of UPF1 that contributes to Alzheimers disease, and the DDX3X binding to cardiomyopathy and muscular dystrophy. Together, TransRBP accurately models the binding of RBP and its interaction with m6A, shedding light on the m6A-RBP dynamics and providing multi-layer mechanistic insights for genetic diseases.

bioinformatics↗

bioRxiv2024

ADAT2/3-mediated tRNA editing promotes cancer cell growth and tumorigenicity

Transfer RNAs (tRNAs) are subject to various chemical modifications that influence their stability or function. Adenosine to Inosine (A-to-I) editing in the tRNA anticodon at position A34 is an important modification that expands anticodon-codon recognition at the wobble position and is required for normal mRNA translation. The relevance of tRNA editing in cancer remains unexplored. Here we show that the genes encoding the ADAT2/3 deaminase complex, responsible for A-to-I tRNA editing in humans, are commonly amplified and/or overexpressed in several tumor types including liposarcoma (LPS). We find that knockdown of the ADAT complex suppresses LPS cell growth and tumorigenicity. Mechanistically, we find that decreased tRNA editing upon ADAT2 depletion leads to defective translation of a subset of mRNAs. Thus, ADAT-mediated tRNA modification promotes oncogenesis by enhancing the translation of growth promoting mRNAs that are enriched in NNC codons that lack cognate tRNAs and therefore depend on A-I tRNA editing for decoding and mRNA translation. Our results uncover an oncogenic role of tRNA editing and identify ADAT2/3 as a potential new cancer therapeutic target.

cancer biology↗

bioRxiv2024

Dynamics of duplicated gene regulatory networks governing cotton fiber development following polyploidy

Cotton fiber development entails complex genome-wide gene regulatory networks (GRN) that remain mostly unexplored. Here we present integrative analyses of fiber GRNs using public RNA-seq datasets, integrated with multi-omics genomic, transcriptomic, and cistromic data. We detail the fiber co-expression dynamics and regulatory connections, validating findings with external datasets and transcription factor (TF) binding site data. We elucidate previously uncharacterized TFs that regulate genes involved in fiber-related functions and cellulose synthesis, and identify the regulatory role of two homoeologous G2-like transcription factors on fiber length. Analysis of duplicated gene expression and network relationships in allopolyploid cotton, which has two co-resident genomes (A, D), revealed novel aspects of asymmetric subgenomic developmental contributions. Whereas D-based homoeolog pairs drive higher overall gene expression from the D subgenome, TFs from the A subgenome play a preferential regulatory role in the fiber gene regulatory network. Following allopolyploid formation, it appears that the trans-regulatory roles of TFs diversified more rapidly between homoeologs than did the cis-regulatory elements of their target genes. Our approach underscores the utility of network analysis for detection of master regulators and provides fresh perspectives on fiber development and polyploid functional genomics, through the lens of co-expression and GRN dynamics.

bioinformatics↗

bioRxiv2024

A high-resolution model of gene expression during Gossypium hirsutum (cotton) fiber development

Cotton fiber development relies on complex and intricate biological processes to transform newly differentiated fiber initials into the mature, extravagantly elongated cellulosic cells that are the foundation of this economically important cash crop. Here we extend previous research into cotton fiber development by employing controlled conditions to minimize variability and utilizing time-series sampling and analyses to capture daily transcriptomic changes from early elongation through the early stages of secondary wall synthesis (6 to 24 days post anthesis; DPA). A majority of genes are expressed in fiber, largely partitioned into two major coexpression modules that represent genes whose expression generally increases or decreases during development. Differential gene expression reveals a massive transcriptomic shift between 16 and 17 DPA, corresponding to the onset of the transition phase that leads to secondary wall synthesis. Subtle gene expression changes are captured by the daily sampling, which are discussed in the context of fiber development. Coexpression and gene regulatory networks are constructed and associated with phenotypic aspects of fiber development, including turgor and cellulose production. Key genes are considered in the broader context of plant secondary wall synthesis, noting their known and putative roles in cotton fiber development. The analyses presented here highlight the importance of fine-scale temporal sampling on understanding developmental processes and offer insight into genes and regulatory networks that may be important in conferring the unique fiber phenotype.

cell biology↗

bioRxiv2024

A duplexed CRISPRi-based approach to predict synergistic antibiotic targets in Escherichia coli.

Co-repressing drug targets with CRISPR interference (CRISPRi) has the potential to predict both synergistic and additive drug combinations. In a screen experiment with Escherichia coli, pairwise gene co-repression assays correctly predicted the interaction type observed for 36 antibiotic pairs in checkerboard assays. Furthermore, we used this approach to identify promising synergistic interactions with a gene currently not targeted by clinical antibiotics. CRISPRi can systematically test the impact of gene inhibitions and identify novel drug targets.

microbiology↗

bioRxiv2024

TopDIA: A Software Tool for Top-Down Data-Independent Acquisition Proteomics

Top-down mass spectrometry is widely used for proteoform identification, characterization, and quantification owing to its ability to analyze intact proteoforms. In the last decade, top-down proteomics has been dominated by top-down data-dependent acquisition mass spectrometry (TD-DDA-MS), and top-down data-independent acquisition mass spectrometry (TD-DIA-MS) has not been well studied. While TD-DIA-MS produces complex multiplexed tandem mass spectrometry (MS/MS) spectra, which are challenging to confidently identify, it selects more precursor ions for MS/MS analysis and has the potential to increase proteoform identifications compared with TD-DDA-MS. Here we present TopDIA, the first software tool for proteoform identification by TD-DIA-MS. It generates demultiplexed pseudo MS/MS spectra from TD-DIA-MS data and then searches the pseudo MS/MS spectra against a protein sequence database for proteoform identification. We compared the performance of TD-DDA-MS and TD-DIA-MS using Escherichia coli K-12 MG1655 cells and demonstrated that TD-DIA-MS with TopDIA increased proteoform and protein identifications compared with TD-DDA-MS.

bioinformatics↗

bioRxiv2024

Deep learning modeling of ribosome profiling reveals regulatory underpinnings of translatome and interprets disease variants

Gene expression involves transcription and translation. Despite large datasets and increasingly powerful methods devoted to calculating genetic variants effects on transcription, discrepancy between mRNA and protein levels hinders the systematic interpretation of the regulatory effects of disease-associated variants. Accurate models of the sequence determinants of translation are needed to close this gap and to interpret disease-associated variants that act on translation. Here, we present Translatomer, a multimodal transformer framework that predicts cell-type-specific translation from mRNA expression and gene sequence. We train Translatomer on 33 tissues and cell lines, and show that the inclusion of sequence substantially improves the prediction of ribosome profiling signal, indicating that Translatomer captures sequence-dependent translational regulatory information. Translatomer achieves accuracies of 0.72 to 0.80 for de novo prediction of cell-type-specific ribosome profiling. We develop an in silico mutagenesis tool to estimate mutational effects on translation and demonstrate that variants associated with translation regulation are evolutionarily constrained, both within the human population and across species. Notably, we identify cell-type-specific translational regulatory mechanisms independent of eQTLs for 3,041 non-coding and synonymous variants associated with complex diseases, including Alzheimers disease, schizophrenia, and congenital heart disease. Translatomer accurately models the genetic underpinnings of translation, bridging the gap between mRNA and protein levels, and providing valuable mechanistic insights toward mapping "missing regulation" in disease genetics.

genomics↗

bioRxiv2024

Ilamycins are dual inhibitors of ClpX and ClpC1 involved in proteostasis in Mycobacteria

The mycobacterial caseinolytic protease (Clp) system has been recognized as a promising therapeutic target. In this study, we identified two novel ilamycin analogs, ilamycin E (ILE) and ilamycin F (ILF), both targeting the ClpC1 component of the ClpC1P1P2 proteasome. ILE potently disrupts ClpC1P1P2-mediated proteolysis, leading to delayed bactericidal activity, while ILF also binds ClpC1, albeit with lower affinity. Notably, we discovered a unique mutation in clpX and a novel insertion in clpC1, both conferring resistance to ILE and ILF in mycobacterium, validated by gene editing. Furthermore, ILE can also inhibit the proteolytic activity of ClpXP1P2 in a manner dependent on the substrates tag sequence and adaptor. This first demonstration of clpX- and clpC1-mediated ilamycin resistance underscores the potential of ilamycins to target multiple components of the Clp protease system, offering a novel dual-target strategy for combating mycobacterial infections.

microbiology↗

bioRxiv2024

Human Motor Cortex Encodes Complex Handwriting Through a Sequence of Primitive Neural States

How the human motor cortex (MC) orchestrates sophisticated fine movements such as handwriting remains a puzzle1-3. Here, we investigate this question through Utah array recordings from human MC hand knob, during imagined handwriting of Chinese characters (306 characters tested, 6.3 {+/-} 2.0 strokes per character). We find MC programs the writing of complicated characters by sequencing a small set of primitive states: The directional tuning of motor neurons remains stable within each primitive state but strongly varies across states. Furthermore, the occurrence of a primitive state is encoded by a separate set of neurons not directly involved in movement control. By automatically identifying the primitive states and corresponding neuronal tuning properties, we can reconstruct a recognizable writing trajectory for each character (84% improvement in reconstruction accuracy compared with baseline). Our findings unveil that skilled, sophisticated movements are decomposed into a sequence of primitive movements that are programmed through state-specific neural configurations, and this hierarchical control mechanism sheds new light on the design of high-performance brain-computer interfaces.

neuroscience↗

bioRxiv2024

Ligand supplementation restores the cancer therapy efficacy of an antirheumatic drug auranofin from serum inactivation

Auranofin, an FDA-approved antirheumatic gold drug, has gained ongoing interest in clinical studies for treating advanced or recurrent tumors. However, gold ions dynamic thiol exchange nature strongly attenuates its bioactivity due to the fast formation of covalent albumin-gold adducts. Here we report that newly-added thiols can modulate the dynamic albumin-gold binding and recover the therapeutic efficacy. Initially, we found that auranofin supplemented with its own thiol ligand, TGTA (1-thio-{beta}-D-glucose tetraacetate), significantly restored the anticancer activities in cells and patient-derived xenograft models. Then, screening a collection of ligand fragments followed by machine learning evaluation unveiled diverse synergizing thiols, including pantethine that effectuates auranofin at a low dosage used for rheumatoid arthritis. Interestingly, the thiol exchange inside cells accounts for a cuproptosis-like phenotype induced by auranofin. Together, we believe the ligand-enabled dynamic modulation strategy is of value to researchers and clinicians contemplating metallodrugs and ligand-like molecules in cancer therapy.

cancer biology↗

bioRxiv2024

A removable and cosmopolitan dsRNA Toti-like virus causes latent infection in a model diatom strain

Diatoms contribute to 20% of global primary productivity. Although some diatom viruses have been identified, the molecular mechanisms underlying their interactions with the host remain poorly understood. In this study, we report the discovery of an RNA molecule in the DNA extracts of the Phaeodactylum tricornutum strain Pt1, which possesses a well-annotated genome and has been used as a diatom model system since 1956. We confirmed this molecule to be a double-stranded linear RNA molecule and, through sequencing, demonstrated it to be a virus in the Totiviridae family that is prevalent among marine stramenopiles. We also detected this virus in Phaeodactylum tricornutum strain Pt3, which was collected in 1930s from a similar geographic location to Pt1, suggesting its prevalence within the region. By employing various inhibitors of the viral RNA-dependent RNA polymerase, we successfully generated a virus-free line isogenic to Pt1, establishing a model system to investigate the impact of RNA viruses on diatom physiology. The virus-free lines did not display obvious growth advantages or defects, indicating a tendency of the virus towards latent infection. Furthermore, we generated a robust antibody against the coat protein of this virus. By performing immunoprecipitation coupled with mass spectrometry, we found that translation-related proteins are enriched as potential interacting partners of the coat protein. Our results suggest that potential viral impacts in molecular research should be considered when Pt1 and Pt3 are used for studying translation-related processes. Additionally, our study unveiled a temperate mode of interaction between viruses and marine algal hosts that differs from the commonly-reported virulent, lytic infections. HighlightsO_LIprevalent dsRNA virus belonging to the Totiviridae family was discovered in the Pt1 and Pt3 strains of the model diatom Phaeodactylum tricornutum. C_LIO_LIvirus is absent in eight other strains of P. tricornutum, highlighting the importance of studying multiple accessions. C_LIO_LIvirus can be eliminated using a nucleotide analogue, resulting in a virus-free isogenic strain that allows us to investigate how viruses may affect diatom physiology. C_LIO_LIrobust antibody against the coat protein of this virus was developed to assist mechanistic studies of diatom-virus interactions. C_LI

microbiology↗

bioRxiv2023

Structural insights into the distinct protective mechanisms of human antibodies targeting ZIKV NS1

Zika virus (ZIKV) Non-structural protein 1 (NS1) plays an essential role in viral replication and immune evasion. Our understanding of the differential protective mechanism of NS1-targeting antibodies is limited. Here, we determined the cryoEM structures of ZIKV NS1 in complex with two group antibodies at 2.6-2.9[A]. Group I antibodies (3G2 and 4B8) potently recognize cell surface form of NS1 and multiple oligomeric forms of NS1 by occupy the epitopes on outer surface of dimeric NS1. IgG and Fab from group I antibodies completely abrogate sNS1-mediated endothelial dysfunction in vitro. Group II antibodies (4F10, 2E11, and 14G5) recognize the previously reported epitopes in distal end of the {beta}-ladder domain of monomeric NS1, and their blockade efficiency depends on the affinity with NS1 protein. These findings elucidate the correlation between the epitope recognition and the protective efficacy of anti-NS1 antibodies and highlight the distinct mechanisms of therapeutic potential of 3G2 and 4B8.

microbiology↗

bioRxiv2023

Antibodies utilizing VL6-57 light chains target a convergent cryptic epitope on SARS-CoV-2 spike protein driving the genesis of Omicron variants

Continued evolution of SARS-CoV-2 generates variants to challenge antibody immunity established by infection and vaccination. A connection between population immunity and genesis of virus variants has long been suggested but its molecular basis remains poorly understood. Here, we identify a class of SARS-CoV-2 neutralising public antibodies defined by their shared usage of VL6-57 light chains. Although heavy chains of diverse genotypes are utilized, convergent HCDR3 rearrangements have been observed among these public antibodies to cooperate with germline VL6-57 LCDRs to target a convergent epitope defined by RBD residues S371-S373-S375. Antibody repertoire analysis identifies that this class of VL6-57 antibodies is present in SARS-CoV-2-naive individuals and is clonally expanded in most COVID-19 patients. We confirm that Omicron specific substitutions at S371, S373 and S375 mediate escape of antibodies of the VL6-57 class. These findings support that this class of public antibodies constitutes immune pressure promoting the introduction of S371L/F-S373P-S375F in Omicron variants. The results provide further molecular evidences to support that antigenic evolution of SARS-CoV-2 is driven by antibody mediated population immunity.

biochemistry↗

bioRxiv2023

Anti-adipogenic properties of clock activator chlorhexidine and a new derivative

BackgroundThe circadian clock exerts temporal control of metabolic pathways to maintain homeostasis, and its disruption leads to the development of obesity and insulin resistance. In adipose tissue, key regulators of clock machinery orchestrate adipogenic processes via the Wnt signaling pathway to impact mature adipocyte development. MethodsBased on the recent finding of chlorhexidine as a new clock activator, we determined its potential anti-adipogenic activities in distinct adipogenic progenitor models. Furthermore, we report the structural optimization of chlorhexidine leading to the discovery of analogs with improved efficacy in inhibiting adipogenesis. ResultsIn adipogenic progenitors with Per2::dLuc luciferase reporter, Chlorhexidine shortened clock period length with induction of core clock components. Consistent with its clock-activating function, Chlorhexidine robustly suppressed the lineage commitment and maturation of adipogenic mesenchymal precursors, with comparable effect on inhibiting preadipocyte terminal differentiation. Mechanistically, we show that Chlorhexidine induces signaling components of the Wnt pathway resulting in activation of Wnt activity. Via modification of its chemical scaffold, we generated analogs of chlorhexidine that led to the identification of CM002 as a new clock- activating molecule with improved anti-adipogenic activity. ConclusionsCollectively, our findings uncovered the anti-adipogenic functions of a new class of small molecule clock activators. These compounds provide novel chemical probes to dissect clock function in maintaining metabolic homeostasis and may have therapeutic implications in obesity and associated metabolic disorders.

cell biology↗

bioRxiv2023

Emergent Antibiotic Persistence in Cross-Feeding Bacteria in Structured Habitats

Antibiotic persistence allows a sub-population of bacteria to survive antibiotic-induced killing and contributes to the evolution of antibiotic resistance. Although bacteria typically live in microbial communities with complex ecological interactions, little is known about how microbial ecology affects antibiotic persistence. Here, we demonstrated that the combination of cross-feeding and community spatial structure can emergently cause high antibiotic persistence in bacteria by increasing the cell-cell heterogeneity in the metabolic state both during growth and antibiotic killing. We tracked the ampicillin-induced death on agar surfaces in a model obligate mutualism of Escherichia coli and Salmonella enterica. We found that E. coli formed [~]100-fold more antibiotic persisters in the cross-feeding coculture than in monoculture. This high persistence could not be explained solely by the presence of S. enterica, the presence of cross-feeding, or the growth rate differences between the mono- and co-cultures. Time-series fluorescent microscopy revealed increased cell-cell variation in E. coli lag time in the mutualistic co-culture. Furthermore, we discovered that an E. coli cell can survive antibiotic killing if the nearby S. enterica cells on which it relies die first--a mechanism we termed "dynamical loss of access to nutrient." In conclusion, we showed that the high antibiotic persistence phenotype can be an emergent phenomenon caused by a combination of cross-feeding and spatial structure. Our work highlights the importance of considering spatially-structured interactions during antibiotic treatment and to understand microbial community resilience more broadly.

microbiology↗

bioRxiv2023

A chromosome-level reference genome of a Fabaceae species yam bean (Pachyrhizus erosus)

Yam bean (Pachyrhizus erosus L.), belonging to the family Fabaceae, is an important but underutilized root crop. Here, we generated a high-quality chromosome-level reference genome of yam bean by PacBio HiFi and Hi-C sequencing, with assembly size of 539.0 Mb, contig N50 of 25.6 Mb, and BUSCO complete rate of 99.3%. Then, we anchored 94.4% of the contig sequences into 11 pseudo-chromosomes, and assembled the telomeres at 86.4% (19/22) of the chromosome-ends. A total of 44,692 protein-coding genes were predicted, with the BUSCO complete rate of 99.3%, comparable to that of the genome assembly. Compared to the previously reported yam bean genome, the current assembly has a 1,388-fold increase in contig N50 size, and 12.2% and 24.3% increase in BUSCO complete rate for the genome sequence and gene set, respectively. Evolutionary analysis revealed that yam bean diverged from the clade of soybean and Pueraria lobata var. montana 22.5 MYA. This high-quality genome assembly will greatly facilitate the breeding of yam bean based on the genetic and genomic methods.

genomics↗

bioRxiv2023

Pervasive mislocalization of pathogenic coding variants underlying human disorders

Widespread sequencing has yielded thousands of missense variants predicted or confirmed as disease-causing. This creates a new bottleneck: determining the functional impact of each variant - largely a painstaking, customized process undertaken one or a few genes or variants at a time. Here, we established a high-throughput imaging platform to assay the impact of coding variation on protein localization, evaluating 3,547 missense variants of over 1,000 genes and phenotypes. We discovered that mislocalization is a common consequence of coding variation, affecting about one-sixth of all pathogenic missense variants, all cellular compartments, and recessive and dominant disorders alike. Mislocalization is primarily driven by effects on protein stability and membrane insertion rather than disruptions of trafficking signals or specific interactions. Furthermore, mislocalization patterns help explain pleiotropy and disease severity and provide insights on variants of unknown significance. Our publicly available resource will likely accelerate the understanding of coding variation in human diseases.

cell biology↗

bioRxiv2023

DeSide: A unified deep learning approach for cellular decomposition of bulk tumors based on limited scRNA-seq data

Cellular decomposition employing bulk RNA-sequencing (RNA-seq) has been consistently under investigation due to its high fidelity, ease of use, and cost-effectiveness compared to single cell RNA-sequencing (scRNA-seq). However, the intricate nature of the tumor microenvironment, and the significant heterogeneity among patients and cells have made it challenging to precisely evaluate the cellular composition of solid tumors using a unified model. In this work, we developed DeSide, a deep learning and single-cell decomposition method for solid tumors, to estimate proportions of cell types presented in tumor samples. Our new deep neural network (DNN) architecture considers only non-cancerous cells during the training process, indirectly calculating the proportion of cancerous cells. This approach avoids directly handling the often more variable heterogeneity of cancerous cells, and instead leverages scRNA-seq data from three different cancer types to empower the DNN model with a robust generalization capability across diverse cancers. Additionally, we used a new sampling method and filtering strategies to simulate the gene expression profiles (GEPs) of solid tumors, creating a carefully controlled training set that could be compared to the bulk RNA-seq data from The Cancer Genome Atlas (TCGA), a database of bulk RNA-seq data collected from cancer patients. Relying on limited yet diverse scRNA-seq data, our approach outperformed current methods in accurately predicting the celluar composition of samples from TCGA and an additional validation set. Furthermore, we demonstrated that the predicted cellular composition can be utilized to stratify cancer patients into different groups with varying overall survival rates. With increased availability of scRNA-seq data for various types of tumors, DeSide holds the potential for a more precise cellular decomposition model using bulk RNA-seq.

bioinformatics↗