Biology subjects

Cheung, J.

Publications and source records attributed to Cheung, J..

At least 19 recordsLinked to original sources

bioRxiv2026

Mapping the Functional Landscape and Viral Diversity of 2A Peptides

2A peptides are short (~18 to 22 amino acid) sequences utilized by a subset of RNA viruses to process their polyproteins in an unusual co-translational event known as StopGo or ribosome skipping. This process involves ribosome pausing at a conserved C-terminal D(V/I)ExNPGP motif, followed by peptidyl-tRNA hydrolysis of the nascent peptide-tRNAGly, and continued translation from the downstream proline codon. However, how nascent 2A peptide interacts with the ribosomal exit tunnel to promote these event(s) remains poorly understood and the molecular determinants governing 2A mediated StopGo translation remain undefined. Here, we developed a mammalian fluorescence resonance energy transfer (FRET) based high-throughput reporter system to quantitatively measure 2A activity across large libraries of natural and engineered 2A peptides. Systematic mutational analysis of canonical viral 2A peptides from porcine tescovirus and foot-and-mouth disease virus (P2A, F2A) identified key residues required for activity and revealed critical contributions from previously uncharacterized N-terminal residues. Extending this approach to metagenomic viral datasets uncovered conserved sequences governing StopGo efficiency leading to functional ranking of diverse viral 2A peptides in mammalian cells, including identification of a novel subclass that supports high stop-go activity (>99%). Phylogenetic analysis of viral genomes revealed clustering of functional 2A peptides and specific motifs within specific viral lineages branches, suggesting an evolutionary trajectory underlying 2A diversification and optimization. These findings define the sequence and functional landscape of viral 2A peptides and provide mechanistic and evolutionary insights into how nascent peptide ribosome exit tunnel interactions mediate StopGo translation.

biochemistry↗

bioRxiv2026

Device-embedded accelerometry complements neural signals for tracking parkinsonian motor states

Adaptive deep brain stimulation (aDBS) relies on physiological biomarkers to infer motor state and guide therapeutic stimulation in Parkinsons disease. However, neural biomarkers may themselves be altered by stimulation, potentially limiting their utility for closed-loop control. We address this limitation by testing whether DBS device-embedded accelerometers can accurately track Parkinsonian motor state across stimulation conditions. We analysed over 1,900 hours of chronic recordings of subthalamic nucleus (STN), sensorimotor cortical and device-embedded accelerometry signals acquired before and during continuous STN stimulation, alongside continuous wearable assessments of bradykinesia and dyskinesia. Across stimulation conditions, accelerometry-derived features robustly tracked motor symptom severity and outperformed neural features for symptom decoding. Mechanistically, total STN beta power - a widely used biomarker for aDBS - proved less informative because it conflates periodic and aperiodic neural processes with opposing relationships to motor state. Under active stimulation, periodic beta activity showed reduced coupling to symptom severity, whereas STN aperiodic activity, cortical periodic activity and cortico-subthalamic coherence remained comparatively stable. Together, these findings demonstrate that neural and behavioural biomarkers exhibit differential robustness during deep brain stimulation and identify device-embedded accelerometry as a robust behavioural biomarker of motor state, motivating its use in next-generation adaptive DBS systems.

bioengineering↗

bioRxiv2026

Rhomboid protease Rhbdl2 regulates macrophage recruitment and wound regeneration in zebrafish

Tissue regeneration requires tight control of immune cell behavior, yet the mechanisms that restrain immune-driven regenerative responses remain poorly defined. Here, we identify the rhomboid intramembrane serine protease Rhbdl2 as a critical regulator of regeneration in zebrafish. We generated rhbdl2 mutants by CRISPR Cas9 and found that it does not affect normal development, but triggers enhanced regenerative growth following injury, accompanied by increased macrophage accumulation at the wound site, which is accompanied by increased early apoptosis and proliferation. Proteomic analyses reveal increased Rac2 protein levels in rhbdl2 mutants, indicating dysregulated immune signaling. Functionally, Rac2 morpholino oligonucleotides mediated knockdown in rhbdl2 mutant larvae suppresses the elevated macrophage recruitment and enhanced tissue regenerative phenotype. Together, these findings uncover Rhbdl2 as an immune checkpoint that constrains macrophage driven enhanced regeneration, with vast implications for inflammatory disease, fibrosis, and tumor immune interactions.

immunology↗

bioRxiv2025

Overcoming air-water interface induced artifacts in Cryo-EM with protein nanocrates

Contact with the air-water interface can bias the orientation of macromolecules during cryo-EM sample preparation, leading to uneven sample distribution, preferred orientation, and damage to the molecules of interest. To prevent this, we describe a method to encapsulate target proteins within highly hydrophilic, structurally homogeneous, and stable protein shells, which we refer to as "nanocrates" for this purpose. Here, we describe packaging, data acquisition, and reconstruction of three proof-of-principle examples, each illuminating a different aspect of the method: apoferritin (ApoF, demonstrating high-resolution), thyroglobulin (Tg, solving a known preferred orientation problem), and 7,8-dihydroneopterin aldolase (DHNA, a structure previously uncharacterized by cryo-EM).

bioengineering↗

bioRxiv2025

A single MHCII neoepitope mRNA vaccine elicits CD4 T- and B- cell responses promoting endogenous CD8 anti-tumor immunity

Recent progress in therapeutic cancer vaccines has shown promising clinical activity, especially when targeting MHC class I (MHCI) neoantigen-specific CD8+ T cell responses in post-surgical patients. To explore the role of CD4+ T cells in vaccine-dependent tumor rejection, we constructed an mRNA lipoplex vaccine encoding a single MHCII-restricted neoantigen. The vaccine elicited Tfh and Th1 cell responses while decreasing Tregs, leading to rejection of established tumors in mice. IL-21 and IFN-{gamma}, crucial for Tfh and Th1 function respectively, contributed to anti-tumor activity. B cells and neoantigen-specific antibodies were also shown to participate in vaccine efficacy. Additionally, conventional type 1 dendritic cells (cDC1s) were essential for eliciting vaccine-induced CD4+ T cells, and both cDC1s and CD4+ T cells were required to enhance endogenous CD8+ responses, which were crucial for tumor control. Our results suggest that immunizing against MHCII neoantigens alone is sufficient to orchestrate a potent and cooperative immune response against cancer.

immunology↗

bioRxiv2025

Intratumoral cDC1-T Cell Clusters Serve as Sites of Local Costimulation to Enhance CTL-Mediated Tumor Rejection

T cells are essential for anti-tumor immunity, but their ability to eliminate tumors depends on coordinated interactions with type 1 conventional dendritic cells (cDC1s). While cDC1s are known for cross-presenting tumor-derived antigens in lymph nodes to prime CD8+ T cells, their role within the tumor itself remains less well understood. Here, we use the Skin Tumor Array by Micro-Poration (STAMP) model to investigate how cDC1-T cell interactions shape immune responses and influence tumor fate. Our data reveal that it is the spatial distribution of both cDC1s and T cells that determines whether a tumor can be rejected. We defined three primary immunotypes based on the spatial distribution of T cells and cDC1s: T cell-inflamed/dendritic cell-inflamed (TC-In/DC-In) tumors, where T cells and cDC1s co-infiltrate the tumor; T cell-inflamed/dendritic cell-excluded (TC-In/DC-Ex) tumors, where T cells infiltrate but cDC1s remain at the periphery; and T cell-excluded/dendritic cell-excluded (TC-Ex/DC-Ex) tumors, which lack both cDC1 and T cell infiltration. Notably, TC-In/DC-In tumors are more likely to undergo rejection, whereas TC-In/DC-Ex tumors persist despite T cell infiltration. Within TC-In/DC-In tumors, cDC1s engage in direct interactions with T cells, upregulate co-stimulatory molecules, and sustain effector T cell responses, while cDC1s in TC-In/DC-Ex tumors express higher migration-associated genes, suggesting a propensity to exit the tumor. We further show that chemokine modulation, particularly through CXCL9, CCL5, and XCL1, can reshape immune infiltration patterns to promote intra-tumoral cDC1-T cell clustering and improve tumor rejection. These findings underscore the unexpectedly important role of cDC1 positioning and function in sustaining effective anti-tumor immunity and highlight spatially organized cDC1-T cell clusters as critical hubs for local T cell activation.

immunology↗

bioRxiv2025

Standardization of postmortem human brainstem along the rostrocaudal axis to accommodate for heterogeneity in samples

Human postmortem brain tissues provide an indispensable resource that is crucial for the understanding of neurological conditions, whether related to pathology subtype, burden, distribution or cell-type specificity. Pathology staging protocols provide guidelines for standardized sampling of brain tissues, but cover only a subset of regions affected by pathologies. Thus, to study how various neuropathologies and cell types in highly specialized circuit nodes correlate with functions specifically served by these nodes, additional protocols are necessary. This especially applies to brainstem tissues due to the small dimension of regions of interest and interindividual variability of specimens, whether due to procurement or intrinsic differences. Here we systematically assessed factors contributing to heterogeneity in the length of whole brainstem samples and then presented a standardized approach to reproducibly assign rostrocaudal levels, with standardization relying upon readily identifiable internal landmarks. We validated this approach using postmortem MRI imaging. Standardized brainstem length correlated positively with subject height and negatively with subject age of death. By providing a reference series, reproducible levels can be assigned to individual histological sections or MRI images, i.e. when full brainstem specimens are not available and irrespective of platform, promoting reproducibility.

neuroscience↗

bioRxiv2024

Dynamic convergence of autism disorder risk genes across neurodevelopment

Over three hundred and seventy-three risk genes, broadly enriched for roles in neuronal communication and gene expression regulation, underlie risk for autism spectrum disorder (ASD) and developmental delay (DD). Functional genomic studies of subsets of these genes consistently indicate a convergent role in neurogenesis, but how these diverse risk genes converge on a smaller number of biological pathways in mature neurons is unclear. To uncover shared downstream impacts between neurodevelopmental disorder (NDD) risk genes, here we apply a pooled CRISPR approach to contrast the transcriptomic impacts of targeting 29 NDD loss-of-function genes across human induced pluripotent stem cell (hiPSC)-derived neural progenitor cells, glutamatergic neurons, and GABAergic neurons. Points of convergence vary between the cell types of the brain and are greatest in mature glutamatergic neurons, where they broadly target not just synaptic and epigenetic, but unexpectedly, mitochondrial biology. The strongest convergent networks occur between NDD genes with common co-expression patterns in the post-mortem brain, biological annotations, and clinical associations, suggesting that convergence may one-day inform patient stratification and treatment. Towards this, ten out of eleven drugs tested that were predicted to reverse convergent signatures in human cells and/or arousal and sensory processing behaviors in zebrafish ameliorated at least one behavioral phenotype in vivo. Altogether, robust convergence in post-mitotic neurons represents a clinically actionable therapeutic window.

genetics↗

bioRxiv2023

Chronic pain mediated changes in the appetitive value of affective touch in mice

The existence of skin-to-brain circuits for rewarding gentle touch highlight its critical nature across species. However, this gentle affective touch is not always appetitive and can produce aversion or negative affect in disorders such as chronic pain. Sensory neurons expressing the protein MrgprB4 detect gentle stroking in mice and their activation of these neurons known to be positively reinforcing. Here we assess whether activation of channelrhodopsin (ChR2) expressing MrgprB4 afferents signal positively valenced tactile information and whether this is altered in models of chronic pain. We further interrogate how this this sensory information is reflected in the downstream circuits recruited. Optogenetic activation of MrgprB4 lineage afferents was found to be appetitive in control and capsaicin sensitized mice but not nerve injured mice, indicating that the appetitive value is diminished in neuropathic pain. Remarkably, this appetitive value was partially recovered in male nerve injured mice by treatment with the analgesic gabapentin. These behavioral changes were also accompanied by different patterns of neuronal activity throughout the brain, including altered activation of sites that receive direct projections from the spinal cord between sham and nerve injured mice. Together, these findings highlight the plastic nature of these affective tactile circuits under pathological conditions.

neuroscience↗

bioRxiv2023

Long-term epidemiology and evolution of swine influenza viruses in Vietnam

Swine influenza virus (SwIV) surveillance in Hanoi, Vietnam from 2013-2019 found gene pool enrichment from imported swine from Asia and North America. Long-term maintenance, persistence and reassortment of SwIV lineages was observed. Co-circulation of H1-{delta}1a viruses with other SwIV genotypes raises concern due to its zoonotic potential.

microbiology↗

bioRxiv2021

A Biodegradable, Tacrolimus-releasing Nerve Wrap Promotes Peripheral Nerve Regeneration

Axonal regeneration following nerve repair is slow and often incomplete, resulting in poor functional recovery and sometimes lifelong disability. Yet, there are no FDA-approved therapies available to promote nerve regeneration. Tacrolimus accelerates axonal regeneration, but systemic side-effects presently outweigh its potential benefits for peripheral nerve surgery. We have developed a biodegradable drug delivery system for the sustained local release of tacrolimus at the nerve repair site, with suitable properties for large-scale manufacturing and clinical application, aiming to promote axonal regeneration and functional recovery with minimal systemic drug exposure. Tacrolimus is encapsulated in polycarbonate-urethane nanofibers and electrospun to generate an implantable nerve wrap that releases therapeutic doses of bioactive tacrolimus over 31 days. Size and drug loading are adjustable for applications in small and large caliber nerves, and the wrap degrades within 120 days into biocompatible byproducts. Tacrolimus released from the nerve wrap promotes axon elongation in vitro and accelerates nerve regeneration and functional recovery in preclinical nerve repair models while systemic drug exposure is reduced by 80% compared to systemic delivery. Given its surgical suitability and preclinical efficacy and safety, this system may provide a readily translatable approach to support axonal regeneration and recovery in patients undergoing nerve surgery.

neuroscience↗

bioRxiv2021

Epigenetic reprogramming of DCCs into dormancy suppresses metastasis via restored TGFβ-SMAD4 signaling.

Disseminated cancer cells (DCCs) identified in secondary organs, sometimes before the primary tumor becomes detectable and treated, can remain dormant for years to decades before manifesting. Microenvironmental and epigenetic mechanisms may control the onset and escape from dormancy, and here we reveal how a combination of the DNA methylation inhibitor 5-azacytidine (AZA) and retinoic acid receptor ligands all-trans retinoic acid (atRA), orchestrate a novel program of stable dormancy. Treatment of HNSCC tumor cells with AZA+atRA induced a SMAD2/3/4 dependent regulation of downstream transcriptional program that restored the anti-proliferative function of TGF{beta} signaling. Significantly, AZA+atRA or AZA+AM80, an RAR specific agonist, strongly suppresses lung metastasis formation. The metastatic suppression occurs via the induction and maintenance of phenotypically homogenous dormant SMAD4+/NR2F1+ non-proliferative DCCs. These findings suggest that strategies that maintain or induce dormancy programs may be a viable alternative strategy to improve patient outcomes by preventing or significantly delaying metastasis development.

cancer biology↗

bioRxiv2021

ALS and FTD-associated missense mutations in TBK1 differentially disrupt mitophagy

TANK-binding kinase 1 (TBK1) is a multi-functional kinase with an essential role in mitophagy, the selective clearance of damaged mitochondria. More than 90 distinct mutations in TBK1 are linked to amyotrophic lateral sclerosis (ALS) and fronto-temporal dementia (FTD), including missense mutations that disrupt the ability of TBK1 to dimerize, associate with the mitophagy receptor optineurin (OPTN), auto-activate, or catalyze phosphorylation. We investigated how ALS-associated mutations in TBK1 affect Parkin-dependent mitophagy using imaging to dissect the molecular mechanisms involved in clearing damaged mitochondria. Some mutations cause severe dysregulation of the pathway, while others induce limited disruption. Mutations that abolish either TBK1 dimerization or kinase activity were insufficient to fully inhibit mitophagy, while mutations that reduced both dimerization and kinase activity were more disruptive. Ultimately, both TBK1 recruitment and OPTN phosphorylation at S177 are necessary for engulfment of damaged mitochondra by autophagosomal membranes. Surprisingly, we find that ULK1 activity contributes to the phosphorylation of OPTN in the presense of either WT- or kinase inactive TBK1. In primary neurons, TBK1 mutants induce mitochondrial stress under basal conditions; network stress is exacerbated with further mitochondrial insult. Our study further refines the model for TBK1 function in mitophagy, demonstrating that some ALS-linked mutations likely contribute to disease pathogenesis by inducing mitochondrial stress or inhibiting mitophagic flux. Other TBK1 mutations exhibited much less impact on mitophagy in our assays, suggesting that cell-type specific effects, cumulative damage, or alternative TBK1-dependent pathways such as innate immunity and inflammation also factor into the development of ALS in affected individuals. SIGNIFICANCE STATEMENTMissense mutations in TANK-binding kinase 1 (TBK1) have various biophysical and biochemical effects on the molecule, and are associated with the neurodegenerative diseases amyotrophic lateral sclerosis (ALS) and fronto-temporal dementia (FTD). TBK1 plays an essential role in clearing damaged mitochondria. Here, we investigate the impact of 10 ALS-linked TBK1 mutations on the critical early stage of mitophagy. We find that both TBK1 recruitment and kinase activity contribute to the clearance of the damaged mitochondria. Furthermore, in neurons, expression of TBK1 mutants alone affects mitochondrial network health. Our investigation utilizes disease-linked mutations to further refine the current model of mitophagy, identifying crosstalk between the regulatory kinases TBK1 and ULK1, and providing new insights into the roles of TBK1 in neurodegenerative pathogenesis.

cell biology↗

bioRxiv2021

EZH2 Abundance Regulated by UHRF1/UBE2L6/UBR4 Ubiquitin System is the Potential Therapeutic Target to Trigger Pigmented Phenotype in Melanoma

Cellular heterogeneity in cancer is linked to disease progression and therapy response, although the mechanisms regulating distinct cellular states within tumours are not well understood. To address this, we identified melanin pigment content as a major source of phenotypic and functional heterogeneity in melanoma and compared RNAseq data from high (HPC) and low pigmented melanoma cells (LPC), revealing the polycomb repressor complex protein, EZH2, as a master regulator of these states. EZH2 protein, but not RNA expression, was found to be upregulated in LPCs and inversely correlated with melanin in pigmented patient melanomas. Surprisingly, conventional EZH2 methyltransferase inhibitors, GSK126 and EPZ6438, had no effect on LPC survival, clonogenicity and pigmentation, despite fully inhibiting methyltransferase activity. In contrast, EZH2 silencing by siRNA strategy or DZNep, MS1943 that reduces EZH2 protein levels, significantly inhibited cell growth in LPCs by hampering ribosome biogenesis. In addition, decline in EZH2 protein level induces pigmented cell phenotype by inducing melanin biosynthesis. Proteasomal inhibitor, MG132 treatment induced EZH2 protein levels in HPCs prompted us to look for differentially regulated ubiquitin system proteins in HPC vs LPCs. UBE2L6, E2 conjugating enzyme has been shown to be downregulated significantly in LPCs by UHRF1-mediated CpG methylation. Both biochemical assays and animal studies demonstrated that UBE2L6 expression decline, in turn, promotes EZH2 protein stability due to lack of ubiquitination on K381 residue in LPCs. UBR4 cooperates with UBE2L6 to facilitate this ubiquitination process. Targeting UHRF1/UBE2L6/UBR4 axis can be a better treatment option to trigger HPC state in melanoma in which conventional EZH2 inhibitors are ineffective.

cancer biology↗

bioRxiv2020

Counterfactual Hypothesis Testing of Tumor Microenvironment Scenarios Through Semantic Image Synthesis

AO_SCPLOWBSTRACTC_SCPLOWRecent multiplexed protein imaging technologies characterize cells, their spatial organization, and interactions within microenvironments at an unprecedented resolution. Although observational data can reveal spatial associations, it does not allow users to infer salient biological relationships and cellular interactions. To address this challenge, we develop a generative model that allows users to test hypotheses about the effect of cell-cell interactions on protein expression through in silico perturbation. Our Cell-Cell Interaction GAN (CCIGAN) model employs a generative adversarial network (GAN) architecture to generate high fidelity synthetic multiplexed images from semantic cell segmentations. Our approach is unique in that it learns relationships between all imaging channels simultaneously and yields biological insights from multiple imaging technologies in silico, capturing known tumor-immune cell interactions missed by other state-of-the-art GAN models.

cancer biology↗

bioRxiv2020

NINJA: an inducible genetic model for creating neoantigens in vivo

Mouse models with inducible neoantigens have been historically difficult to generate because of leaky expression of antigens in the thymus, which causes central tolerance in developing CD8 and CD4 T cells. Attempts to resolve this problem using existing genetic tools have been unsuccessful. We developed the iNversion INducible Joined neoAntigen (NINJA) mouse model that uses RNA splicing, DNA recombination, and three levels of regulation to prevent neoantigen leakiness and allow tight control over the induction of neoantigen expression. We describe the development of these genetic tools and their use for obtaining tumor cell lines with inducible neoantigen expression. Moreover, we show that the genetic regulation in NINJA mice bypasses central and peripheral tolerance mechanisms and allows for robust endogenous CD8 and CD4 T cells responses upon neoantigen induction in peripheral tissues. Thus, NINJA fills a long-standing gap in the field and will enable studies of how T cells respond to defined neoantigens in the context of peripheral tolerance, autoimmune diseases, and developing tumors.

immunology↗

bioRxiv2020

Programmable and portable CRISPR-Cas transcriptional activation in bacteria

Programmable gene activation enables fine-tuned regulation of endogenous and synthetic gene circuits to control cellular behavior. While CRISPR-Cas-mediated gene activation have been extensively developed for eukaryotic systems, similar strategies have been difficult to implement in bacteria. Here, we present a generalizable platform for screening and selection of functional bacterial CRISPR-Cas transcription activators. Using this platform, we identified a novel CRISPR activator, dCas9-AsiA, that could activate gene expression by up to 200-fold across genomic and plasmid targets with diverse promoters after directed evolution. The evolved dCas9-AsiA can simultaneously mediate activation and repression of bacterial regulons in E. coli. We further identified hundreds of promoters with varying basal expression that could be induced by dCas9-AsiA, which provides a rich resource of genetic parts for inducible gene activation. Finally, we show that dCas9-AsiA can be ported to other bacteria of clinical and bioindustrial relevance, thus enabling bacterial CRISPRa in more application areas. This work expands the toolbox for programmable gene regulation in bacteria and provides a useful resource for future engineering of other bacterial CRISPR-based gene regulators.

synthetic biology↗

bioRxiv2019

Evolutionary and biochemical analyses reveal conservation of the Brassicaceae telomerase ribonucleoprotein complex

The telomerase ribonucleoprotein complex (RNP) is essential for genome stability and performs this role through the addition of repetitive DNA to the ends of chromosomes. The telomerase enzyme is composed of a reverse transcriptase (TERT), which utilizes a template domain in an RNA subunit (TER) to reiteratively add telomeric DNA at the ends of chromosomes. Multiple TERs have been identified in the model plant Arabidopsis thaliana. Here we combine a phylogenetic and biochemical approach to understand how the telomerase RNP has evolved in Brassicaceae, the family that includes A. thaliana. Because of the complex phylogenetic pattern of template domain loss and alteration at the previously characterized A. thaliana TER loci, TER1 and TER2, across the plant family Brassicaceae, we bred double mutants from plants with a template deletion at AtTER1 and T-DNA insertion at AtTER2. These double mutants exhibited no telomere length deficiency, a definitive indication that neither of these loci encode a functional telomerase RNA. Moreover, we determined that the telomerase components TERT, Dyskerin, and the KU heterodimer are under strong purifying selection, consistent with the idea that the TER with which they interact is also conserved. To test this hypothesis further, we analyzed the substrate specificity of telomerase from species across Brassicaceae and determined that telomerase from close relatives bind and extend substrates in a similar manner, supporting the idea that TERs in different species are highly similar to one another and are likely encoded from an orthologous locus. Lastly, TERT proteins from across Brassicaceae were able to complement loss of function tert mutants in vivo, indicating TERTs from other species have the ability to recognize the native TER of A. thaliana. Finally, we immunoprecipitated the telomerase complex and identified associated RNAs via RNA-seq. Using our evolutionary data we constrained our analyses to conserved RNAs within Brassicaceae that contained a template domain. These analyses revealed a highly expressed locus whose disruption by a T-DNA resulted in a telomeric phenotype similar to the loss of other telomerase core proteins, indicating that the RNA has an important function in telomere maintenance.

genetics↗