Biology subjects

Ng, K.

Publications and source records attributed to Ng, K..

At least 37 records · Page 2Linked to original sources

bioRxiv2024

Focused learning by antibody language models using preferential masking of non-templated regions

Existing antibody language models (LMs) are pre-trained using a masked language modeling (MLM) objective with uniform masking probabilities. While these models excel at predicting germline residues, they often struggle with mutated and non-templated residues, which are crucial for antigen-binding specificity and concentrate in the complementarity-determining regions (CDRs). Here, we demonstrate that preferential masking of the non-templated CDR3 is a compute-efficient strategy to enhance model performance. We pre-trained two antibody LMs (AbLMs) using either uniform or preferential masking and observed that the latter improves residue prediction accuracy in the highly variable CDR3. Preferential masking also improves antibody classification by native chain pairing and binding specificity, suggesting improved CDR3 understanding and indicating that non-random, learnable patterns help govern antibody chain pairing. We further show that specificity classification is largely informed by residues in the CDRs, demonstrating that AbLMs learn meaningful patterns that align with immunological understanding.

immunology↗

bioRxiv2024

Combined transcriptomic, connectivity, and activity profiling of the medial amygdala using highly amplified multiplexed in situ hybridization (hamFISH)

In situ transcriptomic technologies provide a promising avenue to link gene expression, connectivity, and physiological properties of neural cell types. Commercialized methods that allow the detection of hundreds of genes in situ, however, are expensive and therefore typically used for generating unimodal reference data rather than for resource-intensive multimodal analyses. A major bottleneck is the lack of a routine means to efficiently generate cell type data. Here, we have developed hamFISH (highly amplified multiplexed fluorescence in situ hybridization), which enables the sequential detection of 32 genes using multiplexed branched DNA amplification. We used hamFISH to profile the projection, activity, and transcriptomic diversity of the medial amygdala (MeA), a critical node for innate social and defensive behaviors in mice. In total, we profiled 643,834 cells and classified neurons into 16 inhibitory and 10 excitatory types, many of which were found to be spatially clustered. We then examined the organization of outputs of these cells and activation profiles during different social contexts. Therefore, by facilitating multiplexed detection of single molecule RNAs, hamFISH provides a streamlined and versatile platform for multimodal profiling of specific brain nuclei.

neuroscience↗

bioRxiv2023

Nonlinear readout of spatial cues underlies robustness of asymmetric cell division

A key challenge in the development of an organism is to maintain robust phenotypic outcomes in the face of perturbation. Yet, how such robust outcomes are encoded by developmental networks remains poorly explored. Here we use the C. elegans zygote as a model to understand sources of developmental robustness during PAR polarity-dependent asymmetric cell division. By quantitatively linking alterations in protein dosage to phenotype in individual embryos, we show that spatial information in the zygote is read out in a highly nonlinear fashion and, as a result, phenotypes are highly canalized against substantial variation in input signals. Specifically, our data point towards an intrinsic robustness of the conserved PAR polarity network that renders polarity axis specification resistant to variations in both the strength of upstream symmetry-breaking cues and PAR protein dosage. At the same time, we find that downstream pathways involved in cell size and fate asymmetry are similarly robust to dosage-dependent changes in the local concentrations of PAR proteins, implying non-trivial complexity in translating PAR signals into pathway outputs. We propose that "quantitative decoupling" of symmetry-breaking, polarity, and asymmetric division modules acts to suppress the accumulation of error as embryos move along this developmental trajectory, thereby ensuring that asymmetric division is robust to perturbation. Such modular organization of developmental networks is likely to be a general mechanism to achieve robust developmental outcomes.

developmental biology↗

bioRxiv2023

Age-specific and compartment-dependent changes in mitochondrial homeostasis and cytoplasmic viscosity in mouse peripheral neurons

Mitochondria are dynamic bioenergetic hubs that become compromised with age. In neurons, declining mitochondrial axonal transport has been associated with reduced cellular health. However, it is still unclear to what extent the decline of mitochondrial transport and function observed during ageing are coupled, and if somal and axonal mitochondria display compartment-specific features that make them more susceptible to the ageing process. It is also not known whether the biophysical state of the cytoplasm, thought to affect many cellular functions, changes with age to impact mitochondrial trafficking and homeostasis. Focusing on the mouse peripheral nervous system, we show that age-dependent decline in mitochondrial trafficking is accompanied by reduction of mitochondrial membrane potential and intra-mitochondrial viscosity, but not calcium buffering, in both somal and axonal mitochondria. Intriguingly, we observe a specific increase in cytoplasmic viscosity in the neuronal cell body, where mitochondria are most polarised, which correlates with decreased cytoplasmic diffusiveness. Increasing cytoplasmic crowding in the somatic compartment of DRG neurons grown in microfluidic chambers reduces mitochondrial axonal trafficking, suggesting a mechanistic link between the regulation of cytoplasmic viscosity and mitochondrial dynamics. Our work provides a reference for studying the relationship between neuronal mitochondrial homeostasis and the viscoelasticity of the cytoplasm in a compartment-dependent manner during ageing.

neuroscience↗

bioRxiv2023

Chemotherapeutic regulation of the ROS/MondoA-dependent TXNIP/GDF15 axis; and derivation of a new organoid metric as a predictive biomarker

Chemotherapy, the standard of care treatment for cancer patients with advanced disease, has been increasingly recognised to activate host immune responses to produce durable outcomes. Here, in colorectal adenocarcinoma (CRC) we identify chemotherapy-induced Thioredoxin Interacting Protein (TXNIP), a MondoA-dependent tumor suppressor gene, as a negative regulator of Growth/Differentiation Factor 15 (GDF15). GDF15 is a negative prognostic factor in CRC and promotes the differentiation of regulatory T cells (Tregs), through CD48 ligation. Intriguingly, multiple models including patient-derived tumor organoids demonstrate that loss of TXNIP/GDF15 axis functionality is associated with advanced disease or chemotherapeutic resistance, with transcriptomic or proteomic GDF15/TXNIP ratios showing potential as a prognostic biomarker. These findings illustrate a potentially common pathway where chemotherapy-induced epithelial stress drives local immune remodelling for patient benefit, with disruption of this pathway seen in refractory or advanced cases.

cancer biology↗

bioRxiv2023

Optimized dimerization of the PAR-2 RING domain drives cooperative and selective membrane recruitment for robust feedback-driven cell polarization

The behavior of cell polarity networks is defined by the quantitative features of their constituent feedback circuits, which must be tuned to enable robust and stable polarization, while also ensuring that networks remain responsive to dynamically changing cellular states and/or spatial cues that arise during development. Using the PAR polarity network as a model, we demonstrate that these features are enabled by dimerisation of the polarity protein PAR-2 via ubiquitin-independent function of its N-terminal RING domain. Specifically, we combine theory and experiment to show that dimer affinity is optimized to achieve dynamic, selective, and cooperative recruitment of PAR-2 to the plasma membrane during polarization. Reducing dimerization results in loss of positive feedback and compromises robustness of symmetry-breaking, while enhanced dimerization renders the network less responsive due to kinetic trapping of PAR-2 on internal membranes and reduced sensitivity of PAR-2 to membrane displacement by the polarity kinase, aPKC/PKC-3. Thus, our data reveal how a dynamically oligomeric RING domain results in a cell polarity network that is both robust and responsive and highlight how tuning of oligomerization kinetics can serve as a general strategy for optimizing dynamic and cooperative intracellular targeting.

cell biology↗

bioRxiv2023

A bioorthogonal antibody-based chemically-induced-dimerization switch for therapeutic application

We present the Indinavir Ligand Induced Transient Engagement switch (IDV LITE Switch), a fully synthetic Chemically Induced Dimerization (CID) system wherein two humanized antibody fragments are heterodimerized by the antiviral drug indinavir. The IDV LITE Switch represents the first CID system made from fully humanized protein components and dimerized by a clinically approved small molecule drug lacking a mammalian target, making it an ideal bioorthogonal molecular switch for application in small-molecule controlled therapeutics.

synthetic biology↗

bioRxiv2022

Cleavage furrow-directed cortical flows bias mechanochemical pathways for PAR polarization in the C. elegans germ lineage

During development, the conserved PAR polarity network is continuously redeployed, requiring that it adapts to changing cellular contexts and environmental cues. How it does so and the degree to which these adaptations reflect changes in its fundamental design principles remain unclear. Here, we investigate the process of PAR polarization within the highly tractable C. elegans germline P lineage, which undergoes a series of iterative asymmetric stem cell-like divisions. Compared to the zygote, we observe significant differences in the pattern of polarity emergence, including an inversion of the initial unpolarized state, changes in symmetry breaking cues, and the timings with which anterior and posterior PARs segregate. Beneath these differences, however, polarity establishment remains reliant on the same core pathways identified in the zygote, including conserved roles for cortical actin flows and PAR-dependent self-organization. Intriguingly, we find that cleavage furrow-directed cortical actin flows play a similar symmetry-breaking role for the germline cell P1 as centrosome-induced cortical flows in the zygote. Through their ability to induce asymmetric accumulation of PAR-3 clusters, these furrow-directed flows directly couple the geometry of polarization to cell division, which could be a general strategy for cells to ensure proper organization within dynamically growing systems, such as embryos. In summary, our data suggest that coupling of novel symmetry-breaking cues with highly adaptable core mechanochemical circuits enable robust PAR polarity in response to changing developmental contexts.

cell biology↗

bioRxiv2022

Design principles for selective polarization of PAR proteins by cortical flows

Clustering of membrane-associated molecules is thought to promote interactions with the actomyosin cortex, enabling size-dependent transport by actin flows. Consistent with this model, in the C. elegans zygote, anterior segregation of the polarity protein PAR-3 requires oligomerization. However, through direct assessment of advection of PAR proteins, we not only find no links between PAR-3 advection and oligomer size, but also observe efficient advection of both anterior and posterior PAR proteins. Consequently, differential cortex engagement cannot account for selective size-dependent PAR protein transport. Instead, combining experiment and theory we demonstrate that segregation efficiency of PAR proteins by cortical flow is determined by the stability of membrane association, which is enhanced by clustering and specifies persistence of transport. Indeed, stabilizing membrane association was sufficient to invert polarity of a normally posterior PAR protein. Our data therefore indicate that advection of membrane-associated proteins is more pervasive than anticipated and thus cells must tune membrane association dynamics to achieve differential transport by cortical flows.

cell biology↗

bioRxiv2022

Cross-Modal Autoencoder Framework Learns Holistic Representations of Cardiovascular State

A fundamental challenge in diagnostics is integrating multiple modalities to develop a joint characterization of physiological state. Using the heart as a model system, we develop a cross-modal autoencoder framework for integrating distinct data modalities and constructing a holistic representation of cardio-vascular state. In particular, we use our framework to construct such cross-modal representations from cardiac magnetic resonance images (MRIs), containing structural information, and electrocardiograms (ECGs), containing myoelectric information. We leverage the learned cross-modal representation to (1) improve phenotype prediction from a single, accessible phenotype such as ECGs; (2) enable imputation of hard-to-acquire cardiac MRIs from easy-to-acquire ECGs; and (3) develop a framework for performing genome-wide association studies in an unsupervised manner. Our results provide a framework for integrating distinct diagnostic modalities into a common representation that better characterizes physiologic state.

bioinformatics↗

bioRxiv2022

SAIBR: A simple, platform-independent method for spectral autofluorescence correction

Biological systems are increasingly viewed through the lens of mathematics, physics, and systems approaches that demand accurate quantification of gene expression and local protein concentrations. Such approaches have benefited greatly from the revolution in genetic engineering sparked by CRISPR/Cas9. By facilitating the tagging of genes at their genomic loci, CRISPR/Cas9 allows us to use fluorescence to monitor proteins that are expressed at or near endogenous levels under native regulatory control. However, due to their typically lower expression levels, quantitative experiments using endogenously-tagged genes can run into limits imposed by autofluorescence (AF). AF is often a particular challenge in the illumination bands occupied by the most efficient fluorescent proteins (GFP, mNeonGreen). Stimulated by our work in C. elegans, we describe and validate Spectral Autofluorescence Image correction By Regression (SAIBR), a simple, platform-independent protocol, and associated GUI-based FIJI plugin to correct for autofluorescence using standard filter sets and illumination conditions. Fully validated for use in C. elegans embryos and tested in diverse systems, including starfish oocytes and fission yeast, SAIBR achieves accurate quantitation of fluorophore signal and enables reliable detection and quantification of even weakly expressed proteins. Thus, SAIBR provides a highly accessible, low barrier way to incorporate AF correction as standard for researchers working on a broad variety of cell and developmental systems. Summary StatementImplemented as an easy-to-use Fiji Plugin, SAIBR provides effective autofluorescence correction for cells and tissues using standard imaging conditions.

cell biology↗

bioRxiv2021

Broad human and animal coronavirus neutralisation by SARS-CoV-2 S2-targeted vaccination

Several common-cold coronaviruses (HCoVs) are endemic in humans and several variants of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have emerged during the current Coronavirus disease 2019 (COVID-19) pandemic. Whilst antibody cross-reactivity with the Spike glycoproteins (S) of diverse coronaviruses has been documented, it remains unclear whether such antibody responses, typically targeting the conserved S2 subunit, contribute to or mediate protection, when induced naturally or through vaccination. Using a mouse model, we show that prior HCoV-OC43 S immunity primes neutralising antibody responses to otherwise subimmunogenic SARS-CoV-2 S exposure and promotes S2-targeting antibody responses. Moreover, mouse vaccination with SARS-CoV-2 S2 elicits antibodies that neutralise diverse animal and human alphacoronaviruses and betacoronaviruses in vitro, and protects against SARS-CoV-2 challenge in vivo. Lastly, in mice with a history of SARS-CoV-2 Wuhan-based S vaccination, further S2 vaccination induces stronger and broader neutralising antibody response than booster Wuhan S vaccination, suggesting it may prevent repertoire focusing caused by repeated homologous vaccination. The data presented here establish the protective value of an S2-targeting vaccine and support the notion that S2 vaccination may better prepare the immune system to respond to the changing nature of the S1 subunit in SARS-CoV-2 variants of concern (VOCs), as well as to unpredictable, yet inevitable future coronavirus zoonoses.

immunology↗

bioRxiv2021

Stenoparib, an inhibitor of cellular poly (ADP-ribose) polymerases (PARPs), blocks in vitro replication of SARS-CoV-2 variants

We recently published a preliminary assessment of the activity of a poly (ADP-ribose) polymerase (PARP) inhibitor, stenoparib, also known as 2X-121, which inhibits viral replication by affecting pathways of the host. Here we show that stenoparib effectively inhibits a SARS-CoV-2 wt (BavPat1/2020) strain and four additional variant strains; alpha (B.1.1.7), beta (B.1.351), delta (B.1.617.2) and gamma (P.1) in vitro, with 50% effective concentration (EC50) estimates of 4.1 M, 8.5 M, 24.1 M, 8.2 M and 13.6 M, respectively. A separate experiment focusing on a combination of 10 M stenoparib and 0.5 M remdesivir, an antiviral drug, resulted in over 80% inhibition of the alpha (B.1.1.7) variant, which is substantially greater than the effect achieved with either drug alone, suggesting at least additive effects from combining the different mechanisms of activity of stenoparib and remdesivir.

microbiology↗

bioRxiv2021

T cell receptor and IL-2 signaling strength control memory CD8+ T cell functional fitness via chromatin remodeling

Cognate antigen signal controls CD8+ T cell priming, expansion size and effector versus memory cell fates, but it is not known if and how it modulates the functional features of memory CD8+ T cells. Here we show that the strength of T cell receptor (TCR) signaling determines the requirement for interleukin-2 (IL-2) signals to form a pool of memory CD8+ T cells that competitively re-expand upon secondary antigen encounter. Combining strong TCR and intact IL-2 signaling synergistically induces genome-wide chromatin accessibility in regions targeting a wide breadth of biological processes, consistent with their greater functional fitness. Chromatin accessibility in promoters of genes encoding for stem cell, cell cycle and calcium-related proteins correlated with faster intracellular calcium accumulation, initiation of cell cycle and more robust expansion. High-dimensional flow-cytometry analysis also highlights higher subset diversity and phenotypes. These results formally establish that epitope selection in vaccine design strongly impacts memory CD8+ T cell epigenetic programming and functions. One Sentence SummaryThe strength of antigenic and interleukin 2 signals received by CD8+ T cells during vaccination epigenetically programs their ability to form functional memory.

immunology↗

bioRxiv2021

RUNIMC: An R-based package for imaging mass cytometry data analysis and pipeline validation

We propose a novel pipeline for the analysis of imaging mass cytometry data, comparing an unbiased approach, representing the actual gold standard, with a novel biased method. We made use of both synthetic/ controlled datasets as well as two datasets obtained from FFPE sections of follicular lymphoma, and head and neck patients, stained with a 14 and 29-markers panels respectively. The novel pipeline, denominated RUNIMC, has been completely developed in R and contained in a single package. The novelty resides in the ease with which multi-class random forest classifier can be used to classify image features, making the pathologists and expert classification pivotal, and the use of a random forest regression approach that permits a better detection of cell boundaries, and alleviates the necessity of relying on a perfect nuclear staining.

cancer biology↗

bioRxiv2021

Multiplexed bioluminescence microscopy via phasor analysis

Microscopic bioluminescence imaging has been historically challenging due to a lack of detection methods and easily resolved probes. Here we combine bioluminescence with phasor analysis, an optical method commonly used to distinguish spectrally similar fluorophores. Bioluminescent phasor enabled rapid differentiation of multiple luciferase reporters and resonance energy transfer processes. The merger of bioluminescence and phasor analysis provides a platform for routine, time-lapse imaging of collections of cellular features.

bioengineering↗

bioRxiv2021

Reduced antibody cross-reactivity following infection with B.1.1.7 than with parental SARS-CoV-2 strains

We examined the immunogenicity of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variant B.1.1.7 that arose in the United Kingdom and spread globally. Antibodies elicited by B.1.1.7 infection exhibited significantly reduced recognition and neutralisation of parental strains or of the South Africa B.1.351 variant, than of the infecting variant. The drop in cross-reactivity was more pronounced following B.1.1.7 than parental strain infection, indicating asymmetric heterotypic immunity induced by SARS-CoV-2 variants.

immunology↗

bioRxiv2021

Favourable antibody responses to human coronaviruses in children and adolescents with autoimmune rheumatic diseases

Differences in humoral immunity to coronaviruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), between children and adults remain unexplained and the impact of underlying immune dysfunction or suppression unknown. Here, we examined the antibody immune competence of children and adolescents with prevalent inflammatory rheumatic diseases, juvenile idiopathic arthritis (JIA), juvenile dermatomyositis (JDM) and juvenile systemic lupus erythematosus (JSLE), against the seasonal human coronavirus (HCoV)-OC43 that frequently infects this age group. Despite immune dysfunction and immunosuppressive treatment, JIA, JDM and JSLE patients mounted comparable or stronger responses than healthier peers, dominated by IgG antibodies to HCoV-OC43 spike, and harboured IgG antibodies that cross-reacted with SARS-CoV-2 spike. In contrast, responses to HCoV-OC43 and SARS-CoV-2 nucleoproteins exhibited delayed age-dependent class-switching and were not elevated in JIA, JDM and JSLE patients, arguing against increased exposure. Consequently, autoimmune rheumatic diseases and their treatment were associated with a favourable ratio of spike to nucleoprotein antibodies.

immunology↗