Biology subjects

Hsu, A.

Publications and source records attributed to Hsu, A..

At least 19 recordsLinked to original sources

bioRxiv2026

CD117 epitope-shielded hematopoietic stem cell transplantation with toxin-free conditioning and in vivo selection ameliorates β-thalassemia model

Clinical evidence demonstrates that ex vivo gene therapy and genome engineering of hematopoietic stem and progenitor cells (HSPCs) could represent one-time cures. However, while genome editing itself has become increasingly efficient and precise, the toxic conditioning required for hematopoietic stem cell transplantation remains a major barrier to broad clinical implementation of these otherwise curative therapies. In particular, the use of busulfan for myeloablative conditioning constitutes a major safety concern. While preclinical studies established CD117 as a promising target for antigen-specific therapy, clinical translation faced setbacks balancing efficacy and safety. To overcome current limitations, we generated a new CD117-blocking monoclonal antibody (CIM058) and demonstrate its potency to block wild-type HSPCs. To enable long-term blockade of host HSPCs even after transplantation, we used prime editing to engineer CIM058-resistant human CD34+ HSPCs. When combined, CIM058 and the epitope engineered CD34+ HSPCs ameliorated disease phenotype in a {beta}-thalassemia model. Our results suggest that this approach may overcome the reliance on busulfan or other myeloablative conditioning regimens with their associated morbidities, and by enabling toxin-free conditioning and in vivo selection of edited cells, may facilitate clinical implementation of these highly valuable genetic therapies.

bioengineering↗

bioRxiv2026

An innate immune circuit involving the lectin pathway, complement and lung epithelial cells enforce barrier immunity to inhaled fungal disease

The lung mucosal barrier thwarts many inhaled pathogens, including Coccidioides, the causative agent of Valley Fever. Because the earliest stages of pathogen recognition in the lung remain obscure, we investigated the initial events of barrier immunity in a murine model involving inhalation of Coccidioides sp. arthroconidia. Neutrophils accumulated rapidly, within 24 hours, near fungal spores in the bronchiolar airways. This response was driven by pattern recognition via the lectin complement pathway. Bronchiolar club cells propagated C3a signals and amplified the response via convergent C3aR and P2X7 signaling. We identified several MBL2 and P2RX7 polymorphisms that correlated with progressive disease in humans. Our assays revealed that these mutations caused functional impairments in C3a generation and P2X7 responsiveness. Our findings establish how complement signaling and epithelial sensing coordinate early immune responses to fungal infection, offering insights into essential host defense mechanisms and risk factors for disease progression in coccidioidomycosis.

microbiology↗

bioRxiv2026

Language Modeling Materializes a World Model of Protein Biology

Proteins are fundamental to life. The full extent of their biology is beyond our ability to characterize with experimental approaches in the physical laboratory. Accurate digital representations could accelerate the discovery of protein biology through virtual experiments. We propose language modeling to learn unified and general representations that can be scaled to all of protein biology. Building on these representations, we develop a structure prediction model that exceeds the performance of established methods for biomolecular complex prediction across benchmarks, including for the interactions of antibodies with their targets. A simple search procedure yields high experimental success rates for the discovery of proteins with nanomolar binding affinities for both miniproteins and single-chain antibodies, a modality critical for therapeutic design. Study of the concepts in the language models representation space reveals a systematic organization aligned with the reductionist understanding of proteins developed through empirical science. Leveraging this organization, we generate a comprehensive map of protein biology encompassing over 6.8 billion sequences and 1.1 billion predicted structures, identifying connections across known and unknown biology. As a whole, this shows language modeling as a powerful substrate for representing the biology of proteins, operating across scales from the prediction and design of protein interactions at the atomic level, to identifying properties of proteins at different levels of granularity and abstraction, to the scale of mapping connections between proteins across billions of years of evolution.

bioinformatics↗

bioRxiv2026

Mechanistic machine learning enables interpretable and generalizable prediction of prime editing outcomes

Although prime editing (PE) can effect virtually any specified local change to genomic DNA in living systems, its efficient application currently requires extensive optimization of prime editing guide RNA (pegRNA) sequences. We present OptiPrime, a machine learning model of PE efficiency based on our current understanding of the mechanism of prime editing. OptiPrime achieves state-of-the-art accuracy on PE efficiency prediction and also enables prediction of nicking guide RNA (PE3) and dual pegRNA (twinPE) outcomes. We validated that OptiPrime has learned the determinants of mammalian mismatch repair (MMR), and is therefore well suited for nominating MMR-evasive silent edits that improve PE efficiency. We demonstrate the utility of OptiPrime in a variety of prospective therapeutic contexts, including in primary human and mouse cells. Finally, we show how OptiPrime can be used to achieve highly streamlined and efficient in vivo correction of a pathogenic mutation in the brain of a mouse model of KIF1A-associated neurological disorder.

biochemistry↗

bioRxiv2026

Single-cell profiling of synchronous multi-organ metastasis reveals a systemic CD74+ lipid-associated macrophage niche driving polymetastatic breast cancer

Systemic, multi-organ metastasis is the primary cause of breast cancer mortality, yet the biological mechanisms that allow disseminated tumor cells to simultaneously colonize physiologically diverse tissues remain poorly understood. Current paradigms focus on organ-specific tropism, largely overlooking the potential for systemic, conserved and synchronized programs that facilitate widespread colonization. Here, we present a high-resolution, multi-organ atlas of metastatic ecosystems and their niches using a synchronous model of brain, lung, liver, and bone metastasis combined with in vivo proximal niche labeling and single-cell RNA sequencing. We identify a remarkably conserved proximal niche program defined by the accumulation of CD74+ lipid-associated, metastasis-associated macrophages (LA-MAMs) across all metastatic sites. CD74+ LA-MAMs are characterized by a unique metabolic-immune signature and drive T cell suppression. We show that the cytokine Macrophage Migration Inhibitory Factor (MIF), secreted by metastatic cells, acts as the universal paracrine mediator that instructs the LA-MAM phenotype via the CD74 receptor. Interference of the MIF-CD74 axis effectively disrupts the LA-MAM niche, mitigates T cell exhaustion, and reduces metastatic burden across all organs. Analysis of a 100-patient cohort of metastasis samples from different sites confirms that the MIF-CD74 axis is a hallmark of human multi-organ colonization and independently predicts poor post-metastasis survival. Our findings define a synchronized and systemic metastatic niche that can be targeted, providing a mechanistic rationale for neutralizing the MIF-CD74 axis to treat polymetastatic breast cancer.

cancer biology↗

bioRxiv2025

Autofluorescence imaging reveals the impact of cryopreservation on T cell metabolism and activation response

Cryopreservation or the process of freezing cells is a cornerstone of most cell therapy protocols. Optimization of cryopreservation protocols and cryoprotectant agents to improve cell viability and functionality is under further investigation. However, the impact of cryopreservation on cellular metabolism and function immediately post-thaw is not fully understood. Here, we used label-free, non-invasive optical metabolic imaging (OMI) of NAD(P)H and FAD to characterize the activation response of frozen T cells from healthy donors and lymphoma patients post-thaw. Using OMI, we identified significant metabolic shift, along with delayed and diminished activation response in healthy donor T cells throughout the first 4.5 hours upon thawing. In cryopreserved peripheral T cells from lymphoma patients in our bispecific CD19/CD20 CAR T therapy clinical trial, OMI could identify early metabolic stress and allowed gating of metabolically-fit cells associated with post-thaw viability. Notably, in our pilot study, only metabolically-fit T cells from complete responders exhibited metabolic responses to activating stimuli within the first 4.5 hours post-thaw. Overall, our findings suggest that 4-5 hours post-thaw is a critical time window to assess the impact of cryopreservation and thawing, and support the potential of OMI to optimize cryopreservation protocols and evaluate patient T cell quality for cell therapy.

immunology↗

bioRxiv2025

An automated image analysis pipeline for wide-field optical redox imaging of patient-derived cancer organoids

Wide-field optical redox imaging provides a fast and accessible method to monitor metabolic changes in cells and has recently been developed for drug screening in patient-derived cancer organoids (PDCOs). However, manual analysis of wide-field optical redox images is inefficient and laborious for large-scale drug screens. Here, we developed an automated pipeline for PDCO segmentation, single-PDCO tracking, and background correction in autofluorescence images. This pipeline was tested on two imaging systems over a 3-day time-course with two drug doses to demonstrate generalizability across imaging systems. Segmentation was performed using a fine-tuned Cellpose model, which when compared to manual masks, achieved mean Dice scores >0.8 across systems, indicating high reproducibility. Automated single-PDCO tracking was compared to manual tracking and the accuracy of the tracking algorithm exceeded 94% by two metrics, recall and Jaccard index. For background correction, the automated pipeline uses the full field-of-view to reduce sampling bias. Compared to the manual analysis pipeline, the automated pipeline resolves single-PDCO responses with comparable sensitivity to drug treatment but with over 127x faster processing time. This novel automated image analysis pipeline improves throughput and robustness in PDCO image analysis, which increases the accessibility and scalability of wide-field optical redox imaging for PDCO drug screening.

bioengineering↗

bioRxiv2025

Clade V MLO proteins are bona fide host susceptibility factors required for powdery mildew pathogenesis in Arabidopsis

Obligate biotrophic powdery mildew (PM) fungi strictly require living host to survive. To search for host factors or processes essential for PM pathogenesis, a tailored genetic screen was conducted with the immuno-compromised eds1-2/pad4-1/sid2-2 (eps) triple Arabidopsis mutant. This led to the identification of five allelic disruptive mutations in Mildew Locus O 2 (MLO2) to be responsible for the compromised-immunity-yet-poor infection (cipi) mutant phenotype upon challenge from an adapted PM isolate. Moreover, the eds1/pad4/sid2/mlo2/mlo6/mlo12 (eps3m) sextuple mutant display near complete immunity to the adapted PM fungus without sign of defense activation, demonstrating that these three clade V MLOs in Arabidopsis are bona fide host susceptibility factors of PM fungi. Confocal imaging revealed focal accumulation of MLO2-GFP in the peri-penetration peg membranous space, implicating MLO2 in repairing and stabilizing the damaged host plasma membrane, which may be co-opted by PM fungi for haustorium differentiation. Results from domain-swapping analysis between MLO1 and MLO2 suggest a bipartite functional configuration for MLO2: its C-terminus determines where and when MLO2 functions, while its N-terminal seven transmembrane domain region executes the cellular function that is critical for PM pathogenesis. Genetic studies further demonstrate that, unlike MLO7 in synergids, focal accumulation of MLO2 does not depend on FERONIA (FER) and its five other family members, nor does it require phosphatidylinositol 4,5-bisphosphate produced from phosphatidylinositol 4-phosphate 5-kinase 1 (PIP5K1) and PIP5K2. Together, these findings define clade V MLOs as host factors co-opted by obligate biotrophic PM fungi for successful host colonization.

plant biology↗

bioRxiv2024

MetaLATTE: Metal Binding Prediction via Multi-Task Learning on Protein Language Model Latents

The bioremediation of environments contaminated with heavy metals is an important challenge in environmental biotechnology, which may benefit from the identification of proteins that bind and neutralize these metals. Here, we introduce a novel predictive algorithm that conducts Metal binding prediction via LAnguage model laTenT Embeddings using a multi-task learning approach to accurately classify the metal-binding properties of input protein sequences. Our MetaLATTE model utilizes the state-of-the-art ESM-2 protein language model (pLM) embeddings and a position-sensitive attention mechanism to predict the likelihood of binding to specific metals, such as zinc, lead, and mercury. Importantly, our approach addresses the challenges posed by proteins from understudied organisms, which are often absent in traditional metal-binding databases, without the requirement of an input structure. By providing a probability distribution over potential binding metals, our classifier elucidates specific interactions of proteins with diverse metal ions. We envision that MetaLATTE will serve as a powerful tool for rapidly screening and identifying new metal-binding proteins, from metagenomic discovery or de novo design efforts, which can later be employed in targeted bioremediation campaigns.

bioinformatics↗

bioRxiv2023

An optimised protocol for quality control of gene therapy vectors using Nanopore direct RNA sequencing

Despite recent advances made towards improving the efficacy of lentiviral gene therapies, a sizeable proportion of produced vector contains an incomplete and thus potentially nonfunctional RNA genome. This can undermine gene delivery by the lentivirus as well as increase manufacturing costs and must be improved to facilitate the widespread clinical implementation of lentiviral gene therapies. Here, we compare three long-read sequencing technologies for their ability to detect issues in vector design and determine Nanopore direct RNA sequencing to be the most powerful. We show how this approach identifies and quantifies incomplete RNA caused by cryptic splicing and polyadenylation sites, including a potential cryptic polyadenylation site in the widely used Woodchuck Hepatitis Virus Posttranscriptional Regulatory Element (WPRE). Using artificial polyadenylation of the lentiviral RNA, we also identify multiple hairpin-associated truncations in the analysed lentiviral vectors, which account for most of the detected RNA fragments. Finally, we show that these insights can be used for optimization of lentiviral vector design. In summary, Nanopore direct RNA sequencing is a powerful tool for the quality control and optimisation of lentiviral vectors, which may help to improve lentivirus manufacturing and thus the development of higher quality lentiviral gene therapies.

genomics↗

bioRxiv2022

The Eyes Absent family members EYA4 and EYA1 promote PLK1 activation and successful mitosis through tyrosine dephosphorylation.

The Eyes Absent family of proteins (EYA1-4) are a biochemically unique group of tyrosine phosphatases known to be tumour promoting across a range of cancer types. To date, the molecular targets of EYA phosphatase activity remain largely uncharacterised. Here, we identify Polo-like kinase 1 (PLK1) as a direct interactor and phosphatase substrate of both EYA4 and EYA1, with pY445 on PLK1 being the primary target site. EYA-mediated dephosphorylation of PLK1 in the G2 phase of the cell cycle is required for centrosome maturation, PLK1 localization to centrosomes, and polo-box domain (PBD) dependent interactions between PLK1 and the PLK1-activating proteins BORA and CEP192. Molecular dynamics simulations support the rationale that pY445 confers a structural impairment to PBD-substrate interactions that is relieved by EYA-mediated dephosphorylation. Depletion of EYA4 or EYA1, or chemical inhibition of EYA phosphatase activity, dramatically reduces PLK1 activation, causing mitotic defects and cell death. Overall, we have characterized a novel phosphotyrosine signalling network governing PLK1 and mitosis. This work provides a mechanism of cell killing for EYA phosphatase inhibitors with important therapeutic implications.

cell biology↗

bioRxiv2022

Dual IgA/IgG family autoantibodies from individuals at-risk for rheumatoid arthritis identify an arthritogenic strain of Subdoligranulum

The mucosal origins hypothesis of rheumatoid arthritis (RA) proposes a central role for mucosal immune responses in the initiation and/or perpetuation of the systemic autoimmunity that occurs with disease. However, the connection between the mucosa and systemic autoimmunity in RA remains unclear. Using dual IgA/IgG family plasmablast-derived monoclonal autoantibodies obtained from peripheral blood of individuals at-risk for RA, we identified cross-reactivity between RA-relevant autoantigens and bacterial taxa in the closely related families Lachnospiraceae and Ruminococcaceae. After generating bacterial isolates within Lachnospiraceae/Ruminococcaceae genus Subdoligranulum from the feces of an individual, we confirmed monoclonal antibody binding as well as MHC class II dependent CD4+ T cell activation in RA cases compared to controls. Additionally, when Subdoligranulum isolate 7 but not isolate 1 colonized germ-free mice, it stimulated Th17 cell expansion, serum RA-relevant IgG autoantibodies, and joint swelling reminiscent of early RA with histopathology characterized by antibody deposition and complement activation. Systemic immune responses were likely due to the generation of colon isolated lymphoid follicles (ILFs) driving increased fecal and serum IgA by isolate 7, as B and T cell depletion not only halted intestinal immune responses but also eliminated detectable clinical disease. In aggregate, these findings demonstrate one mucosal mechanism in RA through which an intestinal strain of bacteria can drive systemic autoantibody generation and joint-centered antibody deposition and immune activation. One-sentence summarySubdoligranulum spp. targeted by rheumatoid arthritis (RA)-relevant autoantibodies activate T cells in individuals with RA, and in mice stimulate autoantibodies and joint swelling associated with antibody deposition and complement activation.

immunology↗

bioRxiv2021

The humanized nanobody RBD-1-2G tolerates the spike N501Y mutation to neutralize SARS-CoV-2

Neutralizing antibodies targeting the SARS-CoV-2 spike protein have shown a great preventative/therapeutic potential. Here, we report a rapid and efficient strategy for the development and design of SARS-CoV-2 neutralizing humanized nanobody constructs with sub-nanomolar affinities and nanomolar potencies. CryoEM-based structural analysis of the nanobodies in complex with spike revealed two distinct binding modes. The most potent nanobody, RBD-1-2G(NCATS-BL8125), tolerates the N501Y RBD mutation and remains capable of neutralizing the B.1.1.7 (Alpha) variant. Molecular dynamics simulations provide a structural basis for understanding the neutralization process of nanobodies exclusively focused on the spike-ACE2 interface with and without the N501Y mutation on RBD. A primary human airway air-lung interface (ALI) ex vivo model showed that RBD-1-2G-Fc antibody treatment was effective at reducing viral burden following WA1 and B.1.1.7 SARS-CoV-2 infections. Therefore, this presented strategy will serve as a tool to mitigate the threat of emerging SARS-CoV-2 variants.

synthetic biology↗

bioRxiv2020

An automated homecage system for multiwhisker detection and discrimination learning in mice

Automated, homecage behavioral training for rodents has many advantages: it is low stress, requires little interaction with the experimenter, and can be easily manipulated to adapt to different experimental condition. We have developed an inexpensive, Arduino-based, homecage training apparatus for sensory association training in freely-moving mice using multiwhisker air current stimulation coupled to a water reward. Animals learn this task readily, within 1-2 days of training, and performance progressively improves with training. We examined the parameters that regulate task acquisition using different stimulus intensities, directions, and reward valence. Learning was assessed by comparing anticipatory licking for the stimulus compared to the no-stimulus (blank) trials. At high stimulus intensities (>9 psi), animals showed markedly less participation in the task. Conversely, very weak air current intensities (1-2 psi) were not sufficient to generate rapid learning behavior. At intermediate stimulus intensities (5-6 psi), a majority of mice learned that the multiwhisker stimulus predicted the water reward after 24-48 hrs of training. Both exposure to isoflurane and lack of whiskers decreased animals ability to learn the task. Perceptual learning was assessed and following training at an intermediate stimulus intensity, perception was likely heightened as mice were able to transfer learning behavior when exposed to the lower stimulus intensity. Mice learned to discriminate between two directions of stimulation rapidly and accurately, even when the angular distance between the stimuli was <15 degrees. Switching the reward to a more desirable reward, aspartame, had little effect on learning trajectory. Our results show that a tactile association task in an automated homecage environment can be monitored by anticipatory licking to reveal rapid and progressive behavioral change. These Arduino-based, automated mouse cages enable high-throughput training that facilitate analysis of large numbers of genetically modified mice with targeted manipulations of neural activity.

neuroscience↗

bioRxiv2020

BRD4 Interacts with GATA4 to Govern Mitochondrial Homeostasis in Adult Cardiomyocytes

Gene regulatory networks control tissue plasticity during basal homeostasis and disease in a cell-type specific manner. Ubiquitously expressed chromatin regulators modulate these networks, yet the mechanisms governing how tissue-specificity of their function is achieved are poorly understood. BRD4, a member of the BET (Bromo- and Extra-Terminal domain) family of ubiquitously expressed acetyl-lysine reader proteins, plays a pivotal role as a coactivator of enhancer signaling across diverse tissue types in both health and disease, and has been implicated as a pharmacologic target in heart failure. However, the cell-specific role of BRD4 in adult cardiomyocytes remains unknown. Here, we show that cardiomyocyte-specific deletion of BRD4 in adult mice leads to acute deterioration of cardiac contractile function with mutant animals demonstrating a transcriptomic signature enriched for decreased expression of genes critical for mitochondrial energy production. Genome-wide occupancy data show that BRD4 enriches at many downregulated genes and preferentially co-localizes with GATA4, a lineage determining cardiac transcription factor not previously implicated in regulation of adult cardiac metabolism. Co-immunoprecipitation assays demonstrate that BRD4 and GATA4 form a complex in a bromodomain-independent manner, revealing a new interaction partner for BRD4 that has functional consequences for target transactivation and may allow for locus and tissue specificity. These results highlight a novel role for a BRD4-GATA4 module in cooperative regulation of a cardiomyocyte specific gene program governing bioenergetic homeostasis in the adult heart.

molecular biology↗

bioRxiv2019

Inducible overexpression of zebrafish microRNA-722 suppresses chemotaxis of human neutrophil like cells

Neutrophil migration is essential for battling against infections but also drives chronic inflammation. Since primary neutrophils are terminally differentiated and not genetically tractable, leukemia cells such as HL-60 are differentiated into neutrophil-like cells to study mechanisms underlying neutrophil migration. However, constitutive overexpression or inhibition in this cell line does not allow the characterization of the genes that affect the differentiation process. Here we apply the tet-on system to induce the expression of zebrafish miR-722 in differentiated HL-60. Overexpression of miR-722 reduced the mRNA level of genes in the chemotaxis and inflammation pathways, including RAC2. Consistently, cell migration is significantly inhibited upon induced miR-722 overexpression. Together, miR-722 is an evolutionarily conserved suppressor for neutrophil migration and signaling in zebrafish and human.

immunology↗

bioRxiv2019

Visualizing structural transitions of ligand-dependent gating of the TRPM2 channel

The calcium-permeable transient receptor potential melastatin 2 (TRPM2) channel plays a key role in redox sensation in many cell types 1-3. Channel activation requires binding of both ADP-ribose (ADPR) 2,4-6 and Ca2+ 7. The recently published TRPM2 structures from Danio rerio in the ligand-free and in the ADPR/Ca2+-bound conditions represent the channel in closed and open states, which uncover substantial tertiary and quaternary conformational rearrangements 8. However, it is unclear how these rearrangements occur within the tetrameric channel during channel gating. Here we report two cryo-electron microscopy structures of TRPM2 from the same species in complex with Ca2+ alone, and with both ADPR and Ca2+, determined to an overall resolution of ~3.8 [A] and ~4.2 [A] respectively. In comparison with the published results, our studies capture TRPM2 in two-fold symmetric intermediate states, offering a glimpse of the structural transitions within the tetramer that bridge the closed and open conformations.

biochemistry↗

bioRxiv2018

Structural basis for activation of Dot1L methyltransferase on the nucleosome by histone H2BK120 ubiquitylation

Histone H3 lysine 79 (H3K79) methylation is enriched on actively transcribed genes, and its misregulation is a hallmark of leukemia. Methylation of H3K79, which resides on the structured disk face of the nucleosome, is mediated by the Dot1L methyltransferase. Dot1L activity is part of a trans-histone crosstalk pathway, requiring prior histone H2B ubiquitylation of lysine 120 (H2BK120ub) for optimal activity. However, the molecular details describing both how Dot1L binds to the nucleosome and why Dot1L is activated by H2BK120 ubiquitylation are unknown. Here we present the cryo-EM structure of Dot1L bound to a nucleosome reconstituted with a site-specifically ubiquitylated H2BK120. The structure reveals that Dot1L engages the nucleosome acidic patch using an arginine anchor and occupies a conformation poised for methylation. Ubiquitin directly interacts with Dot1L and is positioned as a clamp on the nucleosome interacting region of Dot1L. Using our structure, we identify point mutations that disrupt the nucleosome-specific and ubiquitin-dependent activities of Dot1L. This study establishes a path to better understand Dot1L function in normal and leukemia cells.

biophysics↗