Biology subjects

Huh, J.

Publications and source records attributed to Huh, J..

11 recordsLinked to original sources

bioRxiv2026

RhoG, Rac1 and Cdc42 cooperation in cell protrusion revealed bymultiplexed optogenetics and biosensor imaging

The small GTPase Rac1 controls cell protrusion for a wide variety of critical cell functions. Its regulation by upstream guanine exchange factors (GEFs) has been the focus of multiple studies, but regulation by the GTPase RhoG remains poorly understood. RhoG is known to activate the ELMO/DOCK180 GEF complex, which in turn interacts with Rac1. It is unclear which aspects of protrusion are controlled by RhoG, and which of RhoGs effects on protrusion are mediated by Rac1. To address these questions, we developed biosensors and optogenetic tools to activate one GTPase while observing another, and to simultaneously visualize the activity of two GTPases. New tools included a photoactivable RhoG, a RhoG biosensor, and red shifted biosensors of RhoG and Rac1. RhoG and Rac1 activation events in protrusions were spatio-temporally correlated with one another and with protrusion velocity. Causal inference indicated that RhoG indeed unidirectionally activated Rac1. Photoactivation of RhoG and Rac1 indicated that specific aspects of protrusion behavior were controlled by RhoG, and only some via Rac1. Further dissection of RhoG to Rac1 signaling through simultaneous GTPase activation and biosensor visualization showed that PA-RhoG activates Rac1 predominantly through DOCK180 and that PA-RhoG can activate Cdc42 independently of Rac1.

cell biology↗

bioRxiv2026

Genetic comparisons of interleukin-17 reveal a framework for complex signaling evolution

The interleukin-17 (IL-17) family of cytokines comprises structurally distinct ligands and receptors which mediate immune responses at mucosal surfaces. The growing understanding of its regulatory functions beyond immunity, together with extensive genetic variation in protein-coding genes, raises the possibility that IL-17 cytokines participate in an even wider network of biologic processes. Despite successes of experimental approaches to chart IL-17 functions, inherent signaling complexities and crosstalk with multiple physiologic pathways obscure a full appreciation of the biological potential of IL-17. Here, we integrated comparative genomics, evolutionary rate covariation (ERC), and signatures of natural selection to resolve phylogenetic relationships between IL-17 ligands and receptors and discovered evidence for hidden signaling interactions. ERC analysis revealed putative ligand-receptor interactions for IL-17D and IL-17RC and suggested uncharacterized potential signaling mediator for the receptor IL-17REL, such as IL-17B. Signals of covariation extended beyond the IL-17 family to other genes encoding neurodevelopmental effectors and growth factors, emphasizing recurrent co-evolutionary patterns that delineate the immune and neuromodulatory roles of IL-17. These connections are underlined by signatures of positive selection in the disordered N-terminal domain of IL-17E and its cognate receptor, IL-17RB, key modulators of both type 2 immune response and neuronal function, suggesting functional consequences of this understudied domain. Together, our findings suggest that IL-17 biology is repeatedly impacted by lineage-specific selective pressures that dictate both immune and non-immune functions. By anchoring the expanding IL-17 field in an evolutionary framework, we propose a model for understanding the diversification and functional expansion of this and other cytokine families.

evolutionary biology↗

bioRxiv2026

immgenT Treg: A Reference Landscape of Regulatory T Cell States in Mice

CD4+FoxP3+ regulatory T cells (Tregs) are central to immunity, tolerance, and tissue homeostasis, yet their extensive heterogeneity lacks a unifying framework. Within the immgenT project, we profiled gene expression, surface markers and TCR clonotypes of mouse Tregs. Using a joint RNA-protein deep generative model, we define the Treg landscape, organized around eight conserved clusters shared across tissues and conditions, with immune context reshaping their relative abundance rather than generating new states, including a prominent circulating effector Treg population enriched in select non-lymphoid tissues. We validate this framework by integrating external datasets from conditions not represented in immgenT and by defining a flow cytometry panel spanning the Treg landscape. Together, immgenT provides a scalable, reusable reference that unifies Treg heterogeneity across tissues and immune challenges.

immunology↗

bioRxiv2025

Formic acid impairs α-synuclein seeding activity in human and mouse brains

-Synuclein (-syn) is an abundant monomeric protein that can aggregate into fibrils and form neuropathological inclusions in the brains of patients with synucleinopathies. New evidence suggests that the mouse-human transmission barrier of -syn is lower than previously reported, emphasizing the need for improved biosafety procedures when working with -syn aggregates. Histopathology of -syn-infected brain represents a significant potential source of occupational exposure, and current methods for tissue fixation do not inactivate the ability of pathologic -syn to seed the conversion of endogenous, monomeric -syn into fibrils. In this study, we tested whether 96% formic acid treatment could reduce the seeding activity of -syn aggregates in paraformaldehyde-fixed brain samples from dementia with Lewy bodies (DLB) patients and -syn pre-formed fibrils (PFF)-injected mouse brains. Using real-time quaking-induced conversion (RT-QuIC), we found that formic acid treatment reduced -syn seeding dose (picograms of -syn seeds per ml of brain homogenate) in DLB and mouse brain by 6 and 8 logarithms, respectively. RT-QuIC reactions seeded with formic acid-treated brain homogenates showed significantly longer lag phase, and decreased total thioflavin T fluorescence compared to untreated samples, indicating that formic acid treatment impairs the ability of pathological -syn to seed monomeric -syn. Importantly, the -syn pathologic features and the immunostaining quality were preserved in formic acid-treated tissues. Our results demonstrate that formic acid treatment is a quick and efficient procedure for reducing -syn seeding activity in fixed brain samples, thereby lowering the risk of accidental exposure in laboratories without compromising the quality of histopathological analysis. SummaryFormic acid treatment drastically reduces -synuclein seeding activity in fixed human and mouse brain samples while preserving histopathological quality.

neuroscience↗

bioRxiv2024

Scalable Guide Tree Construction Using Quantum Annealing for Multiple Sequence Alignment

Multiple sequence alignment (MSA) reveals homology in biological sequences, which is crucial for phylogenetics, medicine, and molecular biology. Many heuristic MSA algorithms use guide trees to determine sequence alignment order, but finding an optimal guide tree is an NP-hard problem. Conventional guide tree algorithms are greedy heuristics designed for better scalability at the cost of accuracy. By utilizing quantum algorithms, such as quantum annealing, we can overcome the problem of local minima that occurs in greedy methods. We propose a scalable guide tree algorithm that achieves scalability through quantum annealing. The theoretical foundations of our method are both minimum evolution and molecular clock. Unlike classical greedy approaches, we directly map the minimum evolution problem to a traveling salesperson problem (TSP), which quantum annealing can solve efficiently. The TSP solution enables the guide tree to be constructed in linear time using the molecular clock. Even with only a single sample from a D-Wave hybrid solver, our guide tree generally performed comparably to classical trees on the BAliBASE 3.0 benchmark. With a single iterative refinement, no statistically significant performance differences were found between our method and classical guide trees. Our scalable guide tree algorithm is practical in the sense that only a single sampling can be enough for constructing a good guide tree. Further performance analysis requires larger-scale benchmark tests. Fortunately, rapid advances in quantum hardware may soon enable these tests. While the practical application of quantum algorithms in bioinformatics has been relatively overlooked, this study highlights the potential of quantum algorithms for targeting computational bottlenecks in the field.

bioinformatics↗

bioRxiv2024

Quantum-classical hybrid approach for codon optimizationand its practical applications

Codon optimization is crucial for gene expression in heterologous hosts with varying genetic codes and codon usage, potentially resulting in enhanced protein expression and stability. Traditionally, the codon optimization problem has been solved using classical numerical techniques; however, with recent advancements, quantum algorithms deployed on quantum computers have been adopted for this purpose. This study proposes a codon sequence search protocol tailored to host preferences. Specifically, codon optimization is formulated as a constrained quadratic binary problem and solved using a quantum-classical hybrid approach, integrating quantum annealing with the Lagrange multiplier method. The proposed methodology is then applied to two real-world scenarios: optimizing the codon sequence of the severe respiratory syndrome coronavirus 2 spike protein in human hosts and insulin in Escherichia coli (E. coli) hosts. Finally, evaluations of several biological metrics demonstrate the effectiveness of our protocol, offering insights into the codon usage patterns governing translational efficiency and adaptation to the genetic code preferences of the host organisms.

bioinformatics↗

bioRxiv2022

A robust deep learning platform to predict CD8+ T-cell epitopes

T-cells play a crucial role in the adaptive immune system by inducing an anti-tumour response, defending against pathogens, and maintaining tolerance against self-antigens, which has sparked interest in the development of T-cell-based vaccines and immunotherapies. Because screening antigens driving the T-cell response is currently low-throughput and laborious, computational methods for predicting CD8+ T-cell epitopes have emerged. However, most immunogenicity algorithms struggle to learn features of peptide immunogenicity from small datasets, suffer from HLA bias and are unable to reliably predict pathology-specific CD8+ T-cell epitopes. Therefore, we developed TRAP (T-cell recognition potential of HLA-I presented peptides), a robust deep learning platform for predicting CD8+ T-cell epitopes from MHC-I presented pathogenic and self-peptides. TRAP uses transfer learning, deep learning architecture and MHC binding information to make context-specific predictions of CD8+ T-cell epitopes. TRAP also detects low-confidence predictions for peptides that differ significantly from those in the training datasets to abstain from making incorrect predictions. To estimate the immunogenicity of pathogenic peptides with low-confidence predictions, we further developed a novel metric, RSAT (relative similarity to autoantigens and tumour-associated antigens), as a complementary to dissimilarity to self from cancer studies. We used TRAP to identify epitopes from glioblastoma patients as well as SARS-CoV-2 peptides, and it outperformed other algorithms in both cancer and pathogenic settings. Thus, this study presents a novel computational platform for accurately predicting CD8+ T-cell epitopes to foster a better understanding of antigen-specific T-cell response and the development of effective clinical therapeutics. HighlightsO_LIHLA bias and out-of-distribution problem are causes of poor performance of current state-of-the-art algorithms C_LIO_LITransfer learning, deep learning architecture, context-specific and HLA-generalised approaches improve CD8+ T-cell epitope prediction C_LIO_LITRAP reports degree of correctness to improve reliability of the prediction C_LIO_LIA novel metric termed RSAT estimates immunogenicity of pathogenic peptides, as a complementary to dissimilarity to self from cancer studies C_LI

bioinformatics↗

bioRxiv2019

An endoribonuclease-based incoherent feedforward loop for decoupling resource-limited genetic modules

Synthetic biology has the potential to bring forth advanced genetic devices for applications in healthcare and biotechnology. However, accurately predicting the behavior of engineered genetic devices remains difficult due to lack of modularity, wherein a devices output does not depend only on its intended inputs but also on its context. One contributor to lack of modularity is competition among genes for shared cellular resources, such as those required for transcription and translation, which can induce coupling among otherwise independently-regulated genes. Here, we quantify the effects of resource sharing on engineered genetic systems in mammalian cells and develop an endoribonuclease-based incoherent feedforward loop (iFFL) to make gene expression levels robust to changes in resource availability. Our iFFL accurately controls gene expression levels in various cell lines and in the presence of significant resource sequestration by transcriptional activators. In addition to mitigating resource sharing, our iFFL also adapts gene expression to multiple log decades of DNA copy number variation, substantially improving upon previously-described miRNA-based iFFLs. Ultimately, our iFFL device will enable predictable, robust, and context-independent control of gene expression in mammalian cells.

synthetic biology↗

bioRxiv2019

A unified role for membrane-cortex detachment during cell protrusion initiation

Cell morphogenesis employs a diversity of membrane protrusions. They are discriminated by differences in force generation. Actin polymerization is the best studied mechanism of force generation, but growing interest in how variable molecular conditions and microenvironments alter morphogenesis has revealed other mechanisms, including intracellular pressure. Here, we show that local depletion of membrane cortex links is an essential step in the initiation of both pressure-based and actin-based protrusions. This observation challenges the quarter-century old Brownian ratchet model of actin-driven membrane protrusion, which requires an optimal balance of actin filament growth and membrane tethering. An updated model confirms membrane-filament detachment is necessary to activate the ratchet mechanism. These findings unify the regulation of different protrusion types, explaining how cells generate robust yet flexible strategies of morphogenesis.

cell biology↗

bioRxiv2018

Bile acid metabolites control Th17 and Treg cell differentiation

Bile acids are abundantly present in the mammalian gut, where they undergo bacteria-mediated transformation, generating a large pool of bioactive molecules. While they have been shown to affect host metabolism, cancer progression and innate immunity, it is unknown whether bile acids affect the function of adaptive immune cells such as T cells expressing IL-17a (Th17 cells) and regulatory T cells (Tregs) that mediate inflammatory and anti-inflammatory responses, respectively. By screening a small-molecule library primarily composed of bile acid metabolites, we identified two distinct derivatives of lithocholic acid (LCA), 3-oxoLCA and isoalloLCA, as specific regulators of Th17 and Treg cells. While 3-oxoLCA inhibited Th17 cell differentiation by directly binding to its key transcription factor ROR{gamma}t (retinoid-related orphan receptor {gamma} t), isoalloLCA enhanced differentiation of Tregs through mitochondrial-dependent metabolic changes, leading to an increased expression of Foxp3. IsoalloLCA-dependent Treg enhancement required an intronic Foxp3 enhancer, the conserved noncoding sequence 3 (CNS3), which acts as an epigenetic switch that confers a poised state to the Foxp3 promoter. Lastly, oral administration of 3-oxoLCA and isoalloLCA to mice led to reduced Th17 and increased Treg cell differentiation in the intestinal lamina propria. Altogether, our data suggest novel mechanisms by which bile acid metabolites control host immune responses by directly modulating the Th17 and Treg balance.

immunology↗

bioRxiv2018

Emergence of the erythroid lineage from multipotent hematopoiesis

Red cell formation begins with the hematopoietic stem cell, but the manner by which it gives rise to erythroid progenitors, and their subsequent developmental path, remain unclear. Here we combined single-cell transcriptomics of murine hematopoietic tissues with fate potential assays to infer a continuous yet hierarchical structure for the hematopoietic network. We define the erythroid differentiation trajectory as it emerges from multipotency and diverges from 6 other blood lineages. With the aid of a new flow-cytometric sorting strategy, we validated predicted cell fate potentials at the single cell level, revealing a coupling between erythroid and basophil/mast cell fates. We uncovered novel growth factor receptor regulators of the erythroid trajectory, including the proinflammatory IL-17RA, found to be a strong erythroid stimulator; and identified a global hematopoietic response to stress erythropoiesis. We further identified transcriptional and high-purity FACS gates for the complete isolation of all classically-defined erythroid burst-forming (BFU-e) and colony-forming progenitors (CFU-e), finding that they express a dedicated transcriptional program, distinct from that of terminally-differentiating erythroblasts. Intriguingly, profound remodeling of the cell cycle is intimately entwined with CFU-e developmental progression and with a sharp transcriptional switch that extinguishes the CFU-e stage and activates terminal differentiation. Underlying these results, our work showcases the utility of theoretic approaches linking transcriptomic data to predictive fate models, providing key insights into lineage development in vivo.

systems biology↗