Biology subjects

Song, S.-H.

Publications and source records attributed to Song, S.-H..

5 recordsLinked to original sources

bioRxiv2026

Pharmacological intervention targeting neuroimmune axis in the aging hypothalamus prevents age-associated physiological decline

Systemic physiological aging is largely driven by disrupted metabolic homeostasis, yet the central mechanisms of this metabolic dysfunction remain poorly defined. Here, we identify the hypothalamus as a critical hub driving systemic aging through neuroimmune-mediated mechanisms. Single-cell transcriptomic and immunohistochemical analyses revealed that aged hypothalami exhibit significant infiltration of CD8 T lymphocytes, beginning in middle age, with signatures of activation and tissue residency. These T cells intimately interact with tanycytes and microglia, promoting neuroinflammation and progressive tanycyte loss, a defining hallmark of hypothalamic aging. T cell receptor profiling revealed a substantial presence of invariant natural killer T (iNKT) and mucosal-associated invariant T (MAIT) cells, likely activated through cytokine-driven, antigen-independent mechanisms. In aged hypothalamus, microglia secrete chemokines CCL3 and CCL4, whose ectopic expression in young mice was sufficient to trigger persistent hypothalamic T cell infiltration and accelerated systemic aging. Circulating oxidized LDLs (OxLDLs) were identified as upstream inducers of this chemokine response. Notably, pharmacological blockade of the CCL3/4-CCR5 axis with Maraviroc and Cenicriviroc prevented T cell recruitment and ameliorated metabolic and physiological impairments. Given the clinical safety of CCR5 antagonists and individuals lacking functional CCR5 remain generally healthy throughout life, our findings highlight midlife CCL3/4-CCR5 inhibition as a translatable therapeutic target for delaying age-related decline and promoting healthspan.

neuroscience↗

bioRxiv2024

Metabolic profiling using benchtop NMR identifies the metabolomic signature of persistent CAR-T cells

Analytical technologies for engineered biological systems hold great promise in addressing various challenges in modern pharmaceuticals and biomedical therapies. These endeavors often follow a design-build-test-learn approach, utilizing biological data from genetic circuits, signal pathways, metabolites, and proteins to optimize biological systems. Deciphering large and intricate datasets can prove to be a formidable task. Principal component analysis (PCA) tool is an invaluable method for reducing dataset complexity and enhancing interpretability while minimizing information loss, simultaneously. PCA tool achieves this by creating new, uncorrelated variables that capture the maximum variance in the data. Herein chimeric antigen receptors (CARs) T cells metabolomic study presents a slightly inverse problem, where PCA is applied to model sparse extracellular metabolites data from CAR-T cells, resulting in a two-component model. Using an benchtop NMR spectrometer only six metabolites could be annotated, nevertheless, the PCA model could identify differences in the metabolites of CAR-T cells based on the design of CARs, specifically the combinations of the intracellular domain (ICDs). Its noteworthy that the behavior and fate of CAR-T cells are distinctly influenced by the type of ICDs used upon antigen recognition.

cell biology↗

bioRxiv2023

Different mechanisms of synapsin-induced vesicle clustering at inhibitory and excitatory synapses

Synapsins cluster synaptic vesicles (SVs) to provide a reserve pool (RP) of SVs that maintains synaptic transmission during sustained activity. However, it is unknown how synapsins cluster SVs. Here we show that either liquid-liquid phase separation (LLPS) or tetramerization-dependent cross-linking can cluster SVs, depending upon whether a synapse is excitatory or inhibitory. Cell-free reconstitution revealed that both mechanisms can cluster SVs, with tetramerization bring more effective. At inhibitory synapses, perturbing synapsin-dependent LLPS impairs SV clustering and synchronization of GABA release, while perturbing synapsin tetramerization does not. At glutamatergic synapses, the opposite is true: synapsin tetramerization enhances clustering of glutamatergic SVs and mobilization of these SVs from the RP, while synapsin LLPS does not. Comparison of inhibitory and excitatory transmission during prolonged synaptic activity revealed that synapsin LLPS serves as a brake to limit GABA release, while synapsin tetramerization enables rapid mobilization of SVs from the RP to sustain glutamate release.

neuroscience↗

bioRxiv2020

DNA micro-disk for the management of DNA-based data storage with index and write-once-read-many (WORM) memory features

DNA-based data storage has attracted attention because of its higher physical density of the data and longer retention time than those of conventional digital data storage1-7. However, previous DNA-based data storage lacked index features and the data quality of storage after a single access is not preserved, obstructing its industrial use. Here, we propose DNA micro-disks, quick response (QR)-coded micro-sized disks that harbour data-encoded DNA molecules for the efficient management of DNA-based data storage. We demonstrate the two major features that previous DNA-based data storage studies could not achieve. One feature is accessing data items efficiently by indexing the data-encoded DNA library. Another is achieving write-once-read-many (WORM) memory through the immobilization of DNA molecules on the disk and their enrichment through in situ DNA production. Through these features, the reliability of DNA-based data storage was increased by allowing multiple accession of data-encoded DNA without data loss.

bioengineering↗

bioRxiv2018

Addition of Degenerate Bases to DNA-based Data Storage for Increased Information Capacity

Introductory paragraphDNA-based data storage has emerged as a promising method to satisfy the exponentially increasing demand for information storage. However, practical implementation of DNA-based data storage remains a challenge because of the high cost of DNA per unit data. Here, we propose the use of eleven degenerate bases as encoding characters in addition to A, C, G, and T, which increases the information capacity (the amount of data that can be stored per length of DNA sequence designed) and reduce the cost of DNA per unit data. Using the proposed method, we experimentally achieved an information capacity of 3.37 bits/character, which is more than twice when compared to the highest information capacity previously achieved. Finally, the platform was projected to reduce the cost of DNA-based data storage by 50%.

synthetic biology↗