Biology subjects

Lin, S. H.

Publications and source records attributed to Lin, S. H..

9 recordsLinked to original sources

bioRxiv2026

Colchicine Directly Targets Aldehyde Dehydrogenase 2 (ALDH2) to Suppress Radiation-Induced Senescence and Atherosclerosis

BackgroundIonizing radiation (IR) accelerates atherosclerosis through induction of cellular senescence, DNA damage, defective efferocytosis, and dysregulation of clonal hematopoiesis (CH) drivers. Although low-dose colchicine reduces ischemic cardiovascular events in coronary artery disease, the precise molecular mechanisms underlying its vasculoprotective effects remain incompletely defined, and whether it mitigates radiation-associated vascular injury is unknown. MethodsBone marrow-derived macrophages (BMDMs) were pretreated with low-dose colchicine and exposed to 2 Gy IR. Molecular effects were assessed by RNA-seq, immunoblotting, and molecular docking. In vivo effects were tested in a partial carotid ligation (PLCL) model using spatial proteomics. Human monocyte-derived macrophages (HMDMs) from thoracic malignancy patients were analyzed before and after radiation therapy (RT). ResultsLow-dose colchicine suppressed IR-induced macrophage senescence signaling while preserving NRF2 activity. In a cell-free assay, colchicine directly activated aldehyde dehydrogenase 2 (ALDH2) in a dose-dependent manner (EC50 1-5 nM), identifying ALDH2 as a direct molecular target of colchicine. Following irradiation, colchicine restored ALDH2, reduced mitochondrial (mt)ROS-dependent p90 ribosomal S6 kinase (p90RSK) activation and lipid peroxidation, preserved TET2 and DNMT3A expression, and rescued impaired efferocytosis while preventing nicotinamide adenine dinucleotide (NAD) and adenosine triphosphate (ATP) depletion. These protective effects were ALDH2-dependent, as they were lost with ALDH2 inhibition or depletion and were mimicked by pharmacologic ALDH2 activation. In vivo, colchicine attenuated radiation-induced atherosclerosis and macrophage senescence-associated stemness (SAS). Consistently, macrophages from patients after RT showed reduced ALDH2 with increased mtROS, lipid peroxidation, and senescence. ConclusionThese findings identify ALDH2 as a previously unrecognized molecular target of colchicine that links mitochondrial redox control to suppression of radiation-induced macrophage senescence and atherosclerosis and may contribute to the efficacy of low-dose colchicine in cardiovascular disease.

cell biology↗

bioRxiv2026

SuperSurv: A Unified Framework for Machine Learning Ensembles in Survival Analysis

This paper introduces SuperSurv, a user-friendly R package for building, evaluating, and interpreting ensemble models for right-censored survival data. Although many survival modeling methods are available, existing tools are often model-specific and lack a unified platform for systematically integrating, comparing, and ensembling heterogeneous learners. SuperSurv addresses this gap by providing a unified interface for diverse survival learners, including models that return full survival curves as well as methods that produce only risk scores. All learner outputs are mapped to calibrated survival probability curves on a common evaluation time grid, enabling direct comparison and ensemble construction across heterogeneous model classes. SuperSurv implements stacking of survival models using inverse-probability-of-censoring weighted (IPCW) Brier risk to estimate ensemble weights in the presence of right censoring. The framework integrates hyperparameter tuning, time-dependent benchmarking metrics, and visualization tools for survival model evaluation. In addition, the package provides post-hoc interpretability utilities based on SHAP(SHapley Additive exPlanations) values and supports covariate-adjusted restricted mean survival time (RMST) contrasts through g-computation. By bridging the gap between theoretical rigor and clinical application, SuperSurv offers researchers a comprehensive ecosystem for modern survival analysis. The SuperSurv package is open-source and available on CRAN at https://CRAN.R-project.org/package=SuperSurv and on GitHub at https://github.com/yuelyu21/SuperSurv. An empirical example using the METABRIC breast cancer dataset demonstrates a complete workflow from model training and benchmarking to explainability and clinically interpretable survival contrasts.

bioinformatics↗

bioRxiv2025

FLASH Radiotherapy Mitigates Radiation-Induced Lymphopenia and Prevents Immunosuppression via Chk1-STAT3 Axis Modulation in a Preclinical Thoracic Irradiation Model

Background and AimsRadiation-induced lymphopenia (RIL) is a frequent side effect of conventional radiation therapy (CONV RT), due to the high radiosensitivity of circulating lymphocytes. Ultra-high dose rate "FLASH" RT may preferentially spare normal tissue while maintaining tumor control. This study evaluates the impact of single-fraction and multi-fraction thoracic FLASH RT on lymphocyte preservation, apoptosis, and immunosuppressive signaling in mice. MethodsWe compared the immunological impact of thoracic FLASH RT and CONV RT in C57BL/6 mice using single-fraction (17 Gy) and multi-fraction (2 Gy x 5) regimens using the Mobetron (IntraOp). Longitudinal blood sampling was performed at multiple time-points post-irradiation through facial vein bleed with flow cytometry analysis for CD4+, CD8+, CD19+, and NK cells to assess lymphocyte counts, apoptotic lymphocytes through Annexin V staining, and immune suppression by examining regulatory T cells (Tregs) and PD-1/PD-L1 expression. Mechanistic studies included immunofluorescence and Western blot analyses of splenic tissues to evaluate Chk1 and STAT3 signaling pathways. ResultsIn single-fraction RT, FLASH significantly reduced lung and heart fibrosis (p < 0.0001) at 28 weeks post-RT. The FLASH effect was also seen acutely on circulating immune cells, with significantly reduced lymphocyte apoptosis and accelerated recovery of CD4, CD8, CD3, NK, and B cell populations compared to CONV RT in both single-fraction and multi-fraction regimens. Conversely, CONV RT induced long-lasting increases in Tregs and sustained PD-1 and PD-L1 expression on T- and B-cells at 2- and 5-months post-irradiation in both fractionation regimens. Within the spleen, we also found CONV RT induced sustained activation of the Chk1-STAT3 pathway in CD45+ immune cells, which correlates with increased PD-1/PD-L1 expression. ConclusionFLASH RT mitigates RIL, reduces lymphocyte apoptosis, and prevents long-term immunosuppression by reduced activation of the Chk1-STAT3 pathway. These findings suggest FLASH RT may confer immunological advantages over CONV RT to enhance therapeutic efficacy.

cancer biology↗

bioRxiv2025

A Hierarchical Robust Linear Model for Cryo-EM Map Analysis

Cryo-electron microscopy (cryo-EM) has become a central tool for determining the atomic structures of biological macromolecules, producing three-dimensional reconstruction maps that guide atomic model building. Current quantitative analyses of cryo-EM maps largely rely on simplified Gaussian signal models with fixed width parameters, limiting their ability to capture atom-specific signal characteristics directly from the maps. We propose a robust hierarchical linear (RHL) model for the statistical analysis of paired cryo-EM maps and atomic structures deposited in the EMDB and PDB. Within this framework, the logarithms of local voxel intensities surrounding each atom are modeled using a linearized Gaussian form, and atom-specific amplitude and width parameters are estimated through a hierarchical structure that pools information across atoms of the same type. The primary objective is to provide a statistically grounded framework for extracting quantitative atomic signal features from cryo-EM maps without imposing rigid physical assumptions on these parameters. To address contamination arising from overlapping atomic signals, spatially varying resolution, and experimental noise, we incorporate a data-adaptive weighting scheme based on minimum density power divergence estimation (MDPDE). This formulation preserves computational efficiency and ensures stable parameter estimation, while naturally facilitating the identification of deviated atomic profiles that may stem from model misspecification or local structural heterogeneity. Simulation studies demonstrate that the proposed method yields stable group-level parameter estimates even under substantial contamination. Applications to cryo-EM maps at multiple resolutions show that an atomic resolution (1.25 Angstrom) apoferritin map exhibits systematic atom-type differences in Gaussian amplitude and width, whereas a near-atomic-resolution map (2.20 Angstrom) shows closer agreement with conventional uniform-width assumptions. Beyondstructural biology, the proposed RHL framework provides a general statistical methodology for hierarchical data integration under heterogeneous noise, with potential applications in multi-center studies and federated learning.

bioinformatics↗

bioRxiv2024

Discordance in acute gastrointestinal toxicity between synchrotron-based proton and linac-based electron ultra-high dose rate irradiation

PurposeProton FLASH has been investigated using cyclotron and synchrocyclotron beamlines but not synchrotron beamlines. We evaluated the impact of dose rate (ultra-high [UHDR] vs. conventional [CONV]) and beam configuration (shoot-through [ST] vs. spread-out-Bragg-peak [SOBP]) on acute radiation-induced gastrointestinal toxicity (RIGIT) in mice. We also compared RIGIT between synchrotron-based protons and linac-based electrons with matched mean dose rates. Methods and MaterialsWe administered abdominal irradiation (12-14 Gy single fraction) to female C57BL/6J mice with an 87 MeV synchrotron-based proton beamline (2 cm diameter field size as a lateral beam). Dose rates were 0.2 Gy/s (S-T pCONV), 0.3 Gy/s (SOBP pCONV), 150 Gy/s (S-T pFLASH), and 230 Gy/s (SOBP pFLASH). RIGIT was assessed by the jejunal regenerating crypt assay and survival. We also compared responses to proton [pFLASH and pCONV] with responses to electron CONV (eCONV, 0.4 Gy/s) and electron FLASH (eFLASH, 188-205 Gy/s). ResultsThe number of regenerating jejunal crypts at each matched dose was lowest for pFLASH (similar between S-T and SOBP), greater and similar between pCONV (S-T and SOBP) and eCONV, and greatest for eFLASH. Correspondingly, mice that received pFLASH SOBP had the lowest survival rates (50% at 50 days), followed by pFLASH S-T (80%), and pCONV SOBP (90%), but 100% of mice receiving pCONV S-T survived (log-rank P = 0.047 for the four groups). ConclusionsOur findings are consistent with an increase in RIGIT after synchrotron-based pFLASH versus pCONV. This negative proton-specific FLASH effect versus linac-based electron irradiation underscores the importance of understanding the physical and biological factors that will allow safe and effective clinical translation.

cancer biology↗

bioRxiv2024

Redefining FLASH RT: the impact of mean dose rate and dose per pulse in the gastrointestinal tract

BackgroundThe understanding of how varying radiation beam parameter settings affect the induction and magnitude of the FLASH effect remains limited. PurposeWe sought to evaluate how the magnitude of radiation-induced gastrointestinal (GI) toxicity (RIGIT) depends on the interplay between mean dose rate (MDR) and dose per pulse (DPP). MethodsC57BL/6J mice were subjected to total abdominal irradiation (11-14 Gy single fraction) under conventional irradiation (low DPP and low MDR, CONV) and various combinations of DPP and MDR up to ultra-high-dose-rate (UHDR) beam conditions. The effects of DPP were evaluated for DPPs of 1-6 Gy while the total dose and MDR were kept constant; the effects of MDR were evaluated for the range 0.3- 1440 Gy/s while the total dose and DPP were kept constant. RIGIT was quantified in non-tumor-bearing mice through the regenerating crypt assay and survival assessment. Tumor response was evaluated through tumor growth delay. ResultsWithin each tested total dose using a constant MDR (>100 Gy/s), increasing DPP led to better sparing of regenerating crypts, with a more prominent effect seen at 12 and 14 Gy TAI. However, at fixed DPPs >4 Gy, similar sparing of crypts was demonstrated irrespective of MDR (from 0.3 to 1440 Gy/s). At a fixed high DPP of 4.7 Gy, survival was equivalently improved relative to CONV for all MDRs from 0.3 Gy/s to 104 Gy/s, but at a lower DPP of 0.93 Gy, increasing MDR produced a greater survival effect. We also confirmed that high DPP, regardless of MDR, produced the same magnitude of tumor growth delay relative to CONV using a clinically relevant melanoma mouse model. ConclusionsThis study demonstrates the strong influence that the beam parameter settings have on the magnitude of the FLASH effect. Both high DPP and UHDR appeared independently sufficient to produce FLASH sparing of GI toxicity, while isoeffective tumor response was maintained across all conditions.

cancer biology↗

bioRxiv2024

High-fidelity enhanced AsCas12a knock-in mice for efficient multiplexed gene editing, disease modeling and orthogonal immunogenetics

The advancement of CRISPR gene editing technology, especially the development of Cas9 knock-in mice, has significantly boosted the functional discovery of various genetics factors in diverse fields including genetics, genomics, immunology, and the biology of cancer. However, the pleiotropic effects on human disease and the complex nature of gene interaction networks require a knock-in mouse model capable of simultaneous multiplexed gene perturbation. Here, we present the generation and applications of Cre-dependent conditional and constitutive high-fidelity, enhanced AsCas12a (enAsCas12a-HF1) Rosa26-knock-in mice in the C57BL/6 background. With these mouse strains, we demonstrate highly efficient and multiplexed in vivo and ex vivo genome engineering as applied to lipid nanoparticle (LNP)-RNA-based liver protein targeting, AAV-based tumor modeling, and retrovirus-based immune cell engineering. By integrating with a dCas9-SPH CRISPR activation transgenic strain, we establish a simultaneous dual gene activation and knockout (DAKO) system that showcases the modular potential of these enAsCas12a-HF1 mice. Importantly, constitutive expression of enAsCas12a-HF1 does not lead to any discernable pathological differences as compared to the C57BL/6 background strain. These knock-in mice and the accompanying delivery methods would empower the deconvolution of complex gene interaction networks in broad areas of research.

bioengineering↗

bioRxiv2020

Cell lines of the same anatomic site and histologic type show large variability in intrinsic radiosensitivity and relative biological effectiveness to protons and carbon ions

PurposeTo show that radiation response across cancer cell lines of the same anatomic site and histologic type varies remarkably for protons and carbon (C) ions. Materials and MethodsWe measured and obtained from the literature clonogenic survival of human cancer cell lines of the lung (n=18), brain (n=10) and pancreas (n=10) exposed to photons, protons, and C-ions to assess their variability in response. We also treated cancer cell lines with DNA repair inhibitors prior to irradiation to assess how DNA repair capacity affects their variability in response. We quantified the variability in response by calculating the relative range (range/mean) and the coefficient of variation (COV) of the dose at 10% survival fraction (D10%) and relative biological effectiveness (RBE10%). ResultsThe relative range of D10% for lung cancer cell lines varied from 55-92% for photons, protons, and C-ions, with the relative range in RBE varying from 16-45% for protons and C-ions. For brain and pancreatic cancer cell lines, the relative range of D10% varied from 95-112%, and 39-75%, respectively, with the relative range in RBE varying from 27-33% and 25-50%, respectively. However, the COVs in D10% were approximately equal across radiation qualities, varying from 0.24{+/-}0.07-0.35{+/-}0.10, 0.35{+/-}0.09-0.69{+/-}0.62 and 0.13{+/-}0.03- 0.21{+/-}0.04 for lung, brain and pancreatic cancer cell lines, respectively. Greater relative ranges in D10% were observed in the cell lines with inhibited DNA repair, varying from 108%-157% for photons, protons, and C-ions, with relative ranges in RBE varying from 29-67%. The COVs in the D10% were also greater for the cell lines treated with inhibitors of DNA repair, varying from 0.34{+/-}0.09-0.41{+/-}0.06. ConclusionCell lines of the same anatomic site and histologic type have a remarkable variability in response, not only to photons but also to protons and C-ions. We attributed this variability to differences in DNA repair capacity. CategoryBiological Physics and Response Prediction

biophysics↗

bioRxiv2020

Gene regulatory networks controlling differentiation, survival, and diversification of hypothalamic Lhx6-expressing GABAergic neurons.

GABAergic neurons of the hypothalamus regulate many innate behaviors, but little is known about the mechanisms that control their development. We previously identified hypothalamic neurons that express the LIM homeodomain transcription factor Lhx6, a master regulator of cortical interneuron development, as sleep-promoting. In contrast to telencephalic interneurons, hypothalamic Lhx6 neurons do not undergo long-distance tangential migration and do not express cortical interneuronal markers such as Pvalb. Here, we show that Lhx6 is necessary for the survival of hypothalamic neurons. Dlx1/2, Nkx2-2, and Nkx2-1 are each required for specification of spatially distinct subsets of hypothalamic Lhx6 neurons, and that Nkx2-2+/Lhx6+ neurons of the zona incerta are responsive to sleep pressure. We further identify multiple neuropeptides that are enriched in spatially segregated subsets of hypothalamic Lhx6 neurons, and that are distinct from those seen in cortical neurons. These findings identify common and divergent molecular mechanisms by which Lhx6 controls the development of GABAergic neurons in the hypothalamus.

neuroscience↗