Deep Learning-Based Morphological Classification of Endoplasmic Reticulum Under Stress.

Endoplasmic reticulum stress (ER stress) is a condition that is defined by abnormal accumulation of unfolded proteins. It plays an important role in maintaining cellular protein, lipid, and ion homeostasis. By triggering the unfolded protein response (UPR) under ER stress, cells restore homeostasis or undergo apoptosis. Chronic ER stress is implicated in many human diseases. Despite extensive studies on related signaling mechanisms, reliable image biomarkers for ER stress remain lacking. To address this deficiency, we have validated a morphological image biomarker for ER stress and have developed a deep learning-based assay to enable automated detection and analysis of this marker for screening studies. Specifically, ER under stress exhibits abnormal morphological patterns that feature ring-shaped structures called whorls (WHs). Using a highly specific chemical probe for unfolded and aggregated proteins, we find that formation of ER whorls is specifically associated with the accumulation of the unfolded and aggregated proteins. This confirms that ER whorls can be used as an image biomarker for ER stress. To this end, we have developed ER-WHs-Analyzer, a deep learning-based image analysis assay that automatically recognizes and localizes ER whorls similarly as human experts. It does not require laborious manual annotation of ER whorls for training of deep learning models. Importantly, it reliably classifies different patterns of ER whorls induced by different ER stress drugs. Overall, our study provides mechanistic insights into morphological patterns of ER under stress as well as an image biomarker assay for screening studies to dissect related disease mechanisms and to accelerate related drug discoveries. It demonstrates the effectiveness of deep learning in recognizing and understanding complex morphological phenotypes of ER.

Pathological hydrogen peroxide triggers the fibrillization of wild-type SOD1 via sulfenic acid modification of Cys-111.

Amyotrophic lateral sclerosis (ALS) involves the abnormal posttranslational modifications and fibrillization of copper, zinc superoxide dismutase (SOD1) and TDP-43. However, how SOD1-catalyzed reaction product hydrogen peroxide affects amyloid formation of SOD1 and TDP-43 remains elusory. 90% of ALS cases are sporadic and the remaining cases are familial ALS. In this paper, we demonstrate that H2O2 at pathological concentrations triggers the fibrillization of wild-type SOD1 both in vitro and in SH-SY5Y cells. Using an anti-dimedone antibody that detects sulfenic acid modification of proteins, we found that Cys-111 in wild-type SOD1 is oxidized to C-SOH by pathological concentration of H2O2, followed by the formation of sulfenic acid modified SOD1 oligomers. Furthermore, we show that such SOD1 oligomers propagate in a prion-like manner, and not only drive wild-type SOD1 to form fibrils in the cytoplasm but also induce cytoplasm mislocalization and the subsequent fibrillization of wild-type TDP-43, thereby inducing apoptosis of living cells. Thus, we propose that H2O2 at pathological concentrations triggers the fibrillization of wild-type SOD1 and subsequently induces SOD1 toxicity and TDP-43 toxicity in neuronal cells via sulfenic acid modification of Cys-111 in SOD1. Our Western blot and ELISA data demonstrate that sulfenic acid modified wild-type SOD1 level in cerebrospinal fluid of 15 sporadic ALS patients is significantly increased compared with 6 age-matched control patients. These findings can explain how H2O2 at pathologic concentrations regulates the misfolding and toxicity of SOD1 and TDP-43 associated with ALS, and suggest that sulfenic acid modification of wild-type SOD1 should play pivotal roles in the pathogenesis of sporadic ALS.

Crystal Structure of the ERp44-Peroxiredoxin 4 Complex Reveals the Molecular Mechanisms of Thiol-Mediated Protein Retention.

ERp44 controls the localization and transport of diverse proteins in the early secretory pathway. The mechanisms that allow client recognition and the source of the oxidative power for forming intermolecular disulfides are as yet unknown. Here we present the structure of ERp44 bound to a client, peroxiredoxin 4. Our data reveal that ERp44 binds the oxidized form of peroxiredoxin 4 via thiol-disulfide interchange reactions. The structure explains the redox-dependent recognition and characterizes the essential non-covalent interactions at the interface. The ERp44-Prx4 covalent complexes can be reduced by glutathione and protein disulfide isomerase family members in the ER, allowing the two components to recycle. This work provides insights into the mechanisms of thiol-mediated protein retention and indicates the key roles of ERp44 in this biochemical cycle to optimize oxidative folding and redox homeostasis.