Unconventional protein secretion-mediated protein quality control in health and diseases

点击翻译按钮获取中文摘要

Proteinaceous inclusions termed Lewy bodies (LBs) are a classical hallmark of Parkinson's disease (PD). The primary component of these inclusions is alpha-synuclein (a-syn), a protein with an intrinsic propensity to misfold and aggregate. In PD patients, alpha-syn inclusions first observed in the olfactory bulb and the dorsal motor nucleus, progressively spread throughout the brain. Further findings that healthy embryonic dopamine neurons transplanted into PD patients developed LBs points suggest the tantalizing possibility of neuron-to-neuron transmission of a-syn. Subsequent work confirmed that synthetic a-syn pre-formed fibrils (PFFs) can be taken up by neurons, eliciting the misfolding of endogenous -syn into insoluble Lewy-like inclusions. Collectively, these studies led to the prion hypothesis of PD, wherein misfolded -syn provides a template for seeding new aggregates, propagating misfolded a-syn, and associated cytotoxicity. Thus, the propagation of alpha-syn is a viable new target to be explored in the development of new PD therapies. The intercellular transmission of synuclein consists of two key steps: the secretion of synuclein from a donor neuron and its uptake by a recipient neuron. Misfolding-associated protein secretion (MAPS) is a recently discovered protein quality control process that selectively exports misfolded cytosolic proteins including a-syn. Secretion through MAPS requires the membrane-localized deubiquitinase USP19, which recruits aberrant polypeptides to the endoplasmic reticulum (ER) surface to facilitate their incorporation into late endosomes that are in tight association with the ER. Misfolded proteins are secreted to the extracellular milieu when late endosomes fuse with the plasma membrane. The fate of the released misfolded proteins is currently unknown. Our recent studies suggest that mammalian cells can internalize misfolded proteins via endocytosis, but it is unclear whether they possess one or more receptors for misfolded proteins. Whether internalized proteins can impose damage prior to degradation by the lysosome is also unclear. The proposal is to elucidate the physiological relevance of the MAPS pathway by characterizing the interplay between secreted misfolded proteins and target cells. Using a proximity-based ligation approach, we have used purified Tau as a bait to identify candidate Protein aggregates formed by microtubule-associated protein Tau can be transmitted in a stereotypic pattern in human brains, which correlates with the progression of several neurodegenerative diseases collectively termed Tauopathies. This process requires Tau, released from neurons, to interact with a cell surface receptor on a target cell, but little is known about the underlying mechanisms and the downstream pathophysiological consequences, particularly for Tau that engages glial cells. Using a spatially resolved proteomic mapping strategy, we identify integrin V/1, a heterodimer of the integrin family, as a receptor that not only mediates Tau fibril entry into astrocytes but also activates integrin signaling upon Tau binding. We show that distinct Tau species differentially activate integrin signaling, with filamentous Tau being a more robust stimulator. This leads to NFB activation and differential upregulation of pro-inflammatory cytokines and chemokines. Additionally, a sub-group of neurotoxic astrocyte markers are also induced in an integrin-dependent manner preferentially by filamentous Tau, causing astrocytes to release a neurotoxic factor(s). Together, these findings establish a paradigm that astrocytes can be directly converted into a neurotoxic state by filamentous Tau via an integrin receptor, which as a sensor transducing unfolded protein pathology, may provide a new therapeutic target for Tauopathies. In a more recent effort, we used RNAseq to analyze the gene expression profile changes in response to a proteotoxic insult using tau fibrils as a model. We find that in addition to integrin signing, tau fibril ligation to an integrin receptor also activates PI3K, which leads to a cellular response that is distinct from the canonical integrin signaling. We further analyzed the secretome of cells treated with tau fibrils and found that many proteins are secreted in a tau-inducible manner. Among the secreted proteins, we identified C3 complement and Mmp3 metalloprotease as key regulators of neuronal cell death. This study reveals why astrocytes stimulated by tau fibrils can generate a pathological response via a physiologically critical cell surface receptor. We recently reported that the infection of human cells by SARS-CoV-2, the virus underlying the ongoing COVID-19 pandemic depends on an endocytosis mechanism akin to the entry of alpha-Syn fibrils. Accordingly, knockdown of cell surface HSPG or cortex actin reduces viral infection in an in vitro cell-based model. A drug repurpose screen has identified several FDA-approved drugs that can effectively block the entry of SARS-CoV-2 in vitro. We further showed that in addition to protein receptors, the SARS-CoV-2 spike (S) protein also interacts with heparan sulfate, a negatively charged glycosaminoglycan (GAG) attached to certain membrane proteins on the cell surface. This interaction facilitates the engagement of Spike with a downstream receptor to promote viral entry. We reported that Mitoxantrone, an FDA-approved topoisomerase inhibitor, targets a heparan sulfate-spike complex to compromise the fusogenic function of Spike in viral entry. As a single agent, Mitoxantrone inhibits the infection of an authentic SARS-CoV-2 strain in a cell-based model and in human lung EpiAirway 3D tissues. Gene expression profiling supports the plasma membrane as a major target of Mitoxantrone but also underscores an undesired activity targeting nucleosome dynamics. We propose that Mitoxantrone analogs bearing similar heparan sulfate-binding activities but with reduced affinity for DNA topoisomerases may offer an alternative therapy to overcome breakthrough infections in the post-vaccine era. To identify additional viral entry inhibitors, we also developed and trained a graph convolutional network (GCN)-based classification model using information extracted from experimentally identified HSPG and actin inhibitors. This method allowed us to virtually screen 170,000 compounds, resulting in 2000 potential hits. A hit confirmation assay with the uptake of a fluorescently labeled HSPG cargo further shortlisted 256 active compounds. Among them, 16 compounds had modest to the strong inhibitory activity