Biology subjects

Go, C. D.

Publications and source records attributed to Go, C. D..

3 recordsLinked to original sources

bioRxiv2023

α-catenin mechanosensitivity as a route to cytokinesis failure through sequestration of LZTS2

Epithelial cells can become polyploid upon tissue injury, but mechanosensitive cues that trigger this state are poorly understood. Using -catenin (-cat) knock-out Madin Darby Canine Kidney (MDCK) cells reconstituted with wild-type and mutant forms of -cat as a model system, we find that an established -cat actin-binding domain unfolding mutant designed to reduce force-sensitive binding to F-actin (-cat-H0-FABD+) can promote cytokinesis failure, particularly along epithelial wound-fronts. Enhanced -cat coupling to cortical actin is neither sufficient nor mitotic cell-autonomous for cytokinesis failure, but critically requires the mechanosensitive Middle-domain (M1-M2-M3) and neighboring cells. Disease relevant -cat M-domain missense mutations known to cause a form of retinal pattern dystrophy (-cat E307K or L436P) are associated with elevated binucleation rates via cytokinesis failure. Similar binucleation rates are seen in cells expressing an -cat salt-bridge destabilizing mutant (R551A) designed to promote M2-M3 domain unfurling at lower force thresholds. Since binucleation is strongly enhanced by removal of the M1 as opposed to M2-M3 domains, cytokinetic fidelity is most sensitive to -cat M2-M3 domain opening. To identify -cat conformation-dependent proximity partners that contribute to cytokinesis, we used a biotin-ligase approach to distinguished proximity partners that show enhanced recruitment upon -cat M-domain unfurling (R551A). We identified Leucine Zipper Tumor Suppressor 2 (LZTS2), an abscission factor previously implicated in cytokinesis. We confirm that LZTS2 enriches at the midbody, but discover it also localizes to tight and tricellular junctions. LZTS2 knock-down promotes binucleation in both MDCK and Retinal Pigmented Epithelial (RPE) cells. -cat mutants with persistent M2-M3 domain opening showed elevated junctional enrichment of LZTS2 from the cytosol compared -cat wild-type cells. These data implicate LZTS2 as a mechanosensitive effector of -cat that is critical for cytokinetic fidelity. This model rationalizes how persistent mechano-activation of -cat may drive tension-induced polyploidization of epithelia post-injury and suggests an underlying mechanism for how pathogenic -cat mutations drive macular dystrophy.

cell biology↗

bioRxiv2020

A SARS-CoV-2 - host proximity interactome

Viral replication is dependent on interactions between viral polypeptides and host proteins. Identifying virus-host protein interactions can thus uncover unique opportunities for interfering with the virus life cycle via novel drug compounds or drug repurposing. Importantly, many viral-host protein interactions take place at intracellular membranes and poorly soluble organelles, which are difficult to profile using classical biochemical purification approaches. Applying proximity-dependent biotinylation (BioID) with the fast-acting miniTurbo enzyme to 27 SARS-CoV-2 proteins in a lung adenocarcinoma cell line (A549), we detected 7810 proximity interactions (7382 of which are new for SARS-CoV-2) with 2242 host proteins (results available at covid19interactome.org). These results complement and dramatically expand upon recent affinity purification-based studies identifying stable host-virus protein complexes, and offer an unparalleled view of membrane-associated processes critical for viral production. Host cell organellar markers were also subjected to BioID in parallel, allowing us to propose modes of action for several viral proteins in the context of host proteome remodelling. In summary, our dataset identifies numerous high confidence proximity partners for SARS-CoV-2 viral proteins, and describes potential mechanisms for their effects on specific host cell functions.

systems biology↗

bioRxiv2019

A proximity biotinylation map of a human cell

Compartmentalization is a defining characteristic of eukaryotic cells, partitioning cellular processes into discrete subcellular locations. High throughput microscopy1 and biochemical fractionation coupled with mass spectrometry2-6 have helped to define the proteomes of a variety of organelles and macromolecular structures. However, many other intracellular compartments have remained refractory to such approaches, due for example to difficulty in purifying non-membrane bound structures. Proximity-dependent biotinylation techniques such as BioID provide an alternative approach for defining the composition of cellular compartments in living cells7-10. Here we present a BioID-based map of a human cell based on 192 markers from 32 different subcellular compartments, comprising 35,902 high confidence proximity interactions, and defining the intracellular locations of 4,145 unique proteins in HEK 293 cells. Our localization predictions meet or exceed previous-approaches, with higher specificity, and enabled the discovery of proteins at the mitochondrial outer membrane-endoplasmic reticulum (ER) interface that are critical for mitochondrial homeostasis. Based on this dataset, we have established humancellmap.org as a community resource that provides online tools for localization analysis of user BioID data, and demonstrate how this resource can be used to better understand BioID datasets.

molecular biology↗