The ESCRT machinery mediates membrane fission in a variety of processes in cells. According to current models, ESCRT-III proteins drive membrane fission by assembling into helical filaments on membranes. Here, we used 3D STORM imaging of endogenous ESCRT-III component IST1 to reveal the evolution of the structural organization of ESCRT-III in mammalian cytokinetic abscission. Using this approach, ESCRT-III ring and spiral assemblies were resolved and characterized at different stages of abscission. Visualization of IST1 structures in cells lacking the microtubule-severing enzyme spastin and in cells depleted of specific ESCRT-III components or the ATPase VPS4 demonstrated the contribution of these components to the organization and function of ESCRTs in cells. This work provides direct evidence that ESCRT-III proteins form helical filaments to mediate their function in cells and raises new mechanistic scenarios for ESCRT-driven cytokinetic abscission.

This work describes an elegant approach for direct, site-specific labeling of proteins with fluorescent-dyes for live cell imaging. By integrating a noncanonical amino acid that is capable of binding a fluorescent dye into tubulin, we directly and specifically labeled tubulin with a fluorescent-dye and imaged microtubules in live mammlian cells.

No abstract

Background: In the high-resolution microscopy era, genetic code expansion (GCE)-based bioorthogonal labeling offers an elegant way for direct labeling of proteins in live cells with fluorescent dyes. This labeling approach is currently not broadly used in live-cell applications, partly because it needs to be adjusted to the specific protein under study. Results: We present a generic, 14-residue long, N-terminal tag for GCE-based labeling of proteins in live mammalian cells. Using this tag, we generated a library of GCE-based organelle markers, demonstrating the applicability of the tag for labeling a plethora of proteins and organelles. Finally, we show that the HA epitope, used as a backbone in our tag, may be substituted with other epitopes and, in some cases, can be completely removed, reducing the tag length to 5 residues. Conclusions: The GCE-tag presented here offers a powerful, easy-to-implement tool for live-cell labeling of cellular proteins with small and bright probes.

In the high-resolution microscopy era, genetic code expansion (GCE)-based bioorthogonal labeling offers an elegant way for direct labeling of proteins in live cells with fluorescent dyes. This labeling approach is currently not broadly used live cell applications, partly because it needs to be adjusted to the specific protein under study. Here, we present a generic, 14-residues long, N-terminal tag for GCE-based labeling of proteins in live mammalian cells. Using this tag, we generated a library of GCE-based organelle markers, demonstrating the applicability of the tag for labeling a plethora of proteins and organelles. Finally, we show that the HA epitope, used as a backbone in our tag, can be substituted with other epitopes and, in some cases, can be completely removed, reducing the tag length to 5 residues. The GCE-tag presented here offers a powerful, easy-to-implement tool for live cell labeling of cellular proteins with small and bright probes. 3 BACKGROUND

Avian oviductal secretions contain an exoATPase which has been purified up to 900-fold. Antibodies to this fraction bind to several cell types and to membrane fractions solubilized by selected detergents. The experiments indicate that a related antigen is present in plasma membranes and that exoATPase is released by a shedding process.

The assembly and budding of newly formed human immunodeficiency virus-1 (HIV-1) particles occur at the plasma membrane of infected cells. Whereas the molecular basis for viral budding has been studied extensively, investigation of its spatiotemporal characteristics has been limited by the small dimensions (< 100 nm) of HIV particles and the fast kinetics of the process (a few minutes from bud formation to virion release). Here we applied ultra-fast atomic force microscopy to achieve real-time visualization of individual HIV-1 budding events from wildtype (WT) cell lines as well as from mutated lines lacking vacuolar protein sorting-4 (VPS4) or visceral adipose tissue-1 protein (VTA1). Using single-particle analysis, we show that HIV-1 bud formation follows two kinetic pathways (fast and slow) with each composed of three distinct phases (growth, stationary, decay). Notably, approximately 30% of events did not result in viral release and were characterized by the formation of short (rather than tall) particles that slowly decayed back into the cell membrane. These non-productive events became more abundant in VPS4 knockout cell lines. Strikingly, the absence of VPS4B, rather than VPS4A, increased the production of short viral particles, suggesting a role for VPS4B in earlier stages of HIV-1 budding than traditionally thought.