Nanoscale cellular organization of viral RNA and proteins in SARS-CoV-2 replication organelles.

The SARS-CoV-2 viral infection transforms host cells and produces special organelles in many ways, and we focus on the replication organelle where the replication of viral genomic RNA (vgRNA) occurs. To date, the precise cellular localization of key RNA molecules and replication intermediates has been elusive in electron microscopy studies. We use super-resolution fluorescence microscopy and specific labeling to reveal the nanoscopic organization of replication organelles that contain vgRNA clusters along with viral double-stranded RNA (dsRNA) clusters and the replication enzyme, encapsulated by membranes derived from the host endoplasmic reticulum (ER). We show that the replication organelles are organized differently at early and late stages of infection. Surprisingly, vgRNA accumulates into distinct globular clusters in the cytoplasmic perinuclear region, which grow and accommodate more vgRNA molecules as infection time increases. The localization of ER labels and nsp3 (a component of the double-membrane vesicle, DMV) at the periphery of the vgRNA clusters suggests that replication organelles are enclosed by DMVs at early infection stages which then merge into vesicle packets as infection progresses. Precise co-imaging of the nanoscale cellular organization of vgRNA, dsRNA, and viral proteins in replication organelles of SARS-CoV-2 may inform therapeutic approaches that target viral replication and associated processes.

Curved adhesions mediate cell attachment to soft matrix fibres in three dimensions.

Integrin-mediated focal adhesions are the primary architectures that transmit forces between the extracellular matrix (ECM) and the actin cytoskeleton. Although focal adhesions are abundant on rigid and flat substrates that support high mechanical tensions, they are sparse in soft three-dimensional (3D) environments. Here we report curvature-dependent integrin-mediated adhesions called curved adhesions. Their formation is regulated by the membrane curvatures imposed by the topography of ECM protein fibres. Curved adhesions are mediated by integrin ɑvß5 and are molecularly distinct from focal adhesions and clathrin lattices. The molecular mechanism involves a previously unknown interaction between integrin ß5 and a curvature-sensing protein, FCHo2. We find that curved adhesions are prevalent in physiological conditions, and disruption of curved adhesions inhibits the migration of some cancer cell lines in 3D fibre matrices. These findings provide a mechanism for cell anchorage to natural protein fibres and suggest that curved adhesions may serve as a potential therapeutic target.

Combining deep learning approaches and point spread function engineering for simultaneous 3D position and 3D orientation measurements of fluorescent single molecules.

Point Spread Function (PSF) engineering is an effective method to increase the sensitivity of single-molecule fluorescence images to specific parameters. Classical phase mask optimization approaches have enabled the creation of new PSFs that can achieve, for example, localization precision of a few nanometers axially over a capture range of several microns with bright emitters. However, for complex high-dimensional optimization problems, classical approaches are difficult to implement and can be very time-consuming for computation. The advent of deep learning methods and their application to single-molecule imaging has provided a way to solve these problems. Here, we propose to combine PSF engineering and deep learning approaches to obtain both an optimized phase mask and a neural network structure to obtain the 3D position and 3D orientation of fixed fluorescent molecules. Our approach allows us to obtain an axial localization precision around 30 nanometers, as well as an orientation precision around 5 degrees for orientations and positions over a one micron depth range for a signal-to-noise ratio consistent with what is typical in single-molecule cellular imaging experiments.

Curved adhesions mediate cell attachment to soft matrix fibres in 3D.

Mammalian cells adhere to the extracellular matrix (ECM) and sense mechanical cues through integrin-mediated adhesions 1, 2 . Focal adhesions and related structures are the primary architectures that transmit forces between the ECM and the actin cytoskeleton. Although focal adhesions are abundant when cells are cultured on rigid substrates, they are sparse in soft environments that cannot support high mechanical tensions 3 . Here, we report a new class of integrin-mediated adhesions, curved adhesions, whose formation is regulated by membrane curvature instead of mechanical tension. In soft matrices made of protein fibres, curved adhesions are induced by membrane curvatures imposed by the fibre geometry. Curved adhesions are mediated by integrin ɑVß5 and are molecularly distinct from focal adhesions and clathrin lattices. The molecular mechanism involves a previously unknown interaction between integrin ß5 and a curvature-sensing protein FCHo2. We find that curved adhesions are prevalent in physiologically relevant environments. Disruption of curved adhesions by knocking down integrin ß5 or FCHo2 abolishes the migration of multiple cancer cell lines in 3D matrices. These findings provide a mechanism of cell anchorage to natural protein fibres that are too soft to support the formation of focal adhesions. Given their functional importance for 3D cell migration, curved adhesions may serve as a therapeutic target for future development.

Multi-color super-resolution imaging to study human coronavirus RNA during cellular infection.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the third human coronavirus within 20 years that gave rise to a life-threatening disease and the first to reach pandemic spread. To make therapeutic headway against current and future coronaviruses, the biology of coronavirus RNA during infection must be precisely understood. Here, we present a robust and generalizable framework combining high-throughput confocal and super-resolution microscopy imaging to study coronavirus infection at the nanoscale. Using the model human coronavirus HCoV-229E, we specifically labeled coronavirus genomic RNA (gRNA) and double-stranded RNA (dsRNA) via multi-color RNA immunoFISH and visualized their localization patterns within the cell. The 10-nm resolution achieved by our approach uncovers a striking spatial organization of gRNA and dsRNA into three distinct structures and enables quantitative characterization of the status of the infection after antiviral drug treatment. Our approach provides a comprehensive imaging framework that will enable future investigations of coronavirus fundamental biology and therapeutic effects.

Multi-color super-resolution imaging to study human coronavirus RNA during cellular infection.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the third human coronavirus within 20 years that gave rise to a life-threatening disease and the first to reach pandemic spread. To make therapeutic headway against current and future coronaviruses, the biology of coronavirus RNA during infection must be precisely understood. Here, we present a robust and generalizable framework combining high-throughput confocal and super-resolution microscopy imaging to study coronavirus infection at the nanoscale. Employing the model human coronavirus HCoV-229E, we specifically labeled coronavirus genomic RNA (gRNA) and double-stranded RNA (dsRNA) via multicolor RNA-immunoFISH and visualized their localization patterns within the cell. The exquisite resolution of our approach uncovers a striking spatial organization of gRNA and dsRNA into three distinct structures and enables quantitative characterization of the status of the infection after antiviral drug treatment. Our approach provides a comprehensive framework that supports investigations of coronavirus fundamental biology and therapeutic effects.

Exploring Cell Surface-Nanopillar Interactions with 3D Super-Resolution Microscopy.

Plasma membrane topography has been shown to strongly influence the behavior of many cellular processes such as clathrin-mediated endocytosis, actin rearrangements, and others. Recent studies have used three-dimensional (3D) nanostructures such as nanopillars to imprint well-defined membrane curvatures (the "nano-bio interface"). In these studies, proteins and their interactions were probed by two-dimensional fluorescence microscopy. However, the low resolution and limited axial detail of such methods are not optimal to determine the relative spatial position and distribution of proteins along a 100 nm-diameter object, which is below the optical diffraction limit. Here, we introduce a general method to explore the nanoscale distribution of proteins at the nano-bio interface with 10-20 nm precision using 3D single-molecule super-resolution (SR) localization microscopy. This is achieved by combining a silicone-oil immersion objective and 3D double-helix point spread function microscopy. We carefully adjust the objective to minimize spherical aberrations between quartz nanopillars and the cell. To validate the 3D SR method, we imaged the 3D shape of surface-labeled nanopillars and compared the results with electron microscopy measurements. Turning to transmembrane-anchored labels in cells, the high quality 3D SR reconstructions reveal the membrane tightly wrapping around the nanopillars. Interestingly, the cytoplasmic protein AP-2 involved in clathrin-mediated endocytosis accumulates along the nanopillar above a specific threshold of 1/R (the reciprocal of the radius) membrane curvature. Finally, we observe that AP-2 and actin preferentially accumulate at positive Gaussian curvature near the pillar caps. Our results establish a general method to investigate the nanoscale distribution of proteins at the nano-bio interface using 3D SR microscopy.

Deep learning in single-molecule microscopy: fundamentals, caveats, and recent developments [Invited].

Deep learning-based data analysis methods have gained considerable attention in all fields of science over the last decade. In recent years, this trend has reached the single-molecule community. In this review, we will survey significant contributions of the application of deep learning in single-molecule imaging experiments. Additionally, we will describe the historical events that led to the development of modern deep learning methods, summarize the fundamental concepts of deep learning, and highlight the importance of proper data composition for accurate, unbiased results.

Accurate and rapid background estimation in single-molecule localization microscopy using the deep neural network BGnet.

Background fluorescence, especially when it exhibits undesired spatial features, is a primary factor for reduced image quality in optical microscopy. Structured background is particularly detrimental when analyzing single-molecule images for 3-dimensional localization microscopy or single-molecule tracking. Here, we introduce BGnet, a deep neural network with a U-net-type architecture, as a general method to rapidly estimate the background underlying the image of a point source with excellent accuracy, even when point-spread function (PSF) engineering is in use to create complex PSF shapes. We trained BGnet to extract the background from images of various PSFs and show that the identification is accurate for a wide range of different interfering background structures constructed from many spatial frequencies. Furthermore, we demonstrate that the obtained background-corrected PSF images, for both simulated and experimental data, lead to a substantial improvement in localization precision. Finally, we verify that structured background estimation with BGnet results in higher quality of superresolution reconstructions of biological structures.

Quantitative Super-Resolution Microscopy of the Mammalian Glycocalyx.

The mammalian glycocalyx is a heavily glycosylated extramembrane compartment found on nearly every cell. Despite its relevance in both health and disease, studies of the glycocalyx remain hampered by a paucity of methods to spatially classify its components. We combine metabolic labeling, bioorthogonal chemistry, and super-resolution localization microscopy to image two constituents of cell-surface glycans, N-acetylgalactosamine (GalNAc) and sialic acid, with 10-20 nm precision in 2D and 3D. This approach enables two measurements: glycocalyx height and the distribution of individual sugars distal from the membrane. These measurements show that the glycocalyx exhibits nanoscale organization on both cell lines and primary human tumor cells. Additionally, we observe enhanced glycocalyx height in response to epithelial-to-mesenchymal transition and to oncogenic KRAS activation. In the latter case, we trace increased height to an effector gene, GALNT7. These data highlight the power of advanced imaging methods to provide molecular and functional insights into glycocalyx biology.

Refining Disordered Peptide Ensembles with Computational Amide I Spectroscopy: Application to Elastin-Like Peptides.

The characterization of intrinsically disordered protein (IDP) ensembles is complicated both by inherent heterogeneity and by the fact that many common experimental techniques function poorly when applied to IDPs. For this reason, the development of alternative structural tools for probing IDP ensembles has attracted considerable attention. Here we describe our recent work in developing experimental and computational tools for characterizing IDP ensembles using Amide I (backbone carbonyl stretch) vibrational spectroscopy. In this approach, the infrared (IR) absorption frequencies of isotope-labeled amide bonds probe their local electrostatic environments and structures. Empirical frequency maps allow us to use this spectroscopic data as a direct experimental test of atomistic structural models. We apply these methods to a family of short elastin-like peptides (ELPs), fragments of the elastin protein based around the Pro-Gly turn motif characteristic of the elastomeric segments of the full protein. Using a maximum entropy analysis that applies constraints from experimental spectra to weighting predicted spectra from molecular dynamics (MD) ensembles, we find that peptides with Ala or Val side chains preceding the Pro-Gly turn unit exhibit a stronger tendency toward extended structures than do Gly-Pro-Gly motifs, suggesting an important role for steric interactions in tuning the molecular properties of elastin.

Isotope-enriched protein standards for computational amide I spectroscopy.

We present a systematic isotope labeling study of the protein G mutant NuG2b as a step toward the production of reliable, structurally stable, experimental standards for amide I infrared spectroscopic simulations. By introducing isotope enriched amino acids into a minimal growth medium during bacterial expression, we induce uniform labeling of the amide bonds following specific amino acids, avoiding the need for chemical peptide synthesis. We use experimental data to test several common amide I frequency maps and explore the influence of various factors on map performance. Comparison of the predicted absorption frequencies for the four maps tested with empirical assignments to our experimental spectra yields a root-mean-square error of 6-12 cm(-1), with outliers of at least 12 cm(-1) in all models. This means that the predictions may be useful for predicting general trends such as changes in hydrogen bonding configuration; however, for finer structural constraints or absolute frequency assignments, the models are unreliable. The results indicate the need for careful testing of existing literature maps and shed light on possible next steps for the development of quantitative spectral maps.