Browse Articles

Enhanced visualization of influenza A virus entry into living cells using virus-view atomic force microscopy

Influenza A virus (IAV) entry into host cells begins with interactions between the viral envelope proteins hemagglutinin (HA)/neuraminidase (NA) and sialic acid moieties on the cell plasma membrane. These interactions drive IAV’s lateral diffusion along the cell membrane and trigger membrane morphological changes required for endocytosis. However, directly visualizing these dynamic processes, which are crucial for IAV entry, has been challenging using conventional microscopy techniques. In this study, we enabled live-cell observation of nanoscale morphological dynamics of IAV and the cell membrane by reducing the mechanical invasiveness of atomic force microscopy (AFM). A customised cantilever with less than half the spring constant of conventional cantilevers enabled virus-view AFM imaging that preserved IAV–membrane interactions. By combining virus-view AFM with confocal microscopy, we performed correlative morphological and fluorescence observations of IAV lateral diffusion and endocytosis in living cells. Variations in diffusion coefficients of single virions suggested heterogeneity in sialic acid density on the cell membrane. NA inhibition decreased diffusion coefficients, while reduced sialic acid density increased them. The timing of clathrin accumulation at virion binding sites coincided with a decrease in diffusion coefficients, a relationship that was maintained independent of NA activity or sialic acid density. As clathrin assembly progressed, ~100-nm-high membrane bulges emerged adjacent to the virus, culminating in the complete membrane envelopment of the virus at peak clathrin accumulation. Our virus-view AFM will deepen our understanding of various virus–cell interactions, facilitate the evaluation of drug effects and promote future translational research.

Protein-mediated stabilization and nicking of the nontemplate DNA strand dramatically affect R-loop formation in vitro

R-loops are an important class of non-B DNA structures that form co-transcriptionally. Using in vitro transcription and unbiased quantitative sequencing readouts, we show that the addition of single-strand DNA binding proteins co-transcriptionally can drive a 3- to 5-fold increase of R-loop frequency without significant changes to R-loop distribution. We propose that this is caused by stabilizing and preventing the collapse of short nascent R-loops. This suggests that R-loop formation is highly dynamic and highlights single strand binding proteins as players in cellular R-loop regulation. We further show that nontemplate strand DNA nicks are powerful initiators of R-loop formation, increasing R-loop frequencies by up to two orders of magnitude. Atomic force microscopy revealed that the nontemplate strand in nick-initiated structures is often flayed away from the RNA:DNA hybrid and engaged in self-pairing, creating unique forked R-loop features. DNA nicks, one of the most frequent DNA lesions in cells, are therefore potential hotspots for opportunistic R-loop initiation and may cause the formation of a distinct class of R-loops. Overall, this work highlights the importance of the displaced single-strand on R-loop initiation and dynamics.

Virion proteomics of genetically intact HCMV reveals a regulator of envelope glycoprotein composition that protects against humoral immunity

Human cytomegalovirus (HCMV) is a clinically important herpesvirus that has coevolved for millions of years with its human host, and establishes lifelong persistent infection. A substantial proportion of its 235 kb genome is dedicated to manipulating host immunity through targeting antiviral host proteins for degradation or relocalization. Quantitative proteomics of the infected cell has extensively characterized these processes, but the cell-free virion has been less well studied. We therefore carried out proteomic analysis of a clinical HCMV strain (Merlin) virion. This revealed 18 novel components, including the viral protein gpUL141, which is recognized as an NK immune-evasin that targets several host proteins (CD155, CD112, and TRAILR) when expressed within the cell. Coimmunoprecipitation of gpUL141 from virions identified interactions with viral entry glycoproteins from the trimer (gH/gL/gO), pentamer (gH/gL/UL128/UL130/UL131A), and gH/gpUL116 complexes, as well as gB. Only interactions with gH/gB occurred in the absence of other viral proteins. Analysis supported a model in which gpUL141 homodimers independently interacted with separate gB/gH-containing complexes. gpUL141 encodes an ER retention domain that restricts trafficking through the ER/Golgi, and limited the transport of glycoprotein complexes bound by gpUL141. As a result, gpUL141 reduced levels of multiple glycoprotein complexes on the infected cell surface as well as in the virion. This reduced syncytium formation, inhibited antibody-dependent cellular cytotoxicity (ADCC), and reduced susceptibility to neutralizing antibodies. Thus, gpUL141 represents an immune-evasin that not only targets host proteins to limit NK-cell attack, but also alters the trafficking of multiple viral glycoprotein complexes in order to evade humoral immunity.

Inverse design of parameter-controlled disclination paths

Topological defects, such as disclination lines in nematic liquid crystals, are fundamental to many physical systems and applications. In this work, we study the behavior of nematic disclinations in thin parallel-plate geometries with strong patterned planar anchoring. Building on prior models, we solve both the forward problem–predicting disclination trajectories from given surface patterns–and an extended inverse problem–designing surface patterns to produce a tunable family of disclination curves under varying system parameters. We present an explicit calculation for pattern construction, analyze parameter limitations and stability constraints, and highlight experimental and technological applications.

Large-scale surveys of woodrats ( <i>Neotoma</i> spp.) reveal constraints on diet breadth in herbivorous mammals

Characterizing niche space is critical for predicting species interactions and responses to environmental change. To enhance our understanding of dietary niche breadth, we used DNA metabarcoding to examine how diets of a widespread, model herbivore (woodrats, genus Neotoma ) respond to changing resources and the extent to which dietary specialization is conserved across space and time. We used diet data from 13 species, 57 populations, and over 500 individuals to examine predictors of niche breadth and interindividual diet variation at a landscape scale. Then, to test whether these patterns are conserved across scales, we explored the same questions using a single population sampled over 5 y and mark-recapture data from individuals sampled at least three times. We found that woodrats exhibited a continuum of dietary specialization that included species-level dietary generalists and specialists. Specialist species maintained narrow population-level niche breadths with little evidence of interindividual diet variation. In contrast, generalist species consisted of populations with varying degrees of dietary specialization and interindividual diet variation. Across sampling scales, increased population-level niche breadth was explained by both increased interindividual diet variation and increased diet richness. These results are consistent with the Niche Variation Hypothesis and suggest that diet breadth is constrained by costs of both specialization and generalization.

C4d, a high-affinity LilrB2 ligand, is elevated in Alzheimer’s disease and mediates synapse pruning

Synapse pruning sculpts neural circuits throughout life. The human Leukocyte immunoglobulin-like receptor type B2 (LilrB2)/murine Paired immunoglobulin receptor B (PirB) receptors expressed in neurons and complement protein C4 have been separately implicated in pruning. Here, we report that C4d, a C4 cleavage product with unknown function, binds LilrB2/PirB with nanomolar affinity. C4d and LilrB2 colocalize at excitatory synapses in the human cerebral cortex as well as with beta amyloid in Alzheimer’s disease (AD). C4d, as well as C4, increase with age and more so in AD. To examine whether C4d-PirB interactions can drive pruning, dendritic spines—the postsynaptic structure of excitatory synapses—were monitored on L5 pyramidal neurons in the mouse cerebral cortex: A significant decrease in dendritic spine density occurred in WT with C4d exposure, but KO of PirB completely prevented this loss. Together, our findings reveal an unexpected physiological role for C4d in pruning and imply that different complement cascade components may collaborate to engage both neuronal and glial-specific effectors of synaptic pruning.

Dimerization propensity of the β ₁ -adrenergic receptor in lipid nanodiscs probed by DEER and single-molecule spectroscopies

G protein–coupled receptors (GPCRs) comprise a large class of membrane proteins that mediate cellular responses to a wide range of external signals and as such constitute major drug targets. While oligomerization has been shown to play a well-established role in modulating signaling for class C GPCRs (e.g., the glutamate and GABA receptors), the functional relevance of oligomerization for class A receptors, such as the β 1 -adrenergic receptor (β 1 AR), remains unclear. Here, we have examined the influence of the membrane mimetic environment on the dimerization propensity of β 1 AR using a combination of pulsed Q-band double electron–electron resonance spectroscopy and single-molecule fluorescence brightness measurements in an Anti-Brownian Elektrokinetic trap. While β 1 AR is predominantly monomeric in docecyl-β-D-maltoside (DDM) micelles, reconstitution of β 1 AR in lipid nanodiscs preferentially favors symmetric parallel dimers. Using nanodiscs of different diameters we observed a clear size-dependent increase in the dimer fraction, reaching over 50% of the β 1 AR molecules in large (~12.5 nm diameter) nanodiscs. Addition of cholesteryl hemisuccinate, an analog of cholesterol, suppresses β 1 AR dimerization in lipid nanodiscs, recapitulating the behavior in DDM micelles. This work provides quantitative evidence that β 1 AR possesses an intrinsic, membrane sensitive predisposition for dimerization, and highlights the importance of spatial membrane constraints in the modulation of class A GPCR dimerization.

Nanoscale restructuring of the immune synapse with an engager enhances NK cell function

Engagers are antibody-based therapies which bind immune cell receptors and a target cell ligand. Next-gen engagers typically bind two activating receptors, but the effect of this on immune synapse formation and signaling is unknown. Here, we coligated activating receptors CD16a and NKG2D on natural killer (NK) cells with a CD33-binding anti-acute myeloid leukemia (AML) engager. Superresolution microscopy revealed that coligating CD16a and NKG2D with a single molecule triggered their nanoscale coclustering. This enhanced phosphorylation of CD3ζ, ZAP70, and SLP-76 which augmented secretion of IFN-γ and TNF-α, by NK cells from healthy donors and AML patients. Thus, in addition to connecting immune cells to target cells, the clinical promise of engagers results from their ability to manipulate the nanoscale architecture of the immune synapse.

Nuclear receptor coregulator NRIP1 R448G modulates T cell gut homing to control intestinal inflammation

Nuclear receptors (NRs) are crucial to integrate metabolite sensing and immune responses in the gut. NR-interacting protein 1 (NRIP1) is an important coregulator of various NRs that has been implicated in inflammatory bowel disease risk, but mechanistic details of how NRIP1 controls NR activities mediating immune homeostasis and inflammation remain elusive. We demonstrate that a missense risk variant, NRIP1 R448G, promotes activated CD4 + T cell gut homing and inflammatory cytokine production, ultimately leading to exacerbated intestinal inflammation. Mechanistically, NRIP1 acts as a corepressor in retinoic acid signaling by expression of a gut-homing transcriptional program. Our study reveals the impacts of NRIP1 on CD4 + T cells in immune regulation during intestinal inflammation, providing insights into mechanisms by which an NR coregulator controls immune homeostasis and tissue inflammation.

Community notes reduce engagement with and diffusion of false information online

Social networks scaffold the diffusion of information on social media. Much attention has been given to the spread of true vs. false content on online social platforms, including the structural differences between their diffusion patterns. However, much less is known about how platform interventions on false content alter the engagement with and diffusion of such content. In this work, we estimate the causal effects of Community Notes, a novel fact-checking feature adopted by X (formerly Twitter) to solicit and vet crowd-sourced fact-checking notes for false content. We gather detailed time series data for 40,078 posts for which notes have been proposed and use synthetic control methods to estimate a range of counterfactual outcomes. We find that attaching fact-checking notes significantly reduces the engagement with and diffusion of false content. We estimate that, on average, the notes resulted in reductions of 46.1% in reposts, 44.1% in likes, 21.9% in replies, and 13.5% in views after being attached. Over the posts’ entire lifespans, these reductions amount to 11.6% fewer reposts, 13.3% fewer likes, 6.9% fewer replies, and 5.5% fewer views on average. In reducing reposts, we observe that diffusion cascades for fact-checked content are less deep and less “viral,” but not less broad, than synthetic control estimates for non-fact-checked content with similar reach. This structural difference contrasts notably with differences between false vs. true content diffusion itself, where false information diffuses farther, but with structural patterns that are otherwise indistinguishable from those of true information, conditional on reach.

Lymphatic dysfunction is linked to disease pathogenesis in Duchenne muscular dystrophy animal models

Duchenne muscular dystrophy (DMD) is a severe muscle-wasting disorder characterized by progressive muscle weakness and inflammation caused by mutations in the DMD gene. Chronic inflammation in DMD exacerbates the complications associated with disease progression. Since the lymphatic system plays a crucial role in regulating and resolving inflammation, our primary goal was to investigate whether lymphatics were dysregulated in skeletal muscle of DMD animals. We used the D2.mdx murine and golden retriever muscular dystrophy (GRMD) canine models, as well as mouse and rat lymphatic muscle cells (LMCs) to determine the role of dystrophin in lymphatic structure and function in skeletal muscles. Single-cell RNA sequencing data from control LMCs showed dystrophin expression, and protein results demonstrated that the 427-, 140-, and 71-kDa dystrophin isoforms were detectable in the LMCs from control mice, whereas the 427 kDa isoform was undetectable in LMCs derived from D2.mdx mice. Microlymphangiography and magnetic resonance lymphangiogram results showed a significant decrease in lymph transport in D2. mdx mice and GRMD dogs, respectively. Isolated flank lymphatic vessels from D2. mdx mice exhibited an increase in tonic contraction and a significant decrease in the phasic contractile frequency and amplitude, supporting lymphatic vessel dysfunction. The gene expression profile and immunofluorescence analyses of dystrophic muscle revealed inflammatory lymphangiogenesis in dystrophic muscle. Skeletal muscle tissues that showed improvement in function after adeno-associated virus-microdystrophin treatment also showed significant improvement in inflammatory lymphangiogenesis in GRMD dogs. Thus, these results show a linkage between lymphatic function and DMD pathogenesis that merits further investigation in DMD patients.

ComFB, a widespread family of c-di-NMP receptor proteins

Cyclic dimeric-GMP (c-di-GMP) is a ubiquitous bacterial second messenger that regulates a variety of cellular processes, including motility, biofilm formation, secretion, cell cycle progression, and development, and also contributes to the virulence of many bacterial pathogens. While the genes encoding c-di-GMP cyclases and hydrolases are readily identifiable in microbial genomes, known c-di-GMP receptor domains are quite few, with only PilZ and MshEN broadly distributed across bacterial phyla. Recently, a new c-di-GMP receptor, named CdgR or ComFB, has been identified in cyanobacteria and shown to regulate cell size and natural competence. We demonstrated that CdgR proteins exhibit sequence and structural similarity to the Bacillus subtilis late competence development protein ComFB, a conserved protein of unknown function associated with bacterial competence. This prompted us to hypothesize that ComFB and ComFB-like proteins could also serve as c-di-GMP receptors. Here, we comprehensively investigated the ComFB protein family and demonstrated that ComFB proteins are evolutionarily widespread among bacteria and function as a novel family of c-di-GMP receptors. We showed that ComFB proteins from Gram-positive bacteria ( B. subtilis , Thermoanaerobacter brockii ) and Gram-negative pathogens ( Vibrio cholerae , Treponema denticola ) bind c-di-GMP with high affinity. Several ComFB proteins also bind cyclic di-adenosine monophosphate (c-di-AMP), suggesting that ComFB represents a widely distributed bacterial protein family with dual specificity for c-di-GMP and c-di-AMP. Our physiological studies further showed that ComFB plays vital roles in controlling motility in a c-di-GMP-dependent manner in two phylogenetically distant bacteria, B. subtilis and the gram-negative Shewanella oneidensis , attesting to the biological relevance of ComFB as a c-di-GMP binding protein.

Reprogrammable sequencing for physically intelligent underactuated robots

Programming physical intelligence into mechanisms holds great promise for machines that can accomplish tasks such as navigation of unstructured environments while utilizing a minimal amount of computational resources and electronic components. In this study, we introduce a design approach for physically intelligent underactuated mechanisms capable of autonomously adjusting their motion in response to environmental interactions. Specifically, multistability is harnessed to sequence the motion of different degrees of freedom in a programmed order. A key aspect of this approach is that this order can be passively reprogrammed through mechanical stimuli arising from interactions with the environment. To showcase our approach, we construct a mechanism that passively sorts objects based on their mass and a four-degree-of-freedom robot capable of autonomously moving away from obstacles. Remarkably, these devices operate without relying on traditional computational architectures and utilize only a single linear actuator.

Cryo-EM structure of the prohibitin complex in open conformation

Prohibitin 1 (PHB1) and Prohibitin 2 (PHB2), two conserved prohibitin members, are primarily localized to the mitochondrial inner membrane (MIM) to form a nanoscale macromolecular prohibitin complex. This prohibitin complex can facilitate the spatial organization of proteins and lipids, thus maintaining cellular metabolism and homeostasis, but its architecture remains largely unknown. Here, we report the cryo-EM structure of a prohibitin complex at 2.8 Å resolution, which contains 11 PHB1–PHB2 heterodimers. This complex displays a bell-like cage, consisting of a lid and a wall, which creates an intermembrane space-facing compartment for the MIM. The lid of the cage is stably assembled, and it is responsible for the prohibitin complex formation. In contrast, the wall of the cage is flexible and exhibits lateral openings, providing a channel for intramembrane exchange of proteins and lipids. These findings provide a structural basis for understanding the scaffold role of the prohibitin complex in organizing intramembrane proteins and lipids.

USP14-mediated metabolic competition impairs CD8+ T cell immunosurveillance in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) frequently develops resistance to CD8+ T cell–based immunotherapy, yet the mechanisms driving this immune evasion remain poorly understood. To identify tumor-intrinsic regulators of immunotherapy resistance and explore therapeutic strategies to restore T cell–mediated tumor control, we employed three functional genomics approaches using in vitro and in vivo CRISPR screening. Cancer USP14 was identified as the critical immune evasion driver. USP14-high HCC patients exhibited poorer anti-PD1 antibody therapy responses and reduced CD8+ T cell infiltration. Inhibition of USP14 suppressed HCC cell growth in coculture with activated CD8+ T cells and restored cocultured CD8+ T cell cytotoxicity. In vivo USP14 targeting synergized with anti-PD1 antibody therapy. Mechanistically, USP14 deubiquitinated and stabilized GLUT1 through the removal of Lys-48-linked ubiquitin chains at Lys-245, which enabled HCC cells to outcompete CD8+ T cells for glucose, generating a glucose-deprived tumor microenvironment that suppressed CD8+ T cell function. Our findings show USP14 in cancer has a proimmunoevasive role in CD8+ T cell–based tumor immunity through GLUT1-mediated glucose competition. These findings position USP14 inhibitors as promising adjuvants to enhance immunotherapy efficacy in HCC, providing actionable insights for overcoming resistance.

The balance between microbial arsenic methylation and demethylation in paddy soils underpins global arsenic risk and straighthead disease in rice

Arsenic contamination in rice poses a global challenge to food safety and agricultural productivity, as toxic methylated arsenic species—dimethylarsinic acid (DMA) and its highly toxic derivative, methylated monothioarsenate (DMMTA)—accumulate in rice grains. These arsenic species endanger human health and trigger rice straighthead disease, a crop disorder that drastically reduces yields. However, the microbial ecological processes driving arsenic speciation in paddy soils, and their link to striking geographic disparities in rice arsenic speciation profiles and disease prevalence, remain poorly understood. Here, we integrate soil chronosequences spanning 1 to 2,000 y of rice cultivation, a global metagenomic survey of 801 paddy soils, controlled incubations, and field surveys to demonstrate that the balance between arsenic-methylating and arsenic-demethylating microbes is the key determinant of rice grain arsenic speciation and straighthead disease susceptibility. We show that young and moderate-age paddy soils (&lt;700 y), common in regions such as the Americas and Europe, are enriched in arsenic-methylating bacteria, leading to elevated DMA and DMMTA in soils and rice grains. In contrast, ancient paddies in Southeast Asia harbor robust populations of DMA-demethylating methanogenic archaea that effectively mitigate the buildup of these toxic arsenic species. We identify core microbial taxa whose abundances serve as predictive biomarkers and construct a global risk map linking a high methylator-to-demethylator ratio in soils with increased straighthead disease incidence. These findings advance our understanding of arsenic biogeochemistry in agroecosystems and establish a predictive framework for identifying regions at elevated risk of arsenic-induced crop disorders and food contamination.

Correction for Agip et al., Structures of <i>Chaetomium thermophilum</i> TOM complexes with bound preproteins

Microbial conservation is essential for sustaining ecosystem functions and services

Does AI help humans make better decisions? A statistical evaluation framework for experimental and observational studies

The use of AI, or more generally data-driven algorithms, has become ubiquitous in today’s society. Yet, in many cases and especially when stakes are high, humans still make final decisions. The critical question, therefore, is whether AI helps humans make better decisions compared to a human-alone or AI-alone system. We introduce a methodological framework to answer this question empirically with minimal assumptions. We measure a decision maker’s ability to make correct decisions using standard classification metrics based on the baseline potential outcome. We consider a single-blinded and unconfounded treatment assignment, in which the provision of AI-generated recommendations is assumed to be randomized across cases, conditional on observed covariates, with final decisions made by humans. Under this study design, we show how to compare the performance of three alternative decision-making systems—human-alone, human-with-AI, and AI-alone. Importantly, the AI-alone system encompasses any individualized treatment assignment, including those not used in the original study. We also show when AI recommendations should be provided to a human-decision maker, and when one should follow such recommendations. We apply the proposed methodology to our own randomized controlled trial evaluating a pretrial risk assessment instrument. We find that the risk assessment recommendations do not improve the classification accuracy of a judge’s decision to impose cash bail. Furthermore, replacing a human judge with algorithms—the risk assessment score and a large language model in particular—yields worse classification performance.

Correction for Bender et al., Differential representations of spatial location by aperiodic and alpha oscillatory activity in working memory