High-resolution Structures Research Articles

Mechanism of negative μ-opioid receptor modulation by sodium ions.

Negative allosteric modulation of G-protein coupled receptors (GPCRs) by Na+ ions was first described in the 1970s for opioid receptors (ORs) and has subsequently been detected for most class A GPCRs. In high-resolution structures of inactive-state class A GPCRs, a Na+ ion binds to a conserved pocket near residue D2.50, whereas active-state structures of GPCRs are incompatible with Na+ binding. Correspondingly, Na+ diminishes agonist affinity, stabilizes the receptors in the inactive state, and reduces basal signaling. We applied a mutual-information based analysis to μs-timescale biomolecular simulations of the μ-opioid receptor (μ-OR). Our results reveal that Na+ binding is coupled to a water wire linking the Na+ binding site with the agonist binding pocket and to rearrangements in polar networks propagating conformational changes to the agonist and G-protein binding sites. These findings provide a new mechanistic link between the presence of the ion, altered agonist affinity, receptor deactivation, and lowered basal signaling levels.

High-resolution cryo-EM analysis of a Streptococcus pyogenes M-protein/human plasminogen complex.

The importance of human plasminogen (hPg)/plasmin (hPm)/cell receptor complexes in invasiveness of cells has been amply established. The objective of this investigation was to determine a high-resolution structure of a major Group A Streptococcus (GAS) bacterial receptor (PAM) for hPg/hPm when bound on a cell surface to its major ligand, hPg. As a model cell surface with endogenous PAM, we employed engineered PAM-expressing lentivirus (LV) particles. We show that the ectodomain of a PAM-type M-Protein (M-Prt), in complex with hPg, is folded through distinct intra- and inter-domain interactions to a more compact form on the cell surface, thus establishing a new paradigm for membrane-bound M-Prt/ligand structures. These studies provide a framework for addressing the need for treatments of GAS disease by providing a molecular platform to solve structures of virulence-determining membrane proteins.

Chitin Translocation Is Functionally Coupled with Synthesis in Chitin Synthase.

Chitin, an extracellular polysaccharide, is synthesized by membrane-embedded chitin synthase (CHS) utilizing intracellular substrates. The mechanism of the translocation of synthesized chitin across the membrane to extracellular locations remains unresolved. We prove that the chitin synthase from Phytophthora sojae (PsCHS) is a processive glycosyltransferase, which can rapidly produce and tightly bind with the highly polymerized chitin. We further demonstrate that PsCHS is a bifunctional enzyme, which is necessary and sufficient to translocate the synthesized chitin. PsCHS was purified and then reconstituted into proteoliposomes (PLs). The nascent chitin is generated and protected from chitinase degradation unless detergent solubilizes the PLs, showing that PsCHS translocates the newly produced chitin into the lumen of the PLs. We also attempted to resolve the PsCHS structure of the synthesized chitin-bound state, although it was not successful; the obtained high-resolution structure of the UDP/Mn2+-bound state could still assist in describing the characterization of the PsCHS's transmembrane channel. Consistently, we demonstrate that PsCHS is indispensable and capable of translocating chitin in a process that is tightly coupled to chitin synthesis.

DeepVID v2: self-supervised denoising with decoupled spatiotemporal enhancement for low-photon voltage imaging.

Voltage imaging is a powerful tool for studying the dynamics of neuronal activities in the brain. However, voltage imaging data are fundamentally corrupted by severe Poisson noise in the low-photon regime, which hinders the accurate extraction of neuronal activities. Self-supervised deep learning denoising methods have shown great potential in addressing the challenges in low-photon voltage imaging without the need for ground-truth but usually suffer from the trade-off between spatial and temporal performances. We present DeepVID v2, a self-supervised denoising framework with decoupled spatial and temporal enhancement capability to significantly augment low-photon voltage imaging. DeepVID v2 is built on our original DeepVID framework, which performs frame-based denoising by utilizing a sequence of frames around the central frame targeted for denoising to leverage temporal information and ensure consistency. Similar to DeepVID, the network further integrates multiple blind pixels in the central frame to enrich the learning of local spatial information. In addition, DeepVID v2 introduces a new spatial prior extraction branch to capture fine structural details to learn high spatial resolution information. Two variants of DeepVID v2 are introduced to meet specific denoising needs: an online version tailored for real-time inference with a limited number of frames and an offline version designed to leverage the full dataset, achieving optimal temporal and spatial performances. We demonstrate that DeepVID v2 is able to overcome the trade-off between spatial and temporal performances and achieve superior denoising capability in resolving both high-resolution spatial structures and rapid temporal neuronal activities. We further show that DeepVID v2 can generalize to different imaging conditions, including time-series measurements with various signal-to-noise ratios and extreme low-photon conditions. Our results underscore DeepVID v2 as a promising tool for enhancing voltage imaging. This framework has the potential to generalize to other low-photon imaging modalities and greatly facilitate the study of neuronal activities in the brain.

A novel N4,N4-dimethylcytidine in the archaeal ribosome enhances hyperthermophily

Ribosome structure and activity are challenged at high temperatures, often demanding modifications to ribosomal RNAs (rRNAs) to retain translation fidelity. LC-MS/MS, bisulfite-sequencing, and high-resolution cryo-EM structures of the archaeal ribosome identified an RNA modification, N4,N4-dimethylcytidine (m42C), at the universally conserved C918 in the 16S rRNA helix 31 loop. Here, we characterize and structurally resolve a class of RNA methyltransferase that generates m42C whose function is critical for hyperthermophilic growth. m42C is synthesized by the activity of a unique family of RNA methyltransferase containing a Rossman-fold that targets only intact ribosomes. The phylogenetic distribution of the newly identified m42C synthase family implies that m42C is biologically relevant in each domain. Resistance of m42C to bisulfite-driven deamination suggests that efforts to capture m5C profiles via bisulfite sequencing are also capturing m42C.

Waterless structures in the Protein Data Bank.

The absence of solvent molecules in high-resolution protein crystal structure models deposited in the Protein Data Bank (PDB) contradicts the fact that, for proteins crystallized from aqueous media, water molecules are always expected to bind to the protein surface, as well as to some sites in the protein interior. An analysis of the contents of the PDB indicated that the expected ratio of the number of water molecules to the number of amino-acid residues exceeds 1.5 inatomic resolution structures, decreasing to 0.25 at around 2.5 Å resolution. Nevertheless, almost 800 protein crystal structures determined at a resolution of 2.5 Å or higher are found in the current release of the PDB without any water molecules, whereas some other depositions have unusually low or high occupancies of modeled solvent. Detailed analysis of these depositions revealed that the lack of solvent molecules might be an indication of problems with either the diffraction data, the refinement protocol, the deposition process or a combination of these factors. It is postulated that problems with solvent structure should be flagged by the PDB and addressed by the depositors.

Structure of G protein-coupled receptor GPR1 bound to full-length chemerin adipokine reveals a chemokine-like reverse binding mode.

Chemerin is an adipokine with chemotactic activity to a subset of leukocytes. Chemerin binds to 3 G protein-coupled receptors, including chemokine-like receptor 1 (CMKLR1), G protein-coupled receptor 1 (GPR1), and C-C chemokine receptor-like 2 (CCRL2). Here, we report that GPR1 is capable of Gi signaling when stimulated with full-length chemerin or its C-terminal nonapeptide (C9, YFPGQFAFS). We present high-resolution cryo-EM structures of Gi-coupled GPR1 bound to full-length chemerin and to the C9 peptide, respectively. C9 insertion into the transmembrane (TM) binding pocket is both necessary and sufficient for GPR1 signaling, whereas the full-length chemerin uses its bulky N-terminal core for interaction with a β-strand located at the N-terminus of GPR1. This interaction involves multiple β-strands of full-length chemerin, forming a β-sheet that serves as a "lid" for the TM binding pocket and is energetically expensive to remove as indicated by molecular dynamics simulations with free energy landscape analysis. Combining results from functional assays, our structural model explains why C9 is an activating peptide at GPR1 and how the full-length chemerin uses a "two-site" model for enhanced interaction with GPR1.

Structural Basis for Chemerin Recognition and Signaling Through Its Receptors

Chemerin is a chemotactic adipokine that participates in a multitude of physiological processes, including adipogenesis, leukocyte chemotaxis, and neuroinflammation. Chemerin exerts biological functions through binding to one or more of its G protein-coupled receptors (GPCRs), namely chemokine-like receptor 1 (CMKLR1), G protein-coupled receptor 1 (GPR1), and CC-motif receptor-like 2 (CCRL2). Of these receptors, CMKLR1 and GPR1 have been confirmed as signaling receptors of chemerin, whereas CCRL2 serves as a chemerin-binding protein without transmembrane signaling. High-resolution structures of two chemerin receptors are now available thanks to recent advancements in structure biology. This review focuses on the structural perspectives of the chemerin receptors with an emphasis on the structure–activity correlation, including key components of the two receptors for ligand recognition and conformational changes induced by chemerin and its derivative peptides for G protein activation. There are also comparisons between the two chemerin receptors and selected GPCRs with peptide ligands for better appreciation of the shared and distinct features of the chemerin receptors in ligand recognition and transmembrane signaling, and in the evolution of this subclass of GPCRs.

Scalable protein design using optimization in a relaxed sequence space.

Machine learning (ML)-based design approaches have advanced the field of de novo protein design, with diffusion-based generative methods increasingly dominating protein design pipelines. Here, we report a "hallucination"-based protein design approach that functions in relaxed sequence space, enabling the efficient design of high-quality protein backbones over multiple scales and with broad scope of application without the need for any form of retraining. We experimentally produced and characterized more than 100 proteins. Three high-resolution crystal structures and two cryo-electron microscopy density maps of designed single-chain proteins comprising up to 1000 amino acids validate the accuracy of the method. Our pipeline can also be used to design synthetic protein-protein interactions, as validated experimentally by a set of protein heterodimers. Relaxed sequence optimization offers attractive performance with respect to designability, scope of applicability for different design problems, and scalability across protein sizes.

Integrating Multimodal Deep Learning with Multipoint Statistics for 3D Crustal Modeling: A Case Study of the South China Sea

Constructing an accurate three-dimensional (3D) geological model is crucial for advancing our understanding of subsurface structures and their evolution, particularly in complex regions such as the South China Sea (SCS). This study introduces a novel approach that integrates multimodal deep learning with multipoint statistics (MPS) to develop a high-resolution 3D crustal P-wave velocity structure model of the SCS. Our method addresses the limitations of traditional algorithms in capturing non-stationary geological features and effectively incorporates heterogeneous data from multiple geophysical sources, including 44 wide-angle seismic crustal structure profiles obtained by ocean bottom seismometers (OBSs), gravity anomalies, magnetic anomalies, and topographic data. The proposed model is rigorously validated against existing methods such as Kriging interpolation and MPS alone, demonstrating superior performance in reconstructing both global and local spatial features of the crustal structure. The integration of diverse datasets significantly enhances the model’s accuracy, reducing errors and improving the alignment with known geological information. The resulting 3D model provides a detailed and reliable representation of the SCS crust, offering critical insights for studies on tectonic evolution, resource exploration, and geodynamic processes. This work highlights the potential of combining deep learning with geostatistical methods for geological modeling, providing a robust framework for future applications in geosciences. The flexibility of our approach also suggests its applicability to other regions and geological attributes, paving the way for more comprehensive and data-driven investigations of Earth’s subsurface.

Modulation of ABCG2 Transporter Activity by Ko143 Derivatives.

ABCG2 is a multidrug transporter that protects tissues from xenobiotics, affects drug pharmacokinetics, and contributes to multidrug resistance of cancer cells. Here, we present tetracyclic fumitremorgin C analog Ko143 derivatives, evaluate their in vitro modulation of purified ABCG2, and report four high-resolution cryo-EM structures and computational analyses to elucidate their interactions with ABCG2. We found that Ko143 derivatives that are based on a ring-opened scaffold no longer inhibit ABCG2-mediated transport activity. In contrast, closed-ring, tetracyclic analogs were highly potent inhibitors. Strikingly, the least potent of these compounds, MZ82, bound deeper into the central ABCG2 cavity than the other inhibitors and it led to partial closure of the transmembrane domains and increased flexibility of the nucleotide-binding domains. Minor structural modifications can thus convert a potent inhibitor into a compound that induces conformational changes in ABCG2 similar to those observed during binding of a substrate. Molecular dynamics simulations and free energy binding calculations further supported the correlation between reduced potency and distinct binding pose of the compounds. We introduce the highly potent inhibitor AZ99 that may exhibit improved in vivo stability.

Crystal structures of Aspergillus oryzae exo-β-(1,3)-glucanase reveal insights into oligosaccharide binding, recognition, and hydrolysis.

Exo-β-(1,3)-glucanases are promising enzymes for use in the biofuel industry as they hydrolyse sugars such as laminarin, a major constituent of the algal cell wall. This study reports structural and biochemical characterizations of Aspergillus oryzae exo-β-(1,3)-glucanase (AoBgl) belonging to the GH5 family. Purified AoBgl hydrolyses β-(1,3)-glycosidic linkages of the oligosaccharide laminaritriose and the polysaccharide laminarin effectively. We have determined three high-resolution structures of AoBgl: (a) the apo form at 1.75 Å, (b) the complexed form with bound cellobiose at 1.73 Å and (c) the glucose-bound form at 1.20 Å. The crystal structures, molecular dynamics simulation studies and site-directed mutagenesis reveal the mode of substrate binding and interactions at the active site. The results also indicate that AoBgl effectively hydrolyses trisaccharides and higher oligosaccharides. The findings from our structural and biochemical studies would aid in rational engineering efforts to generate superior AoBgl variants and similar GH5 enzymes for their industrial use.

Long-Time Scale Simulations Reveal Key Dynamics That Drive the Onset of the N State in the Proteorhodopsin Photocycle.

Proteorhodopsin (PR) is a microbial proton pump that plays a significant role in phototrophy of bacteria in marine environments. Fundamental understanding of the structure-function relationship that drives proton pumping in PR has largely been elusive due to a lack of high-resolution structures of the photointermediates in the PR photocycle. Extending upon previous work, we used long-time scale molecular dynamics (MD) simulations to characterize the M state of the blue variant of PR, which represents the first proton transfer that takes place in the photocycle. Several notable structural changes occur in the M state that are hallmarks of subsequent steps in the PR photocycle, indicating that although this protein is often compared to the canonical microbial rhodopsins, such as bacteriorhodopsin, PR possesses characteristics that make it distinct among the rapidly increasing and widely variable catalog of microbial rhodopsins.

Near-atomic structures of RHDV reveal insights into capsid assembly and different conformations between mature virion and VLP.

Rabbit hemorrhagic disease virus (RHDV) poses a significant threat to rabbits, causing substantial economic losses in rabbit farming. The virus also endangers wild populations of rabbit species and the predatory animals that rely on rabbits as a food source, thereby disturbing the ecological balance. However, the structural understanding of RHDV has been limited due to the lack of high-resolution structures. Here, we present the first high-resolution cryo-EM structures of the mature virion and virus-like particles (VLPs) derived from both full-length and N-terminal arm (NTA)-truncated VP60. These structures reveal intricate structural details of the icosahedral capsid and crucial NTA-mediated interactions essential for capsid assembly. In addition, dramatic conformational differences are unexpectedly observed between the mature virion and VLP. The protruding spikes of the A-B dimers adopt a "raised" state in the mature virion and a "resting" state in the VLP. These findings enhance our understanding of the structure, assembly, and conformational dynamics of the RHDV capsid, laying the essential groundwork for further virological research and therapeutic advancements.IMPORTANCERHDV is a pathogen with significant economic and ecological impact. By presenting the first high-resolution cryo-EM structures of RHDV, we have uncovered detailed interactions among neighboring VP60 subunits of the icosahedral capsid. The NTA of VP60 is uniquely clustered around the threefold axis of the capsid, probably play a critical role in dragging the six VP60 dimers around the threefold axis during capsid assembly. Additionally, we observed dramatic conformational differences between the mature virion and VLPs. VLPs are commonly used for vaccine development, under the assumption that their structure closely resembles that of the mature virion. Our findings significantly advance the understanding of the RHDV capsid structure, which may be used for developing potential therapeutic strategies against RHDV.

The Bor1 elevator transport cycle is subject to autoinhibition and activation

Boron, essential for plant growth, necessitates precise regulation due to its potential toxicity. This regulation is achieved by borate transporters (BORs), which are homologous to the SLC4 family. The Arabidopsis thaliana Bor1 (AtBor1) transporter from clade I undergoes slow regulation through degradation and translational suppression, but its potential for fast regulation via direct activity modulation was unclear. Here, we combine cryo-electron microscopy, mutagenesis, and functional characterization to study AtBor1, revealing high-resolution structures of the dimer in one inactive and three active states. Our findings show that AtBor1 is regulated by two distinct mechanisms: an autoinhibitory domain at the carboxyl terminus obstructs the substrate pathway via conserved salt bridges, and phosphorylation of Thr410 allows interaction with a positively charged pocket at the cytosolic face, essential for borate transport. These results elucidate the molecular basis of AtBor1’s activity regulation and highlight its role in fast boron level regulation in plants.

Control of 3' splice site selection by the yeast splicing factor Fyv6.

Pre-mRNA splicing is catalyzed in two steps: 5' splice site (SS) cleavage and exon ligation. A number of proteins transiently associate with spliceosomes to specifically impact these steps (1st and 2nd step factors). We recently identified Fyv6 (FAM192A in humans) as a 2nd step factor in S. cerevisiae; however, we did not determine how widespread Fyv6's impact is on the transcriptome. To answer this question, we have used RNA-seq to analyze changes in splicing. These results show that loss of Fyv6 results in activation of non-consensus, branch point (BP) proximal 3' SS transcriptome-wide. To identify the molecular basis of these observations, we determined a high-resolution cryo-EM structure of a yeast product complex spliceosome containing Fyv6 at 2.3 Å. The structure reveals that Fyv6 is the only 2nd step factor that contacts the Prp22 ATPase and that Fyv6 binding is mutually exclusive with that of the 1st step factor Yju2. We then use this structure to dissect Fyv6 functional domains and interpret results of a genetic screen for fyv6Δ suppressor mutations. The combined transcriptomic, structural, and genetic studies allow us to propose a model in which Yju2/Fyv6 exchange facilitates exon ligation and Fyv6 promotes usage of consensus, BP distal 3' SS.

The role of synaptic protein NSF in the development and progression of neurological diseases.

This document provides a comprehensive examination of the pivotal function of the N-ethylmaleimide-sensitive factor (NSF) protein in synaptic function. The NSF protein directly participates in critical biological processes, including the cyclic movement of synaptic vesicles (SVs) between exocytosis and endocytosis, the release and transmission of neurotransmitters, and the development of synaptic plasticity through interactions with various proteins, such as SNARE proteins and neurotransmitter receptors. This review also described the multiple functions of NSF in intracellular membrane fusion events and its close associations with several neurological disorders, such as Parkinson's disease, Alzheimer's disease, and epilepsy. Subsequent studies should concentrate on determining high-resolution structures of NSF in different domains, identifying its specific alterations in various diseases, and screening small molecule regulators of NSF from multiple perspectives. These research endeavors aim to reveal new therapeutic targets associated with the biological functions of NSF and disease mechanisms.

Prediction of conformational states in a coronavirus channel using Alphafold-2 and DeepMSA2: Strengths and limitations

The envelope (E) protein is present in all coronavirus genera. This protein can form pentameric oligomers with ion channel activity which have been proposed as a possible therapeutic target. However, high resolution structures of E channels are limited to those of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), responsible for the recent COVID-19 pandemic. In the present work, we used Alphafold-2 (AF2), in ColabFold without templates, to predict the transmembrane domain (TMD) structure of six E-channels representative of genera alpha-, beta- and gamma-coronaviruses in the Coronaviridae family. High-confidence models were produced in all cases when combining multiple sequence alignments (MSAs) obtained from DeepMSA2. Overall, AF2 predicted at least two possible orientations of the α-helices in E-TMD channels: one where a conserved polar residue (Asn-15 in the SARS sequence) is oriented towards the center of the channel, ‘polar-in’, and one where this residue is in an interhelical orientation ‘polar-inter’. For the SARS models, the comparison with the two experimental models ‘closed’ (PDB: 7K3G) and ‘open’ (PDB: 8SUZ) is described, and suggests a ∼60˚ α-helix rotation mechanism involving either the full TMD or only its N-terminal half, to allow the passage of ions. While the results obtained are not identical to the two high resolution models available, they suggest various conformational states with striking similarities to those models. We believe these results can be further optimized by means of MSA subsampling, and guide future high resolution structural studies in these and other viral channels.

Design of a Cereblon construct for crystallographic and biophysical studies of protein degraders

The ubiquitin E3 ligase cereblon (CRBN) is the target of therapeutic drugs thalidomide and lenalidomide and is recruited by most targeted protein degraders (PROTACs and molecular glues) in clinical development. Biophysical and structural investigation of CRBN has been limited by current constructs that either require co-expression with the adaptor DDB1 or inadequately represent full-length protein, with high-resolution structures of degrader ternary complexes remaining rare. We present the design of CRBNmidi, a construct that readily expresses from E. coli with high yields as soluble, stable protein without DDB1. We benchmark CRBNmidi for wild-type functionality through a suite of biophysical techniques and solve high-resolution co-crystal structures of its binary and ternary complexes with degraders. We qualify CRBNmidi as an enabling tool to accelerate structure-based discovery of the next generation of CRBN based therapeutics.

Structural topology optimization based on deep learning

In the mechanical design of structures, traditional topology optimization methods involve numerous finite element iterative analyses, leading to a significant expenditure of computational resources. Therefore, the improved multi-scale gradient generative adversarial networks topology optimization technique is proposed. The topology optimization condition parameters are compressed into a low-dimensional latent space feature representation using the encoder, allowing the model to better extract features from these parameters. To speed up model training, the generator and discriminator networks use lightweight residual convolutional blocks. The hybrid attention mechanism extracts prominent region features from the topology optimization structure map. The model training process is guided by a multi-dimensional fusion loss function to enhance the quality of generated model samples. Finally, transferring the parameters of the low-resolution topology optimization model to the high-resolution model enables complete training on a limited amount of high-resolution topology optimization datasets. The experimental data on the low- and high-resolution topology optimization datasets demonstrate that, when compared to alternative methods, this method produces better-quality topology optimization structure maps. Additionally, it can generate high-resolution topology optimization structure maps in minimal time, enabling real-time topology optimization.