Structural dynamics of receptor recognition and pH-induced dissociation of full-length Clostridioides difficile Toxin B.

Clostridioides difficile secretes Toxin B (TcdB) as one of its major virulence factors, which binds to intestinal epithelial and subepithelial receptors, including frizzled proteins and chondroitin sulfate proteoglycan 4 (CSPG4). Here, we present cryo-EM structures of full-length TcdB in complex with the CSPG4 domain 1 fragment (D1401-560) at cytosolic pH and the cysteine-rich domain of frizzled-2 (CRD2) at both cytosolic and acidic pHs. CSPG4 specifically binds to the autoprocessing and delivery domains of TcdB via networks of salt bridges, hydrophobic and aromatic/proline interactions, which are disrupted upon acidification eventually leading to CSPG4 drastically dissociating from TcdB. In contrast, FZD2 moderately dissociates from TcdB under acidic pH, most likely due to its partial unfolding. These results reveal structural dynamics of TcdB during its preentry step upon endosomal acidification, which provide a basis for developing therapeutics against C. difficile infections.

Hierarchical natural move Monte Carlo refines flexible RNA structures into cryo-EM densities.

Ribonucleic acids (RNAs) play essential roles in living cells. Many of them fold into defined three-dimensional (3D) structures to perform functions. Recent advances in single-particle cryo-electron microscopy (cryo-EM) have enabled structure determinations of RNA to atomic resolutions. However, most RNA molecules are structurally flexible, limiting the resolution of their structures solved by cryo-EM. In modeling these molecules, several computational methods are limited by the requirement of massive computational resources and/or the low efficiency in exploring large-scale structural variations. Here we use hierarchical natural move Monte Carlo (HNMMC), which takes advantage of collective motions for groups of nucleic acid residues, to refine RNA structures into their cryo-EM maps, preserving atomic details in the models. After validating the method on a simulated density map of tRNA, we applied it to objectively obtain the model of the folding intermediate for the specificity domain of ribonuclease P from Bacillus subtilis and refine a flexible ribosomal RNA (rRNA) expansion segment from the Mycobacterium tuberculosis (Mtb) ribosome in different conformational states. Finally, we used HNMMC to model atomic details and flexibility for two distinct conformations of the complete genomic RNA (gRNA) inside MS2, a single-stranded RNA virus, revealing multiple pathways for its capsid assembly.

Structures of Qß virions, virus-like particles, and the Qß-MurA complex reveal internal coat proteins and the mechanism of host lysis.

In single-stranded RNA bacteriophages (ssRNA phages) a single copy of the maturation protein binds the genomic RNA (gRNA) and is required for attachment of the phage to the host pilus. For the canonical Allolevivirus Qß the maturation protein, A2, has an additional role as the lysis protein, by its ability to bind and inhibit MurA, which is involved in peptidoglycan biosynthesis. Here, we determined structures of Qß virions, virus-like particles, and the Qß-MurA complex using single-particle cryoelectron microscopy, at 4.7-Å, 3.3-Å, and 6.1-Å resolutions, respectively. We identified the outer surface of the ß-region in A2 as the MurA-binding interface. Moreover, the pattern of MurA mutations that block Qß lysis and the conformational changes of MurA that facilitate A2 binding were found to be due to the intimate fit between A2 and the region encompassing the closed catalytic cleft of substrate-liganded MurA. Additionally, by comparing the Qß virion with Qß virus-like particles that lack a maturation protein, we observed a structural rearrangement in the capsid coat proteins that is required to package the viral gRNA in its dominant conformation. Unexpectedly, we found a coat protein dimer sequestered in the interior of the virion. This coat protein dimer binds to the gRNA and interacts with the buried α-region of A2, suggesting that it is sequestered during the early stage of capsid formation to promote the gRNA condensation required for genome packaging. These internalized coat proteins are the most asymmetrically arranged major capsid proteins yet observed in virus structures.

Structural insights into species-specific features of the ribosome from the human pathogen Mycobacterium tuberculosis.

Ribosomes from Mycobacterium tuberculosis (Mtb) possess species-specific ribosomal RNA (rRNA) expansion segments and ribosomal proteins (rProtein). Here, we present the near-atomic structures of the Mtb 50S ribosomal subunit and the complete Mtb 70S ribosome, solved by cryo-electron microscopy. Upon joining of the large and small ribosomal subunits, a 100-nt long expansion segment of the Mtb 23S rRNA, named H54a or the 'handle', switches interactions from with rRNA helix H68 and rProtein uL2 to with rProtein bS6, forming a new intersubunit bridge 'B9'. In Mtb 70S, bridge B9 is mostly maintained, leading to correlated motions among the handle, the L1 stalk and the small subunit in the rotated and non-rotated states. Two new protein densities were discovered near the decoding center and the peptidyl transferase center, respectively. These results provide a structural basis for studying translation in Mtb as well as developing new tuberculosis drugs.

Structural basis of Acinetobacter type IV pili targeting by an RNA virus.

Acinetobacters pose a significant threat to human health, especially those with weakened immune systems. Type IV pili of acinetobacters play crucial roles in virulence and antibiotic resistance. Single-stranded RNA bacteriophages target the bacterial retractile pili, including type IV. Our study delves into the interaction between Acinetobacter phage AP205 and type IV pili. Using cryo-electron microscopy, we solve structures of the AP205 virion with an asymmetric dimer of maturation proteins, the native Acinetobacter type IV pili bearing a distinct post-translational pilin cleavage, and the pili-bound AP205 showing its maturation proteins adapted to pilin modifications, allowing each phage to bind to one or two pili. Leveraging these results, we develop a 20-kilodalton AP205-derived protein scaffold targeting type IV pili in situ, with potential for research and diagnostics.

Structure of the Schizosaccharomyces pombe Gtr-Lam complex reveals evolutionary divergence of mTORC1-dependent amino acid sensing.

mTORC1 is a protein kinase complex that controls cellular growth in response to nutrient availability. Amino acid signals are transmitted toward mTORC1 via the Rag/Gtr GTPases and their upstream regulators. An important regulator is LAMTOR, which localizes Rag/Gtr on the lysosomal/vacuole membrane. In human cells, LAMTOR consists of five subunits, but in yeast, only three or four. Currently, it is not known how variation of the subunit stoichiometry may affect its structural organization and biochemical properties. Here, we report a 3.1 Å-resolution structural model of the Gtr-Lam complex in Schizosaccharomyces pombe. We found that SpGtr shares conserved architecture as HsRag, but the intersubunit communication that coordinates nucleotide loading on the two subunits differs. In contrast, SpLam contains distinctive structural features, but its GTP-specific GEF activity toward SpGtr is evolutionarily conserved. Our results revealed unique evolutionary paths of the protein components of the mTORC1 pathway.

Structural Assembly of Qß Virion and Its Diverse Forms of Virus-like Particles.

The coat proteins (CPs) of single-stranded RNA bacteriophages (ssRNA phages) directly assemble around the genomic RNA (gRNA) to form a near-icosahedral capsid with a single maturation protein (Mat) that binds the gRNA and interacts with the retractile pilus during infection of the host. Understanding the assembly of ssRNA phages is essential for their use in biotechnology, such as RNA protection and delivery. Here, we present the complete gRNA model of the ssRNA phage Qß, revealing that the 3' untranslated region binds to the Mat and the 4127 nucleotides fold domain-by-domain, and is connected through long-range RNA-RNA interactions, such as kissing loops. Thirty-three operator-like RNA stem-loops are located and primarily interact with the asymmetric A/B CP-dimers, suggesting a pathway for the assembly of the virions. Additionally, we have discovered various forms of the virus-like particles (VLPs), including the canonical T = 3 icosahedral, larger T = 4 icosahedral, prolate, oblate forms, and a small prolate form elongated along the 3-fold axis. These particles are all produced during a normal infection, as well as when overexpressing the CPs. When overexpressing the shorter RNA fragments encoding only the CPs, we observed an increased percentage of the smaller VLPs, which may be sufficient to encapsidate a shorter RNA.

The Structural Basis of T4 Phage Lysis Control: DNA as the Signal for Lysis Inhibition.

Optimal phage propagation depends on the regulation of the lysis of the infected host cell. In T4 phage infection, lysis occurs when the holin protein (T) forms lesions in the host membrane. However, the lethal function of T can be blocked by an antiholin (RI) during lysis inhibition (LIN). LIN sets if the infected cell undergoes superinfection, then the lysis is delayed until host/phage ratio becomes more favorable for the release of progeny. It has been thought that a signal derived from the superinfection is required to activate RI. Here we report structures that suggest a radically different model in which RI binds to T irrespective of superinfection, causing it to accumulate in a membrane as heterotetrameric 2RI-2T complex. Moreover, we show the complex binds non-specifically to DNA, suggesting that the gDNA from the superinfecting phage serves as the LIN signal and that stabilization of the complex by DNA binding is what defines LIN. Finally, we show that soluble domain of free RI crystallizes in a domain-swapped homotetramer, which likely works as a sink for RI molecules released from the RI-T complex to ensure efficient lysis. These results constitute the first structural basis and a new model not only for the historic LIN phenomenon but also for the temporal regulation of phage lysis in general.

Structural basis for the adsorption of a single-stranded RNA bacteriophage.

Single-stranded RNA bacteriophages (ssRNA phages) infect Gram-negative bacteria via a single maturation protein (Mat), which attaches to a retractile pilus of the host. Here we present structures of the ssRNA phage MS2 in complex with the Escherichia coli F-pilus, showing a network of hydrophobic and electrostatic interactions at the Mat-pilus interface. Moreover, binding of the pilus induces slight orientational variations of the Mat relative to the rest of the phage capsid, priming the Mat-connected genomic RNA (gRNA) for its release from the virions. The exposed tip of the attached Mat points opposite to the direction of the pilus retraction, which may facilitate the translocation of the gRNA from the capsid into the host cytosol. In addition, our structures determine the orientation of the assembled F-pilin subunits relative to the cell envelope, providing insights into the F-like type IV secretion systems.

Predicting Ligand Binding Sites on Protein Surfaces by 3-Dimensional Probability Density Distributions of Interacting Atoms.

Predicting ligand binding sites (LBSs) on protein structures, which are obtained either from experimental or computational methods, is a useful first step in functional annotation or structure-based drug design for the protein structures. In this work, the structure-based machine learning algorithm ISMBLab-LIG was developed to predict LBSs on protein surfaces with input attributes derived from the three-dimensional probability density maps of interacting atoms, which were reconstructed on the query protein surfaces and were relatively insensitive to local conformational variations of the tentative ligand binding sites. The prediction accuracy of the ISMBLab-LIG predictors is comparable to that of the best LBS predictors benchmarked on several well-established testing datasets. More importantly, the ISMBLab-LIG algorithm has substantial tolerance to the prediction uncertainties of computationally derived protein structure models. As such, the method is particularly useful for predicting LBSs not only on experimental protein structures without known LBS templates in the database but also on computationally predicted model protein structures with structural uncertainties in the tentative ligand binding sites.

Protein-protein interaction site predictions with three-dimensional probability distributions of interacting atoms on protein surfaces.

Protein-protein interactions are key to many biological processes. Computational methodologies devised to predict protein-protein interaction (PPI) sites on protein surfaces are important tools in providing insights into the biological functions of proteins and in developing therapeutics targeting the protein-protein interaction sites. One of the general features of PPI sites is that the core regions from the two interacting protein surfaces are complementary to each other, similar to the interior of proteins in packing density and in the physicochemical nature of the amino acid composition. In this work, we simulated the physicochemical complementarities by constructing three-dimensional probability density maps of non-covalent interacting atoms on the protein surfaces. The interacting probabilities were derived from the interior of known structures. Machine learning algorithms were applied to learn the characteristic patterns of the probability density maps specific to the PPI sites. The trained predictors for PPI sites were cross-validated with the training cases (consisting of 432 proteins) and were tested on an independent dataset (consisting of 142 proteins). The residue-based Matthews correlation coefficient for the independent test set was 0.423; the accuracy, precision, sensitivity, specificity were 0.753, 0.519, 0.677, and 0.779 respectively. The benchmark results indicate that the optimized machine learning models are among the best predictors in identifying PPI sites on protein surfaces. In particular, the PPI site prediction accuracy increases with increasing size of the PPI site and with increasing hydrophobicity in amino acid composition of the PPI interface; the core interface regions are more likely to be recognized with high prediction confidence. The results indicate that the physicochemical complementarity patterns on protein surfaces are important determinants in PPIs, and a substantial portion of the PPI sites can be predicted correctly with the physicochemical complementarity features based on the non-covalent interaction data derived from protein interiors.

Prediction of carbohydrate binding sites on protein surfaces with 3-dimensional probability density distributions of interacting atoms.

Non-covalent protein-carbohydrate interactions mediate molecular targeting in many biological processes. Prediction of non-covalent carbohydrate binding sites on protein surfaces not only provides insights into the functions of the query proteins; information on key carbohydrate-binding residues could suggest site-directed mutagenesis experiments, design therapeutics targeting carbohydrate-binding proteins, and provide guidance in engineering protein-carbohydrate interactions. In this work, we show that non-covalent carbohydrate binding sites on protein surfaces can be predicted with relatively high accuracy when the query protein structures are known. The prediction capabilities were based on a novel encoding scheme of the three-dimensional probability density maps describing the distributions of 36 non-covalent interacting atom types around protein surfaces. One machine learning model was trained for each of the 30 protein atom types. The machine learning algorithms predicted tentative carbohydrate binding sites on query proteins by recognizing the characteristic interacting atom distribution patterns specific for carbohydrate binding sites from known protein structures. The prediction results for all protein atom types were integrated into surface patches as tentative carbohydrate binding sites based on normalized prediction confidence level. The prediction capabilities of the predictors were benchmarked by a 10-fold cross validation on 497 non-redundant proteins with known carbohydrate binding sites. The predictors were further tested on an independent test set with 108 proteins. The residue-based Matthews correlation coefficient (MCC) for the independent test was 0.45, with prediction precision and sensitivity (or recall) of 0.45 and 0.49 respectively. In addition, 111 unbound carbohydrate-binding protein structures for which the structures were determined in the absence of the carbohydrate ligands were predicted with the trained predictors. The overall prediction MCC was 0.49. Independent tests on anti-carbohydrate antibodies showed that the carbohydrate antigen binding sites were predicted with comparable accuracy. These results demonstrate that the predictors are among the best in carbohydrate binding site predictions to date.

Rationalization and design of the complementarity determining region sequences in an antibody-antigen recognition interface.

Protein-protein interactions are critical determinants in biological systems. Engineered proteins binding to specific areas on protein surfaces could lead to therapeutics or diagnostics for treating diseases in humans. But designing epitope-specific protein-protein interactions with computational atomistic interaction free energy remains a difficult challenge. Here we show that, with the antibody-VEGF (vascular endothelial growth factor) interaction as a model system, the experimentally observed amino acid preferences in the antibody-antigen interface can be rationalized with 3-dimensional distributions of interacting atoms derived from the database of protein structures. Machine learning models established on the rationalization can be generalized to design amino acid preferences in antibody-antigen interfaces, for which the experimental validations are tractable with current high throughput synthetic antibody display technologies. Leave-one-out cross validation on the benchmark system yielded the accuracy, precision, recall (sensitivity) and specificity of the overall binary predictions to be 0.69, 0.45, 0.63, and 0.71 respectively, and the overall Matthews correlation coefficient of the 20 amino acid types in the 24 interface CDR positions was 0.312. The structure-based computational antibody design methodology was further tested with other antibodies binding to VEGF. The results indicate that the methodology could provide alternatives to the current antibody technologies based on animal immune systems in engineering therapeutic and diagnostic antibodies against predetermined antigen epitopes.