Simultaneous binding of Guidance Cues NET1 and RGM blocks extracellular NEO1 signaling.

During cell migration or differentiation, cell surface receptors are simultaneously exposed to different ligands. However, it is often unclear how these extracellular signals are integrated. Neogenin (NEO1) acts as an attractive guidance receptor when the Netrin-1 (NET1) ligand binds, but it mediates repulsion via repulsive guidance molecule (RGM) ligands. Here, we show that signal integration occurs through the formation of a ternary NEO1-NET1-RGM complex, which triggers reciprocal silencing of downstream signaling. Our NEO1-NET1-RGM structures reveal a "trimer-of-trimers" super-assembly, which exists in the cell membrane. Super-assembly formation results in inhibition of RGMA-NEO1-mediated growth cone collapse and RGMA- or NET1-NEO1-mediated neuron migration, by preventing formation of signaling-compatible RGM-NEO1 complexes and NET1-induced NEO1 ectodomain clustering. These results illustrate how simultaneous binding of ligands with opposing functions, to a single receptor, does not lead to competition for binding, but to formation of a super-complex that diminishes their functional outputs.

Pairing interacting protein sequences using masked language modeling.

Predicting which proteins interact together from amino acid sequences is an important task. We develop a method to pair interacting protein sequences which leverages the power of protein language models trained on multiple sequence alignments (MSAs), such as MSA Transformer and the EvoFormer module of AlphaFold. We formulate the problem of pairing interacting partners among the paralogs of two protein families in a differentiable way. We introduce a method called Differentiable Pairing using Alignment-based Language Models (DiffPALM) that solves it by exploiting the ability of MSA Transformer to fill in masked amino acids in multiple sequence alignments using the surrounding context. MSA Transformer encodes coevolution between functionally or structurally coupled amino acids within protein chains. It also captures inter-chain coevolution, despite being trained on single-chain data. Relying on MSA Transformer without fine-tuning, DiffPALM outperforms existing coevolution-based pairing methods on difficult benchmarks of shallow multiple sequence alignments extracted from ubiquitous prokaryotic protein datasets. It also outperforms an alternative method based on a state-of-the-art protein language model trained on single sequences. Paired alignments of interacting protein sequences are a crucial ingredient of supervised deep learning methods to predict the three-dimensional structure of protein complexes. Starting from sequences paired by DiffPALM substantially improves the structure prediction of some eukaryotic protein complexes by AlphaFold-Multimer. It also achieves competitive performance with using orthology-based pairing.

Improved the heterodimer protein complex prediction with protein language models.

AlphaFold-Multimer has greatly improved the protein complex structure prediction, but its accuracy also depends on the quality of the multiple sequence alignment (MSA) formed by the interacting homologs (i.e. interologs) of the complex under prediction. Here we propose a novel method, ESMPair, that can identify interologs of a complex using protein language models. We show that ESMPair can generate better interologs than the default MSA generation method in AlphaFold-Multimer. Our method results in better complex structure prediction than AlphaFold-Multimer by a large margin (+10.7% in terms of the Top-5 best DockQ), especially when the predicted complex structures have low confidence. We further show that by combining several MSA generation methods, we may yield even better complex structure prediction accuracy than Alphafold-Multimer (+22% in terms of the Top-5 best DockQ). By systematically analyzing the impact factors of our algorithm we find that the diversity of MSA of interologs significantly affects the prediction accuracy. Moreover, we show that ESMPair performs particularly well on complexes in eucaryotes.

Quantification of biases in predictions of protein-protein binding affinity changes upon mutations.

Understanding the impact of mutations on protein-protein binding affinity is a key objective for a wide range of biotechnological applications and for shedding light on disease-causing mutations, which are often located at protein-protein interfaces. Over the past decade, many computational methods using physics-based and/or machine learning approaches have been developed to predict how protein binding affinity changes upon mutations. They all claim to achieve astonishing accuracy on both training and test sets, with performances on standard benchmarks such as SKEMPI 2.0 that seem overly optimistic. Here we benchmarked eight well-known and well-used predictors and identified their biases and dataset dependencies, using not only SKEMPI 2.0 as a test set but also deep mutagenesis data on the severe acute respiratory syndrome coronavirus 2 spike protein in complex with the human angiotensin-converting enzyme 2. We showed that, even though most of the tested methods reach a significant degree of robustness and accuracy, they suffer from limited generalizability properties and struggle to predict unseen mutations. Interestingly, the generalizability problems are more severe for pure machine learning approaches, while physics-based methods are less affected by this issue. Moreover, undesirable prediction biases toward specific mutation properties, the most marked being toward destabilizing mutations, are also observed and should be carefully considered by method developers. We conclude from our analyses that there is room for improvement in the prediction models and suggest ways to check, assess and improve their generalizability and robustness.

Tetramerization of upstream stimulating factor USF2 requires the elongated bent leucine zipper of the bHLH-LZ domain.

Upstream stimulating factors (USFs), including USF1 and USF2, are key components of the transcription machinery that recruit coactivators and histone-modifying enzymes. Using the classic basic helix-loop-helix leucine zipper (bHLH-LZ) domain, USFs bind the E-box DNA and form tetramers that promote DNA looping for transcription initiation. The structural basis by which USFs tetramerize and bind DNA, however, remains unknown. Here, we report the crystal structure of the complete bHLH-LZ domain of USF2 in complex with E-box DNA. We observed that the leucine zipper (LZ) of USF2 is longer than that of other bHLH-LZ family transcription factors and that the C-terminus of USF2 forms an additional α-helix following the LZ region (denoted as LZ-Ext). We also found the elongated LZ-Ext facilitates compact tetramer formation. In addition to the classic interactions between the basic region and DNA, we show a highly conserved basic residue in the loop region, Lys271, participates in DNA interaction. Together, these findings suggest that USF2 forms a tetramer structure with a bent elongated LZ-Ext region, providing a molecular basis for its role as a key component of the transcription machinery.

High-resolution crystal structure of RNA kinase ArK1 from G. acetivorans.

U47 phosphorylation (Up47) is a novel tRNA modification discovered recently; it can confer thermal stability and nuclease resistance to tRNAs. U47 phosphorylation is catalyzed by Archaeal RNA kinase (Ark1) in an ATP-dependent manner. However, the structural basis for tRNA and/or ATP binding by Ark1 is unclear. Here, we report the expression, purification, and crystallization studies of Ark1 from G. acetivorans (GaArk1). In addition to the Apo-form structure, one GaArk1-ATP complex was also determined in atomic resolution and revealed the detailed basis for ATP binding by GaArk1. The GaArk1-ATP complex represents the only ATP-bound structure of the Ark1 protein. The majority of the ATP-binding residues are conserved, suggesting that GaArk1 and the homologous proteins share similar mechanism in ATP binding. Sequence and structural analysis further indicated that endogenous guanosine will only inhibit the activities of certain Ark1 proteins, such as Ark1 from T. kodakarensis.

Protein complex structure prediction powered by multiple sequence alignments of interologs from multiple taxonomic ranks and AlphaFold2.

AlphaFold2 can predict protein complex structures as long as a multiple sequence alignment (MSA) of the interologs of the target protein-protein interaction (PPI) can be provided. In this study, a simplified phylogeny-based approach was applied to generate the MSA of interologs, which was then used as the input to AlphaFold2 for protein complex structure prediction. In this extensively benchmarked protocol on nonredundant PPI dataset, including 107 bacterial PPIs and 442 eukaryotic PPIs, we show complex structures of 79.5% of the bacterial PPIs and 49.8% of the eukaryotic PPIs can be successfully predicted, which yielded significantly better performance than the application of MSA of interologs prepared by two existing approaches. Considering PPIs may not be conserved in species with long evolutionary distances, we further restricted interologs in the MSA to different taxonomic ranks of the species of the target PPI in protein complex structure prediction. We found that the success rates can be increased to 87.9% for the bacterial PPIs and 56.3% for the eukaryotic PPIs if interologs in the MSA are restricted to a specific taxonomic rank of the species of each target PPI. Finally, we show that the optimal taxonomic ranks for protein complex structure prediction can be selected with the application of the predicted template modeling (TM) scores of the output models.

Structural Perspective of NR4A Nuclear Receptor Family and Their Potential Endogenous Ligands.

There are 48 nuclear receptors in the human genome, and many members of this superfamily have been implicated in human diseases. The NR4A nuclear receptor family consisting of three members, NR4A1, NR4A2, and NR4A3 (formerly annotated as Nur77, Nurr1, and NOR1, respectively), are still orphan receptors but exert pathological effects on immune-related and neurological diseases. We previously reported that prostaglandin A1 (PGA1) and prostaglandin A2 (PGA2) are potent activators of NR4A3, which bind directly to the ligand-binding domain (LBD) of the receptor. Recently, the co-crystallographic structures of NR4A2-LBD bound to PGA1 and PGA2 were reported, followed by reports of the neuroprotective effects of these possible endogenous ligands in mouse models of Parkinson's disease. Based on these structures, we modeled the binding structures of the other two members (NR4A1 and NR4A3) with these potential endogenous ligands using a template-based modeling method, and reviewed the similarity and diversity of ligand-binding mechanisms in the nuclear receptor family.

A structural model of the iRhom-ADAM17 sheddase complex reveals functional insights into its trafficking and activity.

Several membrane-anchored signal mediators such as cytokines (e.g. TNFα) and growth factors are proteolytically shed from the cell surface by the metalloproteinase ADAM17, which, thus, has an essential role in inflammatory and developmental processes. The membrane proteins iRhom1 and iRhom2 are instrumental for the transport of ADAM17 to the cell surface and its regulation. However, the structure-function determinants of the iRhom-ADAM17 complex are poorly understood. We used AI-based modelling to gain insights into the structure-function relationship of this complex. We identified different regions in the iRhom homology domain (IRHD) that are differentially responsible for iRhom functions. We have supported the validity of the predicted structure-function determinants with several in vitro, ex vivo and in vivo approaches and demonstrated the regulatory role of the IRHD for iRhom-ADAM17 complex cohesion and forward trafficking. Overall, we provide mechanistic insights into the iRhom-ADAM17-mediated shedding event, which is at the centre of several important cytokine and growth factor pathways.

Damage Severity Assessment of Multi-Layer Complex Structures Based on a Damage Information Extraction Method with Ladder Feature Mining.

Multi-layer complex structures are widely used in large-scale engineering structures because of their diverse combinations of properties and excellent overall performance. However, multi-layer complex structures are prone to interlaminar debonding damage during use. Therefore, it is necessary to monitor debonding damage in engineering applications to determine structural integrity. In this paper, a damage information extraction method with ladder feature mining for Lamb waves is proposed. The method is able to optimize and screen effective damage information through ladder-type damage extraction. It is suitable for evaluating the severity of debonding damage in aluminum-foamed silicone rubber, a novel multi-layer complex structure. The proposed method contains ladder feature mining stages of damage information selection and damage feature fusion, realizing a multi-level damage information extraction process from coarse to fine. The results show that the accuracy of damage severity assessment by the damage information extraction method with ladder feature mining is improved by more than 5% compared to other methods. The effectiveness and accuracy of the method in assessing the damage severity of multi-layer complex structures are demonstrated, providing a new perspective and solution for damage monitoring of multi-layer complex structures.

Estimating protein complex model accuracy based on ultrafast shape recognition and deep learning in CASP15.

This article reports and analyzes the results of protein complex model accuracy estimation by our methods (DeepUMQA3 and GraphGPSM) in the 15th Critical Assessment of techniques for protein Structure Prediction (CASP15). The new deep learning-based multimeric complex model accuracy estimation methods are proposed based on the ensemble of three-level features coupling with deep residual/graph neural networks. For the input multimeric complex model, we describe it from three levels: overall complex features, intra-monomer features, and inter-monomer features. We designed an overall ultrafast shape recognition (USR) to characterize the relationship between local residues and the overall complex topology, and an inter-monomer USR to characterize the relationship between the residues of one monomer and the topology of other monomers. DeepUMQA3 (Group name: GuijunLab-RocketX) ranked first in the interface residue accuracy estimation of CASP15. The Pearson correlation between the interface residue Local Distance Difference Test (lDDT) predicted by DeepUMQA3 and the real lDDT is 0.570, the only method that exceeds 0.5. Among the top 5 methods, DeepUMQA3 achieved the highest Pearson correlation of lDDT on 25 out of 39 targets. GraphGPSM (Group name: GuijunLab-PAthreader) has TM-score Pearson correlations greater than 0.9 on 14 targets, showing a good ability to estimate the overall fold accuracy. The DeepUMQA3 server is available at http://zhanglab-bioinf.com/DeepUMQA/ and the GraphGPSM server is available at http://zhanglab-bioinf.com/GraphGPSM/.

Discovery of a potent inhibitor targeting the cap-binding domain of the PB2 subunit of influenza RNA-dependent RNA polymerase.

Influenza pandemics have emerged as a significant global public health and security concern. PB2, a crucial subunit of the influenza RNA-dependent RNA polymerase (RdRP), has been identified as a promising target for influenza treatment. We herein report the discovery of a potent novel PB2 inhibitor, 7-51A, with a KD value of 1.64 nM as determined by ITC. The high activity of 7-51A was elucidated by the co-crystal structure of the PB2-7-51A complex, and comparative analysis revealed unique interactions that had never been observed before. The preliminary pharmacological evaluation indicated that 7-51A exhibited commendable cellular safety, hepatic microsomal metabolic safety and stability. Collectively, 7-51A was found to be an effective PB2 inhibitor and could be used as a lead compound for further studies.

High-resolution structure of eukaryotic Fibrillarin interacting with Nop56 amino-terminal domain.

Ribosomal RNA (rRNA) carries extensive 2'-O-methyl marks at functionally important sites. This simple chemical modification is thought to confer stability, promote RNA folding, and contribute to generate a heterogenous ribosome population with a yet-uncharacterized function. 2'-O-methylation occurs both in archaea and eukaryotes and is accomplished by the Box C/D RNP enzyme in an RNA-guided manner. Extensive and partially conflicting structural information exists for the archaeal enzyme, while no structural data is available for the eukaryotic enzyme. The yeast Box C/D RNP consists of a guide RNA, the RNA-primary binding protein Snu13, the two scaffold proteins Nop56 and Nop58, and the enzymatic module Nop1. Here we present the high-resolution structure of the eukaryotic Box C/D methyltransferase Nop1 from Saccharomyces cerevisiae bound to the amino-terminal domain of Nop56. We discuss similarities and differences between the interaction modes of the two proteins in archaea and eukaryotes and demonstrate that eukaryotic Nop56 recruits the methyltransferase to the Box C/D RNP through a protein-protein interface that differs substantially from the archaeal orthologs. This study represents a first achievement in understanding the evolution of the structure and function of these proteins from archaea to eukaryotes.

Resolving complex structures at oncovirus integration loci with conjugate graph.

Oncovirus integrations cause copy number variations and complex structural variations (SVs) on host genomes. However, the understanding of how inserted viral DNA impacts the local genome remains limited. The linear structure of the oncovirus integrated local genomic map (LGM) will lay the foundations to understand how oncovirus integrations emerge and compromise the host genome's functioning. We propose a conjugate graph model to reconstruct the rearranged LGM at integrated loci. Simulation tests prove the reliability and credibility of the algorithm. Applications of the algorithm to whole-genome sequencing data of human papillomavirus (HPV) and hepatitis B virus (HBV)-infected cancer samples gained biological insights on oncovirus integrations. We observed four affection patterns of oncovirus integrations from the HPV and HBV-integrated cancer samples, including the coding-frame truncation, hyper-amplification of tumor gene, the viral cis-regulation inserted at the single intron and at the intergenic region. We found that the focal duplicates and host SVs are frequent in the HPV-integrated LGMs, while the focal deletions are prevalent in HBV-integrated LGMs. Furthermore, with the results yields from our method, we found the enhanced microhomology-mediated end joining might lead to both HPV and HBV integrations and conjectured that the HPV integrations might mainly occur during the DNA replication process. The conjugate graph algorithm code and LGM construction pipeline, available at https://github.com/deepomicslab/FuseSV.

Coherent diffractive imaging of proteins and viral capsids: simulating MS SPIDOC.

MS SPIDOC is a novel sample delivery system designed for single (isolated) particle imaging at X-ray Free-Electron Lasers that is adaptable towards most large-scale facility beamlines. Biological samples can range from small proteins to MDa particles. Following nano-electrospray ionization, ionic samples can be m/z-filtered and structurally separated before being oriented at the interaction zone. Here, we present the simulation package developed alongside this prototype. The first part describes how the front-to-end ion trajectory simulations have been conducted. Highlighted is a quadrant lens; a simple but efficient device that steers the ion beam within the vicinity of the strong DC orientation field in the interaction zone to ensure spatial overlap with the X-rays. The second part focuses on protein orientation and discusses its potential with respect to diffractive imaging methods. Last, coherent diffractive imaging of prototypical T = 1 and T = 3 norovirus capsids is shown. We use realistic experimental parameters from the SPB/SFX instrument at the European XFEL to demonstrate that low-resolution diffractive imaging data (q < 0.3 nm-1) can be collected with only a few X-ray pulses. Such low-resolution data are sufficient to distinguish between both symmetries of the capsids, allowing to probe low abundant species in a beam if MS SPIDOC is used as sample delivery.

Deafness-related protein PDZD7 forms complex with the C-terminal tail of FCHSD2.

In cochlea, deafness-related protein PDZD7 is an indispensable component of the ankle link complex, which is critical for the maturation of inner-ear hair cell for sound perception. Ankle links, connecting the different rows of cochlear stereocilia, are essential for the staircase-like development of stereocilia. However, the molecular mechanism of how PDZD7 governs stereociliary development remains unknown. Here, we reported a novel PDZD7-binding partner, FCHSD2, identified by yeast two-hybrid screening. FCHSD2 was reported to be expressed in hair cell, where it co-operated with CDC42 and N-WASP to regulate the formation of cell protrusion. The association between FCHSD2 and PDZD7 was further confirmed in COS-7 cells. More importantly, we solved the complex structure of FCHSD2 tail with PDZD7 PDZ3 domain at 2.0 Šresolution. The crystal structure shows that PDZD7 PDZ3 adopts a typical PDZ domain topology, comprising five ß strands and two α helixes. The PDZ-binding motif of FCHSD2 tail stretches through the αB/ßB groove of PDZD7 PDZ3. Our study not only uncovers the interaction between FCHSD2 tail and PDZD7 PDZ3 at the atomic level, but also provides clues of connecting the ankle link complex with cytoskeleton dynamics for exploiting the molecular mechanism of stereociliary development.

The structural basis of African swine fever virus pA104R binding to DNA and its inhibition by stilbene derivatives.

African swine fever virus (ASFV) is a highly contagious nucleocytoplasmic large DNA virus (NCLDV) that causes nearly 100% mortality in swine. The development of effective vaccines and drugs against this virus is urgently needed. pA104R, an ASFV-derived histone-like protein, shares sequence and functional similarity with bacterial HU/IHF family members and is essential for viral replication. Herein, we solved the crystal structures of pA104R in its apo state as well as in complex with DNA. Apo-pA104R forms a homodimer and folds into an architecture conserved in bacterial heat-unstable nucleoid proteins/integration host factors (HUs/IHFs). The pA104R-DNA complex structure, however, uncovers that pA104R has a DNA binding pattern distinct from its bacterial homologs, that is, the ß-ribbon arms of pA104R stabilize DNA binding by contacting the major groove instead of the minor groove. Mutations of the basic residues at the base region of the ß-strand DNA binding region (BDR), rather than those in the ß-ribbon arms, completely abolished DNA binding, highlighting the major role of the BDR base in DNA binding. An overall DNA bending angle of 93.8° is observed in crystal packing of the pA104R-DNA complex structure, which is close to the DNA bending angle in the HU-DNA complex. Stilbene derivatives SD1 and SD4 were shown to disrupt the binding between pA104R and DNA and inhibit the replication of ASFV in primary porcine alveolar macrophages. Collectively, these results reveal the structural basis of pA104R binding to DNA highlighting the importance of the pA104R-DNA interaction in the ASFV replication cycle and provide inhibitor leads for ASFV chemotherapy.

High-affinity oligoclonal TCRs define effective adoptive T cell therapy targeting mutant KRAS-G12D.

Complete cancer regression occurs in a subset of patients following adoptive T cell therapy (ACT) of ex vivo expanded tumor-infiltrating lymphocytes (TILs). However, the low success rate presents a great challenge to broader clinical application. To provide insight into TIL-based immunotherapy, we studied a successful case of ACT where regression was observed against tumors carrying the hotspot mutation G12D in the KRAS oncogene. Four T cell receptors (TCRs) made up the TIL infusion and recognized two KRAS-G12D neoantigens, a nonamer and a decamer, all restricted by human leukocyte antigen (HLA) C*08:02. Three of them (TCR9a, 9b, and 9c) were nonamer-specific, while one was decamer-specific (TCR10). We show that only mutant G12D but not the wild-type peptides stabilized HLA-C*08:02 due to the formation of a critical anchor salt bridge to HLA-C. Therapeutic TCRs exhibited high affinities, ranging from nanomolar to low micromolar. Intriguingly, TCR binding affinities to HLA-C inversely correlated with their persistence in vivo, suggesting the importance of antigenic affinity in the function of therapeutic T cells. Crystal structures of TCR-HLA-C complexes revealed that TCR9a to 9c recognized G12D nonamer with multiple conserved contacts through shared CDR2ß and CDR3α. This allowed CDR3ß variation to confer different affinities via a variable HLA-C contact, generating an oligoclonal response. TCR10 recognized an induced and distinct G12D decamer conformation. Thus, this successful case of ACT included oligoclonal TCRs of high affinity recognizing distinct conformations of neoantigens. Our study revealed the potential of a structural approach to inform clinical efforts in targeting KRAS-G12D tumors by immunotherapy and has general implications for T cell-based immunotherapies.

Structure of human MDM2 complexed with RPL11 reveals the molecular basis of p53 activation.

The central region of MDM2 is critical for p53 activation and tumor suppression. Upon ribosomal stress, this region is bound by ribosomal proteins, particularly ribosomal protein L11 (RPL11), leading to MDM2 inactivation and subsequent p53 activation. Here, we solved the complex structure of human MDM2-RPL11 at 2.4 Å. MDM2 extensively interacts with RPL11 through an acidic domain and two zinc fingers. Formation of the MDM2-RPL11 complex induces substantial conformational changes in both proteins. RPL11, unable to bind MDM2 mutants, fails to induce the activation of p53 in cells. MDM2 mimics 28S rRNA binding to RPL11. The C4 zinc finger determines RPL11 binding to MDM2 but not its homolog, MDMX. Our results highlight the essential role of the RPL11-MDM2 interaction in p53 activation and tumor suppression and provide a structural basis for potential new anti-tumor drug development.

Dynamic Dimerization of Chemokine Receptors and Potential Inhibitory Role of Their Truncated Isoforms Revealed through Combinatorial Prediction.

Chemokine receptors play crucial roles in fundamental biological processes. Their malfunction may result in many diseases, including cancer, autoimmune diseases, and HIV. The oligomerization of chemokine receptors holds significant functional implications that directly affect their signaling patterns and pharmacological responses. However, the oligomerization patterns of many chemokine receptors remain poorly understood. Furthermore, several chemokine receptors have highly truncated isoforms whose functional role is not yet clear. Here, we computationally show homo- and heterodimerization patterns of four human chemokine receptors, namely CXCR2, CXCR7, CCR2, and CCR7, along with their interaction patterns with their respective truncated isoforms. By combining the neural network-based AlphaFold2 and physics-based protein-protein docking tool ClusPro, we predicted 15 groups of complex structures and assessed the binding affinities in the context of atomistic molecular dynamics simulations. Our results are in agreement with previous experimental observations and support the dynamic and diverse nature of chemokine receptor dimerization, suggesting possible patterns of higher-order oligomerization. Additionally, we uncover the strong potential of truncated isoforms to block homo- and heterodimerization of chemokine receptors, also in a dynamic manner. Our study provides insights into the dimerization patterns of chemokine receptors and the functional significance of their truncated isoforms.