Structural Characterization and Function Prediction of Immunoglobulin-like Fold in Cell Adhesion and Cell Signaling

Prediction of interactions between cell surface proteins by machine learning

Cells detect changes in their external environments or communicate with each other through proteins on their surfaces. These cell surface proteins form a complicated network of interactions in order to fulfill their functions. The interactions between cell surface proteins are highly dynamic and, thus, challenging to detect using traditional experimental techniques. Here, we tackle this challenge using a computational framework. The primary focus of the framework is to develop new tools to identify interactions between domains in the immunoglobulin (Ig) fold, which is the most abundant domain family in cell surface proteins. These interactions could be formed between ligands and receptors from different cells or between proteins on the same cell surface. In practice, we collected all structural data on Ig domain interactions and transformed them into an interface fragment pair library. A high-dimensional profile can then be constructed from the library for a given pair of query protein sequences. Multiple machine learning models were used to read this profile so that the probability of interaction between the query proteins could be predicted. We tested our models on an experimentally derived dataset that contains 564 cell surface proteins in humans. The cross-validation results show that we can achieve higher than 70% accuracy in identifying the PPIs within this dataset. We then applied this method to a group of 46 cell surface proteins in Caenorhabditis elegans. We screened every possible interaction between these proteins. Many interactions recognized by our machine learning classifiers have been experimentally confirmed in the literature. In conclusion, our computational platform serves as a useful tool to help identify potential new interactions between cell surface proteins in addition to current state-of-the-art experimental techniques. The tool is freely accessible for use by the scientific community. Moreover, the general framework of the machine learning classification can also be extended to study the interactions of proteins in other domain superfamilies.

Towards a structural and functional analysis of the immunoglobulin-fold proteome

The immunoglobulin fold (Ig fold) domain is a super-secondary structural motif consisting of a sandwich with two layers of β-sheets that is present in many proteins with very diverse biological functions covering a wide range of physiological processes. This domain presents a modular architecture built with β strands connected by variable length loops that has a highly conserved structural core of four β-strands and quite variable β-sheet extensions in the two sandwich layers that enable both divergent and convergent evolutionary mechanisms in the known Ig fold proteome. The central role of this Ig fold's structural plasticity in the evolutionary success of antibodies in our immune system is well established. Nature has also utilized this Ig fold in all domains of life in many different physiological contexts that go way beyond the immune system. Here we will present a structural and functional overview of the utilization of the Ig fold in different biological processes and in different cellular contexts to highlight some of the innumerable ways that this structural motif can interact in multidomain proteins to enable their diversity of functions. This includes shareable specific protein structure visualizations behind those functions that serve as starting points for further explorations of the biomolecular interactions spanning the Ig fold proteome. This overview also highlights how this Ig fold is being utilized through natural adaptation, engineering, and even building from scratch for a range of biotechnological applications.

Domain structure and cross-linking in a giant adhesin from the Mobiluncus mulieris bacterium

Cell-surface proteins known as adhesins enable bacteria to colonize particular environments, and in Gram-positive bacteria often contain autocatalytically formed covalent intramolecular cross-links. While investigating the prevalence of such cross-links, a remarkable example was discovered in Mobiluncus mulieris, a pathogen associated with bacterial vaginosis. This organism encodes a putative adhesin of 7651 residues. Crystallography and mass spectrometry of two selected domains, and AlphaFold structure prediction of the remainder of the protein, were used to show that this adhesin belongs to the family of thioester, isopeptide and ester-bond-containing proteins (TIE proteins). It has an N-terminal domain homologous to thioester adhesion domains, followed by 51 immunoglobulin (Ig)-like domains containing ester- or isopeptide-bond cross-links. The energetic cost to the M. mulieris bacterium in retaining such a large adhesin as a single gene or protein construct suggests a critical role in pathogenicity and/or persistence.

Pangenomic Analysis of Nucleo-Cytoplasmic Large DNA Viruses. I: The Phylogenetic Distribution of Conserved Oxygen-Dependent Enzymes Reveals a Capture-Gene Process

The Nucleo-Cytoplasmic Large DNA Viruses (NCLDVs) infect a wide range of eukaryotic species, including amoeba, algae, fish, amphibia, arthropods, birds, and mammals. This group of viruses has linear or circular double-stranded DNA genomes whose size spans approximately one order of magnitude, from 100 to 2500 kbp. The ultimate origin of this peculiar group of viruses remains an open issue. Some have argued that NCLDVs' origin may lie in a bacteriophage ancestor that increased its genome size by subsequent recruitment of eukaryotic and bacterial genes. Others have suggested that NCLDVs families originated from cells that underwent an irreversible process of genome reduction. However, the hypothesis that a number of NCLDVs sequences have been recruited from the host genomes has been largely ignored. In the present work, we have performed pangenomic analyses of each of the seven known NCLDVs families. We show that these families' core- and shell genes have cellular homologs, supporting possible escaping-gene events as part of its evolution. Furthermore, the detection of sequences that belong to two protein families (small chain ribonucleotide reductase and Erv1/Air) and to one superfamily [2OG-Fe(II) oxygenases] that are for distribution in all NCLDVs core and shell clusters encoding for oxygen-dependent enzymes suggests that the highly conserved core these viruses originated after the Proterozoic Great Oxidation Event that transformed the terrestrial atmosphere 2.4-2.3 Ga ago.

Public RNA-seq data-based identification and functional analyses reveal that MXRA5 retains proliferative and migratory abilities of dental pulp stem cells

Dental pulp stem cells (DPSC) usually remain quiescent in the dental pulp tissue; however, once the dental pulp tissue is injured, DPSCs potently proliferate and migrate into the injury microenvironment and contribute to immuno-modulation and tissue repair. However, the key molecules that physiologically support the potent proliferation and migration of DPSCs have not been revealed. In this study, we searched publicly available transcriptome raw data sets, which contain comparable (i.e., equivalently cultured) DPSC and mesenchymal stem cell data. Three data sets were extracted from the Gene Expression Omnibus database and then processed and analyzed. MXRA5 was identified as the predominant DPSC-enriched gene associated with the extracellular matrix. MXRA5 is detected in human dental pulp tissues. Loss of MXRA5 drastically decreases the proliferation and migration of DSPCs, concomitantly with reduced expression of the genes associated with the cell cycle and microtubules. In addition to the known full-length isoform of MXRA5, a novel splice variant of MXRA5 was cloned in DPSCs. Recombinant MXRA5 coded by the novel splice variant potently induced the haptotaxis migration of DPSCs, which was inhibited by microtubule inhibitors. Collectively, MXRA5 is a key extracellular matrix protein in dental pulp tissue for maintaining the proliferation and migration of DPSCs.

Prediction of Interactions between Cell Surface Proteins by Machine Learning

Cells detect changes of external environments or communicate with each other through proteins on their surfaces. These cell surface proteins form a complicated network of interactions in order to fulfill their functions. The interactions between cell surface proteins are highly dynamic and thus challenging to detect using traditional experimental techniques. Here we tackle this challenge by a computational framework. The primary focus of the framework is to develop new tools to identify interactions between domains in immunoglobulin (Ig) fold, which is the most abundant domain family in cell surface proteins. These interactions could be formed between ligands and receptors from different cells, or between proteins on the same cell surface. In practice, we collected all structural data of Ig domain interactions and transformed them into an interface fragment pair library. A high dimensional profile can be then constructed from the library for a given pair of query protein sequences. Multiple machine learning models were used to read this profile, so that the probability of interaction between the query proteins can be predicted. We tested our models to an experimentally derived dataset which contains 564 cell surface proteins in human. The cross-validation results show that we can achieve higher than 70% accuracy in identifying the PPIs within this dataset. We then applied this method to a group of 46 cell surface proteins in C elegans. We screened every possible interaction between these proteins. Many interactions recognized by our machine learning classifiers have been experimentally confirmed in the literatures. In conclusion, our computational platform serves a useful tool to help identifying potential new interactions between cell surface proteins in addition to current state-of-the-art experimental techniques. The tool is freely accessible for use by the scientific community. Moreover, the general framework of the machine learning classification can also be extended to study interactions of proteins in other domain superfamilies.

Towards mechanistic models of mutational effects: Deep Learning on Alzheimer’s Aβ peptide

Deep Mutational Scanning (DMS) has enabled multiplexed measurement of mutational effects on protein properties, including kinematics and self-organization, with unprecedented resolution. However, potential bottlenecks of DMS characterization include experimental design, data quality, and depth of mutational coverage. Here, we apply deep learning to comprehensively model the mutational effect of the Alzheimer's Disease associated peptide Aβ₄₂ on aggregation-related biochemical traits from DMS measurements. Among tested neural network architectures, Convolutional Neural Networks and Recurrent Neural Networks are found to be the most cost-effective models with high performance even under insufficiently-sampled DMS studies. While sequence features are essential for satisfactory prediction from neural networks, geometric-structural features further enhance the prediction performance. Notably, we demonstrate how mechanistic insights into phenotype may be extracted from the neural networks themselves suitably designed. This methodological benefit is particularly relevant for biochemical systems displaying a strong coupling between structure and phenotype such as the conformation of Aβ₄₂ aggregate and nucleation, as shown here using a Graph Convolutional Neural Network (GCN) developed from the protein atomic structure input. In addition to accurate imputation of missing values (which here ranged up to 55% of all phenotype values at key residues), the mutationally-defined nucleation phenotype generated from a GCN shows improved resolution for identifying known disease-causing mutations relative to the original DMS phenotype. Our study suggests that neural network derived sequence-phenotype mapping can be exploited not only to provide direct support for protein engineering or genome editing but also to facilitate therapeutic design with the gained perspectives from biological modeling.

In Silico Approach Gives Insights into Ig-like Fold Containing Proteins in Vibrio parahaemolyticus: A Focus on the Fibrillar Adhesins

Immunoglobulin-like (Ig-like) fold domains are abundant on the surface of bacteria, where they are required for cell-to-cell recognition, adhesion, biofilm formation, and conjugative transfer. Fibrillar adhesins are proteins with Ig-like fold(s) that have filamentous structures at the cell surface, being thinner and more flexible than pili. While the roles of fibrillar adhesins have been proposed in bacteria overall, their characterization in Vibrio parahaemolyticus has not been established and, therefore, understanding about fibrillar adhesins remain limited in V. parahaemolyticus. This in silico analysis can aid in the systematic identification of Ig-like-folded and fibrillar adhesin-like proteins in V. parahaemolyticus, opening new avenues for disease prevention by interfering in microbial interaction between V. parahaemolyticus and the host.

Classification of protein-protein association rates based on biophysical informatics

BACKGROUND

Proteins form various complexes to carry out their versatile functions in cells. The dynamic properties of protein complex formation are mainly characterized by the association rates which measures how fast these complexes can be formed. It was experimentally observed that the association rates span an extremely wide range with over ten orders of magnitudes. Identification of association rates within this spectrum for specific protein complexes is therefore essential for us to understand their functional roles.

RESULTS

To tackle this problem, we integrate physics-based coarse-grained simulations into a neural-network-based classification model to estimate the range of association rates for protein complexes in a large-scale benchmark set. The cross-validation results show that, when an optimal threshold was selected, we can reach the best performance with specificity, precision, sensitivity and overall accuracy all higher than 70%. The quality of our cross-validation data has also been testified by further statistical analysis. Additionally, given an independent testing set, we can successfully predict the group of association rates for eight protein complexes out of ten. Finally, the analysis of failed cases suggests the future implementation of conformational dynamics into simulation can further improve model.

CONCLUSIONS

In summary, this study demonstrated that a new modeling framework that combines biophysical simulations with bioinformatics approaches is able to identify protein-protein interactions with low association rates from those with higher association rates. This method thereby can serve as a useful addition to a collection of existing experimental approaches that measure biomolecular recognition.

Exploring Mucin as Adjunct to Phage Therapy

Conventional phage therapy using bacteriophages (phages) for specific targeting of pathogenic bacteria is not always useful as a therapeutic for gastrointestinal (GI) dysfunction. Complex dysbiotic GI disorders such as small intestinal bowel overgrowth (SIBO), ulcerative colitis (UC), or Crohn’s disease (CD) are even more difficult to treat as these conditions have shifts in multiple populations of bacteria within the microbiome. Such community-level structural changes in the gut microbiota may require an alternative to conventional phage therapy such as fecal virome transfer or a phage cocktail capable of targeting multiple bacterial species. Additionally, manipulation of the GI microenvironment may enhance beneficial bacteria–phage interactions during treatment. Mucin, produced along the entire length of the GI tract to protect the underlying mucosa, is a prominent contributor to the GI microenvironment and may facilitate bacteria–phage interactions in multiple ways, potentially serving as an adjunct during phage therapy. In this review, we will describe what is known about the role of mucin within the GI tract and how its facilitation of bacteria–phage interactions should be considered in any effort directed at optimizing effectiveness of a phage therapy for gastrointestinal dysbiosis.

Characterizing the function of domain linkers in regulating the dynamics of multi-domain fusion proteins by microsecond molecular dynamics simulations and artificial intelligence

Multi-domain proteins are not only formed through natural evolution but can also be generated by recombinant DNA technology. Because many fusion proteins can enhance the selectivity of cell targeting, these artificially produced molecules, called multi-specific biologics, are promising drug candidates, especially for immunotherapy. Moreover, the rational design of domain linkers in fusion proteins is becoming an essential step toward a quantitative understanding of the dynamics in these biopharmaceutics. We developed a computational framework to characterize the impacts of peptide linkers on the dynamics of multi-specific biologics. Specifically, we first constructed a benchmark containing six types of linkers that represent various lengths and degrees of flexibility and used them to connect two natural proteins as a test system. We then projected the microsecond dynamics of these proteins generated from Anton onto a coarse-grained conformational space. We further analyzed the similarity of dynamics among different proteins in this low-dimensional space by a neural-network-based classification model. Finally, we applied hierarchical clustering to place linkers into different subgroups based on the classification results. The clustering results suggest that the length of linkers, which is used to spatially separate different functional modules, plays the most important role in regulating the dynamics of this fusion protein. Given the same number of amino acids, linker flexibility functions as a regulator of protein dynamics. In summary, we illustrated that a new computational strategy can be used to study the dynamics of multi-domain fusion proteins by a combination of long timescale molecular dynamics simulation, coarse-grained feature extraction, and artificial intelligence.

Cell Communications among Microorganisms, Plants, and Animals: Origin, Evolution, and Interplays

Cellular communications play pivotal roles in multi-cellular species, but they do so also in uni-cellular species. Moreover, cells communicate with each other not only within the same individual, but also with cells in other individuals belonging to the same or other species. These communications occur between two unicellular species, two multicellular species, or between unicellular and multicellular species. The molecular mechanisms involved exhibit diversity and specificity, but they share common basic features, which allow common pathways of communication between different species, often phylogenetically very distant. These interactions are possible by the high degree of conservation of the basic molecular mechanisms of interaction of many ligand-receptor pairs in evolutionary remote species. These inter-species cellular communications played crucial roles during Evolution and must have been positively selected, particularly when collectively beneficial in hostile environments. It is likely that communications between cells did not arise after their emergence, but were part of the very nature of the first cells. Synchronization of populations of non-living protocells through chemical communications may have been a mandatory step towards their emergence as populations of living cells and explain the large commonality of cell communication mechanisms among microorganisms, plants, and animals.