Protein Science最新文献

Many proteins harbor covalent intramolecular bonds that enhance their stability and resistance to thermal, mechanical, and proteolytic insults. Intramolecular isopeptide bonds represent one such covalent interaction, yet their distribution across protein domains and organisms has been largely unexplored. Here, we sought to address this by employing a large-scale prediction of intramolecular isopeptide bonds in the AlphaFold database using the structural template-based software Isopeptor. Our findings reveal an extensive phyletic distribution in bacterial and archaeal surface proteins resembling fibrillar adhesins and pilins. All identified intramolecular isopeptide bonds are found in two structurally distinct folds, CnaA-like or CnaB-like, from a relatively small set of related Pfam families, including 10 novel families that we predict to contain intramolecular isopeptide bonds. One CnaA-like domain of unknown function, DUF11 (renamed here to "CLIPPER") is broadly distributed in cell-surface proteins from Gram-positive bacteria, Gram-negative bacteria, and archaea, and is structurally and biophysically characterized in this work. Using x-ray crystallography, we resolve a CLIPPER domain from a Gram-negative fibrillar adhesin that contains an intramolecular isopeptide bond and further demonstrate that it imparts thermostability and resistance to proteolysis. Our findings demonstrate the extensive distribution of intramolecular isopeptide bond-containing protein domains in nature and structurally resolve the previously cryptic CLIPPER domain.

The CLC family of membrane proteins consists of chloride channels and anion/proton antiporters. How the same fold accommodates two distinct mechanisms remains poorly understood, and the small set of experimental structures provides limited insight. Here we show that it is possible to scale up CLC structural information using AlphaFold2 predictions and combine this with an ensemble-based machine learning approach to identify subtle structural differences associated with each mechanistic class. We first carried out a phylogenetic analysis on CLC sequences to infer 569 channels and 1051 transporter homologs of eukaryotic origin that were previously unidentified. We then examined AlphaFold2's ability to detect subtle differences among experimentally solved CLC structures using distance difference matrices and validated the use of these models in our study. Next, we trained a random forest classifier on channel versus transporter grouped distance data, generating a structure-based predictor for CLC subtypes. Shapley value analysis was then carried out to calculate importance values, allowing the identification of changes in helix-pair distances most strongly associated with the classifier's ability to predict CLC subtype. These differences are summarized by three concerted changes found in channels: reduced distance between dimerization interface helices and the C-terminal half of the protein, expansion of the anion transport pathway along the membrane, and insertion of an interfacial helix between αJ and αK. These changes overlap with observations from experimental structures, showing that this approach expands structural information across sequence space. This establishes a framework for large-scale structural analysis when experimental data is limited and may be useful for the study of other protein families.

The human ATP-binding cassette (ABC)-transporter ABCB10 is essential for red blood cell development and protects mitochondria against oxidative stress. The transporter's basal ATPase activity is stimulated by the transported substrate (biliverdin), implying long-range communication between the transmembrane substrate binding site(s) and the nucleotide-binding domains that hydrolyze ATP. However, the molecular events that lead to stimulation by substrate are not fully understood. A recent Cryo-EM model of biliverdin-bound ABCB10 in GDN-micelles shows a substrate binding site that differs from a prior sequence alignment prediction, generating ambiguity. Therefore, we conducted a functional analysis to identify ABCB10's residues necessary for substrate-induced ATPase stimulation, as well as relevant biliverdin functional groups. We found that mutation of the highly conserved R232 and R295 decreased the stimulation by biliverdin, suggesting a more relevant role for these residues than what was derived from the Cryo-EM model. GDN abolished ABCB10's stimulation by biliverdin, so it might also affect substrate binding in this model. We also found that ABCB10's stimulation is specific for biliverdin, as biliverdin dimethyl ester was not an effective stimulator, whereas mesobiliverdin inhibits instead of stimulating. Interestingly, some of the arginine mutants display elevated basal ATPase activity. Also, by using Luminescence Resonance Energy Transfer we have detected alterations in the conformational equilibrium of these arginine mutants and a lack of response to biliverdin. In general, our data suggest delicate complementarities between the substrate and its binding pocket on ABCB10, with small modifications largely impacting the basal ATPase activity and the transporter's ability to be stimulated by substrate.

The amino groups of lysine and arginine residues in proteins react with reducing sugars and carbonyl compounds to form advanced glycation endproducts (AGEs). Several AGEs have been detected in human lenses, and their levels have been shown to increase with age and cataract formation. AGEs can lead to structural changes, yellow pigmentation, and cross-linking of lens proteins. Studies have shown a positive correlation between AGEs levels and lens age, stiffness, and cataracts. Recent research suggests that DJ-1 can inhibit the accumulation of AGEs in cellular proteins by reversing the early glycation steps. Our study found that DJ-1 is present in epithelial cells and the outer cortex of the human lens. DJ-1 is catalytically active in human lenses, and its ability to metabolize methylglyoxal (MGO) into D-lactate diminishes as the lens ages. The formation of MGH-1 from MGO was promoted in lens proteins treated with the DJ-1 inhibitor. Recombinant DJ-1 prevents α-dicarbonyl-mediated cross-linking and AGE accumulation in human αB-crystallin. DJ-1 prevents glyoxal-mediated cell death and AGE accumulation in human lens epithelial cells. Taken together, our results suggest that DJ-1 in the human lens hinders AGE accumulation. Enhancing its activity using pharmacological agents can potentially delay or prevent the onset of presbyopia and cataracts.

Kinetic and crystallographic studies reveal that the binding of the thiocholine-thionitrobenzoic acid product, released during the measurement of thioester-analog substrates hydrolysis according to Ellman's method, inhibits cholinesterases by a pure competitive mechanism. This can only be recorded as the progressive accumulation of the product upon subsequent additions of substrate aliquots. A wide affinity variation was observed among several tested enzymes, with the highest values found in human butyrylcholinesterase and Torpedo acetylcholinesterase. Nearly two orders of magnitude lower affinities were determined with human, mouse, and electrophorus acetylcholinesterases, and human atypical butyrylcholinesterase. These findings can be explained by the unexpected accommodation of the thiocholine-thionitrobenzoic acid in the active site of human butyrylcholinesterase, with the positively charged trimethylammonium choline pointing to the enzyme's peripheral site. At the same time, the carboxyl group of the nitrobenzoic moiety interacts with the enzyme's oxyanion hole. This explains the virtual absence of product inhibition in atypical human butyrylcholinesterase (D70G), purified or in plasma.

The nickel-pincer nucleotide (NPN) cofactor catalyzes the racemization/epimerization of α-hydroxy acids in enzymes of the LarA family. The established proton-coupled hydride transfer mechanism requires two catalytic histidine residues that alternately act as general acids and general bases. Notably, however, a fraction of LarA homologs (LarAHs) lack one of the active site histidine residues, replacing it with an asparaginyl side chain that cannot participate in acid/base catalysis. Here, we investigated two such LarAHs and solved their cryo-electron microscopic structures with and without loaded NPN cofactor, respectively. The structures revealed a consistent octameric assembly that is unprecedented in the LarA family and unveiled a new set of active site residues that likely recognize and process substrates differently from those of the well-studied LarAHs. Genomic context analysis suggested their potential involvement in carbohydrate metabolism. Together, these findings lay the groundwork for expanding the breadth of reactions and the range of mechanisms of LarA enzymes.

Ion channel activity is intricately linked to the surrounding lipid environment, yet the molecular effects of lipid-mediated regulation remain largely understudied. Here, we show that membrane-forming phospholipids, which are known to modulate the activity of the cyclic nucleotide-gated channel SthK from Spirochaeta thermophila, exhibit effects that extend well beyond the membrane boundary. Using stopped-flow flux assays, we demonstrate that anionic lipids, which are known to promote channel opening, also affect the fast-to-slow activation ratio and the cAMP potency in SthK. Enzymatic catalysis studies confirm that this occurs by altering the cis/trans equilibrium at Pro300 in the apo state. Additionally, cryogenic electron microscopy structures of SthK reveal lipid-dependent conformational changes that propagate from the bundle crossing into the cytosolic domains. All observed effects correlate with the electronegativity of the lipid headgroup, indicating a common underlying mechanism. Our results highlight membrane-forming phospholipids as allosteric regulators of SthK, controlling multiple functional characteristics of the channel.

Recent advances in topology-based modeling have greatly improved molecular prediction tasks, particularly in protein-ligand binding affinity. However, when the focus shifts to predicting protein-protein interactions (PPIs) binding free energy, the question becomes significantly more challenging due to the ineffective use of topological features and the lack of reliable datasets. In this work, we propose a persistent-Laplacian machine learning framework centered on the Persistent-Laplacian Neural Network (PLNet), which encodes each protein chain at the binding interface using both persistent Laplacian-based features and protein language model embeddings. It can achieve a promising Pearson correlation of 0.80 under leave-out-protein-out cross-validation on our newly assembled benchmark dataset, P2P, which includes 6886 protein complexes drawn from existing sources. For comparison, we also implement a gradient-boosting decision tree model under the same settings. This baseline method highlights the advantage of PLNet in capturing complex topology-aware descriptors in PPI prediction.

Targeted protein degradation using PROTACs (PROteolysis TArgeting Chimeras) has emerged as a transformative therapeutic strategy, largely relying on a small number of E3 ubiquitin ligases such as CRBN and VHL. However, resistance, toxicity, and poor oral bioavailability limit the utility of PROTACs and highlight the need to expand the E3 ligase toolbox. Fem-1 homolog B (FEM1B) is a lesser-known E3 ligase that offers a promising alternative due to its broad expression and ability to recognize diverse degron motifs. Here, we describe the development of a stable construct of FEM1B, the results of a protein-observed NMR-based fragment screen using this construct, and the X-ray structures of some of the fragment hits when bound to the protein. From these results, new PROTACs utilizing FEM1B as the E3 ligase may be discovered, providing an alternative E3 ligase for targeted protein degradation.

The Protein Data Bank (PDB) contains more than 235,000 three-dimensional biostructures and is growing at a rate of nearly 10% per year. The PDB is essential to gain knowledge on how proteins and ligands interact and how these interactions are correlated with the quantitative activity of each ligand/target pair. Unfortunately, the lack of a tool that can classify structures as apo or holo, that is by no means straightforward, and differentiate between covalent and non-covalent ligand-protein complexes makes it difficult to obtain the structures that belong to each set. To address this issue, we present PDB-CAT, a user-friendly tool that facilitates the categorization and extraction of key information from PDBx/mmCIF files through an efficient parallelized implementation. PDB-CAT uses a blacklist-based approach to automatically identify the ligand in a complex. It then classifies the PDB files based on ligand presence: structures without a ligand are classified as apo, whereas those with a ligand are classified as covalently or non-covalently bound, depending on the type of binding. As well as making this classification, the program can verify if there are any mutations in the protein sequence by comparing it to a reference sequence. An example is included to illustrate two different uses: the classification of SARS-CoV-2 Main Protease complexes depending on their variant, and the complete screening of the PDBbindv2020, achieved in <10 min. PDB-CAT is now available on GitHub (https://github.com/URV-cheminformatics/PDB-CAT) and the corresponding tutorial on GitBook (https://ariadnallopps-organization.gitbook.io/pdb-cat).