Biology subjects

Perrakis, A.

Publications and source records attributed to Perrakis, A..

At least 19 recordsLinked to original sources

bioRxiv2026

So similar, yet so different: the paradigm of PARP9 macro domain paralogs

Human PARP9 harbours two tandem macro domains, MD1 and MD2, with distinct roles in ADP-ribosylation signaling. Whereas MD1 is a MacroD-type de-MARylase ("eraser"), MD2 functions as a MacroH2A-like "reader" lacking detectable hydrolase activity. How two domains so similar in sequence and fold achieve such divergent functionality has remained unclear. Our de-MARylation assays confirmed this division of labor, even though crystal structures revealed nearly identical /{beta}/ folds and similar binding pockets, and solution NMR shows broadly comparable dynamics. A remarkable distinction, however, emerges from isothermal titration calorimetry, showing that both domains bind free ADP-ribose with comparable affinity (KD = 5.4 and 8.4 M), but through markedly different thermodynamics: MD1 binding is enthalpy-driven and offset by a larger entropic penalty, whereas MD2 binds with weaker enthalpy and a smaller entropic cost. Our ADPr-bound crystal structures rationalized this showing that the distal ribose is positioned differently in the two pockets, connecting to a glycine-rich catalytic loop and a conserved aromatic residue present in active macro domains like MD1, but altered in MD2, while the catalytic asparagine itself is structurally conserved (Asn140/Asn339). Moreover, we show that a single amino acid substitution in MD2 leads to detectable RNA de-MARylation activity without altering its fold. Together, our findings show that an eraser-versus-reader distinction between the macrodomains in PARP9 is encoded in the dynamics of ligand binding and the exact position the distal ribose, revealing an unexpected catalytic plasticity relevant to PARP9s roles in immunity and cancer.

biochemistry↗

bioRxiv2025

Integrating computational protein structure predictions and genetic dependencies yields an atlas of human multi-protein complexes (AHMPC)

Knowledge of which proteins interact to form functional complexes in cells is essential for understanding molecular mechanisms in biology. Structure prediction methods recently allowed to compute the Human Interactome of likely binary protein interactions. We combine computational predictions with orthogonal functional data from the Dependency Map that estimate the correlation between genetic dependencies and vulnerabilities of the corresponding gene pairs. This revealed groups of proteins that likely form larger complexes. Clustering analysis followed by AlphaFold3 multi-protein complex predictions and AlphaBridge analysis provided the basis to construct and atlas of human multi-protein complexes (AHMPC), currently encompassing 354 high-confidence predicted multi-protein complexes. These include well-known assemblies and new ones - such as a complex involving SYS1, JTB, and ARFRP1 that we validate experimentally, suggesting an unexpected role of JTB in Golgi traficking. To enable the research community to explore the AHMPC and enable further discovery, we cluster all complexes using functional and disease-related embeddings, demonstrate how structured prompts allow validation by large language models (LLMs), and make all structures and analysis available online as an open resource at https://ahmpc.eu/.

biochemistry↗

bioRxiv2025

Halogen bonds between ligands and proteins: can we use them in validation?

Halogen bonds are polar interactions between a halogen atom and its acceptor. They have characteristic geometry induced by the positively charged {sigma}-hole on the halogen atom. Despite their importance in drug development, halogen bonds are often overlooked in the analysis and validation of ligand-protein complexes. We analysed halogen bonds between ligands and proteins in structure models from the PDB-REDO databank and defined key geometric parameters. These are the donor-acceptor distance, and two bond angles ({theta}1 and {theta}2) for interatomic halogen bonds, or in the case of halogen-{pi} interactions, the distance between the halogen and acceptor {pi}-system and a bond angle ({theta}1). Based on the distribution of these geometric parameters, we introduce a score, HalBS, that marks whether halogen bonds are adopting the preferred or allowed geometry or are an outlier and should be examined critically. A reference implementation of this score is now available in PDB-REDO. This is a first step towards improving halogen bond treatment in the analysis of macromolecular structure models.

biochemistry↗

bioRxiv2025

A functionally validated TCR-pMHC database for TCR specificity model development

Accurate prediction of TCR specificity forms a holy grail in immunology and large language models and computational structure predictions provide a path to achieve this. Importantly, current TCR-pMHC prediction models have been trained and evaluated using historical data of unknown quality. Here, we develop and utilize a high-throughput synthetic platform for TCR assembly and evaluation to assess a large fraction of VDJdb-deposited TCR-pMHC entries using a standardized readout of TCR function. Strikingly, this analysis demonstrates that claimed TCR reactivity is only confirmed for 50% of evaluated entries. Intriguingly, the use of TCRbridge to analyze AlphaFold3 confidence metrics reveals a substantial performance in distinguishing functionally validating and non-validating TCRs even though AlphaFold3 was not trained on this task, demonstrating the utility of the validated VDJdb (TCRvdb) database that we generated. We provide TCRvdb as a resource to the community to support training and evaluation of improved predictive TCR specificity models.

immunology↗

bioRxiv2025

DSSP 4: FAIR annotation of protein secondary structure

Protein secondary structure annotation is essential for understanding protein architecture, serving as a cornerstone for structural classification, alignment, visualisation, and machine learning applications. The Define Secondary Structure of Proteins (DSSP) algorithm has long been the standard for assigning secondary structure elements such as -helices, {beta}-sheets, and loops in protein models. Here, we introduce DSSP version 4, which recapitulates DSSP functionality in a modern computational framework, extending also to the detection of left-handed {kappa}-helices (Poly-Proline II helices). To align with the FAIR principles (Findable, Accessible, Interoperable, Reusable), DSSP 4 adopts mmCIF as its primary input and output format, while retaining compatibility with legacy PDB and DSSP formats. We applied this updated tool to analyse the distribution of secondary structure elements across the Protein Data Bank (PDB) differentiating structures from diverse experimental methods, revealing insights into the prevalence and length of secondary structure elements, including the newly annotated {kappa}-helices. The DSSP 4 software, databank, and server are freely accessible from https://pdb-redo.eu/dssp, ensuring broad utility and interoperability in structural biology research.

bioinformatics↗

bioRxiv2024

Disentangling the CHAOS of intrinsic disorder in human proteins

Most proteins consist of both folded domains and Intrinsically Disordered Regions (IDRs). However, the widespread occurrence of intrinsic disorder in human proteins, along with its characteristics, is often overlooked by the broader communities of structural and molecular biologists. Building on the MobiDB database of intrinsic disorder in proteins, here we develop a comprehensive dataset (Comprehensive analysis of Human proteins And their disOrdered Segments - CHAOS). We implement internally consistent definitions of disordered regions, and annotate general characteristics such as cellular location, essentiality, post-translational modifications, and predicted pathogenicity. Further, we cross-reference to structure predictions from AlphaFold. We find that most human proteins contain at least one disordered region, predominantly located at the protein termini. IDRs are less hydrophobic, enriched in post-translational modifications, and mutations in IDRs are predicted to be less pathogenic than in non-IDRs. Additionally, we discovered that proteins residing in different cellular locations possess distinct disorder profiles. Finally, the predicted AlphaFold models of proteins in CHAOS suggest that disordered regions and proteins are often predicted to adopt secondary structure. Hereby we enhance the visibility and understanding of intrinsic disorder in human proteins. Key messagesO_LIFour out of five human proteins contain one or more intrinsically disordered regions (IDRs). C_LIO_LIHalf of the IDRs are located at protein termini, but three quarters of all human proteins contain a terminal IDR. C_LIO_LIThe amount and location of disordered regions differs throughout cellular compartments. C_LIO_LIOne in five missense mutations in IDRs are likely pathogenic. C_LIO_LIAlphaFold predicts secondary structure elements within intrinsically disordered regions and fully disordered proteins. C_LI

bioinformatics↗

bioRxiv2024

AlphaBridge: tools for the analysis of predicted macromolecular complexes

Artificial intelligence (AI)-powered protein structure prediction has transformed how scientists explore macromolecular function. AI-based prediction of macromolecular complexes is increasingly used for evaluating the likelihood of proteins forming complexes with other proteins, nucleic acids, lipids, sugars, or small-molecule ligands. Efficient tools are needed to evaluate these predicted models. We introduce an approach based on combining the confidence metrics of AlphaFold3 to enable clustering of sequence motifs participating in binary interactions and subsequently in 3D interfaces of complexes. Interaction interfaces within confidence limits are finally visualised in 2D using chord diagrams and network graphs. The analysis and visualisation are implemented in a web tool, which links them with interactive graphics and summary tables of predicted interfaces and intermolecular interactions, including confidence scores. Finally, we demonstrate real-life examples of how AlphaBridge is used for providing an efficient way to assess and validate predicted protein complexes and interfaces. The reproducible, objective and automated procedures we present provide a straightforward critical assessment of structure prediction of biomolecular complexes, that should be consulted before conducting more resource-intensive analyses.

bioinformatics↗

bioRxiv2024

Condensin II activation by M18BP1

Condensin complexes promote the drastic spatial rearrangement of the genome upon mitotic entry. Condensin II initiates chromosome condensation in early mitosis. To prevent chromosome condensation during interphase, condensin II is inhibited by MCPH1, but the mechanism is unknown. Through genetic and proteomic approaches, we identify M18BP1, a protein previously associated with centromere identity, as a factor required for condensin II localization to chromatin. M18BP1 directly binds condensin IIs CAP-G2 subunit and competes with MCPH1 for binding. Upon mitotic entry, CDK1 mediated phosphorylation may promote a switch from MCPH1 to M18BP1 binding to activate condensin II. Our results identify a fundamental and evolutionarily conserved mechanism of condensin II activation.

cell biology↗

bioRxiv2024

Structural plasticity of 2A proteins in the Parechovirus family

Parechoviruses, including Parechovirus A that infects humans as well as Parechovirus B (formerly Ljungan virus) and Parechovirus C (formerly Sebokele virus) that infect rodents, belong to a group of picornaviruses whose 2A proteins, instead of being proteases, contain a conserved H-box and NC-motif and are homologous to a small cellular lipid-modifying enzyme (PLAAT3) that acts as a host factor, enabling the picornavirus life cycle. Despite the common evolutionary origin, 2AH/NC proteins and PLAAT3 have no conserved function, as the active site of the viral proteins cannot support catalysis. Here, we set out to find if all Parechovirus species share the structural rearrangement that destroys the active site configuration of the cellular enzyme. This has revealed a remarkable structural plasticity of these 2AH/NC proteins that arises not only from sequence differences between species, but also from differences in the length of the recombinantly expressed proteins, resulting in large structural rearrangements. These include rerouting of a large internal loop and repositioning of the C-terminal helix with respect to the central {beta}-sheet, and these in turn influence the oligomeric state of the protein. We discuss how this structural plasticity could correlate with the function of these proteins in the viral life cycle and how this could recapitulate the possible evolution of this protein from host factor to viral 2AH/NC protein, with new independent functions in RNA replication.

biochemistry↗

bioRxiv2024

Caskin2 is a novel talin and Abi1-binding protein that promotes cell motility

Talin couples the actomyosin cytoskeleton to integrins and transmits tension to the extracellular matrix. Talin also interacts with numerous additional proteins capable of modulating the actin-integrin linkage and thus downstream mechanosignaling cascades. Here, we demonstrate that the scaffold protein Caskin2 interacts directly with the R8 domain of talin through its C-terminal LD motif. Caskin2 also associates with the WAVE Regulatory Complex to promote cell migration in an Abi1-dependent manner. Furthermore, we demonstrate that the Caskin2-Abi1 interaction is regulated by growth factor-induced phosphorylation of Caskin2 on serine 878. In MCF7 and UACC893 cells, which contain an amplification of CASKIN2, Caskin2 localizes in plasma membrane-associated plaques and around focal adhesions in CMSCs. Taken together, our results identify Caskin2 as a novel talin-binding protein that may not only connect integrin-mediated adhesion to actin polymerization, but could also play a role in crosstalk between integrins and microtubules.

cell biology↗

bioRxiv2023

Distant sequence regions of JBP1 contribute to J-DNA binding

Base-J ({beta}-D-Glucopyranosyloxymethyluracil) is a modified DNA nucleotide that replaces 1% of thymine in kinetoplastid flagellates. The biosynthesis and maintenance of base-J depends on the base-J Binding Protein 1 (JBP1), that has a thymidine hydroxylase domain (THD) and a J-DNA binding domain (JDBD). How the THD synergizes with the JDBD to hydroxylate thymine in specific genomic sites, maintaining base-J during semi-conservative DNA replication, remains unclear. Here we present a crystal structure of the JDBD including a previously disordered DNA-contacting loop and use it as starting point for Molecular Dynamics (MD) simulations and computational docking studies to propose recognition models for JDBD binding to J-DNA. These models guided mutagenesis experiments, providing additional data for docking, which reveals a binding mode for JDBD onto J-DNA. This model, together with the crystallographic structure of the TET2 JBP1-homologue in complex with DNA and the AlphaFold model of full-length JBP1, allowed us to hypothesize that the flexible JBP1 N-terminus contributes to DNA-binding, which we confirmed experimentally. A high-resolution JBP1:J-DNA complex, which must involve conformational changes, would however need to be determined experimentally to further understand this unique underlying molecular mechanism that ensures replication of epigenetic information.

biochemistry↗

bioRxiv2022

A type IV Autotaxin inhibitor ameliorates acute liver injury and non-alcoholic steatohepatitis in mice

An important but rather underexplored pathway implicated in liver disease is the lysophosphatidic acid (LPA) signaling axis. LPA acts through G-protein coupled receptors inducing downstream signaling pathways related to cell proliferation, differentiation, and migration, and is predominantly produced by the extracellular phosphodiesterase, Autotaxin (ATX). ATX has gained significant attention lately with an impressive number of ATX inhibitors (type I-IV) reported. Here, we aim to evaluate the therapeutic potential of a (yet unexplored) type IV ATX inhibitor, Cpd17, in liver injury. In this study, we first confirmed the involvement of the ATX/LPA signaling axis in human and murine diseased livers. Thereafter, we evaluated the effects of Cpd17, in comparison with the classic type I ATX inhibitor PF8380, in vitro. While both inhibitors attenuated induced cell injury phenotypes as assessed using various assays and specific readout parameters in hepatocytes, macrophages, and hepatic stellate cells (HSCs), Cpd17 appeared more effective. This prompted us to characterize the mechanism of action of both inhibitors in situ and in vitro in macrophages and HSCs, demonstrating that Cpd17 was more potent in inhibiting relevant signaling pathways, namely RhoA-mediated cytoskeletal remodeling, and phosphorylation of MAPK/ERK and AKT/PKB. Finally, we investigated the therapeutic potential of Cpd17 in two liver disease mouse models, CCl4-induced acute liver injury and diet-induced non-alcoholic steatohepatitis. We demonstrate that Cpd17 has an excellent potential for reducing liver injury in both disease models in vivo. We conclude that ATX inhibition, by type IV inhibitor in particular, has an excellent potential for clinical application in liver diseases.

molecular biology↗

bioRxiv2022

Autotaxin facilitates selective LPA receptor signaling

Autotaxin (ATX; ENPP2) produces the lipid mediator lysophosphatidic acid (LPA) that signals through disparate EDG (LPA1-3) and P2Y (LPA4-6) G protein-coupled receptors. ATX/LPA promote several (patho)physiological processes, including in pulmonary fibrosis, thus serving as attractive drug targets. However, it remains unclear if clinical outcome depends on how different ATX inhibitors modulate the ATX/LPA signaling axis. Here, we show that inhibitors binding to the ATX "tunnel" specifically abrogate key aspects of ATX/LPA signaling. We find that the tunnel is essential for signaling efficacy and dictates cellular responses independent of ATX catalytic activity, with a preference for activation of P2Y LPA receptors. These responses are abrogated by tunnel-binding inhibitors, such as ziritaxestat, but not by inhibitors that exclusively target the active site, as shown in primary lung fibroblasts and a murine model of radiation-induced pulmonary fibrosis. Our results uncover a receptor-selective signaling mechanism for ATX, implying clinical benefit for tunnel-targeting ATX inhibitors. HighlightsATX is a dual-function protein acting as an LPA-producing enzyme and LPA chaperone. Structural integrity of the ATX tunnel is essential to mediate signaling functions. ATX-bound LPA signals preferentially via P2Y family LPA receptors. Occupancy of the ATX tunnel is crucial for ziritaxestat to exert inhibition in vivo.

cell biology↗

bioRxiv2022

Conformational transitions of the mitotic adaptor Spindly underlie its interaction with Dynein and Dynactin

Cytoplasmic Dynein 1, or Dynein, is a microtubule minus-end directed motor. Dynein motility requires Dynactin and a family of activating adaptors that stabilize the Dynein-Dynactin complex and promote regulated interactions with cargo in space and time. How activating adaptors limit Dynein activation to specialized subcellular locales is unclear. Here, we reveal that Spindly, a mitotic Dynein adaptor at the kinetochore corona, exists natively in a closed conformation that occludes binding of Dynein-Dynactin to its CC1 box and Spindly motif. A structure-based analysis identified various mutations promoting an open conformation of Spindly that binds Dynein-Dynactin. A region of Spindly downstream from the Spindly motif and not required for cargo binding faces the CC1 box and stabilizes the intramolecular closed conformation. This region is also required for robust kinetochore localization of Spindly, suggesting that kinetochores promote Spindly activation to recruit Dynein. This mechanism may be paradigmatic for Dynein activation by other adaptors at various cellular locales.

biochemistry↗

bioRxiv2021

AlphaFill: enriching the AlphaFold models with ligands and co-factors

Artificial intelligence (AI) methods for constructing structural models of proteins on the basis of their sequence are having a transformative effect in biomolecular sciences. The AlphaFold protein structure database makes available hundreds of thousands of protein structures. However, all these structures lack cofactors essential for their structural integrity and molecular function (e.g. hemoglobin lacks a bound heme), key ions essential for structural integrity (e.g. zinc-finger motifs) or catalysis (e.g. Ca2+ or Zn2+ in metalloproteases), and ligands that are important for biological function (e.g. kinase structures lack ADP or ATP). Here, we present AlphaFill, an algorithm based on sequence and structure similarity, to "transplant" such "missing" small molecules and ions from experimentally determined structures to predicted protein models. These publicly available structural annotations are mapped to predicted protein models, to help scientists interpret biological function and design experiments.

bioinformatics↗

bioRxiv2021

A tomato prolyl-4-hydroxylase causes relocation of abscission zone and alters abscission kinetics

The detachment of organs is controlled by highly regulated molecular mechanisms. The position of the tomato abscission zone (AZ) is defined by the ratio of the proximal to distal part of the pedicel. In this study, the ratio was altered due to a shift in the position of the AZ which was attributed to shorter and longer pedicels of SlP4H3 RNAi and OEX lines due to changes on cell division and expansion in AZ and distal part. This might be associated with LM2- and JIM8-AGPs which increased in OEX and decreased in RNAi lines throughout the pedicel. The JIM13 AGPs were downregulated in the flower AZ of OEX lines, pointing to a role on abscission regulation. In addition, Co-IP in flower AZ with SlP4H3-GFP fusion proteins showed interaction with LM2-, JIM13- and JIM8-epitopes suggesting proline hydroxylation by SlP4H3. The lower content of methyl-esterified HGs and higher of demethyl-esterified HGs in the AZs of RNAi lines might be responsible for increased rigidity of the AZ cell walls, accounting for the higher force required for AZ tissue detachment to occur. Moreover, ethylene-induced flower abscission was accelerated in the RNAi lines and delayed in OEX lines, while exactly the opposite response was observed in the red ripe fruit AZs. This was partly attributed to alterations in the expression of cell wall hydrolases. Overall, these results indicate that P4Hs might regulate molecular and structural features of cell walls in the AZ as well as abscission progression by regulating the structure and function of AGPs. Sentence SummaryAlterations on the expression of tomato prolyl 4 hydroxylase 3 shifts the position of the pedicel abscission zone and alters ethylene induced abscission progression according to the developmental context of the pedicel.

plant biology↗

bioRxiv2020

The MPS1 kinase NTE region has helical propensity and preferred conformations towards the TPR domain

The mitotic spindle assembly checkpoint (SAC) ensures accurate segregation of chromosomes by preventing onset of anaphase until all chromosomes are properly attached to spindle microtubules. The Monopolar spindle 1 (MPS1) kinase is one of the SAC components, localizing at unattached kinetochores by an N-terminal localization module. This module comprises a flexible NTE module and the TPR domain, which we previously characterized for their contribution to kinetochore binding. Here we discuss the conformations of the highly flexible NTE with respect to the TPR domain, using paramagnetic NMR. The distance restraints derived from paramagnetic relaxation enhancements (PREs) show that the mobile NTE can be found in proximity of a large but specific part of the surface area of the TPR domain. To sample the conformational space of the NTE in the context of the NTE-TPR module, we used the ab initio Rosetta approach supplemented by paramagnetic NMR restraints. We find that many NTE residues have a propensity to form helical structures and that the module localizes at the convex surface of the TPR domain. This work demonstrates the highly dynamic nature of the interactions between the NTE and TPR domains and it shows that the convex rather than the canonical concave TPR surface mediates interactions, leading to the auto-inhibition that the TPR exerts upon the NTE region in the context of SAC signaling.

biochemistry↗

bioRxiv2020

A global Ramachandran score identifies protein structures with unlikely stereochemistry

Ramachandran plots report the distribution of the ({varphi}, {Psi}) torsion angles of the protein backbone and are one of the best quality metrics of experimental structure models. Typically, validation software reports the number of residues belonging to "outlier", "allowed" and "favored" regions. While "zero unexplained outliers" can be considered the current "gold standard", this can be misleading if deviations from expected distributions, even within the favored region, are not considered. We therefore revisited the Ramachandran Z-score (Rama-Z), a quality metric introduced more than two decades ago, but underutilized. We describe a re-implementation of the Rama-Z score in the Computational Crystallography Toolbox along with a new algorithm to estimate its uncertainty for individual models; final implementations are available both in Phenix and in PDB-REDO. We discuss the interpretation of the Rama-Z score and advocate including it in the validation reports provided by the Protein Data Bank. We also advocate reporting it alongside the outlier/allowed/favored counts in structural publications.

biophysics↗