IDPpi: Protein-Protein Interaction Analyses of Human Intrinsically Disordered Proteins

Imaging-Based Quantitative Assessment of Biomolecular Condensates in vitro and in Cells

The formation of biomolecular condensates contributes to intracellular compartmentalization, and plays an important role in many cellular processes. The characterization of condensates is however challenging, requiring advanced biophysical or biochemical methods that are often less suitable for in vivo studies. A particular need for easily accessible yet thorough methods that enable the characterization of condensates across different experimental systems thus remains. To address this, we present PhaseMetrics, a semi-automated FIJI-based image analysis pipeline tailored for quantifying particle properties from microscopy data. Tested using the FG-domain of yeast nucleoporin Nup100, PhaseMetrics accurately assesses particle properties across diverse experimental setups, including particles formed in vitro in chemically defined buffers or in Xenopus egg extracts, and in cellular systems. Comparing the results with biochemical assays, we conclude that PhaseMetrics reliably detects changes induced by various conditions, including the presence of polyethylene glycol, 1,6-hexanediol, or a salt gradient, as well as the activity of the molecular chaperone DNAJB6b and the protein disaggregase Hsp104. Given the flexibility in its analysis parameters, the pipeline should also be applicable to other condensate-forming systems and we show it application for detecting TDP-43 particles. By enabling the accurate representation of the variability within the population and the detection of subtle changes at the single-condensate level, the method complements conventional biochemical assays. Combined, PhaseMetrics is an easily accessible, customizable pipeline that enables imaging-based quantitative assessment of biomolecular condensates in vitro and in cells, providing a valuable addition to the current toolbox.

SENSE-PPI reconstructs interactomes within, across, and between species at the genome scale

Ab initio computational reconstructions of protein-protein interaction (PPI) networks will provide invaluable insights into cellular systems, enabling the discovery of novel molecular interactions and elucidating biological mechanisms within and between organisms. Leveraging the latest generation protein language models and recurrent neural networks, we present SENSE-PPI, a sequence-based deep learning model that efficiently reconstructs ab initio PPIs, distinguishing partners among tens of thousands of proteins and identifying specific interactions within functionally similar proteins. SENSE-PPI demonstrates high accuracy, limited training requirements, and versatility in cross-species predictions, even with non-model organisms and human-virus interactions. Its performance decreases for phylogenetically more distant model and non-model organisms, but signal alteration is very slow. In this regard, it demonstrates the important role of parameters in protein language models. SENSE-PPI is very fast and can test 10,000 proteins against themselves in a matter of hours, enabling the reconstruction of genome-wide proteomes.

Extending MeCP2 interactome: canonical nucleosomal histones interact with MeCP2

MeCP2 is a general regulator of transcription involved in the repression/activation of genes depending on the local epigenetic context. It acts as a chromatin regulator and binds with exquisite specificity to gene promoters. The set of epigenetic marks recognized by MeCP2 has been already established (mainly, cytosine modifications in CpG and CpA), as well as many of the constituents of its interactome. We unveil a new set of interactions for MeCP2 with the four canonical nucleosomal histones. MeCP2 interacts with high affinity with H2A, H2B, H3 and H4. In addition, Rett syndrome associated mutations in MeCP2 and histone epigenetic marks modulate these interactions. Given the abundance and the structural/functional relevance of histones and their involvement in epigenetic regulation, this new set of interactions and its modulating elements provide a new addition to the 'alphabet' for this epigenetic reader.

Conformational changes in the negative arm of the circadian clock correlate with dynamic interactomes involved in post-transcriptional regulation

Biology is tuned to the Earth's diurnal cycle by the circadian clock, a transcriptional/translational negative feedback loop that regulates physiology via transcriptional activation and other post-transcriptional mechanisms. We hypothesize that circadian post-transcriptional regulation might stem from conformational shifts in the intrinsically disordered proteins that comprise the negative arm of the feedback loop to coordinate variation in negative-arm-centered macromolecular complexes. This work demonstrates temporal conformational fluidity in the negative arm that correlates with 24-h variation in physiologically diverse macromolecular complex components in eukaryotic clock proteins. Short linear motifs on the negative-arm proteins that correspond with the interactors localized to disordered regions and known temporal phosphorylation sites suggesting changes in these macromolecular complexes could be due to conformational changes imparted by the temporal phospho-state. Interactors that oscillate in the macromolecular complexes over circadian time correlate with post-transcriptionally regulated proteins, highlighting how time-of-day variation in the negative-arm protein complexes may tune cellular physiology.

Recent advances in predicting and modeling protein-protein interactions

Protein-protein interactions (PPIs) drive biological processes, and disruption of PPIs can cause disease. With recent breakthroughs in structure prediction and a deluge of genomic sequence data, computational methods to predict PPIs and model spatial structures of protein complexes are now approaching the accuracy of experimental approaches for permanent interactions and show promise for elucidating transient interactions. As we describe here, the key to this success is rich evolutionary information deciphered from thousands of homologous sequences that coevolve in interacting partners. This covariation signal, revealed by sophisticated statistical and machine learning (ML) algorithms, predicts physiological interactions. Accurate artificial intelligence (AI)-based modeling of protein structures promises to provide accurate 3D models of PPIs at a proteome-wide scale.

Characterization of intrinsically disordered regions in proteins informed by human genetic diversity

All proteomes contain both proteins and polypeptide segments that don’t form a defined three-dimensional structure yet are biologically active—called intrinsically disordered proteins and regions (IDPs and IDRs). Most of these IDPs/IDRs lack useful functional annotation limiting our understanding of their importance for organism fitness. Here we characterized IDRs using protein sequence annotations of functional sites and regions available in the UniProt knowledgebase (“UniProt features”: active site, ligand-binding pocket, regions mediating protein-protein interactions, etc.). By measuring the statistical enrichment of twenty-five UniProt features in 981 IDRs of 561 human proteins, we identified eight features that are commonly located in IDRs. We then collected the genetic variant data from the general population and patient-based databases and evaluated the prevalence of population and pathogenic variations in IDPs/IDRs. We observed that some IDRs tolerate 2 to 12-times more single amino acid-substituting missense mutations than synonymous changes in the general population. However, we also found that 37% of all germline pathogenic mutations are located in disordered regions of 96 proteins. Based on the observed-to-expected frequency of mutations, we categorized 34 IDRs in 20 proteins (DDX3X, KIT, RB1, etc.) as intolerant to mutation. Finally, using statistical analysis and a machine learning approach, we demonstrate that mutation-intolerant IDRs carry a distinct signature of functional features. Our study presents a novel approach to assign functional importance to IDRs by leveraging the wealth of available genetic data, which will aid in a deeper understating of the role of IDRs in biological processes and disease mechanisms.

JUPPI: A Multi-Level Feature Based Method for PPI Prediction and a Refined Strategy for Performance Assessment

Over the years, several methods have been proposed for the computational PPI prediction with different performance evaluation strategies. While attempting to benchmark performance scores, most of these methods often suffer with ill-treated cross-validation strategies, adhoc selection of positive/negative samples etc. To address these issues, in our proposed multi-level feature based PPI prediction approach (JUPPI), using sequence, domain and GO information as features, a refined evaluation strategy has been introduced. During the evaluation process, we first extract high quality negative data using three-stage filtering, and then introduce a pair-input based cross validation strategy with three difficulty levels for test-set predictions. Our proposed evaluation strategy reduces the component-level overlapping issue in test sets. Performance of JUPPI is compared with those of the state-of-the-art approaches in this domain and tested on six independent PPI datasets. In almost all the datasets, JUPPI outperforms the state-of-the-art not only at human proteome level for PPI prediction, but also for prediction of interactors for intrinsic disordered human proteins. https://figshare.com/projects/JUPPI_A_Multi-level_Feature_Based_Method_for_PPI_Prediction_and_a_Refined_Strategy_for_Performance_Assessment/81656 JUPPI tool and the developed datasets (JUPPId) are available in public domain for academic use along with supplementary materials, which can be found on the Computer Society Digital Library at http://doi.ieeecomputersociety.org/10.1109/TCBB.2020.3004970.

Combining enhanced sampling and deep learning dimensionality reduction for the study of the heat shock protein B8 and its pathological mutant K141E

The biological functions of proteins closely depend on their conformational dynamics. This aspect is especially relevant for intrinsically disordered proteins (IDP) for which structural ensembles often offer more useful representations than individual conformations. Here we employ extensive enhanced sampling temperature replica-exchange atomistic simulations (TREMD) and deep learning dimensionality reduction to study the conformational ensembles of the human heat shock protein B8 and its pathological mutant K141E, for which no experimental 3D structures are available. First, we combined homology modelling with TREMD to generate high-dimensional data sets of 3D structures. Then, we employed a recently developed machine learning based post-processing algorithm, EncoderMap, to project the large conformational data sets into meaningful two-dimensional maps that helped us interpret the data and extract the most significant conformations adopted by both proteins during TREMD. These studies provide the first 3D structural characterization of HSPB8 and reveal the effects of the pathogenic K141E mutation on its conformational ensembles. In particular, this missense mutation appears to increase the compactness of the protein and its structural variability, at the same time rearranging the hydrophobic patches exposed on the protein surface. These results offer the possibility of rationalizing the pathogenic effects of the K141E mutation in terms of conformational changes.

The study provides the first 3D structural characterization of HSPB8 and its K141E mutant: extensive TREMD are combined with a deep learning algorithm to rationalize the disordered ensemble of structures adopted by each variant.

The disordered PCI-binding human proteins CSNAP and DSS1 have diverged in structure and function

Intrinsically disordered proteins (IDPs) regularly constitute components of larger protein assemblies contributing to architectural stability. Two small, highly acidic IDPs have been linked to the so-called PCI complexes carrying PCI-domain subunits, including the proteasome lid and the COP9 signalosome. These two IDPs, DSS1 and CSNAP, have been proposed to have similar structural propensities and functions, but they display differences in their interactions and interactome sizes. Here we characterized the structural properties of human DSS1 and CSNAP at the residue level using NMR spectroscopy and probed their propensities to bind ubiquitin. We find that distinct structural features present in DSS1 are completely absent in CSNAP, and vice versa, with lack of relevant ubiquitin binding to CSNAP, suggesting the two proteins to have diverged in both structure and function. Our work additionally highlights that different local features of seemingly similar IDPs, even subtle sequence variance, may endow them with different functional traits. Such traits may underlie their potential to engage in multiple interactions thereby impacting their interactome sizes.

On the specificity of protein-protein interactions in the context of disorder

With the increased focus on intrinsically disordered proteins (IDPs) and their large interactomes, the question about their specificity — or more so on their multispecificity — arise. Here we recapitulate how specificity and multispecificity are quantified and address through examples if IDPs in this respect differ from globular proteins. The conclusion is that quantitatively, globular proteins and IDPs are similar when it comes to specificity. However, compared with globular proteins, IDPs have larger interactome sizes, a phenomenon that is further enabled by their flexibility, repetitive binding motifs and propensity to adapt to different binding partners. For IDPs, this adaptability, interactome size and a higher degree of multivalency opens for new interaction mechanisms such as facilitated exchange through trimer formation and ultra-sensitivity via threshold effects and ensemble redistribution. IDPs and their interactions, thus, do not compromise the definition of specificity. Instead, it is the sheer size of their interactomes that complicates its calculation. More importantly, it is this size that challenges how we conceptually envision, interpret and speak about their specificity.

Interaction between the guanylate kinase domain of PSD-95 and the proline-rich region and microtubule binding repeats 2 and 3 of tau

The microtubule-associated protein tau is a key factor in neurodegenerative proteinopathies and is predominantly found in the neuronal axon. However, somatodendritic localization of tau occurs for a subset of pathological and physiologic tau. Dendritic tau can localize to post-synapses where it interacts with proteins of the post-synaptic density (PSD) protein PSD-95, a membrane-associated guanylate kinase (MAGUK) scaffold factor for organization of protein complexes within the PSD, to mediate downstream signals. The sub-molecular details of this interaction, however, remain unclear. Here, we use interaction mapping in cultured cells to demonstrate that tau interacts with the guanylate kinase (GUK) domain in the C-terminal region of PSD-95. The PSD-95 GUK domain is required and sufficient for a complex with full-length human tau. Mapping the interaction of the MAGUK core on tau revealed the microtubule binding repeats 2 and 3 and the proline-rich region contribute to this interaction, while the N- and C-terminal regions of tau inhibit interaction. These results reveal intramolecular determinants of the protein complex of tau and PSD-95 and increase our understanding of tau interactions regulating neurotoxic signaling at the molecular level.

β-catenin regulates FOXP2 transcriptional activity via multiple binding sites

The transcription factor forkhead box protein P2 (FOXP2) is a highly conserved key regulator of embryonal development. The molecular mechanisms of how FOXP2 regulates embryonal development, however, remain elusive. Using RNA sequencing, we identified the Wnt signaling pathway as key target of FOXP2‐dependent transcriptional regulation. Using cell‐based assays, we show that FOXP2 transcriptional activity is regulated by the Wnt coregulator β‐catenin and that β‐catenin contacts multiple regions within FOXP2. Using nuclear magnetic resonance spectroscopy, we uncovered the molecular details of these interactions. β‐catenin contacts a disordered FOXP2 region with α‐helical propensity via its folded armadillo domain, whereas the intrinsically disordered β‐catenin N terminus and C terminus bind to the conserved FOXP2 DNA‐binding domain. Using RNA sequencing, we confirmed that β‐catenin indeed regulates transcriptional activity of FOXP2 and that the FOXP2 α‐helical motif acts as a key regulatory element of FOXP2 transcriptional activity. Taken together, our findings provide first insight into novel regulatory interactions and help to understand the intricate mechanisms of FOXP2 function and (mis)‐regulation in embryonal development and human diseases.

Database

Expression data are available in the GEO database under the accession number GSE138938.

Automated feature engineering improves prediction of protein-protein interactions

Over the last decade, various machine learning (ML) and statistical approaches for protein-protein interaction (PPI) predictions have been developed to help annotating functional interactions among proteins, essential for our system-level understanding of life. Efficient ML approaches require informative and non-redundant features. In this paper, we introduce novel types of expert-crafted sequence, evolutionary and graph features and apply automatic feature engineering to further expand feature space to improve predictive modeling. The two-step automatic feature-engineering process encompasses the hybrid method for feature generation and unsupervised feature selection, followed by supervised feature selection through a genetic algorithm (GA). The optimization of both steps allows the feature-engineering procedure to operate on a large transformed feature space with no considerable computational cost and to efficiently provide newly engineered features. Based on GA and correlation filtering, we developed a stacking algorithm GA-STACK for automatic ensembling of different ML algorithms to improve prediction performance. We introduced a unified method, HP-GAS, for the prediction of human PPIs, which incorporates GA-STACK and rests on both expert-crafted and 40% of newly engineered features. The extensive cross validation and comparison with the state-of-the-art methods showed that HP-GAS represents currently the most efficient method for proteome-wide forecasting of protein interactions, with prediction efficacy of 0.93 AUC and 0.85 accuracy. We implemented the HP-GAS method as a free standalone application which is a time-efficient and easy-to-use tool. HP-GAS software with supplementary data can be downloaded from: http://www.vinca.rs/180/tools/HP-GAS.php .

A Unique Family of Neuronal Signaling Proteins Implicated in Oncogenesis and Tumor Suppression

The neuronal proteins GAP43 (neuromodulin), MARCKS, and BASP1 are highly expressed in the growth cones of nerve cells where they are involved in signal transmission and cytoskeleton organization. Although their primary structures are unrelated, these signaling proteins share several structural properties like fatty acid modification, and the presence of cationic effector domains. GAP43, MARCKS, and BASP1 bind to cell membrane phospholipids, a process reversibly regulated by protein kinase C-phosphorylation or by binding to the calcium sensor calmodulin (CaM). GAP43, MARCKS, and BASP1 are also expressed in non-neuronal cells, where they may have important functions to manage cytoskeleton architecture, and in case of MARCKS and BASP1 to act as cofactors in transcriptional regulation. During neoplastic cell transformation, the proteins reveal differential expression in normal vs. tumor cells, and display intrinsic tumor promoting or tumor suppressive activities. Whereas GAP43 and MARCKS are oncogenic, tumor suppressive functions have been ascribed to BASP1 and in part to MARCKS depending on the cell type. Like MARCKS, the myristoylated BASP1 protein is localized both in the cytoplasm and in the cell nucleus. Nuclear BASP1 participates in gene regulation converting the Wilms tumor transcription factor WT1 from an oncoprotein into a tumor suppressor. The BASP1 gene is downregulated in many human tumor cell lines particularly in those derived from leukemias, which display elevated levels of WT1 and of the major cancer driver MYC. BASP1 specifically inhibits MYC-induced cell transformation in cultured cells. The tumor suppressive functions of BASP1 and MARCKS could be exploited to expand the spectrum of future innovative therapeutic approaches to inhibit growth and viability of susceptible human tumors.