A chloroplast protein atlas reveals punctate structures and spatial organization of biosynthetic pathways.

Chloroplasts are eukaryotic photosynthetic organelles that drive the global carbon cycle. Despite their importance, our understanding of their protein composition, function, and spatial organization remains limited. Here, we determined the localizations of 1,034 candidate chloroplast proteins using fluorescent protein tagging in the model alga Chlamydomonas reinhardtii. The localizations provide insights into the functions of poorly characterized proteins; identify novel components of nucleoids, plastoglobules, and the pyrenoid; and reveal widespread protein targeting to multiple compartments. We discovered and further characterized cellular organizational features, including eleven chloroplast punctate structures, cytosolic crescent structures, and unexpected spatial distributions of enzymes within the chloroplast. We also used machine learning to predict the localizations of other nuclear-encoded Chlamydomonas proteins. The strains and localization atlas developed here will serve as a resource to accelerate studies of chloroplast architecture and functions.

Systematic discovery of mutation-directed neo-protein-protein interactions in cancer.

Comprehensive sequencing of patient tumors reveals genomic mutations across tumor types that enable tumorigenesis and progression. A subset of oncogenic driver mutations results in neomorphic activity where the mutant protein mediates functions not engaged by the parental molecule. Here, we identify prevalent variant-enabled neomorph-protein-protein interactions (neoPPI) with a quantitative high-throughput differential screening (qHT-dS) platform. The coupling of highly sensitive BRET biosensors with miniaturized coexpression in an ultra-HTS format allows large-scale monitoring of the interactions of wild-type and mutant variant counterparts with a library of cancer-associated proteins in live cells. The screening of 17,792 interactions with 2,172,864 data points revealed a landscape of gain of interactions encompassing both oncogenic and tumor suppressor mutations. For example, the recurrent BRAF V600E lesion mediates KEAP1 neoPPI, rewiring a BRAFV600E/KEAP1 signaling axis and creating collateral vulnerability to NQO1 substrates, offering a combination therapeutic strategy. Thus, cancer genomic alterations can create neo-interactions, informing variant-directed therapeutic approaches for precision medicine.

Tau interactome maps synaptic and mitochondrial processes associated with neurodegeneration.

Tau (MAPT) drives neuronal dysfunction in Alzheimer disease (AD) and other tauopathies. To dissect the underlying mechanisms, we combined an engineered ascorbic acid peroxidase (APEX) approach with quantitative affinity purification mass spectrometry (AP-MS) followed by proximity ligation assay (PLA) to characterize Tau interactomes modified by neuronal activity and mutations that cause frontotemporal dementia (FTD) in human induced pluripotent stem cell (iPSC)-derived neurons. We established interactions of Tau with presynaptic vesicle proteins during activity-dependent Tau secretion and mapped the Tau-binding sites to the cytosolic domains of integral synaptic vesicle proteins. We showed that FTD mutations impair bioenergetics and markedly diminished Tau's interaction with mitochondria proteins, which were downregulated in AD brains of multiple cohorts and correlated with disease severity. These multimodal and dynamic Tau interactomes with exquisite spatial resolution shed light on Tau's role in neuronal function and disease and highlight potential therapeutic targets to block Tau-mediated pathogenesis.

A Human IgSF Cell-Surface Interactome Reveals a Complex Network of Protein-Protein Interactions.

Cell-surface protein-protein interactions (PPIs) mediate cell-cell communication, recognition, and responses. We executed an interactome screen of 564 human cell-surface and secreted proteins, most of which are immunoglobulin superfamily (IgSF) proteins, using a high-throughput, automated ELISA-based screening platform employing a pooled-protein strategy to test all 318,096 PPI combinations. Screen results, augmented by phylogenetic homology analysis, revealed ∼380 previously unreported PPIs. We validated a subset using surface plasmon resonance and cell binding assays. Observed PPIs reveal a large and complex network of interactions both within and across biological systems. We identified new PPIs for receptors with well-characterized ligands and binding partners for "orphan" receptors. New PPIs include proteins expressed on multiple cell types and involved in diverse processes including immune and nervous system development and function, differentiation/proliferation, metabolism, vascularization, and reproduction. These PPIs provide a resource for further biological investigation into their functional relevance and may offer new therapeutic drug targets.

A Global Map of G Protein Signaling Regulation by RGS Proteins.

The control over the extent and timing of G protein signaling is provided by the regulator of G protein signaling (RGS) proteins that deactivate G protein α subunits (Gα). Mammalian genomes encode 20 canonical RGS and 16 Gα genes with key roles in physiology and disease. To understand the principles governing the selectivity of Gα regulation by RGS, we examine the catalytic activity of all canonical human RGS proteins and their selectivity for a complete set of Gα substrates using real-time kinetic measurements in living cells. The data reveal rules governing RGS-Gα recognition, the structural basis of its selectivity, and provide principles for engineering RGS proteins with defined selectivity. The study also explores the evolution of RGS-Gα selectivity through ancestral reconstruction and demonstrates how naturally occurring non-synonymous variants in RGS alter signaling. These results provide a blueprint for decoding signaling selectivity and advance our understanding of molecular recognition principles.

Mutations in Disordered Regions Can Cause Disease by Creating Dileucine Motifs.

Many disease-causing missense mutations affect intrinsically disordered regions (IDRs) of proteins, but the molecular mechanism of their pathogenicity is enigmatic. Here, we employ a peptide-based proteomic screen to investigate the impact of mutations in IDRs on protein-protein interactions. We find that mutations in disordered cytosolic regions of three transmembrane proteins (GLUT1, ITPR1, and CACNA1H) lead to an increased clathrin binding. All three mutations create dileucine motifs known to mediate clathrin-dependent trafficking. Follow-up experiments on GLUT1 (SLC2A1), the glucose transporter causative of GLUT1 deficiency syndrome, revealed that the mutated protein mislocalizes to intracellular compartments. Mutant GLUT1 interacts with adaptor proteins (APs) in vitro, and knocking down AP-2 reverts the cellular mislocalization and restores glucose transport. A systematic analysis of other known disease-causing variants revealed a significant and specific overrepresentation of gained dileucine motifs in structurally disordered cytosolic domains of transmembrane proteins. Thus, several mutations in disordered regions appear to cause "dileucineopathies."

Mitochondrial Sirtuin Network Reveals Dynamic SIRT3-Dependent Deacetylation in Response to Membrane Depolarization.

Mitochondrial sirtuins, SIRT3-5, are NAD+-dependent deacylases and ADP-ribosyltransferases that are critical for stress responses. However, a comprehensive understanding of sirtuin targets, regulation of sirtuin activity, and the relationships between sirtuins remains a key challenge in mitochondrial physiology. Here, we employ systematic interaction proteomics to elucidate the mitochondrial sirtuin protein interaction landscape. This work reveals sirtuin interactions with numerous functional modules within mitochondria, identifies candidate sirtuin substrates, and uncovers a fundamental role for sequestration of SIRT3 by ATP synthase in mitochondrial homeostasis. In healthy mitochondria, a pool of SIRT3 binds ATP synthase, but upon matrix pH reduction with concomitant loss of mitochondrial membrane potential, SIRT3 dissociates. This release correlates with rapid deacetylation of matrix proteins, and SIRT3 is required for recovery of membrane potential. In vitro reconstitution experiments, as well as analysis of CRISPR/Cas9-engineered cells, indicate that pH-dependent SIRT3 release requires H135 in the ATP5O subunit of ATP synthase. Our SIRT3-5 interaction network provides a framework for discovering novel biological functions regulated by mitochondrial sirtuins.

HydRA: Deep-learning models for predicting RNA-binding capacity from protein interaction association context and protein sequence.

RNA-binding proteins (RBPs) control RNA metabolism to orchestrate gene expression and, when dysfunctional, underlie human diseases. Proteome-wide discovery efforts predict thousands of RBP candidates, many of which lack canonical RNA-binding domains (RBDs). Here, we present a hybrid ensemble RBP classifier (HydRA), which leverages information from both intermolecular protein interactions and internal protein sequence patterns to predict RNA-binding capacity with unparalleled specificity and sensitivity using support vector machines (SVMs), convolutional neural networks (CNNs), and Transformer-based protein language models. Occlusion mapping by HydRA robustly detects known RBDs and predicts hundreds of uncharacterized RNA-binding associated domains. Enhanced CLIP (eCLIP) for HydRA-predicted RBP candidates reveals transcriptome-wide RNA targets and confirms RNA-binding activity for HydRA-predicted RNA-binding associated domains. HydRA accelerates construction of a comprehensive RBP catalog and expands the diversity of RNA-binding associated domains.

E3 ligase autoinhibition by C-degron mimicry maintains C-degron substrate fidelity.

E3 ligase recruitment of proteins containing terminal destabilizing motifs (degrons) is emerging as a major form of regulation. How those E3s discriminate bona fide substrates from other proteins with terminal degron-like sequences remains unclear. Here, we report that human KLHDC2, a CRL2 substrate receptor targeting C-terminal Gly-Gly degrons, is regulated through interconversion between two assemblies. In the self-inactivated homotetramer, KLHDC2's C-terminal Gly-Ser motif mimics a degron and engages the substrate-binding domain of another protomer. True substrates capture the monomeric CRL2KLHDC2, driving E3 activation by neddylation and subsequent substrate ubiquitylation. Non-substrates such as NEDD8 bind KLHDC2 with high affinity, but its slow on rate prevents productive association with CRL2KLHDC2. Without substrate, neddylated CRL2KLHDC2 assemblies are deactivated via distinct mechanisms: the monomer by deneddylation and the tetramer by auto-ubiquitylation. Thus, substrate specificity is amplified by KLHDC2 self-assembly acting like a molecular timer, where only bona fide substrates may bind before E3 ligase inactivation.

The chemokine CCL1 triggers an AMFR-SPRY1 pathway that promotes differentiation of lung fibroblasts into myofibroblasts and drives pulmonary fibrosis.

Recruitment of immune cells to the site of inflammation by the chemokine CCL1 is important in the pathology of inflammatory diseases. Here, we examined the role of CCL1 in pulmonary fibrosis (PF). Bronchoalveolar lavage fluid from PF mouse models contained high amounts of CCL1, as did lung biopsies from PF patients. Immunofluorescence analyses revealed that alveolar macrophages and CD4+ T cells were major producers of CCL1 and targeted deletion of Ccl1 in these cells blunted pathology. Deletion of the CCL1 receptor Ccr8 in fibroblasts limited migration, but not activation, in response to CCL1. Mass spectrometry analyses of CCL1 complexes identified AMFR as a CCL1 receptor, and deletion of Amfr impaired fibroblast activation. Mechanistically, CCL1 binding triggered ubiquitination of the ERK inhibitor Spry1 by AMFR, thus activating Ras-mediated profibrotic protein synthesis. Antibody blockade of CCL1 ameliorated PF pathology, supporting the therapeutic potential of targeting this pathway for treating fibroproliferative lung diseases.

Revealing protein-protein interactions at the transcriptome scale by sequencing.

We describe PROPER-seq (protein-protein interaction sequencing) to map protein-protein interactions (PPIs) en masse. PROPER-seq first converts transcriptomes of input cells into RNA-barcoded protein libraries, in which all interacting protein pairs are captured through nucleotide barcode ligation, recorded as chimeric DNA sequences, and decoded at once by sequencing and mapping. We applied PROPER-seq to human embryonic kidney cells, T lymphocytes, and endothelial cells and identified 210,518 human PPIs (collected in the PROPER v.1.0 database). Among these, 1,365 and 2,480 PPIs are supported by published co-immunoprecipitation (coIP) and affinity purification-mass spectrometry (AP-MS) data, 17,638 PPIs are predicted by the prePPI algorithm without previous experimental validation, and 100 PPIs overlap human synthetic lethal gene pairs. In addition, four previously uncharacterized interaction partners with poly(ADP-ribose) polymerase 1 (PARP1) (a critical protein in DNA repair) known as XPO1, MATR3, IPO5, and LEO1 are validated in vivo. PROPER-seq presents a time-effective technology to map PPIs at the transcriptome scale, and PROPER v.1.0 provides a rich resource for studying PPIs.

Taking Me away: the function of phosphorylation on histone lysine demethylases.

Histone lysine demethylases (KDMs) regulate eukaryotic gene transcription by catalysing the removal of methyl groups from histone proteins. These enzymes are intricately regulated by the kinase signalling system in response to internal and external stimuli. Here, we review the mechanisms by which kinase-mediated phosphorylation influence human histone KDM function. These include the changing of histone KDM subcellular localisation or chromatin binding, the altering of protein half-life, changes to histone KDM complex formation that result in histone demethylation, non-histone demethylation or demethylase-independent effects, and effects on histone KDM complex dissociation. We also explore the structural context of phospho-sites on histone KDMs and evaluate how this relates to function.

Revealing eEF-2 kinase: recent structural insights into function.

The α-kinase eukaryotic elongation factor 2 kinase (eEF-2K) regulates translational elongation by phosphorylating its ribosome-associated substrate, the GTPase eEF-2. eEF-2K is activated by calmodulin (CaM) through a distinctive mechanism unlike that in other CaM-dependent kinases (CAMK). We describe recent structural insights into this unique activation process and examine the effects of specific regulatory signals on this mechanism. We also highlight key unanswered questions to guide future structure-function studies. These include structural mechanisms which enable eEF-2K to interact with upstream/downstream partners and facilitate its integration of diverse inputs, including Ca2+ transients, phosphorylation mediated by energy/nutrient-sensing pathways, pH changes, and metabolites. Answering these questions is key to establishing how eEF-2K harmonizes translation with cellular requirements within the boundaries of its molecular landscape.

Elucidating the assembly of gas vesicles by systematic protein-protein interaction analysis.

Gas vesicles (GVs) are gas-filled microbial organelles formed by unique 3-nm thick, amphipathic, force-bearing protein shells, which can withstand multiple atmospheric pressures and maintain a physically stable air bubble with megapascal surface tension. However, the molecular process of GV assembly remains elusive. To begin understanding this process, we have devised a high-throughput in vivo assay to determine the interactions of all 11 proteins in the pNL29 GV operon. Complete or partial deletions of the operon establish interdependent relationships among GV proteins during assembly. We also examine the tolerance of the GV assembly process to protein mutations and the cellular burdens caused by GV proteins. Clusters of GV protein interactions are revealed, proposing plausible protein complexes that are important for GV assembly. We anticipate our findings will set the stage for designing GVs that efficiently assemble in heterologous hosts during biomedical applications.

Targeting Degradation of the Transcription Factor C/EBPß Reduces Lung Fibrosis by Restoring Activity of the Ubiquitin-Editing Enzyme A20 in Macrophages.

Although recent progress provides mechanistic insights into the pathogenesis of pulmonary fibrosis (PF), rare anti-PF therapeutics show definitive promise for treating this disease. Repeated lung epithelial injury results in injury-repairing response and inflammation, which drive the development of PF. Here, we report that chronic lung injury inactivated the ubiquitin-editing enzyme A20, causing progressive accumulation of the transcription factor C/EBPß in alveolar macrophages (AMs) from PF patients and mice, which upregulated a number of immunosuppressive and profibrotic factors promoting PF development. In response to chronic lung injury, elevated glycogen synthase kinase-3ß (GSK-3ß) interacted with and phosphorylated A20 to suppress C/EBPß degradation. Ectopic expression of A20 or pharmacological restoration of A20 activity by disturbing the A20-GSK-3ß interaction accelerated C/EBPß degradation and showed potent therapeutic efficacy against experimental PF. Our study indicates that a regulatory mechanism of the GSK-3ß-A20-C/EBPß axis in AMs may be a potential target for treating PF and fibroproliferative lung diseases.

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.

p53 protects against alcoholic fatty liver disease via ALDH2 inhibition.

The tumor suppressor p53 is critical for tumor suppression, but the regulatory role of p53 in alcohol-induced fatty liver remains unclear. Here, we show a role for p53 in regulating ethanol metabolism via acetaldehyde dehydrogenase 2 (ALDH2), a key enzyme responsible for the oxidization of alcohol. By repressing ethanol oxidization, p53 suppresses intracellular levels of acetyl-CoA and histone acetylation, leading to the inhibition of the stearoyl-CoA desaturase-1 (SCD1) gene expression. Mechanistically, p53 directly binds to ALDH2 and prevents the formation of its active tetramer and indirectly limits the production of pyruvate that promotes the activity of ALDH2. Notably, p53-deficient mice exhibit increased lipid accumulation, which can be reversed by ALDH2 depletion. Moreover, liver-specific knockdown of SCD1 alleviates ethanol-induced hepatic steatosis caused by p53 loss. By contrast, overexpression of SCD1 in liver promotes ethanol-induced fatty liver development in wild-type mice, while it has a mild effect on p53-/- or ALDH2-/- mice. Overall, our findings reveal a previously unrecognized function of p53 in alcohol-induced fatty liver and uncover pyruvate as a natural regulator of ALDH2.

RRM Transcription Factors Interact with NLRs and Regulate Broad-Spectrum Blast Resistance in Rice.

Nucleotide-binding site leucine-rich repeat (NLR) receptors perceive pathogen effectors and trigger plant immunity. However, the mechanisms underlying NLR-triggered defense responses remain obscure. The recently discovered Pigm locus in rice encodes a cluster of NLRs, including PigmR, which confers broad-spectrum resistance to blast fungus. Here, we identify PIBP1 (PigmR-INTERACTING and BLAST RESISTANCE PROTEIN 1), an RRM (RNA-recognition motif) protein that specifically interacts with PigmR and other similar NLRs to trigger blast resistance. PigmR-promoted nuclear accumulation of PIBP1 ensures full blast resistance. We find that PIBP1 and a homolog, Os06 g02240, bind DNA and function as unconventional transcription factors at the promoters of the defense genes OsWAK14 and OsPAL1, activating their expression. Knockout of PIBP1 and Os06 g02240 greatly attenuated blast resistance. Collectively, our study discovers previously unappreciated RRM transcription factors that directly interact with NLRs to activate plant defense, establishing a direct link between transcriptional activation of immune responses with NLR-mediated pathogen perception.

TUnA: an uncertainty-aware transformer model for sequence-based protein-protein interaction prediction.

Protein-protein interactions (PPIs) are important for many biological processes, but predicting them from sequence data remains challenging. Existing deep learning models often cannot generalize to proteins not present in the training set and do not provide uncertainty estimates for their predictions. To address these limitations, we present TUnA, a Transformer-based uncertainty-aware model for PPI prediction. TUnA uses ESM-2 embeddings with Transformer encoders and incorporates a Spectral-normalized Neural Gaussian Process. TUnA achieves state-of-the-art performance and, importantly, evaluates uncertainty for unseen sequences. We demonstrate that TUnA's uncertainty estimates can effectively identify the most reliable predictions, significantly reducing false positives. This capability is crucial in bridging the gap between computational predictions and experimental validation.

Cracking the black box of deep sequence-based protein-protein interaction prediction.

Identifying protein-protein interactions (PPIs) is crucial for deciphering biological pathways. Numerous prediction methods have been developed as cheap alternatives to biological experiments, reporting surprisingly high accuracy estimates. We systematically investigated how much reproducible deep learning models depend on data leakage, sequence similarities and node degree information, and compared them with basic machine learning models. We found that overlaps between training and test sets resulting from random splitting lead to strongly overestimated performances. In this setting, models learn solely from sequence similarities and node degrees. When data leakage is avoided by minimizing sequence similarities between training and test set, performances become random. Moreover, baseline models directly leveraging sequence similarity and network topology show good performances at a fraction of the computational cost. Thus, we advocate that any improvements should be reported relative to baseline methods in the future. Our findings suggest that predicting PPIs remains an unsolved task for proteins showing little sequence similarity to previously studied proteins, highlighting that further experimental research into the 'dark' protein interactome and better computational methods are needed.