Accurate prediction of inhibitor selectivity across protein paralogues remains a central challenge in computational drug discovery. Here, we perform a comparative assessment of three computational methods─Molecular Mechanics/Poisson-Boltzmann Surface Area (MM/PBSA), Absolute Binding Free Energy (ABFE) and Umbrella Sampling (US) calculations─in their ability to recapitulate PARP1 versus PARP2 selectivity for eight clinically relevant PARP enzyme inhibitors used in ovarian, breast, and prostate tumors, among others. We demonstrate how MM/PBSA calculations offer rapid and qualitative insights but show pronounced sensitivity to the chosen static conformational pose, being particularly challenging for ligands with subtle energetic differences between distinct protein paralogues. In contrast, both ABFE and US calculations using atomistic models with explicit solvent result in substantially improved agreement with experimental binding affinities. The ABFE method exhibits the strongest quantitative correlation with experimental binding free energy differences, remarkably reproducing selectivity trends even among nearly isoenergetic complexes. Notably, our structural contact analysis reveals how contact connectivity controls ligand selectivity, providing valuable mechanistic and molecular insight into the key residues that stabilize each inhibitor in both protein enzymes. Together, our multimethod computational study contributes to elucidating potential chemical modifications across the ligand chemical space to enhance potency and specificity, informing the future design and evaluation of selective inhibitors for precision oncology, including therapies targeting homologous recombination-deficient cancers.

Congenital dyserythropoietic anemia type II (CDAII) is an autosomal recessive disease resulting from loss-of-function mutations in SEC23 homolog B (SEC23B). We have previously shown that increased expression of SEC23A, a paralogous protein for SEC23B, rescues the CDAII erythroid defect. Here, we generated a human erythroid cell line that expresses enhanced green fluorescent protein (eGFP) from the endogenous SEC23A locus and performed a small-molecule screen to identify compounds that increased SEC23A-eGFP abundance. The top compound passing all filters was an inhibitor of lysine-specific demethylase 1 (LSD1). We found that LSD1 inhibition with RN1 resulted in increased SEC23A expression in erythroid cells derived from human hematopoietic stem and progenitor cells (HSPCs) at doses that did not impair erythroid cell growth or differentiation and rescued the erythroid defect resulting from SEC23B deletion. Genetic down-regulation of LSD1 led to a marked increase in SEC23A mRNA expression in HSPC-derived erythroid cells. Deletion of Lsd1 in mouse erythroid cells resulted in increased Sec23a expression, and RN1 treatment ameliorated the erythroid defect observed in a CDAII mouse model. Mechanistically, we found that LSD1 occupied a sequence in the SEC23A promoter, repressing SEC23A transcription. Deletion of the promotor sequence occupied by LSD1 resulted in increased SEC23A expression and amelioration of CDAII. These findings highlight that LSD1 represses SEC23A transcription and that LSD1 inhibition results in de-repression of SEC23A expression and amelioration of the CDAII erythroid defect, suggesting promising therapeutic strategies for CDAII.

The EMBRYO DEFECTIVE 2207 (EMB2207) gene, encoding ribosomal protein UL3Z, is critical for embryonic development in Arabidopsis, with loss of function resulting in embryo lethality. Despite its importance, the role of UL3Z in the complicated protein translation machinery in plants remains poorly understood due to the lack of viable hypomorphic alleles. In this study, we utilized CRISPR/Cas9 to edit the 5' untranslated region (5'UTR) of UL3Z, generating 5 ul3z mutants with varying degrees of reduced expression levels of UL3Z proteins. The mutant with the lowest expression exhibited the most severe developmental defects. In contrast, null mutants of its paralog UL3Y displayed no observable phenotypes. Interestingly, expression of UL3Y driven by the UL3Z/EMB2207 promoter successfully rescued the phenotypes of ul3z, demonstrating that these 2 paralogous ribosomal proteins actually possess functionally interchangeable roles. GUS staining results showed that UL3Z was constitutively expressed in all examined tissues, while UL3Y was only appreciably expressed in specific tissues. Molecular analysis further revealed the accumulation of ribosomal RNA (rRNA) maturation intermediates and increased polysome levels in ul3z mutants, indicating compromised pre-rRNA processing and disturbed global mRNA translation. Interestingly, 3' ends of many rRNA precursors in ul3z had higher frequency of non-encoded tails compared with Col-0. This study demonstrates that CRISPR/Cas9-mediated 5'UTR editing is an effective tool for generating viable hypomorphic alleles of lethal genes and uncovers the critical roles of UL3Z/EMB2207 in pre-rRNA processing and the maintenance of appropriate mRNA translation on ribosomes, underscoring its importance in plant development.

Cell surface proteins are key disease biomarkers and therapeutic targets, yet high-throughput methods for aptamer discovery targeting these proteins in situ remain limited. We introduce single-cell perturbation-driven aptamer recognition and kinetics sequencing (SPARK-seq), a high-throughput platform integrating single-cell messenger RNA and aptamer sequencing with CRISPR-based surface protein perturbation. In a single experiment, SPARK-seq simultaneously mapped 5535 distinct aptamers to eight surface proteins, capturing interactions across more than two orders of magnitude in protein abundance and spanning diverse biophysical classes. The method discriminated closely related paralogous proteins with no detectable cross-reactivity and provided kinetic information that enabled the prioritization of aptamers with slow dissociation rates. Leveraging this kinetic diversity, we engineered variants with improved off-rate properties. SPARK-seq establishes a platform for high-efficiency discovery and rational variant design of aptamers and functional nucleic acids, unlocking possibilities in diagnostics and therapeutics.

Paralogs engage in biological processes through both redundant and nonredundant functions. In the yeast Saccharomyces cerevisiae, approximately one-fifth of the genome consists of paralogs, with their encoded proteins involved in multiple pathways. However, the unique contributions of individual paralogs have remained poorly defined. Here, we undertook a systematic examination of eight paralog pairs in the glucose-sensing pathways, deleting each component and measuring the resulting changes in gene expression. To that end, we established a new transcription reporter system to monitor the response to glucose as well as to nonpreferred sugars in single cells. Focusing on the PKA catalytic subunits, comprised of the paralogs Tpk1 and Tpk3 as well as the isomorphic kinase Tpk2, we employed mass spectrometry to identify their contribution to cellular metabolism, used a GFP-based sensor to follow changes in cytosolic pH, and used BioID to identify unique and shared candidate binding partners. Our data reveal that paralogs in the glucose-sensing pathway contribute in multiple and unique ways to signal transduction, and establish potential mechanisms driving the preservation of these and other duplicated genes throughout long periods of evolution.

Ribosomes catalyze all protein synthesis, and mutations altering their levels and function underlie many developmental diseases and cancer. Historically considered to be invariant machines, ribosomes differ in composition between tissues and developmental stages, incorporating a diversity of ribosomal proteins (RPs) encoded by duplicated paralogous genes. Here, we use Drosophila to systematically investigate the origins and functions of noncanonical RP paralogs. We show that new paralogs mainly originated through retroposition and that only a few new copies retain coding capacity over time. Although transcriptionally active noncanonical RP paralogs often present tissue-specific expression, we show that the majority of those are not required for either viability or fertility in Drosophila melanogaster. The only exception, RpS5b, which is required for oogenesis, is functionally interchangeable with its canonical paralog, indicating that the RpS5b-/- phenotype results from insufficient ribosomes rather than the absence of an RpS5b-specific, functionally specialized ribosome. Altogether, our results provide evidence that instead of new functions, RP gene duplications provide a means to regulate ribosome levels during development.

Slf1 and Sro9 are paralogous RNA-binding proteins in Saccharomyces cerevisiae that belong to the LARP1 (La-related protein 1) subgroup of the greater La family. These proteins function as translational regulators during cellular stress, acting through either direct mRNA binding or interactions with ribosomal factors. In this study, we characterized the structural and RNA-binding properties of the La-motif (LaM) domains of Slf1 and Sro9 using a combination of nuclear magnetic resonance (NMR) spectroscopy, calorimetry, and molecular dynamics (MD) simulations. Both LaM domains exhibited micromolar affinity for RNA ligands, including poly(A). Notably, the Sro9 LaM domain displayed a thermal denaturation midpoint of 36°C suggesting a potential regulatory mechanism for this protein during hyperthermic stress. An NMR analysis of the Slf1 LaM domain revealed that its RNA binding platform undergoes widespread conformational sampling on the micro- to millisecond timescale, even in the presence of RNA. Molecular dynamics simulations corroborated these experimental NMR observations and highlighted the role of transient aromatic stacking during RNA binding. Furthermore, a glutamine substitution mutant (Q278A in Slf1) known to impair RNA binding also destabilized the protein-RNA interaction in molecular simulations. Collectively, our findings confirm that RNA binding by LaM domains is an evolutionarily conserved feature among eukaryotes and provide critical insights into the structural and dynamic mechanisms underlying Slf1 and Sro9 function in yeast.

In vertebrates, two genes, Musashi1 (Msi1) and Musashi2 (Msi2), encode for highly similar Musashi protein paralogs. The Musashi proteins are known to bind to 3’-UTRs and control translation. In photoreceptor cells, the Musashi proteins promote the inclusion of photoreceptor-specific alternative exons by binding to the proximal downstream of their introns. While the Musashi proteins are expressed in various cell types, their role in regulating splicing appears to be confined to photoreceptor cells, where the two proteins have exceptionally high expression levels. To test if the photoreceptor-specific role of MSI1 and MSI2 in splicing is due to their expression levels in photoreceptor cells, we generated combined Msi1 and Msi2 knockouts that progressively reduced the number of Musashi alleles in photoreceptor cells. We analyzed the splicing of photoreceptor-specific exons in the Cc2d2a, Cep290, Prom1, and Ttc8 genes and the function of photoreceptor cells in the knockouts. We found that a single allele from either Msi1 or Msi2 is sufficient to maintain photoreceptor function and support high inclusion levels of the photoreceptor-specific exons.

The CCR4–NOT complex is a multi-subunit assembly found in all eukaryotic cells. Yet, its composition varies across organisms, with a universally conserved core enriched by lineage-specific subunits. Further, heterogeneity results from the occurrence of paralogous proteins, substoichiometric subunits, transient partners, and protein isoforms. Altogether, multiple CCR4–NOT complexes exist, and some even coexist within a single cell. The CCR4–NOT complex is an essential actor of gene expression through its roles in messenger RNA (mRNA) deadenylation, decay, and translation. Over time, support for the originally proposed role of the CCR4–NOT complex in transcription has been waning. Consistent with a role in post-transcriptional regulation, ribosomes appear to be major partners of the CCR4–NOT complex to coordinate translation and mRNA decay. Further, the CCR4–NOT complex is at the center of a network involving RNA-binding proteins and ubiquitin ligases, as well as factors of currently unknown function. Structural and functional analyses indicate that the CCR4–NOT complex integrates different levels of information present in mRNAs to control their stability and translation, thereby contributing to diverse functions including intricate processes such as human brain or pancreas development. It is thus not surprising that genetic alteration of this essential cellular machine, or impairment of its activity by pathogens, contributes to human diseases.

Protein diversity is a fundamental biological process that enhances the functional complexity of cellular signaling pathways. This diversity arises through multiple molecular mechanisms such as gene duplication, alternative splicing, and alternative translation initiation, which together expand the proteome landscape. Calcium signaling showcases this diversity, with several channels, pumps, and regulatory proteins expressed as multiple isoforms and variants. Within the store-operated calcium entry pathway, protein diversity is evident in the existence of distinct paralogs of ORAI channels and STIM proteins. The additional presence of numerous isoforms and variants of ORAI and STIM shapes the store-operated calcium entry pathway, providing flexibility to cellular calcium regulation in various contexts. Deciphering how protein diversity modulates store-operated calcium entry function is essential for advancing our understanding of calcium signaling in both health and disease.

Plants have evolved an extensive repertoire of specialized metabolites to adapt to complex environmental changes. Here, we identify two paralogous dirigent proteins (DPs) in cotton that serve as gatekeepers of extracellular terpenoid phytoalexin production in green organs, directing the transition of hemigossypol away from gossypol synthesis toward a hydroxylation pathway that leads to the biosynthesis of highly toxic hemigossypolone and heliocides. Under oxidative conditions, these proteins function synergistically with aldo-keto reductases to catalyze the hydroxylation of hemigossypol, followed by spontaneous oxidation that yields hemigossypolone, revealing a noncanonical role for aldo-keto reductases in extracellular terpenoid metabolism. Notably, mutants lacking these dirigent proteins produce gossypol but are devoid of hemigossypolone and heliocides in green organs exhibit heightened susceptibility to multiple biotic stresses, underscoring the enhanced protective role of these metabolites. This study describes a DPs-mediated mechanism of extracellular hydroxylation and highlights the potential ecological advantages of redirecting specialized metabolism extracellularly for enhanced defense against varying types of pathogens and herbivores.

Paralogs engage in biological processes through both redundant and non-redundant functions. In the yeast Saccharomyces cerevisiae, approximately one-fifth of the genome consists of paralogs, with their encoded proteins involved in multiple pathways. However, the unique contributions of individual paralogs have remained poorly defined. Here, we undertook a systematic examination of eight paralog pairs in the glucose-sensing pathways, deleting each component and measuring the resulting changes in gene expression. To that end we established a new transcription reporter system to monitor the response to glucose as well as to non-preferred sugars in single cells. Focusing on the PKA catalytic subunits, comprised of the paralogs Tpk1 and Tpk3 as well as the isomorphic kinase Tpk2, we employed mass spectrometry to identify their contribution to cellular metabolism, used a GFP-based sensor to follow changes in cytosolic pH, and used BioID to identify unique and shared binding partners. Our data reveal that paralogs in the glucose-sensing pathway contribute in multiple and unique ways to signal transduction, and establish potential mechanisms driving the preservation of these and other duplicated genes throughout long periods of evolution.

The rise of multidrug-resistant (MDR) Pseudomonas aeruginosa poses a significant threat in clinical settings due to its intricate antimicrobial resistance mechanism, biofilm formation, quorum sensing, and efflux pump-mediated antibiotic tolerance capability. The progressive decline in the efficacy of conventional antibiotics necessitates the development of new treatment strategies. Disrupting the Quorum sensing, a pivotal regulator of virulence and biofilm-associated resistance presents a promising anti-virulence strategy. An integrated Subtractive genomics and in silico drug discovery approach was applied to the complete proteome of P. aeruginosa JJPA01, excluding paralogous, human homologous, and non-essential proteins to identify virulence-associated targets. 27 pathogen-specific pathway proteins were identified, with pqsH (WP_003090354.1), a key monooxygenase in the PQS quorum-sensing system. Potential inhibitors for pqsH were identified using High-Throughput Virtual Screening (HTVS) on natural compounds from the COCONUT, CMNPD, MNPD, Seaweed, and Specs databases. The docking study identified five compounds with the best binding affinities, ranging from -6.6 to -7.7kcal/mol. However, only CNP0000215 and CNP0007440 exhibited higher binding affinity to the pqsH protein than the cofactor Flavine Adenine Dinucleotide. With its established role in Antimicrobial Resistance and Virulence, pqsH has been selected as a therapeutic target and CNP0000215 as a promising PQS inhibitor to disrupt biofilm formation and combat antimicrobial resistance. These findings lay the groundwork for the strategic design of novel anti-therapeutics offering a promising strategy to inhibit persistent infections and resistance mechanisms in P. aeruginosa.

Chloroplast-containing species possess 2 α-(1→6)-glucosidases that share a common ancestor but were independently acquired by horizontal gene transfer from separate eubacterial donors. The pullulanase-type enzyme (CAZy subfamily GH13_13) and the isoamylase-type enzyme (CAZy subfamily GH13_11) both hydrolyze branch linkages in α-polyglucans. Thus, both enzyme types function as debranching enzymes (DBE) in starch metabolism. As both enzyme types are conserved, distinct selectable functions are expected. This study describes the functional interactions between maize (Zea mays L.) pullulanase1 (ZPU1) and the isoamylase-type enzyme complex comprising the paralogous proteins isoamylase1 (ISA1) and isoamylase2 (ISA2). Mutation of ISA1 or ISA2 caused reduced ZPU1 activity in developing endosperm extracts, and the addition of ISA1 to ZPU1-expressing yeast (Saccharomyces cerevisiae) cells caused increased ZPU1 activity. Specific amino acid substitutions in ISA1 resulted in altered ZPU1 mobility in SDS-PAGE. In vivo protein–protein interaction tests and co-immunoprecipitation revealed that ZPU1 and ISA1 interact in multi-subunit complexes. Maize lines harboring ISA1 mutations, exhibiting a classical low-starch, high-phytoglycogen-accumulation phenotype, were altered by recurrent selection so that kernel appearance reverted to near normal. Extragenic suppression indicated the requirement for ISA1/ISA2 activity had been bypassed. These results are consistent with a functional overlap between the GH13_11 and GH13_13 DBE types and raise the possibility that multiple GH13 proteins, namely ZPU1, ISA1 and ISA2, act together to physically coordinate their hydrolytic activities on precursor α-polyglucans.

Chemical Inducers of Proximity DNA-Encoded Library (CIP-DEL) screening enables high-throughput discovery of compounds that induce protein-protein interactions, including Proteolysis-Targeting Chimeras (PROTACs). Simultaneous screening of protein paralogs with CIP-DEL allows profiling of compound selectivity and efficient identification of paralog-selective degraders, but such an application has not been reported. Here, we optimized CIP-DEL screening conditions and conducted a von Hippel-Lindau (VHL)-biased CIP-DEL screen with two million DNA-barcoded PROTAC compounds on eight closely related Bromodomain and Extra Terminal domain (BET) bromodomains: BRD2 BD1, BRD2 BD2, BRD3 BD1, BRD3 BD2, BRD4 BD1, BRD4 BD2, BRDT BD1, and BRDT BD2. We observed a marked tendency of compounds to bind the first bromodomain (BD1) preferentially over the second bromodomain (BD2), which contrasts with the predominantly BD2-selective inhibitors reported in the literature. Specifically, our screening approach yielded compound 21-1, which demonstrated promising BRD2 BD1 selectivity in both sequencing data of DEL screening output and in vitro assays. Additionally, normalized relative enrichment selectivity from sequencing data rather than unnormalized absolute enrichment selectivity correlated more closely with experimentally validated selectivity. Overall, we highlight the value of CIP-DEL in profiling PROTAC selectivity, which should be applicable to other protein families with high sequence homologies, where selective degrader discovery remains challenging.

Many but not all Eukaryotes have protein-enclosed compartments called vaults. Vaults are composed of multiple copies of the major vault protein, symmetrically assembled into a basket-like shell. A human cell contains approximately 100,000 vault particles, the vast majority localized to the cytosol but also observed in the nucleus and at the nuclear pore complex. Whilst there is intriguing structural information of the vault shell, the function of vaults remains largely elusive, apart from a potential contribution to mRNA maturation. We set out to explore the vault interactome in the early branching eukaryote Trypanosoma brucei employing a combination of affinity capture and TurboID proximity labelling. T. brucei encodes three major vault protein (MVP) paralogs, which exhibit a considerable degree of divergence. Unexpectedly, affinity capture proteomics with one MVP as a bait precipitated the other two paralogs, detected with similar intensities, indicating the possibility that all three are incorporated into the same particle. Dual color fluorescence microscopy of MVP pairs fused with different GFP-variants confirmed that all three paralogs are incorporated into a single vault shell. Our combined interactome data, including immune-isolations with varying stringencies, suggest a vault particle core composition of three MVPs homologs and the telomerase-associated protein 1 (TEP1), which has been described as a vault component in various organisms. Further, we demonstrate the association of vtRNA with the particle and suggest a cohort of potential transient vault interactors, dominated by RNA-binding proteins and splicing factors, which were found enriched in both orthogonal interactome approaches.

An evolutionarily conserved ribosomal protein Rps29/uS14 participates in the assembly of late pre-40S particles. In yeast Saccharomyces cerevisiae, duplicate genes RPS29A and RPS29B encode two paralogous proteins with 91% sequence identity. Here, we report that loss of either paralog impairs final step of cytoplasmic processing of 20S pre-rRNA, which is a direct precursor of mature 18S ribosomal RNA (rRNA)—a component of small ribosomal subunit. Consistently, we found that late processing factors remain bound to ribosome particles in Rps29-deficient cells. However, pre-40S particles containing 20S pre-rRNA are largely absent from translation-competent 80S ribosomes in Rps29-deficient cells, suggesting that lower levels of Rps29 protein induce quality control steps during maturation of cytoplasmic pre-ribosomal particles. Moreover, we analyzed Rps29 function during cellular stress conditions and found that cells with decreased levels of Rps29 protein adapted more rapidly to osmotic stress, but the effect was independent of the 20S pre-rRNA maturation.

BackgroundThe majority of missense variants in clinical genetic tests are classified as variants of uncertain significance. Prior research shows that the deleterious effects and the subsequent molecular consequences of variants are often conserved among paralogous protein sequences within a gene family. Here, we systematically quantify on an exome-wide scale whether the existence of pathogenic variants in paralogous genes at a conserved position can serve as evidence for the pathogenicity of a new variant. For the gene family of voltage-gated sodium channels, where variants and expert-curated clinical phenotypes are available, we also assess whether phenotype patterns of multiple disorders for each gene are conserved across variant positions within the gene family.ResultsMapping 590,000 pathogenic and 1.9 million population variants onto 9928 genes grouped into 2054 paralogous families increases the number of residues with classifiable evidence 5.1-fold compared with gene-specific data alone. The presence of a pathogenic variant in a paralogous gene is associated with a positive likelihood ratio of 13.0 for variant pathogenicity. Across ten genes encoding voltage-gated sodium channels and 22 expert-curated disorders, we identify cross-paralog correlated phenotypes based on 3D structure spatial position. For example, multiple established loss-of-function related disorders across SCN1A, SCN2A, SCN5A, and SCN8A show overlapping spatial variant clusters. Finally, we show that phenotype integration in paralog variant selection improves variant classification.ConclusionConserved pathogenic missense variants in paralogous genes provide robust, quantifiable support for clinical variant interpretation, and phenotype-informed mapping further improves predictions.

Neural tube defects (NTDs) are congenital malformations arising when the neural tube (NT), precursor of the brain and spine, fails to properly close during neurulation. Etiology is multifactorial, with environmental and genetic factors variably contributing on a case-by-case basis. Molecular genetic studies of murine NTD genes have been precious in the identification of predisposing NTD genes in humans, highlighting the peculiar role of the planar cell polarity (PCP) pathway in a fraction of human NTD patients. Seventy-eight patients with NTD treated at a pediatric tertiary care center were selected for genetic analysis. A custom next-generation sequencing (NGS) panel of 29 genes encoding for components of the core PCP pathway or for family members and paralogs of proteins (including SHROOM and GRHL) underlying NTDs in well-known animal models was used to re-sequence patients with NTD. A gene-burden analysis was also performed to assess potential enrichment of rare damaging variants in the NTD cohort compared to ethnically matched controls. Thirty-nine of 78 individuals (50%) presented with at least one putatively damaging rare variant, most of which (87%) were missense substitutions. Rare variants of GRHL1 and WNT5A, and among gene families GRHL and SHROOM, were significantly enriched in the patients' cohort compared to controls. This study supports the involvement of human orthologues of mouse genes in human NTD phenotypes. Further re-sequencing or, even better, whole-exome sequencing of a large group of cases will give the clues for a better understanding of NTD etiology, ameliorating the clinical management of patients and their families.

The fidelity of assembly of multiprotein complexes is essential for the formation of stable and functional protein complexes that are critical for cell growth and survival. In this context, TBP-associated factor (TAF) subunits maintain tight specificity for their integration into TFIID and SAGA complexes. In this work, using affinity purification-coupled mass spectrometry of epitope-tagged TFIID subunits TBP and TAF11, and the SAGA subunit TAF12L we identified components of the C. albicans TFIID and SAGA complexes. Whereas TAF12 is a subunit of TFIID, the paralogous TAF12L is a subunit of the SAGA complex, and we further identified each of the TFIID and SAGA complex subunits with high confidence. We found that the steady-state levels of the H2B-H2A-like histone fold domain containing pairs, TAF12-TAF4 and TAF12L-Ada1 proteins, are mutually dependent on the stable expression of each other. Using RNA coimmunoprecipitation from polysome-containing extracts, we found that nascent TAF4 and Ada1 proteins interact with TAF12 and TAF12L, respectively, by a cotranslational mechanism in an ordered, sequential mode of assembly. Thus, our results indicate that heterodimerization of the TAF12 paralogs with cognate partners occur by sequential cotranslational assembly thereby ensuring both selectivity and stability of the H2A-H2B heterodimers in fungal pathogen C. albicans.