Insights into RAG Evolution from the Identification of "Missing Link" Family A RAGL Transposons.

A series of "molecular domestication" events are thought to have converted an invertebrate RAG-like (RAGL) transposase into the RAG1-RAG2 (RAG) recombinase, a critical enzyme for adaptive immunity in jawed vertebrates. The timing and order of these events are not well understood, in part because of a dearth of information regarding the invertebrate RAGL-A transposon family. In contrast to the abundant and divergent RAGL-B transposon family, RAGL-A most closely resembles RAG and is represented by a single orphan RAG1-like (RAG1L) gene in the genome of the hemichordate Ptychodera flava (PflRAG1L-A). Here, we provide evidence for the existence of complete RAGL-A transposons in the genomes of P. flava and several echinoderms. The predicted RAG1L-A and RAG2L-A proteins encoded by these transposons intermingle sequence features of jawed vertebrate RAG and RAGL-B transposases, leading to a prediction of DNA binding, catalytic, and transposition activities that are a hybrid of RAG and RAGL-B. Similarly, the terminal inverted repeats (TIRs) of the RAGL-A transposons combine features of both RAGL-B transposon TIRs and RAG recombination signal sequences. Unlike all previously described RAG2L proteins, RAG2L-A proteins contain an acidic hinge region, which we demonstrate is capable of efficiently inhibiting RAG-mediated transposition. Our findings provide evidence for a critical intermediate in RAG evolution and argue that certain adaptations thought to be specific to jawed vertebrates (e.g. the RAG2 acidic hinge) actually arose in invertebrates, thereby focusing attention on other adaptations as the pivotal steps in the completion of RAG domestication in jawed vertebrates.

Insights into RAG evolution from the identification of "missing link" family A RAGL transposons.

A series of "molecular domestication" events are thought to have converted an invertebrate RAG-like (RAGL) transposase into the RAG1-RAG2 (RAG) recombinase, a critical enzyme for adaptive immunity in jawed vertebrates. The timing and order of these events is not well understood, in part because of a dearth of information regarding the invertebrate RAGL-A transposon family. In contrast to the abundant and divergent RAGL-B transposon family, RAGL-A most closely resembles RAG and is represented by a single orphan RAG1-like (RAG1L) gene in the genome of the hemichordate Ptychodera flava (PflRAG1L-A). Here, we provide evidence for the existence of complete RAGL-A transposons in the genomes of P. flava and several echinoderms. The predicted RAG1L-A and RAG2L-A proteins encoded by these transposons intermingle sequence features of jawed vertebrate RAG and RAGL-B transposases, leading to a prediction of DNA binding, catalytic, and transposition activities that are a hybrid of RAG and RAGL-B. Similarly, the terminal inverted repeats (TIRs) of the RAGL-A transposons combine features of both RAGL-B transposon TIRs and RAG recombination signal sequences. Unlike all previously described RAG2L proteins, PflRAG2L-A and echinoderm RAG2L-A contain an acidic hinge region, which we demonstrate is capable of efficiently inhibiting RAG-mediated transposition. Our findings provide evidence for a critical intermediate in RAG evolution and argue that certain adaptations thought to be specific to jawed vertebrates (e.g., the RAG2 acidic hinge) actually arose in invertebrates, thereby focusing attention on other adaptations as the pivotal steps in the completion of RAG domestication in jawed vertebrates.

Methionine oxidation selectively enhances T cell reactivity against a melanoma antigen.

The impact of the peptide amino acids side-chain modifications on the immunological recognition has been scarcely explored. We investigate here the effect of methionine oxidation on the antigenicity of the melanoma immunodominant peptide 369-YMDGTMSQV-377 (YMD). Using CD8+ T cell activation assays, we found that the antigenicity of the sulfoxide form is higher when compared to the YMD peptide. This is consistent with free energy computations performed on HLA-A∗02:01/YMD/TCR complex showing that this is lowered upon oxidation, paired with a steep increase in order at atomic level. Oxidized YMD forms were identified at the melanoma cell surface by LC-MS/MS analysis. These results demonstrate that methionine oxidation in the antigenic peptides may generate altered peptide ligands with increased antigenicity, and that this oxidation may occur in vivo, opening up the possibility that high-affinity CD8+ T cells might be naturally primed in the course of melanoma progression, as a result of immunosurveillance.

Oncogenic CALR mutant C-terminus mediates dual binding to the thrombopoietin receptor triggering complex dimerization and activation.

Calreticulin (CALR) frameshift mutations represent the second cause of myeloproliferative neoplasms (MPN). In healthy cells, CALR transiently and non-specifically interacts with immature N-glycosylated proteins through its N-terminal domain. Conversely, CALR frameshift mutants turn into rogue cytokines by stably and specifically interacting with the Thrombopoietin Receptor (TpoR), inducing its constitutive activation. Here, we identify the basis of the acquired specificity of CALR mutants for TpoR and define the mechanisms by which complex formation triggers TpoR dimerization and activation. Our work reveals that CALR mutant C-terminus unmasks CALR N-terminal domain, rendering it more accessible to bind immature N-glycans on TpoR. We further find that the basic mutant C-terminus is partially α-helical and define how its α-helical segment concomitantly binds acidic patches of TpoR extracellular domain and induces dimerization of both CALR mutant and TpoR. Finally, we propose a model of the tetrameric TpoR-CALR mutant complex and identify potentially targetable sites.

NLRscape: an atlas of plant NLR proteins.

NLRscape is a webserver that curates a collection of over 80 000 plant protein sequences identified in UniProtKB to contain NOD-like receptor signatures, and hosts in addition a number of tools aimed at the exploration of the complex sequence landscape of this class of plant proteins. Each entry gathers sequence information, domain and motif annotations from multiple third-party sources but also in-house advanced annotations aimed at addressing caveats of the existing broad-based annotations. NLRscape provides a top-down perspective of the NLR sequence landscape but also services for assisting a bottom-up approach starting from a given input sequence. Sequences are clustered by their domain organization layout, global homology and taxonomic spread-in order to allow analysis of how particular traits of an NLR family are scattered within the plant kingdom. Tools are provided for users to locate their own protein of interest in the overall NLR landscape, generate custom clusters centered around it and perform a large number of sequence and structural analyses using included interactive online instruments. Amongst these, we mention: taxonomy distribution plots, homology cluster graphs, identity matrices and interactive MSA synchronizing secondary structure and motif predictions. NLRscape can be found at: https://nlrscape.biochim.ro/.

Tsw - A case study on structure-function puzzles in plant NLRs with unusually large LRR domains.

Plant disease immunity heavily depends on the recognition of plant pathogens and the subsequent activation of downstream immune pathways. Nod-like receptors are often crucial in this process. Tsw, a Nod-like resistance gene from Capsicum chinense conferring resistance against Tomato spotted wilt virus (TSWV), belongs to the small group of Nod-like receptors with unusually large LRR domains. While typical protein domain dimensions rarely exceed 500 amino acids due to stability constraints, the LRR of these unusual NLRs range from 1,000 to 3,400 amino acids and contain over 30 LRR repeats. The presence of such a multitude of repeats in one protein is also difficult to explain considering protein functionality. Interactions between the LRR and the other NLR domains (CC, TIR, NBS) take place within the first 10 LRR repeats, leaving the function of largest part of the LRR structure unexplained. Herein we discuss the structural modeling limits and various aspects of the structure-function relation conundrums of large LRRs focusing on Tsw, and raise questions regarding its recognition of its effector NSs and the possible inhibition on other domains as seen in other NLRs.

NLRexpress-A bundle of machine learning motif predictors-Reveals motif stability underlying plant Nod-like receptors diversity.

Examination of a collection of over 80,000 Plant Nod-like receptors (NLRs) revealed an overwhelming sequence diversity underlying functional specificity of pathogen detection, signaling and cooperativity. The NLR canonical building blocks-CC/TIR/RPW8, NBS and LRR-contain, however, a number of conserved sequence motifs showing a significant degree of invariance amongst different NLR groups. To identify these motifs we developed NLRexpress-a bundle of 17 machine learning (ML)-based predictors, able to swiftly and precisely detect CC, TIR, NBS, and LRR motifs while minimizing computing time without accuracy losses-aimed as an instrument scalable for screening overall proteomes, transcriptomes or genomes for identifying integral NLRs and discriminating them against incomplete sequences lacking key motifs. These predictors were further used to screen a subset of ∼34,000 regular plant NLR sequences. Motifs were analyzed using unsupervised ML techniques to assess the structural correlations hidden underneath pattern variabilities. Both the NB-ARC switch domain which admittedly is the most conserved region of NLRs and the highly diverse LRR domain with its vastly variable lengths and repeat irregularities-show well-defined relations between motif subclasses, highlighting the importance of structural invariance in shaping NLR sequence diversity. The online NLRexpress webserver can be accessed at https://nlrexpress.biochim.ro.

Defining the altered glycoproteomic space of the early secretory pathway by class I mannosidase pharmacological inhibition.

N-glycosylation is a key process for various biological functions like protein folding, maturation and sorting for the conventional secretory compartment, cell-cell communication and immune response. This is usually accomplished by a complex system of mannosidases in which those from class I have an outstanding role, commonly involved in the early protein sorting associated to the Endoplasmic Reticulum (ER) in the N-glycan dependent quality control (ERQC) and ER-associated degradation (ERAD). Although these are vital processes in maintaining cellular homeostasis, large-scale analysis studies for this pool of molecules, further denoted as proteins from the early secretory pathway (ESP), were limited addressed. Here, using a custom workflow employing a combination of glycomics and deglycoproteomics analyses, using lectin affinity and selective Endoglycosidase H (Endo H) digestion, we scrutinize the steady-state oligomannosidic glycoprotein load and delineate ESP fraction in melanoma cells. All of these were assessed by applying our workflow for glycosite relative quantification of both the peptide chain and carbohydrate structure in cells with inhibited activity of class I mannosidases after kifunensine treatment. We found that most of the ESP are transient clients involved in cell communication via extracellular matrix, particularly integrin-mediated communication which adopt Man9 N-glycans in kifunensine-treated cells. Moreover, our results reveal that core-fucosylation is decreased subsequent inhibition of class I mannosidases and this could be explained by a general lower protein level of FUT8, the enzyme responsible for fucosylation. By comparing our data with results obtained following downregulation of a key mannosidase in misfolded protein degradation, we mapped both novel and previously suggested endogenous substrate candidates like PCDH2, HLA-B, LAMB2 or members of the integrin family of proteins such as ITGA1 and ITGA4, thus validating the findings obtained using our workflow regarding accumulation and characterization of ESP transitory members following mannosidase class I inhibition. This workflow and the associated dataset not only allowed us to investigate the oligomannosidic glycoprotein fraction but also to delineate differences mediated at glycosite-level upon kifunensine treatment and outline the potential associated cellular responses.

Affinity Proteomics and Deglycoproteomics Uncover Novel EDEM2 Endogenous Substrates and an Integrative ERAD Network.

Various pathologies result from disruptions to or stress of endoplasmic reticulum (ER) homeostasis, such as Parkinson's disease and most neurodegenerative illnesses, diabetes, pulmonary fibrosis, viral infections, and cancers. A critical process in maintaining ER homeostasis is the selection of misfolded proteins by the ER quality-control system for destruction via ER-associated degradation (ERAD). One key protein proposed to act during the first steps of misfolded glycoprotein degradation is the ER degradation-enhancing α-mannosidase-like protein 2 (EDEM2). Therefore, characterization of the EDEM2-associated proteome is of great interest. We took advantage of using melanoma cells overexpressing EDEM2 as a cancer model system, to start documenting at the deglycoproteome level (N-glycosites identification) the emerging link between ER homeostasis and cancer progression. The dataset created for identifying the EDEM2 glyco clients carrying high mannose/hybrid N-glycans provides a comprehensive N-glycosite analysis mapping over 1000 N-glycosites on more than 600 melanoma glycoproteins. To identify EDEM2-associated proteins, we used affinity proteomics and proteome-wide analysis of sucrose density fractionation in an integrative workflow. Using intensity and spectral count-based quantification, we identify seven new EDEM2 partners, all of which are involved in ER quality-control system and ERAD. Moreover, we defined novel endogenous candidates for EDEM2-dependent ERAD by combining deglycoproteomics, stable isotope labeling with amino acids in cell culture-based proteomics, and biochemical methods. These included tumor antigens and several ER-transiting endogenous melanoma proteins, including integrin alpha-1 and protocadherin 2, the expression of which was negatively correlated with that of EDEM2. Tumor antigens are key in the antigen presentation process, whereas integrin alpha-1 and protocadherin 2 are involved in melanoma metastasis and invasion. EDEM2 could therefore have a regulatory role in melanoma through the modulation of degradation and trafficking in these glycoproteins. The data presented herein suggest that EDEM2 is involved in ER homeostasis to a greater extent than previously suggested.

EDEM3 Domains Cooperate to Perform Its Overall Cell Functioning.

EDEM3 recognizes and directs misfolded proteins to the ER-associated protein degradation (ERAD) process. EDEM3 was predicted to act as lectin or as a mannosidase because of its homology with the GH47 catalytic domain of the Man1B1, but the contribution of the other regions remained unresolved. Here, we dissect the molecular determinants governing EDEM3 function and its cellular interactions. LC/MS analysis indicates very few stable ER interactors, suggesting EDEM3 availability for transient substrate interactions. Sequence analysis reveals that EDEM3 consists of four consecutive modules defined as GH47, intermediate (IMD), protease-associated (PA), and intrinsically disordered (IDD) domain. Using an EDEM3 knock-out cell line, we expressed EDEM3 and domain deletion mutants to address EDEM3 function. We find that the mannosidase domain provides substrate binding even in the absence of mannose trimming and requires the IMD domain for folding. The PA and IDD domains deletions do not impair the trimming, but specifically modulate the turnover of two misfolded proteins, NHK and the soluble tyrosinase mutant. Hence, we demonstrate that EDEM3 provides a unique ERAD timing to misfolded glycoproteins, not only by its mannose trimming activity, but also by the positive and negative feedback modulated by the protease-associated and intrinsically disordered domain, respectively.

Deep Learning in the Quest for Compound Nomination for Fighting COVID-19.

The current COVID-19 pandemic initiated an unprecedented response from clinicians and the scientific community in all relevant biomedical fields. It created an incredible multidimensional data-rich framework in which deep learning proved instrumental to make sense of the data and build models used in prediction-validation workflows that in a matter of months have already produced results in assessing the spread of the outbreak, its taxonomy, population susceptibility, diagnostics or drug discovery and repurposing. More is expected to come in the near future by using such advanced machine learning techniques to combat this pandemic. This review aims to unravel just a small fraction of the large global endeavors by focusing on the research performed on the main COVID-19 targets, on the computational weaponry used in identifying drugs to combat the disease, and on some of the most important directions found to contain COVID-19 or alleviating its symptoms in the absence of specific medication.

A natural diversity screen in Arabidopsis thaliana reveals determinants for HopZ1a recognition in the ZAR1-ZED1 immune complex.

Pathogen pressure on hosts can lead to the evolution of genes regulating the innate immune response. By characterizing naturally occurring polymorphisms in immune receptors, we can better understand the molecular determinants of pathogen recognition. ZAR1 is an ancient Arabidopsis thaliana NLR (Nucleotide-binding [NB] Leucine-rich-repeat [LRR] Receptor) that recognizes multiple secreted effector proteins from the pathogenic bacteria Pseudomonas syringae and Xanthomonas campestris through its interaction with receptor-like cytoplasmic kinases (RLCKs). ZAR1 was first identified for its role in recognizing P. syringae effector HopZ1a, through its interaction with the RLCK ZED1. To identify additional determinants of HopZ1a recognition, we performed a computational screen for ecotypes from the 1001 Genomes project that were likely to lack HopZ1a recognition, and tested ~300 ecotypes. We identified ecotypes containing polymorphisms in ZAR1 and ZED1. Using our previously established Nicotiana benthamiana transient assay and Arabidopsis ecotypes, we tested for the effect of naturally occurring polymorphisms on ZAR1 interactions and the immune response. We identified key residues in the NB or LRR domain of ZAR1 that impact the interaction with ZED1. We demonstrate that natural diversity combined with functional assays can help define the molecular determinants and interactions necessary to regulate immune induction in response to pathogens.

Identification of RAG-like transposons in protostomes suggests their ancient bilaterian origin.

BACKGROUND: V(D) J recombination is essential for adaptive immunity in jawed vertebrates and is initiated by the RAG1-RAG2 endonuclease. The RAG1 and RAG2 genes are thought to have evolved from a RAGL (RAG-like) transposon containing convergently-oriented RAG1-like (RAG1L) and RAG2-like (RAG2L) genes. Elements resembling this presumptive evolutionary precursor have thus far only been detected convincingly in deuterostomes, leading to the model that the RAGL transposon first appeared in an early deuterostome. RESULTS: We have identified numerous RAGL transposons in the genomes of protostomes, including oysters and mussels (phylum Mollusca) and a ribbon worm (phylum Nemertea), and in the genomes of several cnidarians. Phylogenetic analyses are consistent with vertical evolution of RAGL transposons within the Bilateria clade and with its presence in the bilaterian ancestor. Many of the RAGL transposons identified in protostomes are intact elements containing convergently oriented RAG1L and RAG2L genes flanked by terminal inverted repeats (TIRs) and target site duplications with striking similarities with the corresponding elements in deuterostomes. In addition, protostome genomes contain numerous intact RAG1L-RAG2L adjacent gene pairs that lack detectable flanking TIRs. Domains and critical active site and structural amino acids needed for endonuclease and transposase activity are present and conserved in many of the predicted RAG1L and RAG2L proteins encoded in protostome genomes. CONCLUSIONS: Active RAGL transposons were present in multiple protostome lineages and many were likely transmitted vertically during protostome evolution. It appears that RAGL transposons were broadly active during bilaterian evolution, undergoing multiple duplication and loss/fossilization events, with the RAGL genes that persist in present day protostomes perhaps constituting both active RAGL transposons and domesticated RAGL genes. Our findings raise the possibility that the RAGL transposon arose earlier in evolution than previously thought, either in an early bilaterian or prior to the divergence of bilaterians and non-bilaterians, and alter our understanding of the evolutionary history of this important group of transposons.

Robosample: A rigid-body molecular simulation program based on robot mechanics.

BACKGROUND: Compared with all-atom molecular dynamics (MD), constrained MD methods allow for larger time steps, potentially reducing computational cost. For this reason, there has been continued interest in improving constrained MD algorithms to increase configuration space sampling in molecular simulations. METHODS: Here, we introduce Robosample, a software package that implements high-performance constrained dynamics algorithms, originally developed for robotics, and applies them to simulations of biomolecular systems. As in the gMolmodel package developed by Spiridon and Minh in 2017, Robosample uses Constrained Dynamics Hamiltonian Monte Carlo (CDHMC) as a Gibbs sampling move - a type of Monte Carlo move where a subset of coordinates is allowed to change. In addition to the previously described Cartesian and torsional dynamics moves, Robosample implements spherical and cylindrical joints that can be distributed along the molecule by the user. RESULTS: In alanine dipeptide simulations, the free energy surface is recovered by mixing fully flexible with torsional, cylindrical, or spherical dynamics moves. Ramachandran dynamics, where only the two key torsions are mobile, accelerate the slowest transition by an order of magnitude. We also show that simulations of a complex glycan cover significantly larger regions of the configuration space when mixed with constrained dynamics. MAJOR CONCLUSIONS: Robosample is a tool of choice for efficient conformational sampling of large biomolecules. GENERAL SIGNIFICANCE: Robosample is intended as a reliable and user-friendly simulation package for fast biomolecular sampling that does not require extensive expertise in mechanical engineering or in the statistical mechanics of reduced coordinates.

LRRpredictor-A New LRR Motif Detection Method for Irregular Motifs of Plant NLR Proteins Using an Ensemble of Classifiers.

Leucine-rich-repeats (LRRs) belong to an archaic procaryal protein architecture that is widely involved in protein-protein interactions. In eukaryotes, LRR domains developed into key recognition modules in many innate immune receptor classes. Due to the high sequence variability imposed by recognition specificity, precise repeat delineation is often difficult especially in plant NOD-like Receptors (NLRs) notorious for showing far larger irregularities. To address this problem, we introduce here LRRpredictor, a method based on an ensemble of estimators designed to better identify LRR motifs in general but particularly adapted for handling more irregular LRR environments, thus allowing to compensate for the scarcity of structural data on NLR proteins. The extrapolation capacity tested on a set of annotated LRR domains from six immune receptor classes shows the ability of LRRpredictor to recover all previously defined specific motif consensuses and to extend the LRR motif coverage over annotated LRR domains. This analysis confirms the increased variability of LRR motifs in plant and vertebrate NLRs when compared to extracellular receptors, consistent with previous studies. Hence, LRRpredictor is able to provide novel insights into the diversification of LRR domains and a robust support for structure-informed analyses of LRRs in immune receptor functioning.

Structure-function analysis of ZAR1 immune receptor reveals key molecular interactions for activity.

NLR (nucleotide-binding [NB] leucine-rich repeat [LRR] receptor) proteins are critical for inducing immune responses in response to pathogen proteins, and must be tightly modulated to prevent spurious activation in the absence of a pathogen. The ZAR1 NLR recognizes diverse effector proteins from Pseudomonas syringae, including HopZ1a, and Xanthomonas species. Receptor-like cytoplasmic kinases (RLCKs) such as ZED1, interact with ZAR1 and provide specificity for different effector proteins, such as HopZ1a. We previously developed a transient expression system in Nicotiana benthamiana that allowed us to demonstrate that ZAR1 function is conserved from the Brassicaceae to the Solanaceae. Here, we combined structural modelling of ZAR1, with molecular and functional assays in our transient system, to show that multiple intramolecular and intermolecular interactions modulate ZAR1 activity. We identified determinants required for the formation of the ZARCC oligomer and its activity. Lastly, we characterized intramolecular interactions between ZAR1 subdomains that participate in keeping ZAR1 immune complexes inactive. This work identifies molecular constraints on immune receptor function and activation.

Profiling Optimal Conditions for Capturing EDEM Proteins Complexes in Melanoma Using Mass Spectrometry.

Endoplasmic reticulum (ER) resident and secretory proteins that fail to reach their native conformation are selected for degradation through the ER-Associated Degradation (ERAD) pathway. The ER degradation-enhancing alpha-mannosidase-like proteins (EDEMs) were shown to be involved in this pathway but their precise role is still under investigation. Mass spectrometry analysis has contributed significantly to the characterization of protein complexes in the last years. The recent advancements in instrumentation, especially within resolution and speed can provide unique insights concerning the molecular architecture of protein-protein interactions in systems biology. Previous reports have suggested that several protein complexes in ERAD are sensitive to the extraction conditions. Indeed, whilst EDEM proteins can be recovered in most detergents, some of their partners are not solubilized, which further emphasizes the importance of the experimental setup. Here, we define such dynamic interactions of EDEM proteins by employing offline protein fractionation, nanoLC-MS/MS and describe how mass spectrometry can contribute to the characterization of such complexes, particularly within a disease context like melanoma.

Gangliosidome of human anencephaly: A high resolution multistage mass spectrometry study.

Widely dispersed throughout the entire body tissues, gangliosides (GGs) are essential components of neuronal cell membranes, where exhibit a vital role in neuronal function and brain development, directly influencing the neural tube formation, neurogenesis, neurotransmission, etc. Due to several factors, partial or complete closing faults of the fetal neural tube may occur in the first trimester of pregnancy, generating a series of neural tube defects (NTD), among which anencephaly. The absence in anencephaly of the forebrain and skull bones determines the exposure to the amniotic fluid of the remaining brain tissue and the spinal cord, causing the degeneration of the nervous system tissue. Based on the previously achieved information related to the direct alteration of neural development with deficient concentration of several GGs, a systematic and comparative mass spectrometry (MS) mapping assay on GGs originating from fetuses in different intrauterine developmental stages, i.e. the 29th (denoted An29), 35th (An35) and the 37th (An37) gestational weeks was here conducted. Our approach, based on Orbitrap MS under high sensitivity, resolution and mass accuracy conditions, enabled for the first time the nanoelectrospray ionization, detection and identification of over 150 glycoforms, mainly novel, polysialylated species. Such a pattern, specific for incipient developmental stages reliably documents the brain development stagnation, characteristic for anencephaly. Further, the fragmentation MS2-MS3 experiments by collision induced dissociation (CID) confirmed the incidence in all three samples of GT2(d18:1/16:2) as a potential biomarker. Therefore, this fingerprinting of the anencephalic gangliosidome may serve in development of approaches for routine screening and early diagnosis.

Transposon molecular domestication and the evolution of the RAG recombinase.

Domestication of a transposon (a DNA sequence that can change its position in a genome) to give rise to the RAG1-RAG2 recombinase (RAG) and V(D)J recombination, which produces the diverse repertoire of antibodies and T cell receptors, was a pivotal event in the evolution of the adaptive immune system of jawed vertebrates. The evolutionary adaptations that transformed the ancestral RAG transposase into a RAG recombinase with appropriately regulated DNA cleavage and transposition activities are not understood. Here, beginning with cryo-electron microscopy structures of the amphioxus ProtoRAG transposase (an evolutionary relative of RAG), we identify amino acid residues and domains the acquisition or loss of which underpins the propensity of RAG for coupled cleavage, its preference for asymmetric DNA substrates and its inability to perform transposition in cells. In particular, we identify two adaptations specific to jawed-vertebrates-arginine 848 in RAG1 and an acidic region in RAG2-that together suppress RAG-mediated transposition more than 1,000-fold. Our findings reveal a two-tiered mechanism for the suppression of RAG-mediated transposition, illuminate the evolution of V(D)J recombination and provide insight into the principles that govern the molecular domestication of transposons.