Nucleic Acids Research

Many bacteriophages encode anti-CRISPR (Acr) proteins that inhibit the CRISPR-Cas immune systems. Rapid acr gene expression upon phage entry enables CRISPR-Cas neutralisation, but can impact phage fitness if unregulated. Therefore, Acr production is often controlled by distinct families of co-encoded anti-CRISPR-associated (Aca) proteins, which are usually helix-turn-helix (HTH) regulators that bind DNA within acr-aca operon promoters. Previously, we demonstrated that the Aca2 family additionally represses Acr production translationally by binding structured RNA motifs within the 5' untranslated regions (UTRs) of the acr-aca mRNA. Here, through systematic bioinformatic analyses, we provide evidence of structured RNA motifs in the 5' UTRs of operons encoding members of other Aca families, and that Aca1 also specifically binds its cognate RNA motif. Additionally, many Aca proteins are predicted to regulate not only their own but also adjacent operons with potential anti-defence genes. Indeed, we show that Aca14, newly identified in this study, represses two predicted anti-defence operons. Aca14 is a ribbon-helix-helix domain protein, revealing regulatory diversity beyond the canonical HTH Aca family members. Collectively, our findings expand the understanding of acr regulation in mobile genetic elements and reveal novel mechanisms by which phages fine-tune anti-defence gene expression. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=78 SRC="FIGDIR/small/740549v1_ufig1.gif" ALT="Figure 1"> View larger version (21K): org.highwire.dtl.DTLVardef@1ca8b1dorg.highwire.dtl.DTLVardef@73291eorg.highwire.dtl.DTLVardef@90ab45org.highwire.dtl.DTLVardef@1dc739e_HPS_FORMAT_FIGEXP M_FIG C_FIG

Mammalian base excision repair (BER) is the major repair pathway for correcting small DNA base lesions. BER operates via two sub-pathways: short patch (or single nucleotide) and long patch BER (SP-BER and LP-BER, respectively). In LP-BER, DNA polymerase beta (Pol{beta}) performs strand-displacement synthesis generating a 5'-flap that is removed by flap endonuclease 1 (FEN1). To determine the precise DNA-intermediate that "switches on" FEN1 activity, we analyzed the FEN1 cleavage position in DNA-intermediates generated by Pol{beta}. For simultaneous detection of Pol{beta}-catalyzed primer elongation and FEN1's cleavage products, we utilized DNA structures with fluorescent labels on the upstream and downstream primers. Our data indicate that FEN1 operates efficiently on DNAs containing one or more equilibrium nucleotides. FEN1 binds to the 3'-terminal equilibrium nucleotide of the primer and probes it for a 3'-OH group. The cleavage point is located directly opposite this 3'-terminal equilibrium nucleotide of the primer. The subsequent elimination of the equilibrium area leads to restoration of the ligatable structure, thereby limiting further strand-displacement synthesis. Since the product sets generated by the two enzymes are mirror-symmetrical, we propose that pausing of Pol{beta} permits FEN1 to join the Pol{beta}-DNA complex. Our study demonstrates that the FEN1-Pol{beta} interplay occurs via the generation and elimination of equilibrium nucleotide areas rather than through a "gap-translation" mechanism. The length of the equilibrium area depends on dNTP concentration, suggesting that the size of the LP-BER repair patches may differ between S-phase and the remainder of the cell cycle.

Site-directed RNA editing (SDRE) utilizing the adenosine deaminase acting on RNA (ADAR) enzymes is commonly facilitated by guide RNAs (gRNAs) optimized to enhance on-target editing and minimize bystander effects. However, the impact of gRNA binding and ADAR-mediated SDRE on canonical pre-mRNA splicing remains poorly understood. Here, we developed an in vitro transcript-specific editing strategy that enables selective targeting and direct comparison of SDRE in pre-mRNA and mature mRNA. Using splice-relevant variants associated with inherited retinal diseases, we investigated the effects of SDRE on exonic, near-canonical intronic, and deep intronic splice variants. We identified gRNA-induced splice perturbation at exonic and intronic targets and observed that higher editing levels could be associated with increased splice disruption. Conversely, SDRE of two exonic splice variants and a deep intronic variant resulted in increased production of correctly spliced transcripts, demonstrating the potential of SDRE for splice modulation. Finally, by dissecting the effects of ADAR expression and gRNA design on editing and splicing outcomes, we established a system for identifying design principles that reduce splice interference and enhance the generation of correctly spliced, edited transcripts. These findings highlight the importance of considering transcript context and splicing consequences in the development of SDRE-based therapeutic strategies. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=92 SRC="FIGDIR/small/741163v1_ufig1.gif" ALT="Figure 1"> View larger version (28K): org.highwire.dtl.DTLVardef@2e4703org.highwire.dtl.DTLVardef@3834d3org.highwire.dtl.DTLVardef@1409c8org.highwire.dtl.DTLVardef@18e463b_HPS_FORMAT_FIGEXP M_FIG Created in BioRender. Schneider, N. (2026) https://BioRender.com/59i0tiy C_FIG

During replication, nucleolytic processing of apurinic/apyrimidinic (AP) sites in single-stranded (ss)DNA is attenuated by the evolutionarily conserved 5-hydroxymethylcytosine binding, embryonic-specific (HMCES) protein. HMCES forms a covalent thiazolidine linkage with the ring-opened aldehyde form of a ssDNA AP site to stabilize the AP site and suppress the formation of DNA double-strand breaks. The resulting HMCES DNA-protein cross-link (DPC) can then be digested by the SPRTN protease and bypassed through mutagenic translesion synthesis (TLS). Here, we use Xenopus egg extracts and molecular dynamics simulations to investigate how HMCES-DPC formation influences the mutagenicity of AP site bypass. We show that SPRTN processes the HMCES-DPC to a five amino acid peptide adduct prior to TLS. Surprisingly, the mutagenicity of HMCES-DPC bypass is insensitive to the extent of DPC proteolysis and depends only on cross-link formation, which traps the AP site in a more dynamic ring-opened configuration. We further show that the spectrum of mutations produced during bypass of HMCES-adducts strongly depends on the template strand nucleotide immediately 5 of the AP site. Our data support a model in which HMCES-DPC formation increases the conformational flexibility of the DNA template, allowing template slippage and use of the 5 template nucleotide to direct insertion opposite the adducted AP site.

RNA-based medicines rely on modified nucleotides to promote immune evasion and in vivo efficacy. Nucleotides generated from RNA degradation are either exported or recycled through metabolically favorable salvage pathways, though whether modified nucleotides are efficiently recycled remains unclear. N4-acetylcytidine (acC) is a naturally occurring modification in rRNA and tRNA that has shown promise in therapeutic mRNA applications. However, N4-acetylation impairs cytidine deamination, the first step in cytidine salvage. Here, we investigate the endogenous mechanisms that enable acC metabolism. Through sensitive sequence and structural analyses, we identify the uncharacterized human ASCH domain protein EOLA1 as a key acC deacetylase in nucleotide salvage. EOLA1 inactivation leads to free intracellular acC accumulation and increased cytotoxicity upon nucleotide export inhibition. While steady-state acC levels in cellular RNAs remain unchanged, EOLA1-dependent regulation of free acC is evident basally and is exacerbated by exogenous mRNA delivery. Proteomic analyses place EOLA1 in proximity to ribosomal proteins, adjacent to endogenous acC sources. In vitro assays confirm EOLA1 specificity for acC, and structural analysis reveals a narrow nucleotide-binding pocket consistent with mononucleotide selectivity. These findings identify EOLA1 as a bona fide acC eraser and uncover a previously unrecognized pathway for recycling modified nucleotides with relevance to therapeutic RNA design.

The ribosome, long regarded as a passive, uniform machine, has only recently been recognised as a direct regulator of translation. Mass spectrometry and sequencing approaches have shown that heterogeneity in ribosome composition exists, which can actively regulate the translational process. One source of this heterogeneity is the modification of ribosomal RNA (rRNA), primarily pseudouridylation (pseU) and 2'-O-methylation (2OMe), mediated by specific H/ACA and C/D box small nucleolar RNAs (snoRNAs). Here, we investigate how the stoichiometry of rRNA modifications varies during embryonic stem cell differentiation. Using the modification basecalling capability of Nanopore direct RNA sequencing, we have identified distinct stoichiometric changes in modification patterns between pluripotent and differentiated cells, revealing highly dynamic, site-specific regulation. Further, profiling of snoRNA expression during trilineage differentiation revealed differential expression of H/ACA and C/D box snoRNAs responsible for a subset of these dynamic modifications. By integrating rRNA and snoRNA sequencing approaches, we have built a comprehensive profile of rRNA modification dynamics during early embryonic cell fate decisions, highlighting potential regulatory mechanisms for ribosome heterogeneity during development. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=191 HEIGHT=200 SRC="FIGDIR/small/743918v1_ufig1.gif" ALT="Figure 1"> View larger version (26K): org.highwire.dtl.DTLVardef@48ca91org.highwire.dtl.DTLVardef@eb0418org.highwire.dtl.DTLVardef@159fc8corg.highwire.dtl.DTLVardef@d34f19_HPS_FORMAT_FIGEXP M_FIG C_FIG

RNA structure governs the function of non-coding RNAs and influences mRNA stability and translation, making testing of models of experimental RNA structure relevant to fields spanning structural biology, virology, and molecular therapeutics. Here we present FAST-MaP (Fast and Accessible Sequencing Technology for Mutational Profiling), a protocol that enables per-nucleotide RNA structure characterization using standard molecular biology equipment and requiring no sequencing infrastructure or bioinformatics expertise. RNA is chemically modified with orthogonal probes (2A3 and DMS), reverse-transcribed, and PCR-amplified to produce dsDNA amplicons that are submitted directly to a commercial primer-less sequencing service. Returned FASTQ files are processed through a freely accessible web server to generate normalized reactivity profiles within minutes. The complete protocol, from DNA template to structural data, can be completed in approximately one week. We illustrate the workflow on a 659-nucleotide RNA, demonstrating how to test structure preservation across buffers, and how to test specific secondary and tertiary structure predictions of the RNA from computational modeling or cryo-electron microscopy. The protocol requires only standard molecular-biology skills and does not require sequencing or bioinformatics expertise.

Gene complementation is a powerful tool for genetic selection and protein functional studies but typically requires extensive screening for complementary components. Here we engineered a platform for gene complementation based on RNA end-joining (REJ) that precisely and efficiently splices separate RNA units into functional coding mRNAs. REJ is mediated by short structured modular RNA segments of ~250bp that: (a) promote RNA::RNA interaction, (b) coopt the intrinsic cell splicing machinery to facilitate RNA trans-splicing, (c) minimizes translation of protein fragments from un-spliced RNA segments, and (d) encodes scar-free protein. We demonstrate that REJ is a broadly applicable system that can reliably split nearly any gene into complementary segments regardless of protein structure. To enable the use of this system we provide a toolbox of reporters and a web-based design tool to facilitate vector design. Demonstrated REJ applications include genetic complementation, intersectional labeling of cell types, and efficient expression of large proteins.

The telomere-binding protein Cdc13 specifically binds to single-stranded telomeric DNA, playing a critical role in telomere protection and length regulation. While extensive biochemical, molecular biological, and genetic studies have shown that Cdc13 can form dimers or oligomers in solution and bind telomeric DNA with high specificity, the dynamic mechanism of its loading onto telomeres is less well characterized. Using two single-molecule methods, single-molecule fluorescence resonance energy transfer (smFRET) and colocalization single-molecule spectroscopy (CoSMoS), we demonstrate that Cdc13 initially loads onto telomeres as a monomer. This is followed by the recruitment of a second monomer, forming a stable Cdc13 dimer on a 12-nucleotide telomeric DNA segment. Although genetic studies suggest that monomeric Cdc13 binding alone is insufficient to maintain telomere length, it underscores the Cdc13 monomers regulatory importance in coordinating telomere synthesis and protection. This monomer-to-dimer transition provides a mechanistic basis for understanding the multi-tasked roles of Cdc13 in telomere replication and protection.

DNA replication achieves error rates as low as 10-9-10-11 per base pair through the combined contribution of base selection, exonuclease proofreading, and mismatch repair. Among those processes, base selection determines the initial level of accuracy and is strongly modulated by local sequence context. Existing models address this dependence either by fitting individual rate constants for each sequence context or by invoking the global template properties, neither of which derives sequence dependence from the underlying thermodynamics and kinetics of base pair formation. Here we present a mechanism for sequence-dependent base selection fidelity, built from two physical properties: nearest-neighbor stacking thermodynamics and directional kinetic asymmetry. The model fits the experimentally observed mutation spectra from three mismatch repair-deficient organisms well (r=0.74, 0.70, and 0.63), and predicts that base-selection accuracy varies non-monotonically with temperature in a sequence-dependent manner. Our model, therefore, provides a framework that connects sequence-dependent thermodynamic and kinetic effects during nucleotide incorporation to experimentally observed mutation rates.

Translesion DNA synthesis (TLS) enables DNA replication across damaged DNA and promotes stress-induced mutagenesis that contributes to antibiotic resistance in bacteria. The conserved ImuABC mutasome is essential for TLS in many bacterial species, yet the molecular function of its accessory protein, ImuA, has remained elusive. Here we show that Mycolicibacterium smegmatis ImuA assembles into a hexameric complex, likely arranged as a dimer of trimers, with dual enzymatic activities that reshape current models of its role in DNA damage tolerance. We show that ImuA functions as an ATP-dependent helicase that preferentially unwinds DNA substrates containing single-stranded DNA overhangs and identify amino acids required for both hexamer formation and helicase activity. Unexpectedly, ImuA also possesses ATP-independent 5' exonuclease activity, selectively processing ssDNA substrates with free 5' ends. We show a basic patch on the N-terminus is essential for stabilizing both the nuclease motif and oligomerization. Together, these findings identify ImuA as an active DNA-processing enzyme rather than a passive accessory factor and establish oligomerization as a prerequisite for its function. Our work provides a mechanistic framework for understanding how ImuA may function within the ImuABC mutasome to coordinate DNA processing during translesion synthesis.

UvrD-family SF1A helicases play a variety of biological roles in DNA metabolism including replication, recombination, DNA repair, and conjugative plasmid transfer. The family is described by a subdomain architecture consisting of two RecA motor subdomains (1A and 2A) that each contain an auxiliary B-domain insertion (1B and 2B). Monomeric UvrD-family enzymes possess ATPase and 3-5 single-stranded DNA translocase activity but lack helicase activity. An enzyme dimer formed through a 2B-2B domain interface acts as a processive DNA helicase. A partial explanation for this observation, based on structural comparison between monomeric and dimeric DNA bound complexes, is that dimerization removes inhibitory contacts between the 2B domain and the double-stranded DNA, thus activating the helicase. However, biochemical observations reveal additional aspects of the dimeric mechanism that cannot solely be explained by movement of the 2B subdomain. Here, we present data showing that the Mycobacterium tuberculosis UvrD1 dimer interacts with the displaced strand of the duplex (i.e., the 5-3 strand in the direction of unwinding). In particular, bulky lesions on the displaced strand-including a thymine dimer-led to more efficient unwinding over a finite range of duplex lengths. These findings suggest a model for dimeric UvrD-family unwinding wherein one subunit makes intimate contacts with the displaced strand and that this interaction promotes the processive DNA unwinding unique to dimers.

The base excision repair pathway removes small base lesions from DNA and is initiated in humans by one of eleven DNA glycosylases with overlapping substrate specificities. Despite their central role in genome maintenance and transcription, systematic datasets describing small-molecule binders or inhibitors of DNA glycosylases are lacking. Here, through a collaborative effort across Scandinavia, we establish a high-throughput biochemical assay platform for human DNA glycosylases and use it to generate a systematic small-molecule screening dataset. A chemogenomic library was screened against nine glycosylases with assays of sufficient quality for hit identification, yielding primary screening and concentration-response data and identifying multiple previously unreported inhibitors. All primary and processed data, along with detailed assay protocols and metadata, are publicly available to support reuse in chemical biology, further compound optimization, assay development and comparative studies of DNA repair enzymes.

Long, customizable DNA molecules are an essential tool for in vitro single-molecule biophysics. Most studies still rely on the {lambda}-phage genome, which has a fixed sequence, a non-uniform GC distribution, and limited options for modification. To overcome these limitations, we developed a modular DNA (ModDNA) toolkit that integrates in silico sequence design with cloning, to create long (up to 100 kb) DNA molecules with entirely user-defined sequences. We used a computational workflow with curated databases to design a motif-depleted sequence backbone, into which any desired sequence can be inserted at any position. Golden Gate assembly then builds these sequences into constructs, which we extend iteratively in defined increments up to 100 kb. We show precise placement of local features, e.g. a fluorophore, a bound protein, and a DNA-origami structure, and illustrate the construct's utility in single-molecule visualization assays, including DNA loop extrusion by human condensin I. ModDNA offers a scalable, sequence-programmable alternative to {lambda}-DNA and other restrictive substrates. More broadly, the motif-aware feature-free design connects single-molecule biophysics to emerging efforts in machine-guided regulatory-sequence design and synthetic-genome engineering, where controlled programmable sequence context is increasingly essential.

Sliding clamp proteins play central roles in DNA metabolism, including replication and repair. The ring-shaped E. coli beta clamp accommodates double-stranded DNA and serves as a platform for proteins involved in multiple DNA transactions. The inner pore of the beta clamp harbors a series of positively charged and polar residues that can bind to the negatively charged backbone of the DNA. These residues are arrayed so that they do not align with the charged phosphates of the DNA backbone. It is hypothesized that this arrangement of these residues provides for the movement of the clamp on DNA as it alternates which residues are bound to the DNA backbone. In this work, we mutated specific charged and polar residues that project into the inner pore of the beta clamp. The beta clamp variants are dimers and have similar thermal stability and in general a similar ability to complement a temperature sensitive strain for growth. One exception was beta-Q149A, which appeared as higher-order species on a native gel although its hydrogen-deuterium exchange pattern measured by mass spectrometry was overall similar to WT beta. These variants all had decreased binding to DNA after loading. Optical tweezers experiments were used to monitor loading on single DNA molecules and measure the rate of beta clamp sliding on DNA. Consistent with the hypothesized role of positively charged residues in the beta inner pore, mutation of one residue resulted in a faster rate of sliding on DNA.

Summary dictyExpress 3.0 is a React/TypeScript reimplementation of the Dictyostelium discoideum transcriptomics web server, first introduced in 2009, that unifies bulk and single-cell exploration. A long-standing gene-expression resource for the Dictyostelium community, dictyExpress is also accessible through dictyBase, the central organism database for Dictyostelium. The new release preserves the curated bulk RNA-seq dashboards of earlier versions while adding an interactive module for a developmental scRNA-seq time course of wild-type AX4 and two cAMP-signalling mutants. Following visual analytics principles, interactive visualizations are coordinated through brushing and linking, with bulk time courses, differential expression, gene-ontology enrichment and the UMAP workspace sharing a common gene selection, enabling bookmarkable and exportable cross-modal analysis in one interface for Dictyostelium transcriptomics. Availability and implementation dictyExpress is available at https://app.dictyexpress.org/. Source code: https://github.com/biolab/dictyexpress-js. Contact lena.trnovec@fri-uni-lj.si, blaz.zupan@fri.uni-lj.si

Poly(ADP-ribose) polymerases (PARPs) are important regulators of DNA repair and cellular stress responses in eukaryotes. Although mammalian PARPs have been extensively characterized, comparatively little is known about PARPs in pathogenic filamentous fungi. Here, we define biochemical features of MoPARP1 from the plant-pathogenic fungus Magnaporthe oryzae. Mutation of the conserved catalytic glutamate E714 abolished detectable PARylation activity, whereas PAR generated by MoPARP1 was hydrolyzed by human PARG, suggesting synthesis of polymeric ADP-ribose. Several mammalian PARP inhibitors were effective against MoPARP1 in vitro, consistent with conservation of the catalytic inhibitor-binding pocket. DNA substrates containing 5' phosphate enhanced MoPARP1 catalytic activity, while both wild-type MoPARP1 and the catalytically inactive MoPARP1 E714A bound diverse DNA substrates. Quantitative analyses revealed that structurally distinct DNA substrates exhibit similar apparent binding affinities yet yield markedly different catalytic outputs, suggesting that DNA engagement alone does not determine productive PAR synthesis. MoPARP1 lacks the canonical N-terminal zinc-finger domains of human PARP1 and instead depends on its WGR-containing region for DNA association. Domain truncation analyses showed the BRCT-WGR region supports high-affinity DNA binding, whereas the WGR-PARP region retains catalytic competence despite weaker DNA affinity. Together, these findings establish a mechanistic framework for DNA-dependent PARylation in a filamentous fungal plant pathogen.

G-quadruplexes (G4s) are non-canonical nucleic acid structures with critical roles in gene regulation, genomic stability, and disease, making them prime targets for therapeutic and biotechnological applications. However, the absence of a curated, centralized, experimentally validated repository of short (mostly synthetic) G4-forming sequences, paired with appropriate single-stranded or hairpin controls, has limited reproducibility and hindered progress in the field. Here, we introduce G4All, a comprehensive, curated database of G4-forming short DNA and RNA sequences, complemented by rigorously selected non-G4 controls, all studied under roughly similar experimental conditions (around 100 mM potassium ion at near neutral pH). G4All consolidates sequences validated by diverse experimental methods, including circular dichroism, NMR, UV spectroscopy, with standardized annotations for topology, thermal stability, and experimental conditions. By providing both positive and negative datasets, G4All enables rigorous comparative analyses, assay benchmarking, and the development of predictive models. The database supports a broad range of applications, from fundamental studies of G4 biophysics to the rational design of aptamers, as well as benchmarking prediction algorithms. Future developments will expand G4All to include user-submitted datasets and more RNA sequences. Freely accessible, G4All offers a searchable interface and downloadable datasets, establishing a community-driven hub to accelerate discovery and standardization in G4 research. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=81 SRC="FIGDIR/small/743020v1_ufig1.gif" ALT="Figure 1"> View larger version (30K): org.highwire.dtl.DTLVardef@34c9f3org.highwire.dtl.DTLVardef@1b6a6feorg.highwire.dtl.DTLVardef@8d862borg.highwire.dtl.DTLVardef@1637155_HPS_FORMAT_FIGEXP M_FIG C_FIG

DNA helicases preserve genome stability by unwinding DNA during replication, repair, recombination, and transcription, yet their sites of action in vivo remain difficult to define. Here, we describe a sequencing-based strategy to map helicase activity genome-wide by coupling helicases to the single-stranded DNA-specific activation-induced cytidine deaminase (AID). Deamination of cytosines exposed during helicase-mediated DNA unwinding generates strand-specific mutational footprints that can be detected by whole-genome sequencing at near-nucleotide resolution. Using the Saccharomyces cerevisiae RecQ4-family helicase Hrq1, a functional homolog of human RECQL4, we generated the first genome-wide map of Hrq1 activity. Hrq1-dependent deaminations were highly enriched at RNA polymerase III (RNAPIII)-transcribed genes, particularly tRNA genes, where they occurred predominantly on the transcriptional template strand. This localization was reproducible using both overexpressed Hrq1-AID fusions and an inducible dimerization system that recruited AID to endogenously expressed Hrq1, and it was markedly reduced by helicase-inactivating mutation, indicating that active DNA unwinding underlies the observed signal. Hrq1 associated with nearly all tRNA genes irrespective of transcription level, replication orientation, or proximity to transposable elements, yet deletion or overexpression of Hrq1 did not detectably alter pre-tRNA abundance or RNA polymerase III recycling under the conditions tested. Application of the same approach to the PIF1-family helicase Rrm3 recovered its established enrichment at a subset of highly transcribed, head-on tRNA genes, validating the method. Together, these findings establish AID-mediated mutational footprinting as a general approach for mapping helicase activity in vivo and reveal an unexpected, widespread association of the RecQ4-family helicase Hrq1 with RNAPIII-transcribed genes.

Type II topoisomerases (topo IIs) are essential enzymes that regulate DNA topology through a strand-passage mechanism in which a duplex DNA (transfer-segment) is passed through a transiently cleaved second duplex DNA (gate-segment). Biochemical and structural approaches have revealed critical details of the binding and cleavage of the gate-segment DNA. However, capture of the transfer-segment DNA has proven more difficult to resolve due to the transient nature of the interaction. Nonetheless, selection of the transfer segment with a specific conformation or orientation relative to the gate segment is predicted to govern aspects of topo II activity, including chiral discrimination and the ability to reduce topological complexity below equilibrium. To determine the conformation and orientation of the transfer-segment relative to the gate segment, we combined experimental single-molecule measurements of topo II unlinking a single DNA crossing with Brownian dynamics simulations of the DNA crossing. By correlating the unlinking rate with the geometric features of the DNA crossing, we obtain the complete three-dimensional preferred crossing geometry for strand passage. Strikingly, the preferred crossing geometries for Escherichia coli topoisomerase IV and Methanosarcina mazei topoisomerase VI are distinct and provide structural models of the DNA synapse selected for strand passage along with a mechanistic basis for their differing activities and biological functions. The approach we develop is generalizable, providing unique insights into the kinetic selection of DNA synapse structure.