Giulia Palermo

Anti-CRISPRs (Acrs) are small protein inhibitors of CRISPR-Cas effectors that originate from the translated genetic material of bacteriophage. Harnessing the natural ability of Acrs to bind and disrupt CRISPR-Cas editing can provide enhanced spatiotemporal control of gene editing. Recent studies have revealed diverse structures and functions of Acrs, however, atomistic insight into the specific molecular mechanisms behind Acr inhibition is lacking. Here, we reveal how structure, function, and dynamics govern AcrIIC1 inhibition of Cas9 fromGeobacillus stearothermophilus(GeoCas9) via its HNH nuclease domain. An X-ray crystal structure of the GeoHNH-AcrIIC1complex reveals a conserved binding interface at the catalytic site and disruption of crucial electrostatic contacts known to modulate the thermostability of GeoCas9. AcrIIC1 binding also rewires the intrinsic dynamics of the GeoHNH domain, stimulates millisecond motions that are absent from the unliganded nuclease, and attenuates the guide RNA affinity of GeoCas9. Subsequent AcrIIC1 mutations in residues at its crystallographic binding interface uncouple Acr binding from inhibition, providing new insight into mechanism by which AcrIIC1 acts on GeoCas9.

Cas9 is a metal-dependent nuclease used for genome editing across diverse cells and organisms exhibiting distinct ionic environments, yet how metal ions regulate its catalytic function, and consequently its editing efficiency remains unclear. Here, molecular simulations, Markov state models, and mixed quantum–classical approaches reveal that divalent metals promote activation of the catalytic HNH domain through formation of a divalent metal-binding pocket (DBP) at the HNH–RuvC interface. This cryptic pocket emerges dynamically during HNH activation and is supported by NMR measurements revealing metal-dependent structural perturbations within the HNH–L2 region. Mutations targeting the DBP disrupt HNH activation and impair the coupled catalytic activity of both nucleases, identifying the pocket as a key regulator of Cas9’s metal-dependent activity. Together with quantum–classical simulations revealing distinct catalytic behaviours among divalent metals, these findings uncover an ion-dependent regulatory mechanism in Cas9 with insights for genome editing across diverse biological environments.

Graph attention networks (GATs) are emerging as powerful aritficial intelligence (AI) tools for learning biomolecular dynamics, yet extracting mechanistic insight from learned attention remains challenging. Here, we present an interpretable AI approach that combines molecular dynamics prediction with mechanistic interpretation through attention-derived communication networks. We develop three edge-aware GAT models - Edge-Conditioned, Edge-Injected, and Edge-Gated - that differ in how edge information is incorporated during message passing. Relative to a standard GAT, our edge-aware models improve coordinate prediction while recovering complementary aspects of residue communication. Application to Chignolin and HIV-1 protease demonstrates their ability to characterize communication networks underlying protein folding, allostery, and mutation-induced functional remodeling. Analysis of the communication networks using graph-based descriptors of communication throughput, intensity, and relay revealed that the Edge-Conditioned model preferentially emphasizes communication hubs characterized by high communication throughput and intensity, the Edge-Injected model preferentially highlights high-intensity communication within functionally important regions, and the Edge-Gated model most clearly resolves long-range communication relay. Overall, this approach provides an interpretable AI strategy for uncovering the communication mechanisms that underlie folding, allostery, and long-range signal propagation in molecular machines, while also supporting AI-guided modulation of biomolecular function and rational protein engineering.

CRISPR-associated transposons (CASTs) enable programmable DNA integration, yet how the TnsC regulator forms processive filaments on DNA to coordinate RNA-guided transposition in type V-K CAST systems remains unknown. Here, we integrate large-scale molecular simulations, interpretable deep learning using graph attention networks (GATs), and causal inference analyses to define the molecular determinants of TnsC filament nucleation and elongation. We show that TnsC nucleates by inducing localized DNA deformation that propagates along extended filaments, with Granger causality revealing that TnsC motions precede and predict DNA deformation. Interpretable GAT models demonstrate that elongation is determined during early recognition between incoming and DNA-bound subunits, followed by structural reorganization that regenerates the recruitment interface and enables processive assembly. These results elucidate the molecular mechanism of processive TnsC filament assembly and explain why isolated TnsC filaments preferentially elongate in the 5′ → 3′ direction, while accessory transposition factors can reshape the interaction landscape and alter filament growth polarity. Together, these findings advance our understanding of CAST function and inform the engineering of programmable DNA integration platforms. Beyond CAST systems, this work introduces an interpretable GAT approach as a general and transferable deep learning strategy for uncovering molecular mechanisms in biological systems, while demonstrating the power of causal inference for dissecting directional relationships in molecular dynamics.

To develop gene therapy tools based on CRISPR-associated transposons (CASTs), it is essential to define how transposon ends are recognized and paired during transposition. Tn7-like transposons typically contain asymmetric left- and right-end sequences that flank and define DNA cargo. However, how the transposase recognizes these different sequences and assembles them into a paired end complex for cut-and-paste transposition remains unknown. Here we present the cryo-EM structure of type I-F3 (VchCAST) CAST transposase TnsB in complex with transposon DNA ends and host factor IHF, along with biochemistry, molecular dynamics, and in vivo analyses. Our structure reveals the stoichiometry and architecture of the assembly, as well as the DNA distortions required to accommodate transposon end asymmetry. Molecular dynamics suggests that these distortions are required for the coordinated assembly of the complex. Physical pairing of asymmetric left and right ends results in a novel protein-protein interface that is required for transposition efficiency. Our findings explain how transposases regulate the pairing of transposon end sequences for high-fidelity recognition, reveal a model of transposon end synapsis, and suggest future avenues to engineer DNA cargo for genome-editing applications.

CRISPR-Cas12a is a versatile RNA-guided nuclease that has rapidly gained prominence for its dual functionality in genome editing and nucleic acid detection. In this Review, we discuss the structural, biochemical and mechanistic features of Cas12a that underpin its autonomous processing of the guide RNA and indiscriminate cleavage of single-stranded DNA, which enable Cas12a applications ranging from gene therapy to rapid diagnostics. We discuss key allosteric regulators and functional modules that orchestrate Cas12a activity, focusing on the core regulatory structural elements that control maturation of the guide RNA, target specificity, and both cis-cleavage and trans-cleavage activities, including the determinants of off-target cleavage. We provide a comparative analysis of Cas12a and the widely used Cas9, which further illuminates the distinctive attributes of Cas12a, and discuss recent advances in the characterization of its orthologues and in the development of engineered variants that expand its capabilities. Collectively, we present a comprehensive understanding of Cas12a and its increasing impact on biotechnology, therapeutics and molecular diagnostics.