Repeat region engineering of Cas13a crRNA enables conformational gating-based autocatalytic CRISPR biosensing Authors: Wei Ding, Xiaomin Yang, Yutong Yang, Mingxin Liu, Chenghui Liu, Jinyi Wang, Xiaoling Liu Published: 2026 ABSTRACT AbstractCrRNA engineering has emerged as a pivotal strategy for extending CRISPR–Cas13a biosensing. However, structural modulation of the direct repeat (DR) region remains exceptionally challenging due to its intricate architecture and the high energetic barrier of the Cas13a–crRNA interface, which is conventionally viewed as a rigid and immutable scaffold. Here, we demonstrate that the DR region is instead a programmable topological element with unexpected structural plasticity. By systematically engineering the DR through sequence insertion and structural splitting, we identified multiple DR variants that retain robust catalytic activity. Crucially, this topological reconfiguration enables Cas13a activity to be precisely gated by unmodified nucleic acid blockers, a level of regulation unattainable with the wild-type crRNA. Building on this flexible modulation, we developed Dre-CRISPR, a DR-engineered platform that couples target-triggered DR restoration to a self-reinforcing autocatalytic loop. This self-amplifying system provides a 2 × 106-fold sensitivity enhancement over nonamplified systems. Furthermore, the Dre-CRISPR platform extends the diagnostic scope of Cas13a to a broader spectrum of analytes, ranging from microRNAs to enzymatic activities and heavy metal ions. Our findings redefine the crRNA scaffold as a versatile signaling node and provide a generalizable framework for developing high-sensitivity, self-amplifying CRISPR biosensors through topology-driven guide RNA engineering. Introduction The CRISPR–Cas13a system has emerged as a versatile platform for nucleic acid detection, particularly for RNA targeting, owing to its programmable target recognition and unique signal amplification properties [1–3]. Guided by a specific CRISPR RNA (crRNA), Cas13a undergoes conformational rearrangements upon target binding to activate its HEPN nuclease domains, triggering cis-cleavage of the target RNA accompanied with more attractive collateral trans-cleavage of surrounding nontarget RNAs [4–6]. While this target-triggered trans-cleavage activity underlies the broad utility of Cas13a in RNA diagnostics and molecular biosensing [7–10], finely tuning Cas13a activation remains a key challenge for improving sensing performance and extending Cas13a-based biosensing beyond RNA detection. In this regard, programmable modulation of Cas13a crRNA architecture provides a particularly attractive and effective strategy. CrRNA plays a central role in Cas13a function by directing target recognition and promoting catalytic activation [11, 12]. A typical Cas13a crRNA consists of a conserved direct repeat (DR) region that mediates Cas13a protein recruitment and ribonucleoprotein (RNP) assembly, and a variable spacer region responsible for sequence-specific target recognition. Current strategies for regulating Cas13a-based sensing mainly focus on engineering the spacer region [13–15] and its flanking elements [16–20]. For example, secondary structure-based designs, such as hairpin spacers [21], foldback crRNAs [22], and crRNA/anti-tag pairing architectures [17–19], have been extensively proposed for rationally tuning the spacer configuration in order to enhance either sensing specificity or capability for detecting different RNAs. Beyond structural tuning, chemical crRNA modification strategies, such as reversible RNA acylation, have also been employed to cage crRNA activity with stimuli-responsive groups [23–25]. While these designs offer distinct advantages, they typically rely on complex auxiliary RNA architectures that lack universal adaptability. This stems from the rigid requirement for spacer-target complementarity because the sequence of a given target is fixed, which restricts the available design space for topological modifications and thereby hinders the flexible expansion needed for diverse sensing applications. In contrast, since the conserved DR region of the Cas13a crRNA is independent of target RNA sequences, its structural modulation may offer a more versatile pathway toward expanded biosensing applications. However, unlike the compact DR sequence of Cas12a crRNA whose structural tuning has emerged as a hot topic since 2025 [26–29], Cas13a crRNA possesses a more sophisticated DR architecture that can form an exceptionally stable RNP complex with Cas13a. This tight protein–RNA association creates a high energetic barrier that has long deterred the engineering of its core scaffold. Consequently, this highly structured DR hairpin, although indispensable for Cas13a recruitment and RNP assembly, has largely been relegated to the role of a passive structural scaffold. While significant efforts have optimized the peripheral spacer and flanking sequences, the potential of the DR region as an active regulatory handle remains largely unexplored. Herein, we redefine the Leptotrichia wadei (Lwa) Cas13a DR region as a programmable topological switch, effectively transforming it from a simple structural backbone into an active regulatory module. By systematically investigating how structural reconfigurations such as sequence insertion and splitting modulate the catalytic activities of Cas13a, we reveal an unexpected degree of structural plasticity within the tightly bound RNP complex. This discovery shifts the locus of CRISPR regulation from the conventional spacer and target interface to the core protein-binding scaffold. Our topology-controlled approach allows simple and unmodified single-stranded DNA (ssDNA) blockers to predominantly intercept functional RNP assembly, thereby effectively minimizing basal activity. Building on these biophysical principles, we developed the Dre-CRISPR autocatalytic circuit, a self-reinforcing system utilizing the target-responsive restoration of a topologically gated crRNA. By exploiting the programmable topology of the DR scaffold rather than relying solely on spacer and target interactions, we establish a fundamentally new regulatory dimension for Cas13a, providing a universal and high-contrast sensing framework. Ultimately, our findings uncover a hidden regulatory plasticity of the crRNA backbone, transforming the conceptual scope of CRISPR-based sensing from simple sequence detection to programmable molecular logic. Materials and methods Reagents and materials LwaCas13a was purchased from East-Mab Biomedical Technology (Jiangsu, China). Recombinant RNase inhibitor and the in vitro Transcription T7 Kit were obtained from Takara Bio (Dalian, China). Streptavidin magnetic beads, 20 × TE buffer (pH 8.0), and 4S GelRed nucleic acid stain were supplied by Sangon Biotech (Shanghai, China). Uracil-DNA glycosylase (UDG), uracil glycosylase inhibitor (UGI), apurinic/apyrimidinic endonuclease 1 (APE1), T4 polynucleotide kinase (T4 PNK), and recombinant albumin (BSA) were purchased from New England BioLabs (MA, USA). All DNA and RNA oligonucleotides were synthesized by Tsingke Biotech (Beijing, China) or Sangon Biotech (Shanghai, China), and are listed in Supplementary Table S1. Fluorescence analysis of crRNA variants To evaluate the activities of different crRNA variants, Cas13a trans-cleavage fluorescence assays were performed in the presence of RNA Activator I. Each reaction mixture (10 μl) contained 50 nM of the respective crRNA variant, 50 nM Activator I, 50 nM Cas13a, 1 μM FQ reporter, and 1 × Cas13a buffer (40 mM Tris–HCl, 60 mM NaCl, 1.5 mM MgCl2, pH 7.5). After incubation at 37°C for 60 min, the reactions were terminated by heating at 95°C for 2 min to inactivate Cas13a. The reaction mixtures were then diluted with 90 μl of 1 × TE buffer (20 mM Tris–HCl, 1 mM EDTA, pH 8.0), and the fluorescence signals were recorded on a Hitachi F-7000 fluorescence spectrophotometer (Japan) with an excitation wavelength of 488 nm and an emission range of 500–600 nm. Evaluation of ssDNA blocker-mediated inhibition of CRISPR–Cas13a trans-cleavage activity To evaluate the inhibitory effects of different ssDNA blockers on CRISPR–Cas13a trans-cleavage activity, a pre-hybridization step was first performed to form blocker-bound crRNA complexes. Briefly, 500 nM crRNA and 750 nM ssDNA blocker were mixed in 1 × Cas13a buffer in a total volume of 10 μl (crRNA:blocker molar ratio = 1:1.5). For the control sample, no blocker was added. The mixture was annealed by incubation at 90°C for 10 min, followed by 40°C for 10 min. The resulting blocker-bound crRNA complexes were then used for fluorescence measurements. The trans-cleavage reaction mixture (10 μl) contained 50 nM blocked crRNA (or 50 nM unblocked crRNA for the control), 50 nM Activator I, 50 nM Cas13a, 1 μM FQ reporter, and 1 × Cas13a buffer. Following incubation at 37°C for 60 min, the reactions were terminated by heating at 95°C for 2 min, and subsequently diluted with 90 μl of 1 × TE buffer for fluorescence quantification. Quantitative detection of miR-21 To evaluate the in vitro detection of miR-21, a pre-hybridization step was first performed to assemble the proposed TRAC module within the Dre-CRISPR platform. Briefly, 250 nM crRNA-d’-5′ and 375 nM ssDNA gating blocker (see Supplementary Table S1) were mixed in 1 × Cas13a buffer in a total volume of 10 μl (molar ratio = 1:1.5). The mixture was annealed by incubation at 90°C for 10 min, followed by 40°C for 10 min. The resulting preassembled ssDNA gating blocker–crRNA-d’-5′ complex was then used for fluorescence measurements. The reaction mixture (10 μl) contained 25 nM crRNAmiR-21, 25 nM preassembled ssDNA gating blocker–crRNA-d’-5′ complex, 25 nM crRNA-d-3′, 50 nM Activator I, 50 nM Cas13a, 1 μM FQ reporter, and different concentrations of miR-21 in 1 × Cas13a buffer. After incubation at 37°C for 60 min, the reactions were terminated by heating at 95°C for 2 min, and subsequently diluted with 90 μl of 1 × TE buffer for fluorescence quantification. For miR-21 detection in the absence of the TRAC circuit, the reaction mixture (10 μl) contained 25 nM crRNAmiR-21, 25 nM Activator I, 25 nM Cas13a, 1 μM FQ reporter, and varying concentrations of miR-21 in 1 × Cas13a buffer. Following incubation at 37°C for 60 min, the reactions were terminated by heating at 95°C for 2 min and subsequently diluted with 90 μl of 1 × TE buffer for fluorescence quantification. Quantification of UDG activity To detect UDG activity, a pre-hybridization step was first performed to assemble the B3-dU–crRNA-d-5′ complex. Briefly, 250 nM crRNA-d-5′ and 375 nM B3-dU (see details in Supplementary Table S1) were mixed in 1 × UDG buffer [20 mM Tris–HCl, 10 mM (NH₄)₂SO₄, 10 mM KCl, 2 mM MgSO₄, 0.1% Triton X-100, pH 8.8] in a total volume of 10 μl (molar ratio = 1:1.5). The mixture was annealed by incubation at 90°C for 10 min, followed by 40°C for 10 min. For the UDG reaction, the reaction mixture (5 μl) contained 50 nM preassembled B3-dU–crRNA-d-5′ complex, and different concentrations of UDG in 1 × UDG buffer. The reaction was incubated at 37°C for 2 h. Then, prepare the reaction solution for fluorescence measurement. The reaction system (10 µl) contains 5 µl of UDG reaction product, 25 nM pre-assembled ssDNA gating blocker–crRNA-d’-5′, 50 nM crRNA-d-3′, 50 nM Activator I, 50 nM Cas13a, and 1 µM FQ reporter in 1 × UDG buffer. After incubation at 37°C for 60 min, the reactions were terminated at 95°C for 2 min, and subsequently diluted with 90 μl of 1 × TE buffer for fluorescence quantification. Pb²⁺ sensing via DNAzyme-coupled Dre-CRISPR For Pb²⁺ detection, 500 nM crRNA-i-5′ and 750 nM H1 were annealed in 1 × Cas13a buffer (10 μl of total volume) at 90°C for 10 min followed by 40°C for 10 min. In parallel, 6 μM DNAzyme and 3 μM H0 were annealed in 1 × DNAzyme buffer (50 mM Tris–HCl, 100 mM NaCl, 0.05% Tween 20, pH 7.5) using the same program. The preassembled DNAzyme–H0 complex was then coupled to streptavidin magnetic beads. Briefly, 5 μl of 10 mg/ml streptavidin magnetic beads were washed three times with 1 × DNAzyme buffer and diluted to 5 mg/ml with the same buffer. Next, 10 μl of 5 mg/ml streptavidin magnetic beads and 10 μl of preassembled DNAzyme–H0 were mixed and incubated at room temperature for 60 min on a rotator. After coupling, the supernatant was discarded, and the beads were washed three times with 1 × DNAzyme buffer and finally resuspended in 20 μl of the same buffer. For the Pb²⁺-triggered DNAzyme cleavage reaction, the reaction mixture (5 μl) contained 3 μl of DNAzyme–H0-coupled magnetic beads, different concentrations of Pb²⁺, 5 mM MgCl₂, and 1 × DNAzyme buffer. The reaction was incubated at 37°C for 2 h on a rotator. After magnetic separation, the supernatant was collected for the subsequent Cas13a assay. For fluorescence measurement, the reaction mixture (20 μl) contained 5 μl of the collected supernatant, 25 nM preassembled ssDNA gating blocker–crRNA-d’-5′ complex, 25 nM crRNA-d-3′, 25 nM preassembled H1–crRNA-i-5′ complex, 25 nM crRNA-i-3′, 50 nM Activator I, 50 nM Cas13a, and 500 nM FQ reporter in 1 × Cas13a buffer. After incubation at 37°C for 60 min, the reactions were terminated at 95°C for 2 min, and subsequently diluted with 90 μl of 1 × TE buffer for fluorescence quantification. Results Structural plasticity and functional tolerance of the Cas13a crRNA scaffold to topological reconfiguration The Cas13a crRNA typically comprises a conserved DR region, functioning as a structural handle or scaffold, and a variable spacer sequence responsible for target recognition [30]. The hairpin/stem-loop architecture of the DR is indispensable for Cas13a recognition, RNP assembly, and subsequent catalytic activation. Structurally, the DR features two distinct loop regions located at the lateral and apical positions [6]. Despite their apparent structural significance, the degree to which these specific motifs can tolerate topological intervention without compromising enzymatic activity remains poorly understood. To address this, we systematically engineered the crRNA scaffold into two primary topological classes (Fig. 1A). The first class, continuous scaffolds, maintains the covalent integrity of the crRNA while introducing sequence insertions into the DR loops (a, b). The second class, split scaffolds, involves the structural fragmentation of the crRNA at strategic positions. This category ranges from a fundamental decoupling of the DR and the spacer sequence, to internal binary splitting or splicing of the DR hairpin (c–i), and ultimately to ternary split architectures where the DR is fragmented into multiple independent segments (j, k). Impact of topological modifications within the crRNA scaffold on Cas13a cleavage activities. (A) Schematic representation of crRNA variants (crRNA-a∼crRNA-k) engineered through sequence insertion or structural splitting at diverse positions within the DR region. (B) Trans-cleavage fluorescence signals for the indicated crRNA variants. The left and right bars represent signals recorded at 10 and 60 min, respectively. (C) Denaturing polyacrylamide gel electrophoresis (PAGE) analysis of the corresponding cis-cleavage products at 10 and 60 min. For each crRNA variant, the second and third “+” symbols denote the results at 10 and 60 min, respectively. (D and E) Time-course analysis of the crRNA-i variant, highlighting the progressive increase in trans-cleavage fluorescence (D) and the corresponding accumulation of cis-cleavage products (E). Data are presented as mean ± S.D., n = 3, three technical replicates. We evaluated the effects of such scaffold-engineered crRNA variants using the wild-type (WT) crRNA as a reference. The trans- and cis-cleavage activities of the Cas13a system were assessed at 10 and 60 min. Trans-cleavage was quantified via end-point fluorescence measurements using Activator I as the triggering RNA oligonucleotide, while cis-cleavage was validated by denaturing PAGE monitoring the degradation of Activator II, an extended sequence designed to facilitate the visualization of cleavage products (Fig. 1). As shown in the fluorescence data in Fig. 1B, variants featuring loop insertions (crRNA-a, b) and several DR-split scaffold designs (crRNA-c, d, e, f) exhibited robust signals comparable to the WT system. Notably, while the 5′-truncated variant (crRNA-f) preserved high activity, further splitting at its apical region (crRNA-g, h) led to a significant reduction in signaling. We observed the most radical structural constraints in the ternary split architectures (crRNA-j, k), where partitioning the crRNA into three independent segments resulted in a near-complete loss of catalytic activity. Importantly, these trans-cleavage activity profiles were consistently corroborated by the denaturing PAGE analysis of the cis-cleavage products (Fig. 1C and Supplementary Fig. S1). These results suggest that although Cas13a can tolerate certain backbone discontinuities of the DR region, the excessive fragmentation of the DR likely disrupts the critical structural integrity required for enzymatic activation. Among the functional variants, the Spacer-detached design (crRNA-i) exhibited a distinctive kinetic profile, characterized by a marked discrepancy between the 10- and 60-min measurements (Fig. 1B and C). To further dissect the mechanisms underlying this time-dependent activation, we conducted a multi-point time-course fluorescence analysis (Fig. 1D). The resulting profile clearly reveals a significant lag phase during the initial 10 min, followed by a rapid surge in signal intensity between 10 and 20 min. The corresponding denaturing PAGE analysis in Fig. 1E elucidates the molecular basis of this delayed activation, where the spacer and Activator II initially formed a stable hybridization complex. The gradual depletion of this intermediate over 60 min, synchronized with the progressive emergence of cleavage products, suggests that the spontaneous reassembly of the active RNP from three separate components consisting of Cas13a, the DR fragment, and the pre-formed Spacer-Activator II complex is a viable yet rate-limiting process. To further elucidate the global conformational mechanisms underlying the varied catalytic activities of our engineered scaffolds, we performed structural modeling of the ternary complexes (Cas13a-crRNA variant-Activator I) using AlphaFold 3 [31]. Structural alignment against the catalytically active WT ternary complex (WT CRISPR) revealed a clear correlation between structural deviation and enzymatic activity (Supplementary Fig. S2 and Supplementary Fig. S3). Functional variants (crRNA-a to f, and i) maintained high structural fidelity, with root-mean-square deviation (RMSD) values below 1.3 Å, demonstrating the inherent structural plasticity of the DR region to accommodate topological remodeling while preserving the catalytically competent conformation [32]. In contrast, the partially active variant (crRNA-g) and inactive variants (crRNA-h, j, and k) exhibited significantly higher RMSD values (>1.8 Å) relative to the active WT CRISPR complex. Strikingly, when we aligned the ternary complexes of these inactive variants against the unactivated Cas13a–crRNA, they displayed remarkably low RMSD values (<1.0 Å) (Supplementary Fig. S4). These computational insights indicate that excessive fragmentation of the DR scaffold specifically impedes the activator-induced allosteric transition, effectively locking the Cas13a enzyme in a pre-activation conformation. Together, these results delineate the structural threshold of DR plasticity required for Cas13a activation. Ultimately, this redefines the Cas13a crRNA as a topologically reconfigurable scaffold instead of a covalently constrained backbone, providing a mechanistic foundation for the rational design of dynamically regulated crRNA architectures toward diverse biosensing applications. Topological reconfiguration of the DR enables robust control of Cas13a via unmodified nucleic acids The topological plasticity of the DR scaffold observed above suggests that its structural integrity can be modified without compromising core enzymatic function. We hypothesized that the defined split sites in several active variants, specifically the internal-split (crRNA-d) and binary-split (crRNA-i) configurations, would act as entry points that lower the energetic barriers for competitive conformational gating-based biosensing. Toward this goal, we wish to present in this study the successful transition of the Cas13a system from a constitutively active state to a controllable platform by just using simple, unmodified ssDNA blockers to interfere with the reassembly of the fragmented scaffold. To validate this regulatory strategy, we first examined the susceptibility of the internal-split variant (crRNA-d) to ssDNA-mediated inhibition alongside a WT crRNA control (Fig. 2A). As illustrated in the screening results, crRNA-d was highly susceptible to exogenous interference, where blockers targeting different segments of its split DR region consistently reduced trans-cleavage activity by 63.1%–91.6%. Notably, the WT crRNA remained completely unaffected by these same blockers. A similar inhibitory profile was observed for crRNA-i, where blocking either the full-length DR or its individual 5′ and 3′ segments, resulted in potent suppression of the signal, with inhibition efficiencies ranging from 78.4% to 98.3%, while the WT control again showed no signal reduction (Fig. 2B). Furthermore, we verified that this ssDNA-mediated inhibition remained robust across a wide range of Mg2+ concentrations, indicating that the regulatory effect is independent of divalent cation fluctuations (Supplementary Fig. S5). Conversely, none of the ssDNA blockers effectively inhibited the WT crRNA or the 5′-truncated variant (crRNA-f), even when the entire DR region was targeted (Fig. 2C). These results suggest that the high intrinsic affinity between Cas13a and the canonical crRNA, combined with the extreme stability of the active RNP–target complex, renders the wild-type system largely refractory to simple nucleic acid interference. In contrast, topological reconfiguration of the DR generates structurally permissive states that are accessible to regulation via simple hybridization with unmodified ssDNA. Programmable regulation of Cas13a trans-cleavage activity via engineered crRNA scaffolds. (A–C) Trans-cleavage fluorescence intensities and corresponding inhibitory efficiencies for (A) crRNA-d, (B) crRNA-i, and (C) crRNA-f and WT systems in the presence of diverse ssDNA blockers. (D) Comparative analysis of trans-cleavage fluorescence and inhibitory efficiencies for WT and engineered variants (crRNA-d, f, and i) regulated by anti-tag RNA. Schematic diagrams indicating the corresponding blocker binding sites are provided for all variants. Data are presented as mean ± S.D., n = 3, three technical replicates. We further investigated whether anti-tag pairing could serve as an alternative regulatory mechanism for the Cas13a system. This strategy, which relies on the hybridization between an extended sequence on the activator (the anti-tag) and its complementary region within the crRNA DR to sterically hinder RNP activation, has been successfully applied to other Cas13a orthologs [17–19]. However, our results indicate that even after pre-annealing the anti-tag RNA with its corresponding crRNAs, inhibition of the WT system remained minimal (∼25.1%), while crRNA-f reached only moderate inhibition of 56.0%. In striking contrast, the engineered crRNA-d and crRNA-i variants were inhibited by over 90% under identical conditions (Fig. 2D). Collectively, these findings demonstrate that engineering DR-mediated interactions converts Cas13a from a rigid, poorly regulatable enzyme into a flexible system that can be precisely controlled by simple nucleic acid strands. Molecular mechanism of ssDNA-mediated inhibition of DR-split crRNA-guided Cas13a activation To elucidate the molecular mechanism underlying the enhanced controllability of our engineered scaffolds, we investigated whether the ssDNA blocker acts by intercepting the de novo assembly of the Cas13a–crRNA complex or by disassembling a pre-formed, active RNP. Anticipating that the exceptional stability of the Cas13a RNP might render it refractory to post-assembly interference, we systematically varied the order of component addition. As illustrated in Fig. 3A and B, fluorescence analysis revealed a pronounced dependence on the order of addition for the split system. When the ssDNA blocker (B3) was pre-hybridized with the split crRNA components prior to the introduction of Cas13a (Regimes I and II), trans-cleavage activity was nearly abolished. In striking contrast, the addition of B3 to a pre-assembled Cas13a-split crRNA-Activator I complex resulted in only marginal inhibition, with the system maintaining high catalytic turnover (Regime III). As expected, a WT crRNA control remained completely refractory to B3 blockers; neither pre-hybridization (Regime II) nor post-assembly addition (Regime III) could meaningfully perturb its robust trans-cleavage activity (Supplementary Fig. S7A). These results suggest that ssDNA-mediated regulation occurs through an interception mechanism that must take place before the formation of the functionally mature RNP. Mechanistic characterization of ssDNA-mediated interception of DR-split crRNA assembly and Cas13a activation. (A) Schematic of the three order-of-addition experimental regimes (I, II, and III) used to evaluate the inhibitory mechanism. (B) Trans-cleavage fluorescence intensities corresponding to the regimes in panel (A). (C) Native PAGE analysis of the assembly states for the crRNA-d system. Gold and red boxes indicate the different order-of-addition regimes as defined in panel (A). Data are presented as mean ± S.D., n = 3, three technical replicates. Native PAGE analysis provided further structural evidence for this mechanism (Fig. 3C). When B3 was allowed to bind the split crRNA components before RNP assembly, we observed the clear formation of a B3–split crRNA heteroduplex (lanes 6 and 8). Parallel analysis of the WT crRNA further highlighted the structural resilience of the canonical scaffold. Ultimately, the B3 blocker was incapable of preventing the formation of the active WT RNP complex upon target introduction (Supplementary Fig. S7B). Conversely, for the split system, when Cas13a-containing complexes were pre-assembled prior to B3 addition, only limited changes in band mobility were observed (lanes 5, 7, 9, and 10). Although a weak B3-associated shifted band remained detectable in lane 9, this phenomenon likely reflected the dynamic behavior of the complex. Specifically, in the absence of the target Activator I, a minor fraction of the pre-assembled Cas13a–crRNA complexes may exist in a flexible and loose conformational state, temporarily exposing the binding region to B3. However, the overall pattern confirmed that this limited post-assembly interaction was insufficient to substantially perturb the pre-assembled Cas13a-bound complexes. Similar findings were also obtained for the crRNA-i variant (Supplementary Fig. S6, Supplementary Fig. S7C, and Supplementary Fig. S7D), confirming the generalizability of this inhibitory mechanism across diverse split topologies. To gain atomistic insights into how these ssDNA blockers structurally impede Cas13a activation, we performed AlphaFold 3 modeling of the blocker-bound complexes (Supplementary Fig. S8). For the canonical WT crRNA, the predicted binding of ssDNA blockers (B1 or B3) failed to meaningfully perturb the active ternary conformation, yielding low RMSD values (1.270 and 0.898 Å, respectively) relative to the unblocked active WT CRISPR complex (Supplementary Fig. S9). This structural rigidity strongly corroborates the experimental resistance of the WT system to ssDNA interference. Conversely, for the engineered split variants, the formation of blocker-crRNA heteroduplexes (B3-crRNA-d and B1-crRNA-i) severely distorted their catalytically active conformations (RMSD > 2.1 Å). Strikingly, structural alignment revealed that these blocker-bound split complexes mirror the unactivated Cas13a–crRNA state (RMSD 1.059 Å and 1.270 Å, respectively). Collectively, these results demonstrate that ssDNA blockers inhibit Cas13a primarily by intercepting the assembly pathway rather than by disrupting the exceptionally stable, pre-formed active complexes. By sequestering the split crRNA segments in a non-productive hybridized state prior to RNP formation, the blocker effectively traps the Cas13a enzyme in a locked, pre-activation conformation. This mechanism creates a robust kinetic window for regulation, preventing the complex from reaching its catalytically active conformation. These mechanistic insights provide a solid foundation for designing stimuli-responsive biosensors, where the controlled release of a blocker can be used to trigger de novo RNP assembly and subsequent signal propagation. Design and optimization of the Dre-CRISPR autocatalytic amplification strategy Among the diverse engineered crRNA variants, crRNA-d was of particular interest due to its ability to maintain high catalytic activity while remaining highly responsive to ssDNA-mediated inhibition. Leveraging these properties, we have developed Dre-CRISPR, a DR-engineered CRISPR–Cas13a amplification platform. The core of this system is the Trans-cleavage-triggered Reassembly Amplification Circuit (TRAC), which utilizes an engineered ssDNA gating blocker to conditionally regulate the assembly of the internal-split crRNA-d (Fig. 4A). In its initial state, the gating blocker consisting of a cleavable polyU bulge hybridizes with the crRNA-d’-5′ fragment. This sequestration prevents the functional assembly of the 5′ fragment with crRNA-d-3′ and Cas13a, which effectively locks the system in a dormant state. The amplification cycle is initiated by a target-triggered RNP complex. The trans-cleavage activity of this initial RNP specifically targets and cleaves the polyU bulge within the gating blocker. Such a cleavage event triggers the dissociation of the blocker from the 5′ fragment and subsequently releases the caged crRNA components. Once released, these fragments spontaneously reassemble with Cas13a and the activator to form secondary active RNPs. Crucially, these newly formed RNPs can then cleave additional gating blockers to establish a self-reinforcing autocatalytic circuit, which leads to the exponential amplification of fluorescence signals. Design and optimization of the Dre-CRISPR autocatalytic amplification platform. (A) Schematic illustration of the Dre-CRISPR mechanism. Target-responsive DR-gated architectures (e.g. RNA triggers, enzymes, or metal ions) initiate the TRAC, leading to self-reinforcing signal amplification. (B) Systematic optimization of gating blocker structural parameters. The inhibitory efficiency was evaluated by varying the complementarity length (left) from 18 to 26 nt and the polyT bulge size (right) from 7T to 11T. (C) Influence of the gating blocker molar ratio. Fluorescence responses were recorded at different molar ratios of the crRNA-d′-5′ fragment relative to the blocker, ranging from 1:1.0 to 1:2.0, to determine the optimal suppression of background activation. Data are presented as mean ± S.D., n = 3, three technical replicates. To suppress background activation in the TRAC circuit, we optimized the gating blocker by tuning its hybridization length and bulge size (Fig. 4B). Initial screening using polyT bulges showed that inhibition efficiency improved significantly as the complementarity length increased from 18 to 26 nt. The 26-nt configuration effectively maintains the system in a dormant state by forming a stable duplex. Within this framework, we found that a 9-nt loop provided the ideal balance between minimal background activation and efficient gating. For the final TRAC implementation, we substituted the central five nucleotides of the 9T loop with uridine (4T + 5U) to leverage the higher trans-cleavage affinity of Cas13a toward U-rich substrates. We then fine-tuned the molar ratio of the crRNA-d’-5′ fragment to the blocker, reaching a point of minimal background interference at a ratio of 1:1.5 (Fig. 4C). Collectively, these results establish Dre-CRISPR as a streamlined, one-enzyme cascade strategy that successfully balances tight gating control with rapid activation kinetics for sensitive biomarker detection. Ultrasensitive microRNA detection enabled by the Dre-CRISPR platform with cascade amplification We next established a nucleic acid-responsive Dre-CRISPR platform to systematically evaluate its analytical performance in recognizing low-abundance targets, using miR-21 as a model analyte [33–35]. By integrating 5′ truncation and apical loop insertion, we engineered a target-responsive DR-gated crRNA architecture (Fig. 5A) optimized for high-contrast sensing. In this design, a miR-21 recognition sequence was introduced at the 5′ terminus, while a partially complementary sequence was embedded within the apical loop to maintain a stable, self-inhibited conformation (crRNAmiR-21). Upon target binding, miR-21 initiates a toehold-mediated strand displacement reaction that restores the Cas13a-compatible scaffold, which subsequently activates the TRAC loop to produce an amplified fluorescence output. To maximize detection efficiency, we optimized the blocking architecture of the crRNAmiR-21 scaffold by systematically adjusting the number of occluded bases within both the miR-21 binding sequence embedded in the apical loop (Bn) and the DR region (Dn). It was found that a total occlusion pattern of D5 + B15 yielded the highest signal-to-background (S/B) ratio, thereby providing a robust dynamic range for subsequent molecular recognition (Fig. 5B). Engineering a self-caged Dre-CRISPR platform for ultrasensitive nucleic acid recognition. (A) Schematic of the one-pot, miR-21-responsive activation mechanism. Target binding triggers a toehold-mediated strand displacement reaction within the self-caged crRNAmiR-21 architecture, thereby restoring the functional scaffold and initiating the TRAC amplification loop. (B) Systematic optimization of the self-caged crRNAmiR-21 architecture to maximize the signal-to-background (S/B) ratio in the absence of the TRAC circuit. The fluorescence response and corresponding S/B values are plotted for various occlusion patterns, where Dn and Bn denote the number of occluded bases in the DR and apical loop regions, respectively. The D5 + B15 configuration was identified as the optimal design for subsequent analytical evaluations. (C) Sensitivity of the Non-TRAC sensing system. Fluorescence increments (ΔF) in response to varying concentrations of miR-21 (from 1.0 to 100 nM) without the TRAC circuit. (D) Ultrasensitive detection of miR-21 using the integrated Dre-CRISPR platform. Fluorescence signals were recorded across a dynamic range from 0.5 to 500 fM, demonstrating the enhancement provided by cascade amplification. (E) Modularity and sensitivity evaluation of the WT-control system. miR-21 directly activated its cognate WT-crRNA coupled to the identical TRAC circuit, with detection spanning from 0.5 to 250 pM. Data are presented as mean ± S.D., n = 3, three technical replicates. Prior to systematically evaluating the analytical performance, we sought to validate the stepwise molecular mechanism of the Dre-CRISPR strategy for miR-21 detection. To this end, we performed a sequential reaction and analyzed the products using native PAGE (Supplementary Fig. S10). In the absence of the target miR-21, when the downstream elements (crRNA-d-3′, Blocked-crRNA-d’-5′, and Activator I) were co-incubated with various upstream components (lanes 12, 13, and 14), only a faint background band was observed, which likely results from trace amounts of unblocked crRNA-d’-5′ or transient structural fluctuations within the untriggered state. Upon the introduction of the target miR-21, a fast-migrating band corresponding to the cleaved ssDNA gating blocker clearly emerged (lane 11), indicating that the Blocked-crRNA-d’-5′ was efficiently digested by the trans-cleavage activity of the target-activated Cas13a/crRNAmiR-21/Activator I complex. Furthermore, in the complete system where crRNA-d-3′ was also introduced alongside miR-21 (lane 15), the gating blocker was similarly cleaved, yielding a prominent and distinct band of the fully hybridized crRNA-d-3′/crRNA-d’-5′/Activator I complex. Together, these biochemical results clearly demonstrate that the target-triggered release of the split crRNA fragment successfully drives the de novo assembly of the active complex, thereby initiating the downstream TRAC circuit. Under optimized reaction conditions, we first evaluated the performance of the engineered crRNAmiR-21 in the absence of the TRAC circuit. This Non-TRAC system yielded a minimum miR-21 detection threshold of 1.0 nM (Fig. 5C). Upon integration of the TRAC circuit (Dre-CRISPR with TRAC), a substantial enhancement in fluorescence signal was observed compared to the Non-TRAC baseline. As shown in Fig. 5D, the fluorescence intensity increased progressively across the tested miR-21 concentration gradient. Notably, the Dre-CRISPR platform achieved exceptional sensitivity, as miR-21 concentrations as low as 0.5 fM remained readily identifiable. This represents a 2 × 106-fold improvement in detection capability relative to the 1.0 nM threshold observed without TRAC coupling (Fig. 5C and D). A robust linear relationship was established within the 1.0 to 100 fM range, yielding the regression equation ΔF = 10.35CmiR-21 + 960.5 with a correlation coefficient of R² = 0.9905, where ΔF represents the blank-subtracted fluorescence intensity (Supplementary Fig. S11A). For comparison, we established a control system utilizing WT crRNA coupled to the identical TRAC circuit. In this configuration, the lowest detectable concentration was 0.5 pM (Fig. 5E and Supplementary Fig. S11B). Compared with this WT-coupled system, the Dre-CRISPR platform exhibited a nearly 1000-fold enhancement in sensitivity. Such a dramatic improvement demonstrates that our engineered self-caged topology facilitates a far more efficient activation pathway for short miRNA targets than the native scaffold (Fig. 5D and E). In addition, two single-nucleotide-mismatched synthetic RNAs (miR-21-M1 and miR-21-M2) and the non-target miRNA were used to assess the selectivity of the Dre-CRISPR system for miR-21 detection. As shown in Supplementary Fig. S11C, only miR-21, whether alone or within a mixture, elicited a pronounced fluorescence response. In contrast, the single-base-mismatched variants and the interfering sequence produced negligible signals, indicating high sequence selectivity and the effective discrimination of single-nucleotide variations. Furthermore, spike-recovery experiments in heat-treated serum yielded recoveries ranging from 100.7% to 107.3%, supporting the robustness of the method in complex biological matrices (Supplementary Fig. S11D and Supplementary Table S2). Consequently, the Dre-CRISPR assay provides a streamlined, highly sensitive, and specific approach for the quantitative detection of miRNA. Enzyme-driven molecular switching and quantitative profiling of UDG activity To demonstrate the versatility of the Dre-CRISPR platform beyond direct nucleic acid detection, we configured it as a universal and modular sensing framework. The core design logic relies on rationally coupling upstream target-specific recognition events with the downstream CRISPR cascade. Specifically, the catalytic activity or specific binding of diverse non-nucleic acid analytes can be translated into the structural alteration of a masking module. This rationally designed alteration liberates a universal initiator strand, which then seamlessly feeds into our downstream DR-gated crRNA autocatalytic loop to produce an amplified fluorescent output. To demonstrate this universality, we first integrated a base-excision triggered switching mechanism to sense uracil–DNA glycosylase (UDG) activity. UDG, a pivotal enzyme in the base excision repair (BER) pathway, initiates DNA repair by recognizing and excising uracil bases [36–39]. We hypothesized that the site-specific generation of an apurinic/apyrimidinic (AP) site could serve as a molecular trigger to modulate the stability of the crRNA-blocking architecture. As illustrated in Fig. 6A, we designed a modified B3 blocker containing a deoxyuridine (dU) residue (B3–dU). Optimization of the stoichiometric ratio between the crRNA-d-5′ fragment and the B3–dU ensured maximum suppression of the background signal in the dormant state (Supplementary Fig. S12A). The presence of the dU base maintains the structural integrity of the caged crRNA-d-5′, effectively sequestering the Cas13a–crRNA assembly. Upon UDG-mediated excision, the resulting AP site introduces a localized thermodynamic instability that promotes blocker dissociation and subsequent RNP reassembly. This enzyme-driven topological transition initiates the TRAC-mediated autocatalytic cascade amplification, producing a robust fluorescence response across a concentration gradient from 2.0 × 10−5 to 6.0 × 10−4 U/ml (Fig. 6B). The system achieved a LOD of 6.7 × 10⁻6 U/ml (3σ criterion), with a high degree of linearity maintained in the low-activity regime (Fig. 6B, inset). To evaluate the reliability of this enzymatically triggered switch in biological contexts, we performed spike-recovery experiments in 10% FBS. The measured fluorescence intensities in the serum matrix (blue diamonds, Fig. 6B, inset) were highly congruent with the standard calibration curve, yielding recovery rates between 99.2% and 105.6% (Supplementary Table S3). These results confirm the robustness of the Dre-CRISPR platform against complex matrix interference. Modular expansion of the Dre-CRISPR platform for enzymatically triggered molecular switching. (A) Schematic illustration of the UDG-responsive activation mechanism. Site-specific excision of the dU residue by UDG generates an AP site, which destabilizes the caged crRNA-d-5′ architecture and initiates the TRAC-mediated autocatalytic signal propagation. (B) Concentration-dependent fluorescence response of the Dre-CRISPR platform to UDG. The system demonstrates high sensitivity across a broad range of enzymatic activities. The inset highlights the linear correlation in the low-concentration regime (2.0 × 10−5 to 1.0 × 10−4 U/ml), where blue diamonds represent the measured fluorescence intensities of UDG spiked into 10% FBS. The high congruence between the serum data and the standard calibration curve confirms the platform’s matrix robustness. (C) Fluorescence spectroscopic profiles illustrating precise regulatory control. The robust fluorescence signal initiated by UDG (red) is markedly suppressed upon the addition of the specific inhibitor UGI (blue). The system maintains a quiescent baseline in the presence of UGI alone (purple), confirming that the inhibitor itself does not induce non-specific activation. Data are presented as mean ± S.D., n = 3, three technical replicates. The specificity of the UDG-responsive logic was further validated against a panel of noncognate enzymes, where no significant activation was observed (Supplementary Fig. S12B). Furthermore, we explored the regulatory deactivation of the system using the uracil glycosylase inhibitor (UGI). As evidenced by the fluorescence spectroscopic profiles (Fig. 6C), the addition of UGI effectively blunted the enzymatic initiation, thereby arresting the signal propagation even in the presence of UDG. While a slight elevation in fluorescence was observed in the inhibited group (+ UDG + UGI) compared to the inhibitor-only control (purple line), it remained considerably lower than the robust signal triggered by UDG alone. This sharp contrast between the active and suppressed states demonstrates the high conformational fidelity of the gated DR-split crRNA architecture, confirming that the catalytic cascade is strictly gated by the enzymatic excision event. Collectively, these findings establish Dre-CRISPR as a generalizable platform where specific enzymatic activities can be seamlessly transduced into programmable molecular outputs via responsive topological reconfiguration. Modular expansion of the Dre-CRISPR platform for Pb²⁺ sensing To further demonstrate the broad applicability of the Dre-CRISPR framework, we integrated a Pb²⁺-specific DNAzyme module to serve as a heterologous molecular trigger [40–44]. As illustrated in Fig. 7A, the DNAzyme–substrate complex was pre-immobilized on streptavidin-coated magnetic beads, effectively sequestering the initiator sequence (H0) within a solid-phase scaffold. This spatial segregation ensures that the H0 initiator is only liberated into the liquid phase upon Pb²⁺-activated cleavage, thereby unlocking the downstream strand-displacement and CRISPR-mediated signaling cascade. Modular expansion of Dre-CRISPR for Pb²⁺ sensing. (A) Schematic of the Pb²⁺-responsive cascade involving DNAzyme-mediated initiator release and downstream TRAC amplification. (B) Fluorescence spectroscopic profiles showing the concentration-dependent response to Pb²⁺ concentrations (0–500 pM). (C) Fluorescence intensity changes (ΔF) as a function of Pb²⁺ concentration. (Inset) Linear calibration curve (1.0–100 pM) with spike-recovery data from tap water and 10% FBS. Data are presented as ΔF = F − Fblank, wher Fblank and F represent the fluorescence intensities in the absence and presence of Pb²⁺, respectively. Data are presented as mean ± S.D., n = 3, three technical replicates. The structural parameters of the DNAzyme–substrate complex were systematically optimized to balance system stability and cleavage efficiency. Specifically, the length of the 5′ pairing arm (n) was found to critically influence the S/B ratio, with 7 nt yielding the optimal performance (Supplementary Fig. S13A). With the optimal configuration established, the platform’s performance was characterized via fluorescence spectroscopic profiles (Fig. 7B). In the absence of Pb²⁺, the platform maintained a stable and reproducible baseline, a direct consequence of the stringent conformational gating inherent to the gated DR-split crRNA architecture. Upon incubation with Pb²⁺, a concentration-dependent increase in fluorescence intensity was observed, demonstrating a high S/B ratio and excellent responsiveness of the reconstituted RNP complex. Quantitative analysis revealed that the fluorescence response (ΔF) increased dose-dependently with Pb²⁺ concentration (Fig. 7C). A robust linear correlation was established in the range of 1.0–100 pM (R² = 0.9937, Fig. 7C, inset), with a calculated LOD of 890 fM (3σ criterion). To evaluate the practical utility of the platform, we performed spike-recovery experiments in tap water and 10% FBS. As shown in Fig. 7C (inset), the experimental data points for both matrices closely aligned with the standard calibration curve, yielding recoveries from 99.5% to 105.5% for tap water and from 93.2% to 103.1% for 10% FBS (Supplementary Table S4). Finally, the exquisite selectivity of this recognition module was confirmed through competitive assays against various interfering metal ions. The system exhibited negligible responses to nontarget analytes, maintaining robust specificity for Pb²⁺ even when tested in complex mixtures (Supplementary Fig. S13B). These findings collectively establish Dre-CRISPR as a generalizable and reliable platform for the high-contrast sensing of diverse biomarkers in complex biological and environmental samples. Discussion In this work, for the first time we demonstrate that the DR region of Cas13a crRNA is not merely a passive structural scaffold, but a programmable regulatory element that can tolerate substantial reconfiguration while retaining catalytic competence. Through systematic engineering of the lateral and apical loop regions, including sequence insertion and site-specific splitting, we found that multiple crRNA variants preserved robust cis- and trans-cleavage activities comparable to those of the wild-type (WT) guide. These findings challenge the conventional view that the Cas13a crRNA scaffold must remain structurally intact to sustain activity. Instead, our results reveal that Cas13a activity depends on the adoption of a functional conformation rather than the preservation of a rigid scaffold architecture, thereby redefining the DR as a conformationally and topologically programmable gate for Cas13a activation. A central mechanistic insight of this study is that topological reconfiguration of the DR provides a previously inaccessible regulatory handle for Cas13a. In WT systems, the exceptionally high binding affinity between Cas13a and its cognate crRNA, coupled with the rapid kinetics of RNP assembly render precise intervention inherently difficult. Under such conditions, simple nucleic acid blockers are typically ineffective. By contrast, our results demonstrate that the engineered scaffold variants, especially the split architectures, exhibit superior accessibility to regulation by unmodified nucleic acid blockers. Fluorescence and PAGE analyses support a model in which ssDNA blockers act primarily by intercepting productive crRNA assembly before formation of the active Cas13a complex, rather than by efficiently dismantling preassembled complexes, although limited post-assembly engagement may occur due to dynamic conformational flexibility. Thus, DR engineering does not merely preserve Cas13a activity; it transforms the guide RNA into a topology-controlled molecular switch, enabling a level of signal suppression and target-activated gain that is unattainable with the rigid WT scaffold. Building on this mechanism, we established Dre-CRISPR, a DR-engineered CRISPR–Cas13a amplification platform in which target-responsive restoration of a topologically gated DR-crRNA is coupled to a self-reinforcing amplification circuit. Unlike existing strategies that often rely on auxiliary enzymes or complex multi-component integration [45–49], the present system exploits the intrinsic catalytic plasticity of Cas13a to drive a structural cascade amplification, significantly reducing reaction complexity. This integration allows the guide RNA itself to function as a modular signal-conversion interface capable of coupling diverse upstream triggers, such as nucleic acids, enzyme activities, and metal ions to a unified CRISPR readout. To systematically evaluate the analytical performance of Dre-CRISPR, we compared our platform with existing state-of-the-art CRISPR-based biosensors (Supplementary Table S5). Without requiring target pre-amplification, Dre-CRISPR achieves highly competitive sensitivity and rapid reaction kinetics that are on par with current representative direct-detection methods. Crucially, by minimizing exogenous components, this architecture ensures a high-contrast response with minimal background leakage, a direct benefit of the tightly regulated OFF state inherent to our topologically gated DR-split crRNA design. Beyond the specific targets demonstrated here, the Dre-CRISPR framework addresses critical challenges in molecular diagnostics by establishing DR topology as a primary design parameter. The topological programmability of the scaffold enables the selection of architectures that are thermodynamically compatible with diverse analytes, thereby integrating target recognition, signal transduction, and amplification within a singular modular framework. While future integration with microfluidic platforms could enable multiplexed parallel detection, further advancement of this system relies on elucidating the interplay between local structural context and hybridization stability. High-resolution biophysical characterization will be essential to define the precise structural determinants that render engineered scaffolds regulatable while the wild-type remains refractory. Such insights will facilitate the expansion of this programmable gating strategy toward an even broader range of molecular targets and more sophisticated diagnostic applications. In conclusion, our findings redefine the crRNA scaffold as a programmable signaling node and establish a generalizable framework for constructing ultra-sensitive, self-amplifying biosensors characterized by minimal leakage and modular scalability. By leveraging the intrinsic plasticity of the DR region, this work broadens the design space for CRISPR-based technologies and offers a path toward simpler, more integrated analytical platforms for diverse bioanalytical applications. Supplementary Material Acknowledgements None. Author contributions: W.D. and X.L.L. designed the research. W.D., X.M.Y., Y.T.Y., and M.X.L. performed the research. W.D., C.H.L., and X.L.L. analyzed data. X.L.L. wrote the manuscript. C.H.L., J.Y.W., and X.L.L. revised the manuscript. All authors reviewed and approved the manuscript. Contributor Information Wei Ding, College of Chemistry and Pharmacy, Northwest A&F University, Yangling, Shaanxi 712100, P.R. China. Xiaomin Yang, College of Chemistry and Pharmacy, Northwest A&F University, Yangling, Shaanxi 712100, P.R. China. Yutong Yang, College of Chemistry and Pharmacy, Northwest A&F University, Yangling, Shaanxi 712100, P.R. China. Mingxin Liu, College of Chemistry and Pharmacy, Northwest A&F University, Yangling, Shaanxi 712100, P.R. China. Chenghui Liu, Key Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education; Key Laboratory of Analytical Chemistry for Life Science of Shaanxi Province; School of Chemistry & Chemical Engineering, Shaanxi Normal University, Xi’an, Shaanxi 710119, P.R. China. Jinyi Wang, College of Chemistry and Pharmacy, Northwest A&F University, Yangling, Shaanxi 712100, P.R. China. Xiaoling Liu, College of Chemistry and Pharmacy, Northwest A&F University, Yangling, Shaanxi 712100, P.R. China. Supplementary data Supplementary data is available at NAR online. Conflict of interest None declared. Funding This work was supported by the National Natural Science Foundation of China [22474111 and 22104077]; the Natural Science Basic Research Program (2026JC-YXQN-058) and Innovation Capability Support Program of [2025RS-CXTD-058] of Shaanxi Province; the Xi’an Science and Technology Project [25ZQRC00009] and the Northwest A&F University. Funding to pay the Open Access publication charges for this article was provided by the National Natural Science Foundation of China and the Northwest A & F University. Data availability All data supporting the findings of this study are available within the article and its online supplementary information. References Associated Data Supplementary Materials Data Availability Statement All data supporting the findings of this study are available within the article and its online supplementary information.