English

• 1. Introduction

  The term "circuit" traditionally brings to mind silicon computers powered by electricity. However, the emergence of DNA has revolutionized this concept with its programmatic, good specificity, and biocompatibility [1-3]. DNA-based circuits have demonstrated remarkable advancements in environmental monitoring [4, 5], drug delivery [6, 7], and medical diagnosis [8-10].

  DNA circuits are mainly engineered by toehold-mediated strand displacement reaction (TSDR) for reasonable programming and operation [11, 12]. TSDR begins with a toehold, initiating competitive hybridization between complementary strands to form a more stable dsDNA structure [13]. In this process, a long single-stranded DNA (ssDNA) hybridizes with a short ssDNA, and then another ssDNA complementary to the long ssDNA replaces the short ssDNA by binding to the toehold. This thermodynamically driven reaction occurs quickly and is less affected by external conditions. These universal schemes enable the construction of DNA circuits capable of performing various functions [14].

  On one hand, they provide amplicons to amplify signals termed amplification functions, such as hybridization chain reaction (HCR) [15, 16], catalytic hairpin self-assembly (CHA) [17-19], entropy-driven catalysis (EDC) [20, 21], and cascade circuits [22]. These efficient and direct amplification functions have been proven to amplify the target approximately 100 times [23-25]. To achieve superior amplification effects, the integration of multiple layers of amplification circuits is necessary, albeit this invariably results in the complexity of DNA strand design. Moreover, mitigating leakage induced by the orthogonality between probes remains a pressing challenge to address.

  On the other hand, programmable and addressable DNA circuits can be used as logic gates and thresholds for decision-making functions in molecular computation, which means encoding the algorithm into DNA molecules and completing the calculation process through a series of biochemical reactions [8, 26-29]. In addition, the interplay between nucleic acid circuits and proteins, ions, and small molecules further enables the conversion of input molecules, thus broadening the scope of applications. While previous studies have demonstrated the scalability and accuracy of DNA circuits in logic gate construction and molecular computation, they still suffer from efficiency and DNA orthogonality leakage issues. Various studies have proposed the integration of nanomaterials, enzyme-driven, multi-tube operation, chemical modification, and other strategies into circuits to enhance efficiency and mitigate the risk of leakage [30-34]. However, achieving a balance between short-term consumption and low cost remains elusive.

  In recent years, the clustered regularly interspaced short palindromic repeat (CRISPR) system has emerged as a highly effective tool in the field of molecular biology [35, 36]. CRISPR and CRISPR-associated (Cas) nucleases form a vital component of the adaptive immune system found in bacteria and archaea. Cas nucleases, in conjunction with CRISPR RNA (crRNA), constitute the core of the CRISPR-Cas system. Upon being induced by crRNA, Cas nucleases are capable of specifically identifying and cleaving nucleic acids. The unique advantages of CRISPR-Cas, including single-base specificity, rapid turnover time, and trans-cleavage activity have enabled its expansion from genetic engineering to applications of amplification or decision-making function [37-43]. Notably, given the sequence independence between activators of CRISPR-Cas and reporter, instead of crosstalk in the inevitable sequence orthogonality between upstream probes and reporter in normal DNA circuits, the signal-to-noise ratio of CRISPR-Cas system's reporter cleavage pattern surpasses the traditional TSDR system's reporter displacement pattern. Additionally, CRISPR-Cas can respond to blunt-ended dsDNA to expand the application of clinical nucleic acid input, which is typically unattainable in conventional DNA circuits [44]. Nonetheless, systems that rely exclusively on CRISPR-Cas encounter hurdles in inadequate sensitivity and limited types of inputs, which hinders its widespread application in the field of molecular biology [45-49].

  To address the individual limitations above, scientists have integrated the CRISPR-Cas system with DNA circuits to achieve amplification [50] or decision-making functions [51]. In the past, there have been numerous reviews about amplification or amplification-free applications in CRISPR-Cas systems. Specifically, the applications of CRISPR-Cas in signal amplification strategy have been discussed, focusing on a certain type of biomarker, such as food safety and pathogen, and been compared in terms of amplification efficiency [52, 53]. However, the applications for logic gates and molecular computation of CRISPR-Cas have not been mentioned. In addition, the role of CRISPR in amplification circuits and decision-making circuits requires innovative elucidation to delineate the functional positioning of different Cas systems in various circuits, which has been lacking in previous reviews. Consequently, novel perspectives still need to be proposed to advance our understanding in the field. In this review, we attempt to summarize the advanced applications of the CRISPR-Cas system in DNA circuits, encompassing amplification and decision-making functional circuits. This review commences with different CRISPR-Cas classifications, emphasizes their significance for DNA circuits, and proposes their potential in the future. Furthermore, this review also discusses the existing limitations and challenges associated with current applications, aiming to inspire the development of more streamlined and effective DNA circuits based on CRISPR-Cas, and provide a preliminary direction for advancing the next generation of DNA circuits.

  2. Types of CRISPR-Cas

  The CRISPR-Cas system encompasses two main categories, namely class 1 and class 2, which differ in structure and sequence. Class 1 systems are characterized by a multi-protein complex that functions in a coordinated cascade [54], whereas class 2 systems rely on a single protein with multiple functional domains. Class 2 systems, known for their versatility and broader applications, have garnered more extensive utilization in various fields. Notably, Cas9, Cas12a, and Cas13a are among the most commonly employed variants, each capable of performing distinct tasks based on their unique mechanisms (Fig. 1) [55, 56].

  Figure 1

  

  Figure 1.  Schematic of class 2 CRISPR-Cas system. Cleavage mechanisms of (A) Cas9, (B) Cas12a, and (C) Cas13a. Reproduced with permission [37]. Copyright 2023, The Royal Society of Chemistry.

  DownLoad: Full-Size Img PowerPoint

  Cas9, widely utilized for genome editing in humans, plants, and eukaryotic cells, functions by forming an R-loop through the specific complementary pairing of crRNA or single guide RNA (sgRNA) with a target DNA strand. This process is initiated with the strand invasion of Cas9 with the assistance of protospacer adjacent motifs (PAMs) (5′-NGG-3′) [57, 58]. Subsequently, the variable spacer region of sgRNA selectively binds to the target DNA, resulting in the cleavage of dsDNA approximately 3–4 nucleotides upstream of the PAM sequence [59].

  In contrast to Cas9, Cas12a and Cas13a exhibit both cis-cleavage and trans-cleavage activities when guided by crRNA. Cas12a can be activated by dsDNA containing a PAM sequence (5′-TTTN-3′) or ssDNA without the assistance of a PAM sequence. It is worth mentioning that some reports have demonstrated that RNA can also serve as a direct activator for Cas12a [60, 61]. Alternatively, the combined action of RNA and DNA can be achieved through split activators, deviating from conventional methods [62]. Once activated, Cas12a cleaves the activator, referred to as cis-cleavage [63]. Intriguingly, Cas12a exhibits robust random cleavage activity once activated, fragmenting arbitrary ssDNA in the system into 2–4 nucleotide fragments, a process known as trans-cleavage [64]. While primarily known for its trans-cleavage activity on ssDNA, there are emerging reports that suggest dsDNA can also serve as trans-cleavage substrates for Cas12a [65]. In recent years, researchers have focused on overcoming the limitations of PAM sequences and proposed PAM-free Cas12a strategies by converting dsDNA lacking PAM sequences into ssDNA [66, 67] or embedding bubble structures in dsDNA to mimic the unwinding of PAM sequences for activation [68].

  Cas13a can only be activated by single-stranded RNA (ssRNA) with the guidance of crRNA. It is important to note that the process relies on the H (non-G) protospacer flanking site (PFS). This mechanism differs from the activation of Cas12a by ssDNA, which has no sequence requirements. Once activated, Cas13a exhibits cis-cleavage activity, enabling it to cleave activator ssRNA. For targeted ssRNAs, different Cas13a homologs exhibit different cleavage preferences, in which most prefer uridine substrates and few prefer adenosine substrates. Simultaneously, its trans-cleavage activity non-specifically cleaves other collateral ssRNAs in the system [69-71].

  3. Application of different CRISPR-Cas systems in different function DNA circuits

  In recent years, several amplification and decision-making DNA circuits based on the unique characteristics of CRISPR-Cas systems have been proposed [72-74]. However, many of these methods require the assistance of other enzymes. In contrast, DNA circuits that harness CRISPR-Cas systems through strand displacement reactions have emerged as a popular alternative [75]. These circuits are notable for their dual utilization of CRISPR-Cas systems; not only do they employ these systems for the ultimate transduction of signals, but they also integrate them as integral components of the circuit that are responsive to input signals and capable of producing output signals. Therefore, they can fulfill the requirements of both amplification and decision-making functions. The requirement for detection equipment is significantly increased by the programmed heating and cooling process of PCR, which restricts promotion and field testing at the grass-roots level. Although temperature amplification technologies such as RPA/LAMP/RCA have solved the dependence on thermal cycling, they still require the assistance of enzymes, making it difficult for them to overcome the high costs to produce engineered enzymes and the challenges of storage and transportation of detection reagents [76]. In contrast, enzyme-free DNA circuits including HCR, CHA, and EDC present a more desirable signal amplification method, offering great potential for the development of low-cost and robust point-of-care testing (POCT), due to the entropy or thermodynamics-driven process without the assistance of other enzymes. Therefore, this review will specifically focus on the advantages and disadvantages of different CRISPR-Cas systems in various functional circuits without other enzymes.

  3.1 Circuit based on CRISPR-Cas9

  In addition to its common application in genome editing, Cas9 can be employed in circuits similar to the function of an endonuclease. Cas9 can bind to the dsDNA and cleave both the target ssDNA strand (through the HNH domain) and the non-target ssDNA strand (through the RuvC domain) upon the binding of sgRNA to the target. The specific recognition of the target by sgRNA can be precise up to a single base mismatch within the PAM-proximal region [77, 78]. As mentioned earlier, CRISPR-Cas9 lacks trans-cleavage activity, which poses challenges in directly utilizing it for the amplification step in DNA circuits. This property without trans-activity also makes CRISPR-Cas9 more suitable for upstream circuits as a partial input and potential for decision-making circuits. The enzymatic deactivated Cas9 (dCas9) has also been employed in functional circuits due to its sequence-specific binding capability.

  3.1.1   Cas9 in amplification circuits

  In amplification circuits, Cas9's cleavage is frequently employed at upstream of the amplification circuit to initiate amplification or at downstream of the amplification circuit to transduce amplification products. Specifically, Xu's group devised an EDC circuit activated by CRISPR-Cas9 cleavage. By employing sgRNA, Cas9 recognized and cleaved the target DNA, generating precise primers that are essential for the subsequent operation of the EDC circuit and completing a cyclic amplification circuit. For signal transduction, the EDC circuit product was captured by a 3D GR/AuPtPd nanoflower, resulting in an increased negative charge and reduced peak current. Consequently, the detection sensitivity for circular tumor DNA targets was significantly improved (Fig. 2A) [79]. Similarly, there are also applications in Cas9's signal reporting for amplification circuits, where the amplification products serve as cleavage substrates for Cas9. Huang leveraged the properties of CRISPR-Cas9 to engineer a multi-channel series piezoelectric quartz crystal (MSPQC) sensor. Within the innovative sensor design, the capture probe, located on the Au interdigital electrodes, served a dual role as the target recognition agent and initiator of HCR. The long dsDNA product of HCR bridged the Au interdigital electrodes, and the hybridization between the capture probe and target was recognized and cleaved by CRISPR-Cas9, disrupting the long DNA connection that was established by HCR between the electrodes. Ultimately, silver staining technology distinguished the MSPQC sensor's sensitivity to tuberculosis caused by Mycobacterium tuberculosis, even exhibiting good specificity for single base mismatches (Fig. 2B) [80]. Interestingly, amplification circuits based on CRISPR-Cas9 often incorporate nanomaterials [81]. It is attributed to the limited amplification efficiency of Cas9, whose turnover rate is relatively low. Another signal reporting pattern relies on the binding force of dCas9, where the long amplification products possess multiple dCas9/sgRNA binding sites, two split compounds (i.e., split luciferin and split HRP) were engineered onto the dCas9 protein, and dCas9's specific binding to upstream amplification products brought the split compounds closer together for reconstruction to generate the signal [82].

  Figure 2

  

  Figure 2.  Applications of Cas9-based circuit. (A) CRISPR-Cas9 cleavage triggered the EDC amplification circuit. Reproduced with permission [79]. Copyright 2020, Elsevier B.V. (B) An MSPQC nucleic acid amplified sensor by the combination of CRISPR-Cas9 and HCR. Reproduced with permission [80]. Copyright 2022, American Chemical Society. (C) CRISPR-Cas9 mediated strand displacement logic circuits with toehold-free DNA. Reproduced with permission [44]. Copyright 2021, American Chemical Society. (D) The central processing unit for program complex logic computation in cells based CRISPR-Cas9. Reproduced with permission [89]. Copyright 2019, according to Creative Comments Attribution License 4.0 (CC BY).

  DownLoad: Full-Size Img PowerPoint

  3.1.2   Cas9 in decision-making circuits

  For Cas9-based decision-making circuits, the sgRNA or target dsDNA was designed as input to control the circuit. The sgRNA guides the Cas9 to cleave the specific DNA, thus serving as a signal transduction point and the cleavage product of Cas9 serves as output for subsequent response and decision-making purposes. The unique properties of Cas9 in such circuits allow it to directly respond to dsDNA input without requiring toehold regions or redundant processing, thereby expanding the application range, simplifying the design, enhancing the efficiency, and enriching the functionality of decision-making circuits. For instance, Montagud-Martínez et al. introduced a CRISPR/Cas9-mediated strand displacement technique. The approach involves a series of logic gates that utilize the dsDNA generated within the circuit as integral components instead of generating dsDNA waste as in conventional circuits [44, 83]. In the case of an AND gate, the complex of sgRNA and Cas9 can respond to dsDNA input. Subsequently, the activated Cas9 cleaves the dsDNA, generating output products. Another ssDNA input triggers strand displacement to replace the pre-hybridized ssDNA, exposing the toehold region in the probe for the output dsDNA to initiate another strand displacement reaction, restoring the fluorescence signal and completing the decision-making function of the AND gate (Fig. 2C).

  By introducing mutations into the nuclease protein domain of Cas9, dCas9 can be obtained [84-88]. This engineered version of Cas9 retains its DNA-binding capabilities but loses its DNA-cleaving activity, which makes it a valuable tool for gene expression control without DNA damage. Kim et al. developed a processor that programs transcription regulators through different sgRNA inputs and measures fluorescent protein levels as output through transfection, enabling various logic gate operations, including half adder operations (Fig. 2D) [89]. In detail, in the OFF system, the reporter gene contains an input gRNA (igRNA) binding site between the hCMV promoter and the transcription initiation site, and dCas9-KRAB hybridizes the reporter gene in the presence of igRNA, inhibiting the transcription of the reporter gene. In contrast, the ON system incorporates an additional regulatory gRNA (rgRNA), which contains a binding site for igRNA between the hU6 promoter and the transcription initiation site. Conversely, the reporter gene contains a binding site for rgRNA. The presence of igRNA inhibits the transcription of rgRNA, resulting in the activation of the reporter gene. Based on foundational principles, logic gates for NOR, NIMPLY, AND, and XOR have been conceptualized and implemented in biological systems. The half adder was designed with a binary operation circuit with two inputs and two outputs (sum and carry) [90, 91]. The sum output is the result of an XOR operation on inputs 1 and 2, while the carry output is the result of an AND operation on the same inputs 1 and 2. In this work, the XOR operation involves setting the binding sites of two igRNAs between the promoter and transcription initiation site of the reporter gene, allowing inhibition of the transcription of the reporter gene in the presence of igRNA-1 or −2. The absence of both igRNA-1 and −2 enables the activation of the XOR gate. Additionally, the AND operation involves setting the binding sites of two rgRNAs between the promoter and transcription initiation site of the reporter gene, which are individually suppressed by two igRNAs. The combination of the AND gate and XOR gate enables half adder operations in cells.

  Overall, Cas9 can directly process blunt dsDNA for signal transduction, but the challenges remain on sensitivity for signal reporting. Although dCas9 also has advantages in gene expression control, providing a foundation for cell-free biosensors, regarding sgRNA input, the circuit regulation necessitates considering the cost, stability, and potential misfolding issues of RNA that arise from the introduction of multiple sgRNAs.

  3.2 Circuit based on CRISPR-Cas12a

  Cas12a, also known as Cpf1, is a widely used tool in the CRISPR system. Upon complexation with crRNA, Cas12a recognizes the PAM sequence of the target dsDNA and induces the DNA unwinding process. The crRNA hybridizes with TS to form an RNA/DNA heteroduplex, leading to conformational changes in the Cas12a protein. This exposes the catalytic site of the RuvC domain, which can cleave NTS and TS with cis-cleavage activity. Cas12a also exhibits trans-cleavage activity, which allows it to cleave random ssDNA with the activated RuvC domain [92]. Due to the non-specific trans-cleavage activity, Cas12a is predominantly employed for signal reporting rather than signal transduction in circuits, which makes the circuits susceptible to interference but brings excellent signal reporting ability [93].

  3.2.1   Cas12a in amplification circuits

  As Cas12a can be activated by both dsDNA containing PAM sequence and ssDNA lacking PAM sequences, there are two primary categories of Cas12a-based amplification circuits. The first category involves generating ssDNA complementary to the crRNA through amplification to activate Cas12a. This amplification process typically requires the assistance of enzymes, particularly polymerases, to produce a long ssDNA with repetitive activators [94-98].

  The second category involves generating dsDNA with PAM sequences through enzyme-free reactions to activate Cas12a [99]. The assembled dsDNA with PAM sequences in the circuit is universal and less susceptible to interference from erroneous ssDNA amplicons. For example, Lv et al. combined the CRISPR-Cas12a system with an aptamer and a cascaded circuit using Fe₃O₄@hollow-TiO₂@MoS₂ nanochains. The aptamer recognizes tetracycline (TC), triggering the subsequent EDC, which then triggers the downstream HCR circuit. This cascade effect leads to the production of numerous dsDNA activators, subsequently inducing the trans-cleavage for ssDNA reporters on nanomaterials. The synthetic nanomaterials exhibit multiple functionalities, serving as quenchers for reporters while also demonstrating exceptional capabilities in magnetic separation, photocatalytic degradation, TC removal, self-cleaning, and reusability (Fig. 3A) [100]. Jia et al. successfully integrated an HCR circuit to convert microRNA into multiple dsDNA molecules that contain the necessary PAM sequences to activate Cas12a and generate a detectable signal. The amplification effect of HCR and Cas12a enables target detection at a concentration as low as 1 pmol/L (Fig. 3B) [101]. In Wang's approach, the terminal overhangs of the CHA-derived dsDNA product, triggered by the target, effectively recognize and remove the probe immobilized on the surface of the Bi/g-C₃N₄ photoelectrochemical biosensing platform. This migration process facilitates the formation of dsDNA with PAM sequences, which in turn activates Cas12a, enhancing subsequent trans-cleavage activity. Additionally, the migration process releases the cleavage substrate that incorporates the sensitizer MB, from the dsDNA on the electrode surface. The liberated substrate then becomes available for Cas12a to cleave, resulting in changes in the photoelectrochemical (PEC) signals (Fig. 3C) [102].

  Figure 3

  

  Figure 3.  Applications of Cas12a-based circuits. (A) Cascade circuits. Reproduced with permission [100]. Copyright 2023, Wiley-VCH GmbH. (B) HCR. Reproduced with permission [101]. Copyright 2022, Elsevier B.V. (C) CHA. Reproduced with permission [102]. Copyright 2022, Elsevier B.V. Decision-making circuits-based CRISPR-Cas12a for (D) Dual miRNAs AND logic gate. Reproduced with permission [104]. Copyright 2022, American Chemical Society. (E) Molecular logic gate of heavy metal ions. Reproduced with permission [105]. Copyright 2022, Elsevier B.V. (F) Guanine nanowire boolean logic tree. Reproduced with permission [106]. Copyright 2022, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  3.2.2   Cas12a in decision-making circuits

  Similar to amplification circuits, the decision-making circuit is positioned upstream to generate a specific ssDNA or dsDNA. This output can activate Cas12a activity, leading to the cleavage of the reporter and completing signal reporting [103]. For example, Gong et al. developed an AND gate on magnetic beads capable of recognizing dual inputs of microRNAs. Employing the concept of a seesaw circuit, the introduction of dual microRNAs triggers the trans-cleavage activity of Cas12a, resulting in the release of glucose oxidase (GOx) from the magnetic bead surface. Following magnetic separation, GOx catalyzes substrate discoloration in the solution, thereby achieving a colorimetric AND logic gate circuit (Fig. 3D) [104]. In addition, Pan et al. constructed a cascade logic gate utilizing metal ion-dependent DNAzymes, and the metal ions included Hg²⁺, Mn²⁺, Cd²⁺, and Pb²⁺. Substrate cleavage by DNAzymes led to different cascade designs on the output substrate, designed to respond to particular metal ion stimuli. Hg²⁺ and Mn²⁺-DNAzyme cleave substrates to release triggers that initiate the construction of an AND gate. This process promotes the formation of a dsDNA complex that can act as an initiator for CHA via proximity hybridization, generating dsDNA with PAM sequences that activate Cas12a. Parallelly, the cleavage product of Pb²⁺-DNAzyme can also directly function as the initiator of CHA, achieving an equivalent outcome to that of AND gate of Hg²⁺ and Mn²⁺-DNAzyme. In contrast, the cleavage product of Cd²⁺-DNAzyme acts as a complementary sequence to the initiator of CHA mentioned earlier and serves as an INHIBIT gate, inhibiting the occurrence of CHA (Fig. 3E) [105].

  Furthermore, Gao et al. utilized Resonant Rayleigh Scattering (RRS) to establish a molecular-level logic computing circuit using a guanine nanowire-assisted non-cross-linking hybridization chain reaction (GWancHCR)-based CRISPR-Cas12 system. This system employed combinations of specific complexes or molecules as inputs and generated products (matter level) or changes in RRS (energy level) as outputs. The logic output was determined by setting an RRS threshold. It relied on the interaction between lipopolysaccharide (LPS) and probes containing both an aptamer and a Cas12a activation sequence. The inclusion of LPS locked the activation sequence, rendering the CRISPR-Cas system inactive. In the signal output phase, the trigger reacted with a G-rich hairpin through HCR reactions, generating multiple G-quadruplexes. The abundance of G-quadruplexes significantly enhanced the spectral intensity of RRS. Conversely, the components of HCR would be digested by the activated CRISPR-Cas system, preventing the generation of G-quadruplexes and resulting in low peaks in the RRS spectrum. Multiple factors, including LPS, aptamer, Cas12a, sgRNA of the CRISPR-Cas system, as well as the trigger and G-rich hairpins in the signal output section, and K⁺ and Mg²⁺ related to the formation of G-quadruplexes, regulated the entire process (Fig. 3F) [106]. Based on these molecular interactions, a Boolean logic tree was constructed, enabling basic logic calculations and complex integrated logic circuits. In this context, Cas12a functioned similarly to an INHIBIT gate, where the logical output remains at '0' as long as Cas12a is active within the system, irrespective of other input variables.

  Based on the preceding discussions, it is apparent that the CRISPR-Cas12a system serves primarily as a terminal signal reporting tool and lacks the capability to function effectively as an intermediate component. While the trans-cleavage activity offers advantages, it also introduces interference and disrupts the preceding circuit. Consequently, a stepwise operational approach becomes imperative. Activation of downstream CRISPR-Cas systems occurs only upon completion of the upstream reaction, thereby resulting in experimental complexity or inherent limitations.

  3.3 Circuit based on CRISPR-Cas13a

  In contrast to the previous two DNA-targeting enzymes, CRISPR-Cas13a specifically targets and cleaves ssRNA. Upon hybridization between the target RNA and crRNA, dsRNA is formed, bringing the HEPN domain of Cas13a into proximity. Consequently, conformational changes occur in the protein, resulting in cleavage activity towards the target and collateral ssRNA. Despite its specificity and utility, the broader application of Cas13a is relatively limited compared to Cas9 and Cas12a, possibly due to the higher cost and potential stability issues associated with RNA, as mentioned previously. However, its primary advantage lies in directly recognizing RNA, complementing the other Cas enzymes that exclusively target DNA [107, 108], which makes Cas13a particularly useful in applications where RNA manipulation or regulation is desired.

  3.3.1   Cas13a in amplification circuits

  Typically, the necessity of RNA targets for Cas13a has spurred the development of a series of T7 RNA polymerase-based amplification circuits, which can transcribe RNA targets that can activate Cas13a for detection purposes [109-111]. In the circuits that this review focuses on, Cas13a is predominantly utilized as a signal transduction tool for downstream circuits, retaining the ability to directly recognize RNA, without the need for RNA reporters and other enzymes, making the system more accessible and economically viable. Concretely, the target RNA activates Cas13a/crRNA to cleave the uridine ribonucleotide in the DNA hairpin probe, releasing the stem of the DNA hairpin probe to initiate downstream amplification circuits. For instance, Zhang integrated a branched HCR amplification circuit onto SERS-active silver nanorod (AgNR) array sensing chips. Upon target RNA activation of Cas13a, it cleaves the uridine domain within the hairpin loop, releasing the trigger and initiating the HCR circuit. The circuit products, containing numerous ROX fluorophores, are captured by the SERS sensing chip, significantly enhancing the Raman signal (Fig. 4A) [112]. Similarly, Yang et al. employed a Cas13a-HCR circuit strategy to detect the RNA target, with a key difference in the signal output mechanism. In their approach, the signal output is facilitated on an optical fiber evanescent wave fluorescence biosensor platform. The RNA target leads to the generation of HCR products with biotin and Cy5.5 fluorophore, which are then captured by streptavidin-functionalized optical fiber, enabling the optical detection of the target (Fig. 4B) [113].

  Figure 4

  

  Figure 4.  Applications of Cas13a-based circuits. (A) Amplification circuit integrating CRISPR-Cas13a and HCR for SERS detection. Reproduced with permission [112]. Copyright 2022, Elsevier B.V. (B) Amplification circuit integrating CRISPR-Cas13a and HCR for portable evanescent wave sensor. Reproduced with permission [113]. Copyright 2021, Elsevier B.V. (C) Orthogonal inducible control of Cas13a for programmable AND circuit. (D) An ultralocalized Cas13a assay for digital computing. Reproduced with permission [115]. Copyright 2020, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  3.3.2   Cas13a in decision-making circuits

  The current discussion about decision-making circuits utilizing Cas13a remains inadequate. To address this gap, Ding developed a platform known as CRISTAL (Control of RNA with Inducible SpliT Cas13 Organisms and Exogenic Ligands) based on the RNA regulatory circuit of Cas13a. This platform enables dynamic regulation of transcription engineering. The strategy involves identifying cleavage sites in the Cas13a protein and inducing the re-aggregation of these sites to restore Cas13 protein activity. Cleavage site screening involves the utilization of gibberellic acid (GA)-inducible dimerization domains (GID and GAI). RNA transcripts within the selected cleavage sites can only be cleaved when both the endogenous signal and exogenous small molecule are present simultaneously, thereby implementing an AND gate logic circuit (Fig. 4C) [114]. In another investigation, Tian et al. employed the droplet microfluidic method to enrich the concentration of target RNA by reducing the reaction volume and achieved single-molecule level RNA detection. Combined with a fluorescence microscope, fluorescence signals were accumulated in the droplet volume, illuminating droplets containing the target RNA, while droplets without target RNA remained unilluminated. This approach enabled numerical counting, where '1' represented positive droplets, and '0' represented negative droplets (Fig. 4D) [115]. In the CRISPR-Cas13a system, the interference issue appears to be less significant as Cas13a could not damage DNA-based circuits like Cas12a [116].

  The main advantage is that it directly acts on RNA, which is conducive to the detection of relevant RNA biomarkers without the requirement for conversion. This also solves the interference problem of trans-cleavage activity in Cas12a-based DNA circuits, but its characteristic of only affecting RNA brings about more rigorous issues for cost and stability from RNA substrates.

  3.4 Circuit based on cascaded CRISPR

  Apart from the previously discussed circuits constructed based on DNA or RNA for amplification or decision-making functions, the CRISPR-Cas system can also be cascaded to construct circuits primarily for amplification. This can be achieved by cascading multiple crRNAs with a single Cas enzyme or by cascading multiple Cas enzymes. Cascade configurations within these circuits enable them to operate effectively without the need for supplementary procedures, such as target pre-amplification. This not only streamlines the design of the circuit but also circumvents the potential for cross-contamination that can arise from the introduction of an excess of nucleic acid components. In essence, the cascading architecture facilitates a more streamlined and efficient amplification process, enhancing the overall performance of the circuit. For instance, Fozouni et al. utilized the CRISPR-Cas13a system to directly detect SARS-CoV-2 RNA. Two crRNAs were designed and screened along the sequence of SARS-CoV-2 RNA, and the comparative analysis of Cas13a activity across multiple Cas13a/crRNA complexes versus a single Cas13a/crRNA complex was conducted (Fig. 5A) [117]. The results indicated that the intervention of multiple crRNAs effectively enhanced sensitivity by approximately 100 times. Additionally, the use of multiple crRNAs can also prevent potential losses caused by natural mutations of the virus when using a single crRNA.

  Figure 5

  

  Figure 5.  Circuits based on cascade CRISPR. (A) Cascade of Cas13a/crRNA complexes for amplification-free detection of SARS-CoV-2. Reproduced with permission [117]. Copyright 2020, Elsevier Inc. (B) Multiplex crRNA in CRISPR-Cas12a system for DNA diagnostic. Reproduced with permission [118]. Copyright 2022, American Chemical Society. (C) Cascade CRISPR-Dx system for detection of SARS-CoV-2. Reproduced with permission [119]. Copyright 2023, Elsevier B.V.

  DownLoad: Full-Size Img PowerPoint

  A similar approach was employed in the CRISPR-Cas12a system to detect the B646L gene directly from the African swine virus (ASFV). By introducing multiple crRNAs, each Cas12a/crRNA complex was targeted to distinct sites on the same target dsDNA, allowing a single dsDNA substrate to activate multiple Cas12a/crRNA complexes (Fig. 5B) [118].

  In another type of CRISPR-cascading circuit, Zhang et al. aimed to improve the sensitivity of the CRISPR-Cas system without the need for pre-amplification procedures to meet the requirements of clinical detection. Similar to the previous approach, multiple Cas13a/crRNA complexes activated simultaneously by SARS-CoV-2 led to better trans-cleavage activity, enabling the trans-cleavage for ssRNA, corresponding RNA bubbles on the RNA/DNA heteroduplex. Subsequently, the Cas12a/crRNA complex recognized the released DNA from RNA/DNA heteroduplex, activating the trans-cleavage activity of Cas12a to cleave the DNA reporter (Fig. 5C) [119]. By implementing a cascade involving two distinct types of high-turnover Cas enzymes, the sensitivity can reach the amol/L level, surpassing the sensitivity limit compared to other strategies without pre-amplification and avoiding the potential risk of leakage associated with pre-amplification strategies.

  Multiple crRNA/Cas complexes significantly enhance sensitivity, even without the need for other amplification methods or integration of nanomaterials. Similarly, this undoubtedly requires multiple crRNAs or multiple Cas enzymes, which is not cost-effective.

  3.5 Circuit based on other CRISPR-Cas

  While Cas9, Cas12a, and Cas13a have been extensively studied, the CRISPR revolution is ongoing with the development of other CRISPR systems, such as Cas3 and Cas14. Cas3 is an ATP-dependent nuclease and belongs to the class 1 enzymes that require multiple protein cascades [120-122]. Previous studies have elucidated that its activation effect relies on the collaborative function of crRNA and five Cas enzymes (Cas5, Cas6, Cas7, Cas8, Cas11) [123]. Once activated, Cas3 exhibits both nuclease and helicase activities [124]. The CRISPR-Cas9/12a/13a system utilizes RNA as a guide to recognize and pair with target sequences. The guiding RNA directs the Cas protein to the precise target site, cleaving the target DNA at the designated position. Similarly, CRISPR-Cas3 employs this mechanism to locate specific DNA sequences. However, instead of cleaving DNA into two fragments, Cas3's nuclease activity continuously degrades DNA, allowing for erasing up to 100 kb. Specifically, it creates a cleavage at the position 11 bases away from the PAM in the spacer, degrades ssDNA from 3' to 5' sequentially, and exhibits trans-cleavage activity for random ssDNA [125]. Compared to previous Cas enzymes, Cas3 shows more significant potential in genome editing therapy with lower off-target effects [126]. Besides, Cas14, also known as Cas12f, shares a similar function with Cas12a. When bound to the target nucleic acid, Cas14 exhibits trans-cleavage activity towards ssDNA [127]. Initially, it was believed that Cas14 could only target ssDNA [128], but subsequent studies found that dsDNA with T-rich PAM sequences (5'-TTAT/TTTR-3', where R represents A or G bases) could also be recognized [129]. As the smallest Cas protein discovered so far, Cas14 has a lower tolerance to nucleic acid mismatches between crRNA and the target than the Cas12a system. This property benefits the detection of single nucleotide polymorphisms due to better discrimination [130, 131]. However, despite these intriguing characteristics, these Cas enzymes have not yet been applied in enzyme-free amplification or decision-making circuits. The exploration of their properties and potential applications is still ongoing, and further research is necessary to fully understand and harness their capabilities in various fields. The development and application of these systems could significantly expand the scope of CRISPR-based technologies, offering new avenues for DNA circuits [132, 133].

  4. Summary and perspectives

  This article provides a comprehensive review of the application of the CRISPR-Cas system in DNA functional circuits, specifically focusing on amplification and decision-making functions. Initially, the classification and mechanisms of common CRISPR-Cas systems were introduced, followed by a discussion of their applications, advantages, and potential disadvantages in functional circuits. Compared with traditional DNA circuits, the CRISPR-Cas system has made significant progress in amplification and decision-making circuits. Among them, CRISPR-Cas9 has emerged as a more suitable candidate for decision-making circuits, whereas CRISPR-Cas12a/13a plays a more significant role in amplification functional circuits. In amplification circuits, the rapid turnover rate of the CRISPR-Cas system further enables high amplification benefits. The cascaded CRISPR-Cas system, in particular, simplifies the design of the amplification circuit and minimizes the risk of cross-contamination between nucleic acids. In decision-making circuits, the dsDNA-activated features of Cas9 and Cas12a expand the range of DNA input to inert dsDNA containing PAM sequences. Additionally, the trans-cleavage activity of Cas12a and Cas13a serves as a natural 'INHIBIT' gate in circuit components. Even when solely used as the final signal reporting module, they can maximize the differences in decision-making, especially in the biological regulation of gene expression. Despite these promising attributes, the field faces several challenges that require resolution.

  (1) Whether in amplification circuits or decision-making circuits, the persistent challenge of off-target effects in the CRISPR-Cas system remains urgent, with the probability of occurrence increasing over time [134]. Mitigating these off-target effects poses a significant obstacle. The primary issue contributing to off-target effects is the challenge of precise positioning, given the relatively short recognition sequence of crRNA in CRISPR-Cas, which constrains the matching probability based on complementary base pairs. Currently, efforts to address off-target effects involve investigation of the structure of Cas proteins and artificial processing to exhibit fewer off-target effects [135]. Another approach to tackling off-target issues involves programming target sequences more effectively. Recent studies have shown that mismatches located at PAM sequence sites or proximal upstream and downstream sites of PAM sequences can enhance specificity [136]. However, specificity for non-seed regions distant from PAM sequences remains limited. In the future, improving the precise positioning of the CRISPR-Cas system through chemical modifications, such as the introduction of locked nucleic acids or peptide nucleic acids to enhance the affinity between crRNA and the target, may offer an ideal solution for improving specificity.

  (2) One limitation that hinders simpler, one-step operation is the circuit interference caused by the trans-cleavage activity of Cas12a and Cas13a. It is particularly problematic that a signal switch can only be turned on and not off. Currently, the primary approach to avoid interference is to introduce the CRISPR-Cas system through a step-by-step operation after the completion of the pathway reaction. Another strategy involves substituting the probes used throughout the pathway with ones that are resistant to trans-cleavage activity. However, these methods only provide an indirect anti-digestion strategy rather than fundamental control over the activity of the switch. Therefore, developing more convenient and controllable methods to regulate trans-cleavage activity may be a future trend. Currently, the analysis and comprehension of their cis-cleavage activity have been derived from the crystal structure of proteins, while their trans-cleavage activity lacks a satisfactory explanation. Progress in this direction may address interference issues, allow for more design flexibility in controlling the switches, and potentially resolve cross-contamination from CRISPR-Cas systems targeting multiple targets, leading to breakthroughs in decision-making circuits for Cas12a and Cas13a [137].

  (3) Using multiple crRNAs to construct multiple Cas/crRNA complexes that generate signals can further enhance the sensitivity of the amplification circuit. However, this approach encounters several challenges such as higher costs, potential RNA storage problems, and cross-contamination. A potential direction for improvement is to replace multiple crRNAs with multiple activation sequences in amplification circuits to enhance sensitivity. Cas12a and Cas13a can be directly activated by ssDNA or ssRNA complementary to crRNA, and strategies such as rolling circle amplification, transcription, and other approaches have been proposed to generate multiple activation sequences that can activate more Cas enzymes. These approaches typically require additional polymerase enzymes. Recent advancements in the field have seen the innovative use of trans-cleavage substrates as activators to create a self-catalysis mechanism for amplification [138, 139]. However, the design of trans-cleavage substrates must be meticulous to ensure that the trans-cleavage substrates can avoid inadvertently activating the Cas enzymes in the absence of the intended target. Enzyme-free strategies emphasized in the review including CHA, HCR, and EDC, while advantageous in certain respects, are currently restricted to a single activation sequence or linear production of multiple sequences. The cleavage activity of nonlinear branch junctions remains largely unexplored. By employing thoughtful design, it may be possible to generate planar structures with multiple branch junctions, which could increase the concentration of PAM sequences and improve the recognition probability of CRISPR-Cas.

  (4) Currently, decision-making circuit methods based on the CRISPR-Cas system are limited and primarily used as the final circuit signal reporting component without fully utilizing their advantages. The direct use of CRISPR-Cas as a circuit component is insufficient, indicating a clear demand for an expansion in their application to enhance the sophistication of decision-making circuits.

  Despite some progress made with CRISPR-Cas tools, their comprehensive potential is still significantly untapped. Their prospects in amplification circuits and decision-making circuits hold promise for augmenting the overall performance and functionality of these circuits, comprehensively dedicating to biological sensing and genetic engineering.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Xiaolong Li: Writing – original draft, Visualization. Changjiang Li: Writing – original draft. Chaopeng Shi: Writing – original draft. Jiarun Wang: Visualization. Bei Yan: Writing – review & editing. Xianjin Xiao: Writing – review & editing. Tongbo Wu: Writing – review & editing, Funding acquisition.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (Nos. 82172372 and 82260290), and the Opening Research Fund of State Key Laboratory of Digital Medical Engineering (No. 2023-M04).

  
  
• 1. [1]

     T.C. Xie, Y.H. Deng, J.R. Zhang, et al., Nucleic Acids Res. 50 (2022) 8431–8440. doi: 10.1093/nar/gkac650

  2. [2]

     S. Okumura, G. Gines, N. Lobato-Dauzier, et al., Nature 610 (2022) 496–501. doi: 10.1038/s41586-022-05218-7

  3. [3]

     M.P. Nikitin, Nat. Chem. 15 (2023) 70–82. doi: 10.1038/s41557-022-01111-y

  4. [4]

     Y. Sun, S. Qi, X. Dong, et al., J. Hazard. Mater. 443 (2023) 130252.

  5. [5]

     G. Shi, C. Yan, J. Chen, Anal. Chem. 93 (2021) 3273–3279. doi: 10.1021/acs.analchem.0c05173

  6. [6]

     Q. Hu, H. Li, L. Wang, H. Gu, C. Fan, Chem. Rev. 119 (2019) 6459–6506. doi: 10.1021/acs.chemrev.7b00663

  7. [7]

     S. Yang, J. Luo, L. Zhang, et al., Adv. Sci. 19 (2023) 2301814.

  8. [8]

     J. Chen, S. Fu, C. Zhang, H. Liu, X. Su, Small 18 (2022) 2108008.

  9. [9]

     C. Zhang, Y. Zhao, X. Xu, et al., Nat. Nanotechnol. 15 (2020) 709–715. doi: 10.1038/s41565-020-0699-0

  10. [10]

      Q. Chao, Y. Zhang, Q. Li, et al., Anal. Chem. 95 (2023) 7723–7734. doi: 10.1021/acs.analchem.3c00898

  11. [11]

      A.P. Lapteva, N. Sarraf, L. Qian, J. Am. Chem. Soc. 144 (2022) 12443–12449. doi: 10.1021/jacs.2c04325

  12. [12]

      T. Wang, H. Hellmer, F.C. Simmel, Curr. Opin. Biotechnol. 79 (2023) 102867.

  13. [13]

      D.Y. Zhang, E. Winfree, J. Am. Chem. Soc. 131 (2009) 17303–17314. doi: 10.1021/ja906987s

  14. [14]

      C. Jung, A.D. Ellington, Acc. Chem. Res. 47 (2014) 1825–35. doi: 10.1021/ar500059c

  15. [15]

      R.M. Dirks, N.A. Pierce, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 15275–15278.

  16. [16]

      B.W. Zhang, T.R. Tian, D.X. Xiao, et al., Adv. Funct. Mater. 32 (2022) 2109728.

  17. [17]

      Z.W. Luo, Y.X. Li, P. Zhang, et al., Trends Anal. Chem. 151 (2022) 116582.

  18. [18]

      H.M. Wang, Y.Q. He, J. Wei, et al., Angew. Chem. Int. Ed. 61 (2022) 202115489.

  19. [19]

      P. Yin, H.M. Choi, C.R. Calvert, N.A. Pierce, Nature 451 (2008) 318–22. doi: 10.1038/nature06451

  20. [20]

      D.Y. Zhang, A.J. Turberfield, B. Yurke, E. Winfree, Science 318 (2007) 1148532.

  21. [21]

      Q. Zhang, R. Zhao, C.C. Li, et al., Anal. Chem. 94 (2022) 13978–13986. doi: 10.1021/acs.analchem.2c03223

  22. [22]

      S.Z. He, S.S. Yu, R.M. Li, et al., Angew. Chem. Int. Ed. 61 (2022) 202206529.

  23. [23]

      R.M. Li, Y.X. Zhu, X. Gong, et al., J. Am. Chem. Soc. 145 (2023) 2999–3007. doi: 10.1021/jacs.2c11504

  24. [24]

      J. Wei, M.D. Yu, K.Y. Tan, et al., Small 19 (2023) 2207961.

  25. [25]

      P. Dong, R.M. Li, S.Z. He, et al., Chem. Sci. 14 (2023) 2159–2167. doi: 10.1039/d2sc05568f

  26. [26]

      C. Zhang, L.L. Ge, Y.C. Zhuang, et al., Sci. China Inf. Sci. 62 (2019) 061301.

  27. [27]

      F. Wang, H. Lv, Q. Li, et al., Nat. Commun. 11 (2020) 121.

  28. [28]

      X. Song, J. Reif, ACS Nano 13 (2019) 6256–6268. doi: 10.1021/acsnano.9b02562

  29. [29]

      J.K. Jung, C.M. Archuleta, K.K. Alam, J.B. Lucks, Nat. Chem. Biol. 18 (2022) 385–393. doi: 10.1038/s41589-021-00962-9

  30. [30]

      S. Piranej, A. Bazrafshan, K. Salaita, Nat. Nanotechnol. 17 (2022) 514–523. doi: 10.1038/s41565-022-01080-w

  31. [31]

      N. Xie, M.Q. Li, Y. Wang, et al., J. Am. Chem. Soc. 144 (2022) 9479–9488. doi: 10.1021/jacs.2c03258

  32. [32]

      T. Song, A. Eshra, S. Shah, et al., Nat. Nanotechnol. 14 (2019) 1075–1081. doi: 10.1038/s41565-019-0544-5

  33. [33]

      H. Hu, L.Q. Liu, L. Zhang, et al., Nano Res. 16 (2022) 865–872.

  34. [34]

      K.M. Cherry, L. Qian, Nature 559 (2018) 370–376. doi: 10.1038/s41586-018-0289-6

  35. [35]

      K. Shi, S.Y. Xie, R.Y. Tian, et al., Sci. Adv. 7 (2021) 7802.

  36. [36]

      S.Y. Chen, B. Gong, C. Zhu, C.Y. Lei, Z. Nie, Trends Anal. Chem. 159 (2023) 116931.

  37. [37]

      H.M. Li, Y. Xie, F.M. Chen, et al., Chem. Soc. Rev. 52 (2023) 361–382. doi: 10.1039/d2cs00594h

  38. [38]

      W.R. Su, J.R. Li, C. Ji, et al., Nano Res. 16 (2023) 9940–9953. doi: 10.1007/s12274-023-5567-4

  39. [39]

      Y. Li, S.Y. Li, J. Wang, G.Z. Liu, Trends Biotechnol. 37 (2019) 730–743. doi: 10.3390/catal9090730

  40. [40]

      Z.Y. Weng, Z. You, J. Yang, et al., Angew. Chem. Int. Ed. 62 (2023) 202214987.

  41. [41]

      R.J. Fu, Y.L. Xianyu, Small 19 (2023) 2300057.

  42. [42]

      J.D. Habimana, R. Huang, B. Muhoza, et al., Biosens. Bioelectron. 203 (2022) 114033.

  43. [43]

      X.K. Cheng, Y.R. Li, J. Kou, et al., Biosens. Bioelectron. 215 (2022) 114559.

  44. [44]

      R. Montagud-Martinez, M. Heras-Hernandez, L. Goiriz, J.A. Daros, G. Rodrigo, A.C.S. Synth. Biol. 10 (2021) 950–956. doi: 10.1021/acssynbio.0c00649

  45. [45]

      Y. Dai, R.A. Somoza, L. Wang, et al., Angew. Chem. Int. Ed. 58 (2019) 17399–17405. doi: 10.1002/anie.201910772

  46. [46]

      H. Li, J. Yang, G. Wu, et al., Angew. Chem. Int. Ed. 61 (2022) 202203826.

  47. [47]

      Z.Y. Li, X. Ding, K. Yin, et al., Biosens. Bioelectron. 192 (2021) 113498.

  48. [48]

      D. Samanta, S.B. Ebrahimi, N. Ramani, C.A. Mirkin, J. Am. Chem. Soc. 144 (2022) 16310–16315. doi: 10.1021/jacs.2c07625

  49. [49]

      C.M. Green, J. Spangler, K. Susumu, et al., ACS Nano 16 (2022) 20693–20704. doi: 10.1021/acsnano.2c07749

  50. [50]

      W. Feng, A.M. Newbigging, J. Tao, et al., Chem. Sci. 12 (2021) 4683–4698. doi: 10.1039/d0sc06973f

  51. [51]

      B.D. Huang, T.M. Groseclose, C.J. Wilson, Nat. Commun. 13 (2022) 3901.

  52. [52]

      D.D. Zeng, J.L. Jiao, T.L. Mo, Front. Microbiol. 15 (2024) 1355234.

  53. [53]

      J.L. Qiao, Z.Y. Zhao, Y.R. Li, et al., Crit. Rev. Food Sci. Nutr. (2023) 1–22. https://doi.org/10.1080/10408398.2023.2246558. doi: 10.1080/10408398.2023.2246558

  54. [54]

      Y. Dhingra, S.K. Suresh, P. Juneja, D.G. Sashital, Mol. Cell 82 (2022) 4353–4367.

  55. [55]

      L. Yu, M.A. Marchisio, Nucleic Acids Res. 51 (2023) 1473–1487. doi: 10.1093/nar/gkac1270

  56. [56]

      J.S. Gootenberg, O.O. Abudayyeh, M.J. Kellner, et al., Science 360 (2018) 439–444. doi: 10.1126/science.aaq0179

  57. [57]

      G. Petris, A. Casini, C. Montagna, et al., Nat. Commun. 8 (2017) 15334.

  58. [58]

      J. Burian, V.K. Libis, Y.A. Hernandez, et al., Nat. Biotechnol. 41 (2023) 626–630. doi: 10.1038/s41587-022-01531-8

  59. [59]

      J.C. Cofsky, K.M. Soczek, G.J. Knott, E. Nogales, J.A. Doudna, Nat. Struct. Mol. Biol. 29 (2022) 395–402. doi: 10.1038/s41594-022-00756-0

  60. [60]

      T. Luo, J.C. Li, Y. He, et al., Anal. Chem. 94 (2022) 6566–6573. doi: 10.1021/acs.analchem.2c00401

  61. [61]

      J.Y. Zhang, Z.Y. Li, C. Guo, et al., Angew. Chem. Int. Ed. (2024) 202403123.

  62. [62]

      P. Jain, S. Rananaware, E. Vesco, et al., Nat. Commun. 14 (2023) 5409.

  63. [63]

      S.Y. Li, Q.X. Cheng, J.K. Liu, et al., Cell Res. 28 (2018) 491–493. doi: 10.1038/s41422-018-0022-x

  64. [64]

      J.S. Chen, E. Ma, L.B. Harrington, et al., Science 360 (2018) 436–439. doi: 10.1126/science.aar6245

  65. [65]

      N. Mohammad, L. Talton, Z. Hetzler, M. Gongireddy, Q.S. Wei, Nucleic Acids Res. 51 (2023) 9894–9904. doi: 10.1093/nar/gkad715

  66. [66]

      W. Zhang, Y.Q. Mu, K.J. Dong, et al., Nucleic Acids Res. 50 (2022) 12674–12688. doi: 10.1093/nar/gkac1144

  67. [67]

      Z.J. Liu, J. Xu, S. Huang, et al., Biosens. Bioelectron. 247 (2024) 115936.

  68. [68]

      Y. Wu, W. Luo, Z. Weng, et al., Nucleic Acids Res. 50 (2022) 11727–11737. doi: 10.1093/nar/gkac886

  69. [69]

      O.O. Abudayyeh, J.S. Gootenberg, S. Konermann, et al., Science 353 (2016) 5573.

  70. [70]

      N.K. Fan, X.T. Bian, M. Li, et al., Sci. Adv. 8 (2022) 7382.

  71. [71]

      A. East-Seletsky, M.R. O'Connell, S.C. Knight, et al., Nature 538 (2016) 270–273. doi: 10.1038/nature19802

  72. [72]

      K. Clement, J.Y. Hsu, M.C. Canver, J.K. Joung, L. Pinello, Mol. Cell 79 (2020) 11–29.

  73. [73]

      H.R. Kempton, L.E. Goudy, K.S. Love, L.S. Qi, Mol. Cell 78 (2020) 184–191.

  74. [74]

      S. Kawasaki, H. Ono, M. Hirosawa, et al., Nat. Commun. 14 (2023) 2243.

  75. [75]

      L. Oesinghaus, F.C. Simmel, Nat. Commun. 10 (2019) 2092.

  76. [76]

      W.J. Wang, Q.Y. Ge, X.W. Zhao, Trend. Anal. Chem. 160 (2023) 116960.

  77. [77]

      I.M. Slaymaker, L.Y. Gao, B. Zetsche, et al., Science 351 (2016) 84–88. doi: 10.1126/science.aad5227

  78. [78]

      Y. Liu, F. Pinto, X.Y. Wan, et al., Nat. Commun. 13 (2022) 1937. doi: 10.1109/icsp54964.2022.9778427

  79. [79]

      M. Chen, D.M. Wu, S.H. Tu, et al., Biosens. Bioelectron. 173 (2020) 112821.

  80. [80]

      J. Huang, Z. Liang, Y. Liu, J.D. Zhou, F.J. He, Anal. Chem. 94 (2022) 11409–11415. doi: 10.1021/acs.analchem.2c02538

  81. [81]

      D.G. Zhang, L.J. Cai, X.W. Wei, et al., Nano Today 40 (2021) 101268.

  82. [82]

      Y.H. Zhang, L. Qian, W.J. Wei, et al., ACS Synth. Biol. 6 (2017) 211–216. doi: 10.1021/acssynbio.6b00215

  83. [83]

      B. Yurke, A.J. Turberfield, A.P. Mill, F.C. Simmel, J.L. Neumann, Nature 406 (2000) 605–608.

  84. [84]

      X.B. Huang, Q. Zhou, M.X. Wang, et al., Front. Mol. Biosci. 7 (2020) 627848.

  85. [85]

      Y.C. Gao, X. Xiong, S. Wong, et al., Nat. Methods 13 (2016) 1043. doi: 10.1038/nmeth.4042

  86. [86]

      M. Bellato, A.F. Chiacchiera, E. Salibi, et al., Front. Bioeng. Biotechnol. 9 (2021) 743950.

  87. [87]

      J. Santos-Moreno, Y. Schaerli, Biochem. Soc. Trans. 48 (2020) 1979–1993. doi: 10.1042/bst20200020

  88. [88]

      A.K. Shaytan, R.V. Novikov, R.S. Vinnikov, A.K. Gribkova, G.S. Glukhov, Front. Mol. Biosci. 9 (2022) 1070526.

  89. [89]

      H. Kim, D. Bojar, M. Fussenegger, Proc. Natl. Acad. Sci. U.S.A. 116 (2019) 7214–7219. doi: 10.1073/pnas.1821740116

  90. [90]

      J. Elbaz, O. Lioubashevski, F. Wang, et al., Nat. Nanotechnol. 5 (2010) 417–422. doi: 10.1038/nnano.2010.88

  91. [91]

      H. Pei, L. Liang, G.B. Yao, et al., Angew. Chem. Int. Ed. 51 (2012) 9020–9024. doi: 10.1002/anie.201202356

  92. [92]

      D.C. Swarts, M. Jinek, Mol. Cell 73 (2019) 589–600.

  93. [93]

      M. Rossetti, R. Merlo, N. Bagheri, et al., Nucleic Acids Res. 50 (2022) 8377–8391. doi: 10.1093/nar/gkac578

  94. [94]

      S.D. Wang, H.J. Li, K.J. Dong, et al., Biosens. Bioelectron. 226 (2023) 115139.

  95. [95]

      J. You, H. Park, H. Lee, et al., Biosens. Bioelectron. 224 (2023) 115078.

  96. [96]

      M. Qing, S.L. Chen, Z. Sun, et al., Anal. Chem. 93 (2021) 7499–7507. doi: 10.1021/acs.analchem.1c00805

  97. [97]

      H. Yan, Y.J. Wen, Z.M. Tian, et al., Nat. Biomed. Eng. 7 (2023) 1583–1601. doi: 10.1038/s41551-023-01033-1

  98. [98]

      L. Liao, T.T. Gong, B.Y. Jiang, R. Yuan, Y. Xiang, Sensor. Actuator. B Chem. 379 (2023) 133221.

  99. [99]

      S. Peng, Z. Tan, S.Y. Chen, C.Y. Lei, Z. Nie, Chem. Sci. 11 (2020) 7362–7368. doi: 10.1039/d0sc03084h

  100. [100]

       Y. Lv, Y.H. Sun, Y. Zhou, et al., Small 19 (2023) 2206105.

  101. [101]

       H.Y. Jia, H.L. Zhao, T. Wang, et al., Biosens. Bioelectron. 211 (2022) 114382.

  102. [102]

       X.F. Wang, F.Y. Wang, J.R. Wang, et al., Sensor. Actuator. B: Chem. 370 (2022) 132480.

  103. [103]

       C. Yan, G. Shi, J.H. Chen, J. Agric. Food Chem. 70 (2022) 12700–12707. doi: 10.1021/acs.jafc.2c04548

  104. [104]

       S.H. Gong, X. Wang, P. Zhou, et al., Anal. Chem. 94 (2022) 15839–15846. doi: 10.1021/acs.analchem.2c03666

  105. [105]

       J.F. Pan, F. Deng, Z. Liu, L.W. Zeng, J.H. Chen, Talanta 255 (2023) 124210.

  106. [106]

       Z.F. Gao, L.L. Zheng, L.M. Dong, et al., Anal. Chem. 94 (2022) 6371–6379. doi: 10.1021/acs.analchem.2c00848

  107. [107]

       X.M. Hang, P.F. Liu, S. Tian, et al., Biosens. Bioelectron. 211 (2022) 114393.

  108. [108]

       C. Zhang, P.H. Zhang, H. Ren, et al., Chem. Eng. J. 446 (2022) 136864.

  109. [109]

       J.J. Shen, X.M. Zhou, Y.Y. Shan, et al., Nat. Commun. 11 (2020) 267.

  110. [110]

       Q. Han, X.L. Bian, Y. Chen, B.Z. Li, Sens. Actuators B: Chem. 393 (2023) 134265.

  111. [111]

       Z.Y. Zhang, J.L. Li, C.L. Chen, et al., Anal. Chim. Acta 1300 (2024) 342409.

  112. [112]

       J.J. Zhang, C.Y. Song, Y.F. Zhu, et al., Biosens. Bioelectron. 219 (2022) 114836.

  113. [113]

       Y.H. Yang, J.C. Liu, X.H. Zhou, Biosens. Bioelectron. 190 (2021) 113418.

  114. [114]

       Y.G. Ding, C. Tous, J. Choi, J. Chen, W.W. Wong, Nat. Commun. 15 (2024) 1572.

  115. [115]

       T. Tian, B.W. Shu, Y.Z. Jiang, et al., ACS Nano 15 (2021) 1167–1178. doi: 10.1021/acsnano.0c08165

  116. [116]

       Y. Sheng, T.H. Zhang, S.H. Zhang, et al., Biosens. Bioelectron. 178 (2021) 113027.

  117. [117]

       P. Fozouni, S. Son, M.D.L. Derby, et al., Cell 184 (2021) 323–333.

  118. [118]

       M.C. Zeng, Y.Q. Ke, Z.Y. Zhuang, et al., Anal. Chem. 94 (2022) 10805–10812. doi: 10.1021/acs.analchem.2c01588

  119. [119]

       Y.B. Zhang, Y. Chen, Q.L. Zhang, Y.Z. Liu, X.J. Zhang, Biosens. Bioelectron. 230 (2023) 115248.

  120. [120]

       S. Nimkar, B. Anand, Nucleic Acids Res. 48 (2020) 2486–2501. doi: 10.1093/nar/gkz1218

  121. [121]

       C.Y. Hu, D.C. Ni, K.H. Nam, et al., Mol. Cell 82 (2022) 2754–2768. doi: 10.3390/plants11202754

  122. [122]

       K.E. Dillard, M.W. Brown, N.V. Johnson, et al., Cell 175 (2018) 934–946.

  123. [123]

       K. Yoshimi, K. Takeshita, N. Kodera, et al., Nat. Commun. 13 (2022) 4917.

  124. [124]

       Y.B. Xiao, M. Luo, A.E. Dolan, M.F. Liao, A. Ke, Science 361 (2018) eaat0839.

  125. [125]

       Y.W. Huo, K.H. Nam, F. Ding, et al., Nat. Struct. Mol. Biol. 21 (2014) 771–777. doi: 10.1038/nsmb.2875

  126. [126]

       H. Morisaka, K. Yoshimi, Y. Okuzaki, et al., Nat. Commun. 10 (2019) 5302.

  127. [127]

       R.J. Xiao, Z. Li, S.K. Wang, R.J. Han, L.F. Chang, Nucleic Acids Res. 49 (2021) 4120–4128. doi: 10.1093/nar/gkab179

  128. [128]

       L.B. Harrington, D. Burstein, J.S. Chen, et al., Science 362 (2018) 839–842. doi: 10.1126/science.aav4294

  129. [129]

       T. Karvelis, G. Bigelyte, J.K. Young, et al., Nucleic Acids Res. 48 (2020) 5016–5023. doi: 10.1093/nar/gkaa208

  130. [130]

       X.S. Xu, A. Chemparathy, L.P. Zeng, et al., Mol. Cell 81 (2021) 4333–4345.

  131. [131]

       P. Wu, X.S. Ye, D.Q. Wang, et al., J. Hazard. Mater. 424 (2022) 127690.

  132. [132]

       M.M. Chen, J.Y. Zhang, Y. Peng, et al., Biosens. Bioelectron. 218 (2022) 114792.

  133. [133]

       B. Csörgő, L.M. León, I.J. Chau-Ly, et al., Nat. Methods 17 (2020) 1183–1190. doi: 10.1038/s41592-020-00980-w

  134. [134]

       C. Kuscu, S. Arslan, R. Singh, J. Thorpe, M. Adli, Nat. Biotechnol. 32 (2014) 677–683. doi: 10.1038/nbt.2916

  135. [135]

       H. Mangkhwar, B. Li, X. Ding, et al., Adv. Sci. 7 (2020) 1902312.

  136. [136]

       H.Y. Xiao, J.Y. Hu, C. Huang, et al., Trends Anal. Chem. 161 (2023) 117000.

  137. [137]

       M. Karlikow, E. Amalfitano, X. Yang, et al., Nat. Commun. 14 (2023) 1505.

  138. [138]

       F. Deng, Y. Li, B.Y. Yang, et al., Nat. Commun. 15 (2024) 1818.

  139. [139]

       K. Sun, L. Pu, C. Chen, et al., Nucleic Acids Res. 52 (2024) 1–13.

• 

  Figure 1  Schematic of class 2 CRISPR-Cas system. Cleavage mechanisms of (A) Cas9, (B) Cas12a, and (C) Cas13a. Reproduced with permission [37]. Copyright 2023, The Royal Society of Chemistry.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Applications of Cas9-based circuit. (A) CRISPR-Cas9 cleavage triggered the EDC amplification circuit. Reproduced with permission [79]. Copyright 2020, Elsevier B.V. (B) An MSPQC nucleic acid amplified sensor by the combination of CRISPR-Cas9 and HCR. Reproduced with permission [80]. Copyright 2022, American Chemical Society. (C) CRISPR-Cas9 mediated strand displacement logic circuits with toehold-free DNA. Reproduced with permission [44]. Copyright 2021, American Chemical Society. (D) The central processing unit for program complex logic computation in cells based CRISPR-Cas9. Reproduced with permission [89]. Copyright 2019, according to Creative Comments Attribution License 4.0 (CC BY).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Applications of Cas12a-based circuits. (A) Cascade circuits. Reproduced with permission [100]. Copyright 2023, Wiley-VCH GmbH. (B) HCR. Reproduced with permission [101]. Copyright 2022, Elsevier B.V. (C) CHA. Reproduced with permission [102]. Copyright 2022, Elsevier B.V. Decision-making circuits-based CRISPR-Cas12a for (D) Dual miRNAs AND logic gate. Reproduced with permission [104]. Copyright 2022, American Chemical Society. (E) Molecular logic gate of heavy metal ions. Reproduced with permission [105]. Copyright 2022, Elsevier B.V. (F) Guanine nanowire boolean logic tree. Reproduced with permission [106]. Copyright 2022, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Applications of Cas13a-based circuits. (A) Amplification circuit integrating CRISPR-Cas13a and HCR for SERS detection. Reproduced with permission [112]. Copyright 2022, Elsevier B.V. (B) Amplification circuit integrating CRISPR-Cas13a and HCR for portable evanescent wave sensor. Reproduced with permission [113]. Copyright 2021, Elsevier B.V. (C) Orthogonal inducible control of Cas13a for programmable AND circuit. (D) An ultralocalized Cas13a assay for digital computing. Reproduced with permission [115]. Copyright 2020, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Circuits based on cascade CRISPR. (A) Cascade of Cas13a/crRNA complexes for amplification-free detection of SARS-CoV-2. Reproduced with permission [117]. Copyright 2020, Elsevier Inc. (B) Multiplex crRNA in CRISPR-Cas12a system for DNA diagnostic. Reproduced with permission [118]. Copyright 2022, American Chemical Society. (C) Cascade CRISPR-Dx system for detection of SARS-CoV-2. Reproduced with permission [119]. Copyright 2023, Elsevier B.V.

  下载: 全尺寸图片 幻灯片
•