Type I-F vs. Type V-K CAST Systems: A Comparative Analysis of Editing Efficiency and Cargo Capacity for Therapeutic Genome Integration

Abstract

This article provides a comparative analysis for researchers and drug development professionals on two primary CRISPR-associated transposase (CAST) systems: the multi-subunit Type I-F and the compact Type V-K. We explore their foundational mechanisms, contrasting the Cascade-based targeting of Type I-F with the Cas12k-driven approach of Type V-K. The content details methodological advances that enable therapeutic gene integration in human cells, including evolved transposases (evoCAST) and nuclear localization strategies. We troubleshoot key challenges such as low human cell efficiency and off-target integration, highlighting protein engineering and structural insights as solutions. Finally, we present a validated, side-by-side comparison of their editing efficiency, cargo size handling, product purity, and specificity, synthesizing the current landscape to inform tool selection for next-generation gene therapy applications.

Architectural Blueprints: Deconstructing the Molecular Machinery of Type I-F and Type V-K CAST Systems

CRISPR-associated transposases (CASTs) represent a powerful new class of genome engineering tools that enable RNA-guided integration of large DNA payloads without creating double-strand breaks. These systems combine programmable DNA targeting with transposase-mediated integration, offering significant advantages over conventional CRISPR-Cas systems that rely on host DNA repair mechanisms. Among the various CAST systems characterized to date, type I-F and type V-K have emerged as the leading architectures, each with distinct structural organizations and functional characteristics. This guide provides a comprehensive comparison of these two systems, focusing on their core components, operational mechanisms, and performance metrics to inform selection for research and therapeutic development.

System Architectures and Core Components

The fundamental distinction between type I-F and type V-K CAST systems lies in their DNA-targeting machinery. Type I-F systems employ a multi-subunit Cascade complex, while type V-K systems utilize a single-effector Cas12k protein, resulting in significant differences in system complexity and functional properties.

Table 1: Core Components of Type I-F and Type V-K CAST Systems

Component Type	Function	Type I-F	Type V-K
Targeting Module	RNA-guided DNA recognition	Multi-subunit Cascade complex (Cas5, Cas6, Cas7, Cas8)	Single effector Cas12k
Adaptor Protein	Couples targeting to transposition	TniQ (homodimer)	TniQ
Transposase	Catalyzes DNA integration	TnsB	TnsB
ATPase	Regulates transposition assembly	TnsC	TnsC
Auxiliary Nuclease	Controls integration outcome	TnsA (present)	TnsA (absent)
Host Factors	Enhances specificity/function	Variable	Ribosomal protein S15

Type I-F Cascade Architecture

Type I-F CAST systems employ a multi-protein CRISPR complex known as Cascade (CRISPR-associated complex for antiviral defense) for DNA targeting. This complex typically comprises several Cas proteins (Cas5, Cas6, Cas7, and Cas8) arranged in a pseudo-helical structure that coats the crRNA molecule [1]. The Cas8 protein contains two domains: a bulky domain that interacts with Cas7.1 and binds the crRNA 5' end and PAM sequence, and a second α-helical domain that exhibits dynamic behavior [1]. The TniQ protein forms a stable homodimer that associates with the Cascade complex, creating the complete TniQ-Cascade (QCascade) targeting module [1]. Recent structural studies of the PseCAST QCascade complex using cryo-EM revealed that the TniQ dimer exhibits significant flexibility, populating a range of positions relative to the rest of the complex that pivot around Cas6 and Cas7.6 [1].

Type V-K Cas12k Architecture

In contrast to the multi-subunit Cascade, type V-K CAST systems utilize a single effector protein—Cas12k—for DNA targeting. Cas12k is a ~637-residue protein that adopts a bi-lobed structure connected by a loop, with the N-terminal lobe composed of WED, REC1, and PI domains [2]. Despite belonging to the Cas12 family, Cas12k features a naturally inactivated RuvC nuclease domain, which precludes DNA cleavage activity while preserving DNA binding capability [2]. Structural analyses reveal that Cas12k recognizes a specific GGTT protospacer adjacent motif (PAM) sequence and forms a complex with TniQ and the ribosomal protein S15, which engages the tracrRNA component to facilitate stable R-loop formation [3]. The entire targeting module is significantly more compact than the type I-F Cascade system.

Performance Comparison and Experimental Data

Extensive characterization of both CAST systems has revealed distinct performance profiles in terms of integration efficiency, specificity, and insertion patterns.

Table 2: Performance Characteristics of Type I-F and Type V-K CAST Systems

Performance Metric	Type I-F	Type V-K	Experimental Evidence
Integration Efficiency in E. coli	High (up to 80%)	Variable	[4]
Specificity (On-target Integration)	High (≥98%)	Moderate to Low (12-76%)	[5]
Insertion Orientation	Bidirectional	Predominantly Unidirectional	[6]
Integration Product	Simple Insertion (Cut-and-Paste)	Co-integrate (Copy-and-Paste)	[4] [5]
Activity in Human Cells	Demonstrated (Low Efficiency)	Demonstrated (Requires Engineering)	[1] [4]
PAM Preference	5'-CC-3'	5'-GGTT-3' or 5'-GTN-3'	[2] [1] [4]

Integration Specificity and Fidelity

Type I-F CAST systems generally demonstrate higher integration specificity compared to type V-K systems. Studies of the VchCAST (I-F) system revealed predominantly on-target activity in bacterial cells, with specificities often exceeding 98% [5]. In contrast, type V-K systems such as ShCAST exhibit significant off-target integration, with on-target efficiencies ranging from 12% to 76% depending on the guide RNA used [5].

This fidelity difference stems from distinct mechanistic pathways. Type V-K CASTs maintain both RNA-guided and RNA-independent transposition pathways, with the latter driven by spontaneous TnsC filament formation on AT-rich DNA regions [5]. Biochemical and single-molecule experiments have confirmed that a minimal transpososome comprising TnsB, TnsC, and TniQ (without Cas12k) can catalyze untargeted integration, with TnsC acting as the primary driver of this promiscuous activity [5].

Insertion Patterns and Products

The presence of TnsA in type I-F systems enables a "cut-and-paste" transposition mechanism resulting in simple insertion products [6]. Type V-K systems lack TnsA and consequently operate via a "copy-and-paste" mechanism that generates co-integrate products where the entire donor plasmid integrates alongside the transposon cargo [4] [5]. Quantitative analysis of the MG64-1 and MG64-6 type V-K systems revealed that 70-80% of integration events were co-integrations, with only 20-30% representing simple insertions [4].

Insertion orientation also differs between the systems. Type I-F CASTs can produce bidirectional insertions, though I-F systems favor insertion in a specific orientation (T-RL, with the right homology end closest to the target site) [6]. Type V-K CASTs exhibit predominantly unidirectional insertion behavior, which can be advantageous for applications requiring precise control over insertion orientation [6].

Experimental Protocols and Methodologies

In Vitro DNA Transposition Assay

The foundational assay for characterizing CAST system activity involves reconstituting the integration machinery with purified components [2]. The standard protocol includes:

• Protein Purification: Recombinant expression and purification of all CAST components (Cas12k or Cascade proteins, TnsB, TnsC, TniQ).
• Nucleic Acid Preparation: In vitro transcription of sgRNA and preparation of target DNA containing the PAM sequence and donor DNA with appropriate terminal inverted repeats.
• Integration Reaction: Assembling the reaction mixture containing CAST proteins, sgRNA, target DNA, and donor DNA in appropriate buffer conditions with magnesium.
• Product Detection: PCR amplification of donor-target junctions using orientation-specific primers to detect successful integration events.

This assay has been instrumental in defining component requirements, with studies showing that TnsB, TnsC, and magnesium are strictly required for DNA transposition, while Cas12k, sgRNA, and TniQ are necessary for RNA-guided specificity [2].

Genomic Integration Efficiency Assay

For quantifying CAST activity in cellular environments, researchers employ a conjugation-based chromosomal transposition assay [7]. Key steps include:

• Vector Construction: Assembling CAST genes, CRISPR array, and antibiotic resistance marker flanked by transposon inverted repeats into a conditionally replicative plasmid.
• Bacterial Mating: Conjugative transfer of the CAST plasmid from donor to recipient strains.
• Selection and Screening: Plating recipient cells on selective media containing antibiotics and X-gal for colorimetric screening of integration events.
• Efficiency Calculation: Quantifying transposition efficiency as the ratio of antibiotic-resistant colonies to total viable recipient cells.
• Specificity Assessment: Whole-genome sequencing of resistant colonies to map integration sites and determine on-target versus off-target frequencies.

This assay enabled the discovery that type V-K CASTs can perform RNA-independent transposition and identified TnsC as the primary determinant of this promiscuous activity [5].

Structural Insights and Engineering Strategies

Recent advances in cryo-electron microscopy have provided high-resolution structures of both CAST systems, enabling structure-guided engineering approaches to improve their performance.

Type I-F Engineering Opportunities

Structural analysis of the PseCAST QCascade complex revealed that the Cas8 α-helical domain and TniQ dimer exhibit considerable conformational flexibility, suggesting potential engineering targets for stabilizing specific functional states [1]. Combining structural data with library screening has yielded engineered QCascade variants with increased integration efficiencies and modified PAM specificities [1]. Additionally, computational predictions of transpososome architecture using AlphaFold-Multimer have enabled the design of hybrid CAST systems that combine orthogonal DNA binding and integration modules [1].

Type V-K Specificity Enhancement

For type V-K systems, structural insights have informed strategies to suppress RNA-independent transposition. The cryo-EM structure of the Cas12k-transposon recruitment complex revealed how TniQ contacts TnsC protomers at the Cas12k-proximal filament end, likely nucleating its polymerization [3]. Building on this knowledge, researchers have developed enhanced specificity variants by modulating cytoplasmic TnsC levels and engineering DNA-binding residues in TnsC (such as K103) to reduce AT-rich sequence preference [5]. These approaches have increased type V-K CAST specificity to 98.1% in E. coli without compromising on-target integration efficiency [5].

Research Reagent Solutions

Table 3: Essential Research Reagents for CAST System Investigation

Reagent	Function	Examples/Specifications
Expression Vectors	Protein production	Plasmid systems for recombinant expression of CAST components
sgRNA Constructs	Guide RNA delivery	Templates for in vitro transcription or direct expression
Donor Plasmids	Transposon cargo source	Contains terminal inverted repeats (TIRs) and cargo DNA
Target Substrates	Integration site validation	Plasmid or genomic targets with defined PAM sequences
Host Factor Supplements	Enhance integration	S15 ribosomal protein for type V-K, ClpX for type I-F in human cells
Engineering Toolkits	System optimization	CRISPR-Cas components for genome editing in host organisms

System Selection Guidelines

The choice between type I-F and type V-K CAST systems depends on the specific research requirements:

• For applications demanding high specificity: Type I-F systems are preferable due to their superior fidelity and well-controlled targeted integration.
• For applications requiring compact system size: Type V-K systems offer advantages with their single-effector targeting module and smaller coding sequence.
• For therapeutic applications in human cells: Both systems require substantial engineering, with type I-F demonstrating initial success in human cells and type V-K showing promise with proper nuclear localization and host factor co-expression.
• For fundamental transposition mechanism studies: Type V-K systems provide insights into both RNA-guided and RNA-independent pathways.

Future directions in CAST system development will likely focus on enhancing specificity for type V-K systems, improving efficiency in eukaryotic environments for both systems, and creating hybrid architectures that combine beneficial properties from multiple CAST subtypes.

Visualization: Architecture and functional comparison of Type I-F and Type V-K CAST systems, highlighting their distinct component organizations and performance characteristics.

CRISPR-associated transposases (CASTs) represent a revolutionary class of genome editing tools that combine RNA-guided DNA targeting with programmable transposition. Unlike conventional CRISPR-Cas systems that create double-strand breaks, CAST systems facilitate double-strand break-free integration of large DNA cargoes, making them particularly valuable for therapeutic applications where genomic stability is paramount [8]. These systems are categorized into two primary classes based on their targeting architectures: Type I-F systems utilizing multi-protein Cascade complexes and Type V-K systems employing single-effector Cas12k proteins [8] [4]. The fundamental differences in their targeting modules directly impact their mechanism of action, editing efficiency, and practical applications in human genome engineering.

This review comprehensively compares the RNA-guided DNA recognition and R-loop formation mechanisms between Type I-F and Type V-K CAST systems, examining how their structural differences influence editing efficiency, cargo size capacity, and suitability for therapeutic development. We analyze recent structural insights and experimental data to provide researchers with a foundation for selecting appropriate CAST systems for specific genome editing applications.

Molecular Architecture of Targeting Modules

Type I-F Cascade: A Multi-Subunit Surveillance Complex

The Type I-F CRISPR-associated complex for antiviral defense (Cascade) represents a sophisticated multi-protein assembly that coordinates DNA recognition and R-loop formation through intricate subunit specialization:

• Structural Composition: The PseCAST QCascade complex comprises six Cas7 monomers forming a pseudo-helical backbone, one Cas8 protein containing PAM-recognition and α-helical domains, one Cas6 protein stabilizing the crRNA 3′ end, and a TniQ homodimer that recruits downstream transposition components [9]. This elaborate assembly creates a 405-kDa ribonucleoprotein complex that orchestrates DNA surveillance [10].

• Mechanism of Action: Target recognition initiates with PAM identification by the Cas8 subunit, which triggers local DNA melting and enables crRNA hybridization with the target strand [9]. The six Cas7 subunits form a continuous binding surface that facilitates directional R-loop propagation through stepwise structural rearrangements [11]. Recent cryo-EM structures reveal that the TniQ dimer exhibits significant conformational flexibility, populating both "open" and "closed" states relative to the Cas8 helical domain, suggesting a dynamic recruitment interface for transposition machinery [9].

• R-loop Formation: Structural analyses indicate that R-loop initiation requires only 6 nucleotides of complementarity in the PAM-proximal seed region, with Tyr450 stacking interactions providing a checkpoint against promiscuous binding [11]. As hybridization extends beyond the seed region, REC2 and REC3 domains undergo substantial rearrangements to accommodate the expanding R-loop, with complete heteroduplex formation triggering nuclease domain activation in functional Cas systems [11].

Type V-K Cas12k: Compact Single-Effector Targeting

In contrast to the multi-subunit Cascade, Type V-K CAST systems employ a streamlined targeting architecture centered around a single Cas12k effector protein:

• Structural Composition: Type V-K targeting modules comprise Cas12k, TniQ, and occasionally the bacterial host factor S15, forming a considerably more compact complex than their Type I-F counterparts [8] [12]. This minimalist architecture simplifies heterologous expression while maintaining programmability.

• Mechanism of Action: Cas12k independently handles both PAM recognition and R-loop formation without requiring additional Cas proteins [4]. Structural studies of the holo transpososome reveal that Cas12k undergoes significant conformational changes upon target binding, organizing the integration complex through direct interactions with TniQ and TnsB [12].

• R-loop Formation: Despite its simplified architecture, Cas12k facilitates R-loop formation through mechanisms that share fundamental similarities with Cascade systems, including PAM-dependent DNA melting and directional heteroduplex propagation [8]. The compact nature of the Cas12k complex may limit its ability to stabilize extensive R-loops, potentially influencing targeting flexibility and editing efficiency [4].

Table 1: Structural Comparison of CAST Targeting Modules

Feature	Type I-F Cascade	Type V-K Cas12k
Core Targeting Components	Cas8, Cas7×6, Cas6, TniQ×2	Cas12k, TniQ, S15
Molecular Weight	~405 kDa [10]	~160 kDa (Cas12k only)
crRNA Handling	Cas6 processes pre-crRNA; Cas7 backbone presents guide [9]	Pre-processed sgRNA with conserved tracrRNA structures [4]
PAM Recognition	Cas8 subunit recognizes 5′-CC-3′ PAM [9]	Cas12k recognizes 5′-GTN-3′ or 5′-rGTN-3′ PAM [4]
TniQ Recruitment	TniQ dimer flexibly associates with Cas6/Cas7.6 [9]	TniQ directly interacts with Cas12k [12]
Structural Flexibility	High conformational flexibility in TniQ and Cas8 domains [9]	Moderate conformational changes upon DNA binding [12]

Structural Determinants of R-loop Formation

Universal Principles of R-loop Formation

R-loop formation represents a fundamental biological process wherein RNA invades the DNA duplex, displacing the non-template strand to form an RNA-DNA heteroduplex [13]. While artificial R-loops were initially generated in vitro using denaturing conditions, natural R-loop formation occurs during transcription and CRISPR-guided DNA recognition through threadback invasion mechanisms [13]. The superior thermodynamic stability of RNA-DNA hybrids compared to DNA-DNA duplexes provides the driving force for R-loop formation, particularly in GC-rich sequences where rG/dC base pairs offer exceptional stability [13].

Several factors universally influence R-loop stability across CAST systems:

• DNA Topology: Negative supercoiling facilitates R-loop initiation by reducing the energy barrier for DNA unwinding, while positive supercoiling promotes resolution [13].
• Sequence Composition: GC skew (guanine enrichment in the non-template strand) strongly correlates with R-loop formation propensity due to the exceptional stability of rG/dC hybrids [13].
• Steric Considerations: R-loop formation efficiency decreases with increasing distance between the initiation sequence and free RNA end, as long RNA tails create steric hindrance for efficient strand invasion [13].

System-Specific R-loop Formation Mechanisms

Despite these shared principles, significant differences exist in how Type I-F and Type V-K systems implement R-loop formation:

In Type I-F systems, R-loop formation proceeds through a bipartite seed mechanism initiated by PAM-proximal hybridization [11]. The REC2 and REC3 domains form a positively charged cleft that accommodates the distal DNA duplex during early R-loop formation, with stepwise domain rearrangements coupled to heteroduplex extension [11]. This elaborate mechanism provides multiple checkpoints for off-target discrimination but requires precise coordination between numerous protein subunits.

Type V-K systems employ a more direct R-loop formation pathway where Cas12k alone coordinates both PAM recognition and heteroduplex formation [4]. The simpler architecture may enable faster R-loop formation but potentially with reduced discrimination against mismatched targets. Structural analyses indicate that Cas12k stabilizes shorter heteroduplex regions compared to Cascade complexes, which may influence target site selection and editing efficiency [12].

Experimental Assessment of Targeting Efficiency

Quantitative Comparison of Editing Performance

Recent advances in CAST engineering have enabled direct comparison of Type I-F and Type V-K system performance in human cells. The following experimental data illustrate key differences in their editing capabilities:

Table 2: Performance Comparison of Engineered CAST Systems in Human Cells

Parameter	Type I-F (evoCAST)	Type V-K (MG64-1)
Integration Efficiency	10-25% across 14 genomic loci [14]	~15% at safe harbor locus [4]
Cargo Size Demonstrated	Kilobase-sized cargos [14]	Full therapeutic genes (Factor IX) [4]
PAM Specificity	5′-CC-3′ [9]	5′-GTN-3′ or 5′-rGTN-3′ [4]
Byproduct Formation	Predominantly unidirectional products [14]	20-30% single integration, 70-80% co-integration [4]
Host Factor Requirements	Enhanced activity with ClpX unfoldase [14]	Requires bacterial S15 protein [4]
Off-target Integration	Low detected levels [14]	Rare, localized to specific genomic regions [4]

Key Methodologies for Assessing Targeting Efficiency

Standardized experimental approaches have been developed to quantitatively evaluate CAST system performance:

• In Vitro Integration Assays: Purified CAST components are incubated with target plasmid libraries containing randomized PAM sequences, followed by PCR amplification of integration junctions and next-generation sequencing to determine PAM preferences and integration precision [4]. This approach identified the 5′ GTN PAM for MG64-1 and 5′ rGTN PAM for MG64-6 systems with 90% of integrations occurring 57-67 bp from the PAM [4].

• Genomic Integration in E. coli: Multi-plasmid systems encoding CAST proteins, guide RNAs, and donor DNA are transformed into engineered E. coli strains with integration efficiency quantified via qPCR and whole genome sequencing [4]. This method demonstrated up to 80% integration efficiency at endogenous loci for optimized Type V-K systems [4].

• Human Cell Engineering: CAST components are engineered with nuclear localization signals and codon-optimized for mammalian expression, with integration efficiency measured at safe harbor loci (e.g., AAVS1) using targeted sequencing [14] [4]. Recent optimizations have achieved 10-25% integration efficiencies for Type I-F evoCAST systems across multiple genomic sites [14].

Visualization of Targeting Mechanisms

The fundamental differences in targeting module architecture between Type I-F and Type V-K systems can be visualized through the following mechanistic diagrams:

The Researcher's Toolkit: Essential Reagents and Applications

Table 3: Key Research Reagents for CAST Targeting Studies

Reagent/Category	Function in Targeting	Example Applications
Nuclear Localization Signals (NLS)	Enables nuclear import in human cells [4]	Critical for mammalian CAST engineering
Codon-Optimized Genes	Enhances protein expression in heterologous systems [14]	Improved editing efficiency in human cells
Host Factors (ClpX, S15)	Increases integration activity [14] [4]	Overcoming human cell bottleneck
Engineered sgRNAs	Optimized guide designs with conserved structural motifs [4]	Maintaining function with reduced size
Terminal Inverted Repeats (TIR)	Defines transposon boundaries [4]	Can be reduced by 50% without losing activity
Metagenomic CAST Libraries	Source of novel natural variants [4]	Identification of systems with improved properties

The fundamental differences in RNA-guided DNA recognition and R-loop formation between Type I-F and Type V-K CAST systems present researchers with complementary tools for genome engineering applications. Type I-F systems offer sophisticated multi-layer regulation through their complex Cascade architecture, providing superior control over R-loop formation and enhanced specificity at the cost of delivery complexity. In contrast, Type V-K systems provide a streamlined targeting approach with simpler delivery requirements, making them particularly amenable to therapeutic applications where packaging constraints are paramount.

Recent engineering breakthroughs, including evoCAST for Type I-F systems [14] and metagenomically-discovered Type V-K variants [4], have dramatically improved editing efficiencies in human cells. The choice between these systems ultimately depends on application-specific requirements: Type I-F systems may be preferable for applications demanding maximal specificity and unidirectional integration, while Type V-K systems offer advantages for therapeutic cargo integration where simpler architecture facilitates delivery. As structural insights continue to guide engineering efforts, both platforms are poised to expand the therapeutic frontier of genome editing.

CRISPR-associated transposons (CASTs) represent a revolutionary breakthrough in genome engineering, combining the programmability of CRISPR systems with the DNA integration capabilities of bacterial Tn7-like transposons [15]. Unlike conventional CRISPR-Cas tools that create double-strand breaks (DSBs), CASTs enable DSB-free integration of large DNA cargos, offering a promising solution to the challenges of precision gene insertion [16] [17]. The integration module of these systems is governed by a sophisticated protein machinery centered on TnsA, TnsB, TnsC, and TniQ, which work in concert to ensure specific and efficient transposition [18]. Understanding the distinct roles of these components is crucial for appreciating the functional differences between major CAST subtypes, particularly Type I-F and Type V-K systems, which exhibit significant variations in their protein composition, editing efficiency, and cargo capacity [9] [19]. This comparative analysis examines the structural and functional characteristics of these core components, providing researchers with experimental insights and methodological frameworks for leveraging CAST systems in therapeutic and synthetic biology applications.

Protein Components and Their Functional Roles

The integration module of CAST systems comprises highly specialized proteins that coordinate transposon excision, target site selection, and DNA integration through a series of precisely regulated protein-protein and protein-DNA interactions.

Table 1: Core Components of the CAST Integration Module

Protein	Primary Function	Structural Features	Key Interactions
TnsA	5' end cleavage during transposon excision [20]	Endonuclease-like fold; forms heterotetramer with TnsC (TnsA₂C₂) [20]	Interacts with TnsB and TnsC; positioned at transposon ends by TnsB [20]
TnsB	DDE transposase catalyzing 3' end cleavage and strand transfer [19]	RNase H fold catalytic domain; NTD1/2 helical domains for DNA recognition [19]	Binds transposon ends; interacts directly with TnsC and TnsA [20] [18]
TnsC	ATP-dependent regulator of transposition [20]	AAA+ ATPase motifs; forms hexameric rings on DNA [20] [18]	Interacts with TnsB, TnsA, and TniQ/TnsD; activated by target selection complex [20]
TniQ	Adaptive target selector bridging CRISPR complex to transposition machinery [18]	Conserved zinc-binding TniQ domain; dimerizes in Type I-F systems [9] [18]	Integrates with Cascade effector; recruits TnsC to target DNA [9] [18]

TnsA: The 5' End Processing Enzyme

TnsA functions as a specialized nuclease responsible for cleaving the 5' ends of the transposon during excision, working coordinately with TnsB to completely liberate the transposon from its donor site [20]. In the well-characterized bacterial Tn7 system, TnsA and TnsB form a heteromeric transposase where both proteins are interdependent—catalytically inactive mutants of either protein abolish all breakage and joining activities, even when the other component remains functional [20]. Structural analyses reveal that TnsA interacts directly with TnsC, forming a TnsA₂C₂ heterotetramer that positions the excision machinery near the transposon ends [20]. The C-terminal region of TnsC (residues 504-555) is particularly important for this interaction, with the lysine-rich TnsC495-501 region potentially facilitating contacts with donor DNA near the transposon end [20]. Notably, Type V-K CAST systems naturally lack TnsA, which fundamentally alters their transposition mechanism and leads to the formation of cointegrate structures that require host-mediated resolution [19].

TnsB: The Core Transposase Engine

TnsB represents the catalytic heart of the transposition process, a DDE transposase belonging to the retroviral integrase superfamily that catalyzes the DNA breakage and joining reactions essential for transposon integration [19]. The protein exhibits sequence-specific DNA-binding activity, recognizing and binding to multiple sites at both ends of the transposon [20] [19]. Cryo-EM structures of the Scytonema hofmannii TnsB in complex with DNA reveal an intertwined pseudo-symmetrical architecture where four protomers assemble around the transposon ends, with two catalytically competent subunits positioned for strand transfer and two structural subunits maintaining complex integrity [19]. The N-terminal NTD1/2 helical domains mediate transposon end recognition, while a unique in trans association between domains reinforces the assembly [19]. Beyond its catalytic functions, TnsB plays a crucial regulatory role through its direct interaction with TnsC, with mutations in the C-terminal region of TnsB (particularly P686S, V689M, and P690L) resulting in reduced effectiveness of transposition immunity [20].

TnsC: The ATP-Dependent Transposition Regulator

TnsC serves as the central regulatory ATPase that coordinates the assembly of the transposition machinery and communicates between the target selection complex and the transposase [20] [18]. As a member of the AAA+ ATPase family, TnsC exhibits ATP-dependent DNA binding and ATPase activity that are not required for the chemical steps of transposition but rather regulate the assembly of functional transpososomes [20]. The protein forms hexameric rings on target DNA, creating a platform for recruiting the TnsAB transposase [18]. Structural studies of the Peltigera membranacea cyanobiont CAST system reveal that TnsC interacts directly with both TnsB and the target selector TnsD/TniQ, positioning it as the critical bridge between target recognition and DNA integration [18]. The C-terminal tail of TnsC plays particularly important roles in both transposase recruitment and the mechanism of target immunity, which prevents multiple insertions into the same DNA molecule [20] [18]. ATP hydrolysis by TnsC enables its dissociation from target DNA, providing a clearance mechanism that underlies this immunity phenomenon [20].

TniQ: The Adaptive Target Selector

TniQ functions as the molecular bridge that connects the CRISPR-guided target recognition complex to the transposition machinery, replacing the sequence-specific DNA binding protein TnsD found in non-CRISPR Tn7 systems [18]. While TnsD recognizes specific att sites through helix-turn-helix motifs, TniQ depends on the CRISPR effector complex (Cascade or Cas12k) for target localization [18]. Structural analyses of Type I-F systems reveal that TniQ forms a stable homodimer that associates with the Cas6 and Cas7.6 subunits at the crRNA 3' end of the Cascade complex [9]. Cryo-EM studies of the PseCAST QCascade complex demonstrate significant flexibility in the TniQ dimer, which samples a range of positions relative to the rest of the complex, suggesting dynamic interactions with TnsC during transpososome assembly [9]. In Type V-K systems, which lack TnsA, TniQ associates with Cas12k and the bacterial ribosomal protein uS15 to form a simplified targeting module [19].

Comparative Analysis: Type I-F vs. Type V-K CAST Systems

The architectural differences between Type I-F and Type V-K CAST systems significantly impact their experimental performance, cargo capacity, and suitability for different genome engineering applications.

Table 2: Performance Comparison of CAST Subtypes in Genome Engineering

Parameter	Type I-F CAST	Type V-K CAST
System Complexity	Multi-subunit Cascade (Cas6/7/8) + TniQ dimer [9]	Single-protein Cas12k + TniQ [17]
Transposase Composition	TnsA + TnsB + TnsC (cut-and-paste) [20]	TnsB + TnsC (copy-and-paste, cointegrate formation) [19]
Coding Size	~8 kb [9]	~5 kb [9]
Editing Efficiency in Human Cells	10-25% (evoCAST evolved variant) [14]	Low efficiency in heterologous contexts [9]
Product Purity	High specificity, homogeneous unidirectional products [14]	Reduced specificity, heterogeneous byproducts [9]
Cargo Capacity	Multi-kilobase inserts (demonstrated >1 kb) [14]	Large cargo capability (10-30 kb) [19]

Structural and Mechanistic Divergence

Type I-F and Type V-K CAST systems employ fundamentally different architectural strategies for target recognition and DNA integration. Type I-F systems utilize a multi-subunit Cascade complex comprising Cas6, Cas7, and Cas8 proteins that assemble with a crRNA molecule to form an extended structure that surveys DNA for complementary target sequences [9]. This complex associates with a TniQ homodimer that recruits TnsC to the target site [9]. In contrast, Type V-K systems rely on a single Cas12k protein complexed with TniQ and the bacterial host factor uS15 for target recognition, creating a more compact but functionally limited targeting module [19]. The integration modules also differ substantially, with Type I-F systems employing the complete TnsABC transposase that mediates clean cut-and-paste transposition, while Type V-K systems naturally lack TnsA, resulting in cointegrate formation that requires resolution by host recombination machinery [19].

Editing Efficiency and Product Purity

Recent engineering efforts have dramatically improved the performance of CAST systems in human cells, with Type I-F systems demonstrating particularly promising advancements. The development of evoCAST through phage-assisted continuous evolution (PACE) generated TnsABC variants with approximately 200-fold improved integration activity in human cells compared to wild-type systems [14]. These evolved systems achieve 10-25% integration efficiencies with kilobase-sized DNA cargos across multiple genomic loci while generating predominantly unidirectional transposition products without detectable indel formation [14]. In contrast, Type V-K systems exhibit multiple undesirable biochemical properties in heterologous cellular contexts, including reduced specificity, low overall editing efficiencies, and poor product purity [9]. The enhanced performance of engineered Type I-F systems positions them as particularly promising platforms for therapeutic applications requiring precise, DSB-free integration of large DNA sequences.

Experimental Approaches and Methodologies

Structural Characterization Techniques

The molecular understanding of CAST integration modules has been revolutionized by advances in structural biology, particularly cryo-electron microscopy (cryo-EM).

Diagram 1: Cryo-EM Workflow for CAST Complex Structure Determination. This generalized workflow illustrates the key steps in determining high-resolution structures of CAST integration complexes, from sample preparation to model building.

Structural studies of CAST components typically begin with recombinant protein expression in E. coli, followed by multi-step purification using affinity and size-exclusion chromatography [19] [18]. For the TnsB transposase, DNA binding properties are often characterized using electrophoretic mobility shift assays (EMSAs) with oligonucleotides containing terminal repeats from transposon ends [19]. To capture specific functional states, researchers design oligonucleotide substrates that mimic intermediate stages of transposition, such as the strand transfer complex (STC) that represents the post-catalysis integration state [19]. For analyzing larger assemblies like the complete TnsABCD transpososome, biochemical reconstitution with purified components enables visualization of the intact machinery [18].

Functional Assays for Transposition Activity

Functional characterization of CAST integration modules employs both bacterial and mammalian cell-based assays to quantify transposition efficiency and specificity.

Table 3: Key Research Reagents for CAST Integration Studies

Reagent/Solution	Composition	Experimental Function
Reconstituted TnsABCD Transpososome	TnsA, TnsB, TnsC, TnsD, att site DNA [18]	Structural and biochemical analysis of complete integration machinery
Strand Transfer Complex (STC)	TnsB transposase + DNA oligonucleotides with transposon ends [19]	Capture post-catalysis integration state for structural studies
QCascade Complex	Cas8:Cas7:Cas6:TniQ:crRNA (1:6:1:2:1 stoichiometry) [9]	Target recognition module for Type I-F CAST systems
Phage-Assisted Continuous Evolution (PACE)	E. coli host cells, selection phage, accessory plasmid [14]	Directed evolution of transposase variants with enhanced activity
Transposon Donor Plasmid	Plasmid containing transposon with terminal repeats and cargo DNA [14]	Substrate for assessing integration efficiency and cargo capacity

In bacterial systems, transposition efficiency is typically measured using selection-based assays where successful integration of a transposon-encoded marker gene (e.g., antibiotic resistance) into a target plasmid or chromosome confers a selectable phenotype [14]. The development of PACE (phage-assisted continuous evolution) has enabled rapid optimization of CAST components by linking transposition activity to phage propagation through a selection circuit where targeted insertion of a transposon-encoded promoter activates expression of an essential phage gene [14]. For mammalian cell applications, integration efficiency is quantified using digital droplet PCR or next-generation sequencing to measure precise insertion of transgene cargos at designated genomic loci [14]. These functional assays have been instrumental in engineering enhanced CAST variants like evoCAST, which achieves therapeutic-level integration efficiencies in human cells [14].

Applications and Future Directions

The unique capabilities of CAST integration modules have enabled innovative applications across genome engineering, with particular promise for therapeutic development. Engineered CAST systems have successfully inserted therapeutic transgenes at clinically relevant loci, including Factor IX cDNA for hemophilia B treatment and chimeric antigen receptor (CAR) genes for cancer immunotherapy [14]. The evoCAST system demonstrates particularly robust performance, enabling 10-25% integration efficiencies of kilobase-sized DNA cargos across 14 tested human genomic sites without detectable indel formation or significant off-target activity [14]. In plant systems, transposase-assisted target-site integration (TATSI) technologies based on rice Pong transposase fused to programmable nucleases have achieved precise insertion of gene expression cassettes in Arabidopsis and soybean, outperforming conventional HDR-based approaches in both efficiency and accuracy [21].

Future optimization of CAST integration modules will likely focus on enhancing efficiency and specificity in therapeutically relevant primary cells, reducing system size for improved deliverability, and expanding targeting flexibility through engineered PAM specificities [9] [14]. The continued structural characterization of transposition intermediates, coupled with advanced engineering approaches like continuous evolution, promises to unlock the full potential of CAST systems as next-generation tools for precision genome engineering [14] [18].

Diagram 2: Research and Therapeutic Applications of CAST Integration Modules. The precise, DSB-free integration capability of CAST systems enables diverse applications across therapeutic development, basic research, and agricultural biotechnology.

The discovery and adaptation of CRISPR-associated transposase (CAST) systems represent a significant advancement in genome-editing technology, offering a unique mechanism for programmable, double-strand break-free DNA integration. Unlike conventional CRISPR-Cas systems that rely on creating double-strand breaks and exploiting host repair mechanisms, CAST systems combine the precision of RNA-guided targeting with the DNA integration capability of transposases [17]. This enables precise insertion of large DNA cargo without relying on host DNA repair pathways, making these systems particularly valuable for therapeutic applications requiring gene-sized insertions [4] [17].

Among the diverse CAST systems identified, type I-F and type V-K have emerged as leading candidates for human genome engineering applications. These systems differ fundamentally in their molecular architecture, with type I-F utilizing a multi-protein Cascade complex for DNA targeting, while type V-K employs a single Cas12k effector [4] [22]. This comparative guide examines the natural diversity, experimental performance, and therapeutic potential of these distinct CAST architectures, providing researchers with objective data to inform their experimental designs.

Phylogenetic Distribution and Metagenomic Discovery

The identification of novel CAST systems through metagenomic analysis has revealed remarkable phylogenetic diversity, particularly among type V-K systems. Recent analysis of thousands of high-quality metagenomic assemblies has identified over 70 phylogenetically diverse Cas12k effectors encoded in genomic fragments containing complete and partial type V-K CAST systems [4]. These systems were characterized by conserved features, including a conserved motif (5′-GNNGGNNTGAAAG-3′) at the 3′ end of CRISPR repeats and a conserved "CCYCC(n4-n6)GGRGG" stem-loop structure upstream from the antirepeat in the tracrRNA [4].

Table 1: Classification of CAST Systems by Type and Components

System Feature	Type I-F CAST	Type V-K CAST
Targeting Component	Multi-subunit Cascade complex (Cas8, Cas7, Cas6)	Single Cas12k effector
Transposase Components	TnsA, TnsB, TnsC	TnsB, TnsC
Targeting Complexity	High (3+ proteins)	Low (single protein)
Integration Mechanism	Cut-and-paste	Hybrid replicative
Natural Abundance	Less diverse from metagenomic data	Highly diverse (>70 Cas12k variants identified)

Notably, self-targeting spacers adjacent to pseudo CRISPR repeats were identified within a subset of these metagenomic-derived systems, suggesting functional CAST transposons [4]. From this diversity, 13 predicted complete type V-K CAST systems were selected for functional screening, with MG64-1 and MG64-6 demonstrating programmable, sgRNA-dependent integration in vitro [4]. These systems exhibited distinct protospacer adjacent motif (PAM) preferences—5′ GTN for MG64-1 and 5′ rGTN for MG64-6—with 90% of integration events occurring between 57-67 base pairs from the PAM sequence [4].

Type I-F systems show their own diversity, with distinct subtypes including I-F3a (VchCAST/Tn6677), I-F3b (AsaCAST/Tn6900), and the more distantly related PseCAST (Tn7016), which has demonstrated particular promise for human cell engineering [22]. The structural characterization of PseCAST QCascade complex has revealed novel subtype-specific interactions and RNA-DNA heteroduplex features that distinguish it from other type I-F systems [22].

Comparative Analysis of Editing Efficiency and Cargo Size

Integration Efficiency Across Host Systems

CAST systems demonstrate markedly different integration efficiencies across bacterial and human cell environments. In bacterial systems, both type I-F and type V-K CASTs can achieve high integration rates, but their performance diverges significantly in human cells.

Table 2: Editing Efficiency and Cargo Capacity of CAST Systems

Performance Metric	Type I-F CAST	Type V-K CAST	Experimental Context
Integration Efficiency in E. coli	Up to 80%	Up to 80%	Genomic loci [4]
Integration Efficiency in Human Cells	10-30% (evoCAST)	Single-digit percentages	Genomic safe harbor sites [23] [17]
Cargo Capacity	Up to 15 kb	Therapeutically relevant genes (e.g., Factor IX)	Demonstrated insertion [23] [4]
Product Purity	High homogeneity	Mixed (co-integration events)	Plasmid donor delivery [4] [22]
Multiplexing Capability	Not demonstrated	Up to 50% at secondary loci	Dual targeting in E. coli [4]

Type V-K CAST systems have shown remarkable efficiency in bacterial systems, with demonstrated integration rates of up to 80% at engineered and endogenous loci in E. coli [4]. These systems also support multiplexed integration, with simultaneous insertion at two loci achieving up to 50% efficiency at secondary target sites [4]. However, in human cells, natural type V-K systems show significantly reduced efficiency, typically in the single-digit percentages [22] [17].

Type I-F systems, particularly the engineered PseCAST variant, have demonstrated human cell editing efficiencies that reached single-digit percentages, representing approximately a 100-fold improvement over the original VchCAST candidate [22]. Further engineering of type I-F systems has yielded even more substantial improvements, with the laboratory-evolved evoCAST system achieving integration efficiencies of 10-30% in human cells [23] [17].

Product Purity and Specificity

The purity of integration products represents a significant differentiator between CAST systems. Type V-K CAST systems, which lack TnsA for second-strand donor cleavage, typically produce a mixture of integration events when using circular plasmid donors [4]. Approximately 20-30% of integrations represent simple transposon insertion, while 70-80% are co-integration events containing two copies of the cargo along with plasmid backbone sequences [4].

In contrast, type I-F CAST systems containing TnsA enable cut-and-paste integration, resulting in highly specific and homogeneous integration products [22] [4]. These systems demonstrate markedly fewer off-target events, with one study reporting fewer than 7% off-target integrations across all conditions in multiplexed experiments [4].

Recent advances in screening technology have enabled comprehensive characterization of CAST specificity. Researchers at St. Jude Children's Research Hospital developed a high-throughput method to simultaneously measure the activity and specificity of thousands of CAST variants [24]. This approach identified specific mutations that improved both specificity and activity without compromise, with combined mutations increasing activity fivefold [24].

Molecular Mechanisms and Experimental Workflows

Mechanism of Type V-K CAST Integration

The type V-K CAST system utilizes a relatively simple architecture centered on the Cas12k effector. The following diagram illustrates the key components and their interactions in this system:

(Type V-K CAST Molecular Mechanism)

The type V-K CAST system functions through a coordinated mechanism wherein the Cas12k protein, guided by RNA and assisted by the S15 host factor, identifies target DNA sequences bearing a compatible PAM (GTN or rGTN) [4] [22]. The Cas12k effector complex, including TniQ, then recruits the transposition machinery through TnsC, leading to TnsB-mediated integration of donor DNA approximately 57-67 base pairs downstream of the PAM sequence [4].

Mechanism of Type I-F CAST Integration

Type I-F CAST systems employ a more complex multi-protein approach for targeted DNA integration, as illustrated below:

(Type I-F CAST Molecular Mechanism)

In type I-F systems, the multi-subunit Cascade complex (comprising Cas8, Cas7, and Cas6 proteins) identifies target DNA sequences through gRNA complementarity [22]. The TniQ homodimer, stably associated with Cascade, recruits TnsC to the target site [22]. TnsC then orchestrates the recruitment of TnsA and TnsB transposases, which catalyze cut-and-paste integration of donor DNA, resulting in more homogeneous products compared to type V-K systems [22].

High-Throughput CAST Screening Workflow

Recent advances in CAST engineering have been accelerated by the development of sophisticated screening methodologies:

(High-Throughput CAST Screening Workflow)

This screening approach enables comprehensive profiling of CAST activity and specificity, allowing researchers to systematically evaluate thousands of CAST variants in parallel [24]. The method involves generating CAST mutant libraries, delivering them to host cells, selecting for successful integration events, and using next-generation sequencing to quantitatively assess both on-target efficiency and off-target effects [24]. This workflow has enabled identification of specific mutations that enhance both activity and specificity, facilitating the engineering of improved CAST systems for therapeutic applications [24].

Research Reagent Solutions for CAST Engineering

The following table outlines essential research reagents and their applications in CAST system engineering and evaluation:

Table 3: Essential Research Reagents for CAST System Engineering

Research Reagent	Function	Application Context
Metagenomic DNA Libraries	Source of novel CAST system diversity	Identification of phylogenetically diverse Cas effectors [4]
Nuclear Localization Signal (NLS) Tags	Directs prokaryotic proteins to mammalian nucleus	Engineering CAST function in human cells [4]
Single-Guide RNA (sgRNA)	Programs DNA targeting specificity	Defining genomic integration sites [4] [22]
Bacterial Chaperone Proteins (e.g., ClpX)	Enhances proper protein folding	Improving CAST activity in human cells [4]
Host Factors (e.g., S15)	Supports complex assembly	Enabling episomal integration in human cells [4] [22]
Linear Donor DNA Templates	Provides cargo for integration	Reduces co-integration events in type V-K systems [4]
AAV Safe Harbor Targeting Vectors	Enables therapeutic gene integration	Testing CAST-mediated gene insertion at genomic safe harbor sites [4] [17]
Lipid Nanoparticles (LNPs)	Delivery vehicle for CAST components	In vivo delivery of CAST machinery [25]

Discussion and Future Perspectives

The comparative analysis of type I-F and type V-K CAST systems reveals a fundamental trade-off between simplicity and precision in genome editing applications. Type V-K systems offer a compact architecture with single-protein targeting that facilitates delivery, particularly in therapeutic contexts where vector capacity is limited [17]. However, this simplicity comes at the cost of product heterogeneity due to co-integration events and generally lower efficiency in human cells [4] [22].

Conversely, type I-F systems provide superior product purity through their cut-and-paste integration mechanism and demonstrate higher editing efficiencies in human cells following engineering [22] [23]. The structural insights gained from cryoEM analyses of PseCAST QCascade have enabled rational engineering approaches to further enhance DNA binding and integration efficiency [22].

Future directions in CAST system development will likely focus on combining advantageous features from both systems through chimeric engineering, enhancing delivery efficiency through improved viral and non-viral vectors, and expanding the targeting scope through PAM engineering [22] [25]. The continued application of high-throughput screening methodologies will accelerate this optimization process, enabling systematic evaluation of CAST variants to identify mutations that simultaneously improve activity, specificity, and compatibility with human cellular environments [24].

As CAST systems continue to evolve, they hold particular promise for therapeutic applications requiring large DNA insertions, such as the integration of full-length therapeutic genes for monogenic disorders [4] [17]. With clinical development already underway, including Metagenomi's planned first-in-human studies for 2026, CAST systems are poised to complement existing genome-editing technologies and potentially address limitations of conventional CRISPR-Cas systems in therapeutic contexts [17].

From Bacteria to Human Cells: Engineering and Applying CAST Systems for Therapeutic Integration

The transition of CRISPR-associated transposase (CAST) systems from prokaryotic origins to efficient function in human cells represents a fundamental challenge in genome editing. These systems, which enable RNA-guided integration of large DNA cargo without creating double-strand breaks, must overcome the physical barrier of the nuclear envelope to access chromosomal DNA. This comparison guide examines the strategic approaches developed for Type I-F and Type V-K CAST systems to achieve nuclear localization and therapeutic levels of editing efficiency in human cells. The fundamental architectural differences between these systems—Type I-F employs a multi-subunit Cascade complex for DNA targeting, while Type V-K utilizes a single Cas12k effector—necessitate distinct engineering solutions for nuclear entry and function. Understanding how these divergent strategies impact final editing outcomes provides critical insights for researchers selecting appropriate CAST platforms for therapeutic development.

Table 1: Fundamental Characteristics of CAST Systems for Human Cell Engineering

Characteristic	Type I-F CAST	Type V-K CAST
Targeting Complex	Multi-subunit Cascade (Cas8, Cas7, Cas6, TniQ)	Single Cas12k effector with TniQ
Transposase Components	TnsA, TnsB, TnsC	TnsB, TnsC
Integration Mechanism	Cut-and-paste	Mixture of simple and co-integration events
Coding Size	~8 kb	~5 kb
Natural Product Purity	High, predominantly unidirectional	Lower, mixed integration events

Quantitative Performance Comparison in Human Cells

Recent advances in CAST engineering have yielded substantial improvements in human cell editing efficiency. For Type I-F systems, the development of an evolved CAST (evoCAST) through phage-assisted continuous evolution (PACE) has demonstrated integration efficiencies of 10-25% for kilobase-sized DNA cargos across 14 tested genomic loci in HEK293T cells. This represents an ~200-fold improvement over the wild-type PseCAST system, which initially showed less than 0.1% efficiency. The evolved system maintains favorable properties including undetectable genomic indels, predominately unidirectional integration, and low off-target activity [14].

For Type V-K systems, engineering for nuclear localization and function has enabled integration of therapeutically relevant transgenes at safe-harbor sites in multiple human cell types. While specific efficiency percentages are not provided in the available literature, these compact systems demonstrate significantly fewer off-target events that are reproducibly found in specific genomic regions, highlighting their precision despite challenges with product purity [4].

Table 2: Experimental Performance Metrics in Human Cells

Performance Metric	Type I-F CAST (evoCAST)	Type V-K CAST (Engineered)
Integration Efficiency	10-25% (kb-sized cargo)	Not quantitatively specified
Improvement Over Wild-type	~200-fold	Not specified
Indel Formation	Undetectable levels	Not specified
Off-target Integration	Low levels	Rare, localized to specific regions
Product Purity	High, predominantly unidirectional	Mixed simple and co-integration events
Therapeutic Demonstration	Factor IX cDNA in ALB intron 1; CAR in TRAC	Factor IX at safe-harbor locus

Nuclear Localization Strategies and System Engineering

Nuclear Localization Signal Implementation

Achieving efficient nuclear import represents the first critical step for CAST function in human cells. Both systems require the addition of nuclear localization signals (NLS) to their protein components, though the implementation strategies differ:

• Terminal vs. Internal NLS: Conventional NLS fusion at protein termini has been widely adopted for CAST systems, similar to other CRISPR tools. However, recent advances with Cas9 systems demonstrate that hairpin internal NLS sequences (hiNLS) installed at rationally selected sites within the protein backbone can improve editing efficiency in primary human lymphocytes while maintaining high protein yield and purity [26].

• NLS Optimization Findings: Research on Cas12a systems reveals that nuclear localization levels don't always directly correlate with genome editing efficiencies, particularly contrasting in vitro versus in vivo performance. While adding multiple NLSs significantly enhanced nuclear localization in cultured cells and tissues, the optimized NLS modification for maximum editing efficiency differed between cell culture and mouse liver models [27].

System-Specific Engineering Approaches

Type I-F CAST Engineering: The PseCAST system has been engineered through both evolution and structure-guided approaches. PACE of the transposase module (TnsABC) involved hundreds of generations of mutation, selection, and replication in E. coli, with selection linking transposition activity to bacteriophage propagation [14]. Additionally, structure-guided engineering of the DNA-targeting QCascade complex, informed by cryoEM structures, has identified variants with increased integration efficiencies and modified PAM specificities [9].

Type V-K CAST Engineering: The compact Type V-K systems from metagenomic sources have been engineered for nuclear localization and human cell function through NLS tagging and optimization of the single Cas12k effector. Their simpler composition—requiring only Cas12k rather than multiple Cascade subunits—reduces the number of components requiring nuclear import, potentially simplifying the engineering process [4].

Experimental Protocols for Key Studies

Phage-Assisted Continuous Evolution for Type I-F CAST

The PACE protocol that generated hyperactive CAST variants involved:

• Selection Phage Design: TnsA, TnsB, and TnsC genes were encoded on the selection phage in place of the essential gIII gene.
• Host Cell Configuration: Host E. coli contained complementary plasmids expressing the PseCAST QCascade targeting components and a transposon-encoded promoter sequence.
• Selection Mechanism: Successful transposition inserted the promoter upstream of a promoter-less gIII on an accessory plasmid, activating gIII expression and enabling phage propagation.
• Evolution Parameters: SP populations were mutagenized via an inducible mutagenesis plasmid and diluted with fresh cells continuously in fixed-volume lagoons for hundreds of generations.
• Stringency Tuning: CP2 constructs with a range of weaker promoter strengths required increasing numbers of integration events to trigger sufficient gIII expression [14].

Type V-K CAST Engineering and Validation

The experimental workflow for developing functional Type V-K CAST in human cells included:

• Metagenomic Discovery: Analysis of thousands of metagenomic assemblies to identify diverse Cas12k effectors in genomic contexts with transposon machinery.
• System Validation: In vitro integration assays using expressed CAST proteins, guide RNA, linear donor fragment, and target plasmid libraries to confirm programmable, sgRNA-dependent activity.
• PAM Determination: Next-generation sequencing of integration products to identify enriched recognition motifs (5'-GTN PAM for MG64-1; 5'-rGTN PAM for MG64-6).
• Human Cell Engineering: NLS tagging for nuclear localization and optimization for function in human cells, demonstrating integration of therapeutic Factor IX gene at safe-harbor loci [4].

Visualization of CAST Nuclear Localization Pathways

The following diagram illustrates the nuclear localization challenges and engineering solutions for CAST systems in human cells:

The Scientist's Toolkit: Essential Research Reagents

Table 3: Key Research Reagents for CAST Engineering Studies

Reagent / Solution	Function in CAST Research	Example Application
Nuclear Localization Signals (NLS)	Facilitate nuclear import of CAST proteins	Tagging Cas12k or Cascade components
Phage-Assisted Continuous Evolution (PACE)	Accelerated protein evolution platform	Evolving TnsABC transposase with ~200-fold improved activity
CryoEM Structural Analysis	Determine high-resolution complex structures	Guiding PAM-interacting domain engineering
Metagenomic CAST Libraries	Source of novel natural CAST variants	Identifying diverse Type V-K systems from uncultivated microbes
Golden Gate Assembly Systems	Modular cloning of CAST components	Building UltraCAST vectors for bacterial editing
Single-Guide RNA Designs	Programmable targeting of CAST systems	Optimizing truncated sgRNAs for improved performance

The comparative analysis of Type I-F and Type V-K CAST systems reveals distinct strategic advantages for different research applications. Type I-F systems, particularly the evolved evoCAST platform, currently demonstrate superior editing efficiencies and product purity in human cells, making them favorable for therapeutic development where reliability and predictability are paramount. The multi-component complexity presents delivery challenges but offers more engineering handles for optimization. Conversely, Type V-K systems provide a compact architecture with simpler nuclear localization requirements and demonstrate rare, predictable off-target patterns, potentially advantageous for applications where vector size constraints exist. As both systems continue to evolve through protein engineering and structural insights, the strategic selection between these platforms will depend on specific application requirements including cargo size, target cell type, delivery method, and precision needs.

The targeted insertion of large DNA cargos into the human genome is a cornerstone of advanced gene therapy and functional genomic research. While CRISPR-Cas systems revolutionized the editing of small sequences, efficient, targeted integration of kilobase-sized therapeutic genes has remained a formidable challenge. CRISPR-associated transposases (CASTs) emerged as promising solutions, yet their initial low activity in human cells limited therapeutic application. This review examines how Phage-Assisted Continuous Evolution (PACE) has overcome these limitations by generating hyperactive evolved CAST (evoCAST) systems. We objectively compare the performance of these novel systems against traditional alternatives, with a specific focus on the structural and functional distinctions between Type I-F and Type V-K CAST systems that underpin their divergent editing efficiencies and cargo capacities.

CAST Systems: Type I-F vs. Type V-K

CAST systems are natural bacterial systems that use RNA-guided, nuclease-deficient CRISPR-Cas systems to direct site-specific insertion of kilobase-scale transposons by Tn7-like transposases. The two most extensively characterized subtypes for genome editing applications are Type I-F and Type V-K, which differ significantly in their architecture and performance.

Table 1: Comparison of CAST System Subtypes

Feature	Type I-F CAST	Type V-K CAST
DNA Targeting Module	Multi-subunit QCascade complex (Cas8, Cas7, Cas6, TniQ, crRNA) [9]	Simpler Cas12k-TniQ complex [9]
Integration Module	TnsA, TnsB, TnsC (TnsABC) [14]	TnsB, TnsC (TnsBC) [9]
Coding Size	~8 kb [9]	~5 kb [9]
Product Purity & Specificity	High specificity and homogeneous integration products [9]	Reduced specificity, lower product purity [9]
Editing Efficiency in Human Cells	10-25% (evoCAST) [14]	Typically <~0.1% [9]
Key Representative	PseCAST (from Tn7016 transposon), evoCAST [14] [9]	ShCAST (from Scytonema hoffmannii) [12]

The structural divergence between these systems has direct functional consequences. The multi-subunit QCascade complex of Type I-F systems like PseCAST contributes to their high specificity and homogeneous integration products [9]. In contrast, the more compact Type V-K systems, while advantageous for delivery, exhibit reduced specificity and lower product purity, limiting their therapeutic utility [9].

The PACE Breakthrough: Evolving Hyperactive evoCAST

Phage-Assisted Continuous Evolution (PACE) is a powerful directed evolution technology that maps Darwinian evolution onto the life cycle of the M13 bacteriophage within a fixed-volume vessel called a "lagoon" [28]. This system enables hundreds of generations of mutation, selection, and replication to occur in just days, dramatically accelerating the improvement of protein function with minimal researcher intervention [29] [28].

To overcome the bottleneck of low transposase activity in human cells, researchers developed a specialized PACE selection that linked CAST-mediated integration directly to phage propagation [14]. The selection required targeted insertion of a transposon-encoded promoter sequence upstream of a promoter-less gene III (gIII), an essential gene for phage replication. Successful transposition activated gIII expression, enabling propagation of the selection phage (SP) encoding the transposase variant [14].

After performing hundreds of rounds of evolution, researchers identified transposase variants (TnsABC) from the PseCAST system with an average ~200-fold improved integration activity in human cells compared to wild-type [14]. These evolved variants were combined with structure-guided engineering of the DNA-targeting QCascade module to create an optimized, evolved CAST system dubbed evoCAST.

Performance Comparison: evoCAST vs. Alternative Systems

The development of evoCAST represents a significant milestone in large DNA insertion technology. The table below provides a quantitative comparison of its performance against other contemporary genome editing systems.

Table 2: Performance Comparison of Genome Editing Systems for Large DNA Integration

System	Integration Efficiency	Cargo Size Capacity	Key Advantages	Key Limitations
evoCAST (Type I-F)	~10-25% across 14 tested loci [14]	Multi-kilobase [9]	DSB-free; high product purity; low indels [14]	Large coding size (~8 kb) [9]
Wild-type PseCAST	<~0.1% (up to ~1% with ClpX) [14] [9]	Multi-kilobase [9]	DSB-free; high specificity [9]	Very low efficiency in human cells [14]
Type V-K CAST (ShCAST)	Minimal activity [9]	Multi-kilobase [9]	Compact system (~5 kb) [9]	Low efficiency; poor product purity [9]
HDR with DSB	Highly variable; decreases with cargo size [30]	Theoretically large, but efficiency drops [30]	Established method	Requires dividing cells; induces DSBs [30]
Prime Editing (PE)	High for small edits [30]	<~100-200 bp [14]	Precise; minimal DSBs [30]	Limited cargo capacity [14]
PASSIGE	High [14]	Large [14]	High efficiency [14]	Multiple enzymatic steps [14]

DSB-Free Editing and Product Purity

A critical advantage of evoCAST over traditional nuclease-dependent methods (HDR) is its ability to operate without creating double-strand breaks (DSBs) [14]. DSBs can lead to uncontrolled formation of indels, large deletions, chromosomal rearrangements, and p53 activation [14]. evoCAST generates predominately unidirectional cut-and-paste transposition products and does not induce detected indels at the target site [14]. Furthermore, while HDR efficiency drops significantly for larger cargos and is inefficient in non-dividing cells, CAST systems maintain their activity across cell types [30].

Comparison with Prime Editing and PASSIGE

Prime editing can efficiently install sequences up to ~100-200 bp but cannot currently install gene-sized sequences (≥1 kb) [14]. The PASSIGE (Prime Editing Assisted Site-Specific Integrase Gene Editing) system combines prime editing with site-specific recombinases to enable efficient targeted installation of large cargos [14]. However, PASSIGE requires coordinated prime editing and recombinase systems to catalyze multiple successive enzymatic steps, some of which can generate undesired byproducts [14]. In contrast, evoCAST achieves targeted insertion in a single enzymatic step, simplifying the editing process [14].

Experimental Protocols and Validation

PACE Evolution of CAST Systems

The PACE experiment for evolving CAST systems utilized host E. coli cells containing three plasmid components [14]:

• Selection Phage (SP): M13 phage genome with gIII removed and replaced with TnsABC genes (the evolving transposase)
• Accessory Plasmid (AP): Contains promoter-less gIII essential for phage replication
• Complementary Plasmids (CP1 & CP2): CP1 expresses QCascade for DNA targeting; CP2 provides transposon with promoter for integration

The flow rate in the lagoon was set such that dilution was faster than E. coli reproduction but slower than phage replication, creating selective pressure for phages encoding transposases with enhanced integration activity [28]. Over hundreds of generations, this setup enabled the accumulation of beneficial mutations in the TnsABC genes [14].

Validation in Human Cells

Evolved CAST variants were validated in HEK293T cells using a reporter assay that measured precise integration of a donor plasmid containing a cargo gene [14]. The top evoCAST variant supported ~10-25% integration efficiencies of kilobase-sized DNA cargos across 14 tested genomic loci in HEK293T cells without requiring the bacterial unfoldase ClpX [14]. This represented a substantial improvement over wild-type PseCAST, which showed <~0.1% efficiency in human cells without ClpX supplementation [14].

Further validation demonstrated evoCAST's therapeutic relevance through several key applications [14]:

• Installation of human factor IX cDNA into ALB intron 1
• Insertion of a CD19-targeted chimeric antigen receptor into TRAC
• Integration of wild-type cDNAs of four genes implicated in loss-of-function genetic diseases into intron 1 of their respective endogenous loci

Essential Research Reagents and Tools

The development and application of evoCAST requires several key reagents and methodologies that constitute the core toolkit for researchers in this field.

Table 3: Essential Research Reagent Solutions for CAST System Engineering

Reagent/Resource	Function/Description	Key Features
PACE System	Continuous directed evolution platform [14] [28]	Enables hundreds of rounds of evolution in days; minimal researcher intervention [29]
PseCAST System	Type I-F CAST from Pseudoalteromonas sp. Tn7016 [14] [9]	Parent system for evolution; demonstrated superior activity in human cells vs. other CASTs [9]
CryoEM Structural Data	High-resolution structure determination [9]	Enabled structure-guided engineering of QCascade DNA binding module [9]
Error-Prone Mutagenesis Plasmid (MP)	Introduces genetic variation during PACE [28]	Provides mutational diversity for evolution without manual intervention [28]
QCascade Engineering	Structure-guided optimization of DNA targeting [9]	Improved DNA binding efficiency; modified PAM stringencies [9]

The application of PACE to CAST system evolution represents a transformative advance in genome engineering. The resulting evoCAST system achieves therapeutic-level efficiencies of 10-25% for kilobase-sized cargo integration across multiple genomic loci, outperforming previous CAST systems and offering distinct advantages over nuclease-dependent approaches. While Type V-K systems benefit from compact architecture, Type I-F systems, particularly evolved variants like evoCAST, demonstrate superior editing efficiency, product purity, and specificity. The continued integration of structural insights, library screening, and directed evolution promises to further enhance these powerful tools, potentially enabling new therapeutic paradigms for addressing loss-of-function genetic diseases through one-time, mutation-agnostic gene integration.

The integration of large transgenes, such as those encoding Factor IX (FIX) or Chimeric Antigen Receptors (CARs), represents a formidable challenge in therapeutic genome editing. Conventional tools like CRISPR-Cas9 rely on DNA double-strand breaks (DSBs) and host repair mechanisms, which are inefficient for multi-kilobase insertions and often result in a heterogeneous mixture of undesirable outcomes, including indel mutations and chromosomal rearrangements [9]. CRISPR-associated transposases (CASTs) have emerged as a next-generation solution, enabling DSB-free, RNA-guided integration of large genetic payloads with high specificity and product homogeneity [9]. Two major CAST systems, type I-F and type V-K, are at the forefront of this technological revolution, each with distinct advantages and limitations for therapeutic workflow development. This guide provides a detailed, step-by-step comparison of these systems, focusing on their application in integrating therapeutically relevant transgenes like FIX, and includes supporting experimental data and protocols to inform their use in research and drug development.

System Comparison: Type I-F vs. Type V-K CASTs

The choice between type I-F and type V-K CAST systems is fundamental to experimental design. The table below summarizes their core characteristics based on current research.

Table 1: Key Characteristics of Type I-F and Type V-K CAST Systems

Feature	Type I-F CAST (e.g., PseCAST, VchCAST)	Type V-K CAST (e.g., ShCAST)
CRISPR Effector	Multi-subunit Cascade complex (Cas8, Cas7, Cas6, TniQ) [9]	Single-protein Cas12k [5]
Transposase Proteins	TnsA, TnsB, TnsC [9]	TnsB, TnsC, TniQ [5]
System Size	Larger, more complex (~8 kb coding size) [9]	More compact (~5 kb coding size) [9]
Integration Mechanism	"Cut-and-paste" (TnsA-dependent second-strand cleavage) [5]	"Copy-and-paste" (TnsA-independent) [5]
Editing Efficiency in Human Cells	Single-digit efficiencies, demonstrated in human cells [9] [5]	Low but detectable activity on plasmid targets; lower genomic efficiency in human cells [5]
Integration Specificity (Fidelity)	Highly specific, homogeneous integration products [9] [5]	Prone to RNA-independent "untargeted" transposition; lower fidelity [5]
Cargo Size Capacity	Multi-kilobase insertions demonstrated [9]	Multi-kilobase insertions demonstrated [9]

Analysis of System Selection

The data in Table 1 indicates a critical trade-off. Type I-F systems (e.g., PseCAST) are preferable for applications demanding high specificity, as they exhibit predominantly on-target integration and produce homogeneous products [9] [5]. Their proven, albeit modest, activity in human cells makes them a leading candidate for therapeutic development [9]. In contrast, Type V-K systems (e.g., ShCAST) offer the advantage of a compact coding sequence, which is beneficial for delivery via size-limited viral vectors like adeno-associated virus (AAV) [9]. However, their significant drawback is a propensity for RNA-independent, "untargeted" transposition, driven by the spontaneous formation of TnsC filaments on AT-rich DNA regions, which can lead to a high rate of off-target integration [5]. A key engineering strategy to improve ShCAST fidelity involves modulating cytoplasmic TnsC levels to suppress this pathway, which has been shown to increase on-target specificity up to 98.1% in E. coli without compromising on-target efficiency [5].

Experimental Protocols for CAST-Based Integration

The following section outlines a general workflow for deploying CAST systems, with notes on system-specific variations.

Protocol 1: Plasmid-Based Transposition Assay in Human Cells

This protocol is used to initially validate CAST activity and compare the efficiency of different systems or engineered variants.

Table 2: Key Reagents for Plasmid-Based Transposition Assay

Reagent / Material	Function / Description
CAST Expression Plasmid(s)	Plasmid(s) encoding all necessary CAST components (e.g., TnsA,B,C and QCascade for I-F; Cas12k, TnsB,C, TniQ for V-K) [9].
Donor Plasmid	Contains the transgene (e.g., FIX, CAR) flanked by the cognate transposon ends (e.g., left-end (LE) and right-end (RE)) recognized by TnsB [9].
Target Site Plasmid	A plasmid containing the target genomic DNA sequence with a Protospacer Adjacent Motif (PAM) compatible with the CAST system.
Human Cell Line	Typically HEK293T or other readily transfectable lines for initial testing.
Transfection Reagent	For delivery of plasmid DNA into human cells.

Step-by-Step Methodology:

• Vector Design and Preparation:

  • CAST Machinery: Clone the genes for the CAST system (e.g., PseCAST or ShCAST) into one or more mammalian expression plasmids. For type I-F, this involves the Cas8/7/6/TniQ QCascade complex and TnsA, TnsB, TnsC [9].
  • Donor Construction: Clone the transgene of interest (e.g., a FIX transgene such as the wild-type or the hyperactive Padua variant (R338L) [31] or the FIX-Triple variant (V86A/E277A/R338A) [32]) into the donor plasmid, ensuring it is flanked by the specific transposon ends. The use of a reporter gene (e.g., GFP) in the donor is recommended for initial validation.
  • Targeting Guide RNA: Design a crRNA expression construct that encodes a guide sequence targeting the desired genomic locus.

• Cell Transfection: Co-transfect the human cells with the three plasmid components: the CAST expression plasmid(s), the donor plasmid, and the crRNA plasmid. Include appropriate controls (e.g., missing a key CAST component).

• Incubation and Analysis: Incubate cells for 48-72 hours to allow for expression, DNA integration, and transgene expression.

  • Genomic DNA Extraction: Harvest cells and isolate genomic DNA.
  • Integration Efficiency Analysis: Use a combination of PCR-based assays (e.g., junction PCR) and next-generation sequencing to detect and quantify on-target integration events [9]. This also allows for the assessment of product purity and the detection of off-target events, which is crucial for type V-K systems [5].

Protocol 2: AAV-Mediated Delivery forIn VivoGene Therapy

This protocol is relevant for pre-clinical testing of CAST-mediated gene integration for disorders like hemophilia B.

Table 3: Key Reagents for AAV-Mediated Delivery

Reagent / Material	Function / Description
Recombinant AAV	AAV serotype (e.g., AAV8) engineered to package the CAST machinery and/or donor DNA. The limited cargo capacity of AAV (~4.7 kb) is a key constraint [32].
Donor Template	For in vivo use, this could be a single-stranded DNA (ssDNA) template or a dual-AAV system may be required for large transgenes.
Animal Model	Hemophilia B mouse model or non-human primate (NHP) for pre-clinical studies [32] [31].

Step-by-Step Methodology:

• Vector Production:

  • Payload Design: Given AAV's cargo limit, the strategy depends on the CAST system and transgene size. For a compact transgene like FIX-Padua, it may be possible to fit a single-vector system in a single AAV. For larger payloads or the bulkier type I-F system, a dual-AAV approach is necessary, splitting the CAST genes and the donor.
  • Virus Production: Produce high-titer, purified recombinant AAV vectors encoding the system components.

• Animal Administration: Systemically administer the AAV vector(s) via tail-vein (mice) or intravenous injection (NHPs). A study using AAV8 to deliver a hyperactive FIX (FIX-Triple) in hemophilia B mice demonstrated a 7-fold higher specific clotting activity compared to wild-type FIX [32].

• Efficacy and Safety Assessment:

  • Blood Collection: Periodically collect plasma samples.
  • Transgene Expression Analysis: Quantify FIX protein levels using ELISA [31].
  • Functional Activity: Measure FIX clotting activity using a one-stage APTT (activated partial thromboplastin time) assay. In NHPs, the FIX-Padua variant (AMT-061) showed a 6.5-fold increase in clotting activity over wild-type FIX (AMT-060) at the same protein dose [31].
  • Off-Target Analysis: Perform whole-genome sequencing on target tissues (e.g., liver) to assess the specificity of integration, a critical step for evaluating the safety of type V-K systems [5].

Diagram 1: CAST System Workflow for Large Transgene Integration. This diagram outlines the key decision points and experimental steps, from system selection to final analysis.

Case Study: Integrating Hyperactive Factor IX Transgenes

The integration of FIX for hemophilia B therapy serves as an excellent case study for comparing workflows and demonstrating the potential of CAST systems.

Therapeutic Goal: Achieve sustained, therapeutic levels of FIX activity in patient plasma through targeted genomic integration of a FIX transgene.

Transgene Engineering: Research has focused on using hyperactive FIX variants to achieve greater clotting activity from lower levels of protein expression, potentially allowing for lower and safer vector doses.

• FIX-Padua (R338L): This variant has 8- to 9-fold increased specific activity over wild-type FIX. An AAV-based gene therapy (AMT-061) encoding this variant demonstrated a 6.5-fold increase in FIX activity compared to wild-type FIX at equal doses in non-human primates [31].
• FIX-Triple (V86A/E277A/R338A): This engineered variant exhibits a 13-fold higher specific clotting activity in vitro and a 7-fold higher activity in vivo in mouse models due to tighter binding to FVIIIa. It has been delivered via AAV8 in hemophilia B mice, resulting in superior hemostasis compared to wild-type FIX [32].

Workflow Integration: The workflow for integrating these FIX transgenes would follow the protocols in Section 3. The donor plasmid for CAST systems would be designed to carry the hyperactive FIX cDNA (e.g., Padua or Triple variant) flanked by the appropriate transposon ends. The success of the integration would be measured not only by the presence of the transgene in the genome but, more importantly, by the resulting plasma FIX activity levels measured by APTT assay [32] [31].

Diagram 2: Factor IX Integration Case Study Workflow. Integrating a hyperactive FIX variant via CAST systems leads to a disproportionate increase in functional clotting activity compared to antigen level, enhancing therapeutic efficacy.

The Scientist's Toolkit: Essential Reagents and Methods

This section catalogs key reagents, tools, and methods essential for developing and executing CAST-based integration workflows.

Table 4: Research Reagent Solutions for CAST Engineering

Tool / Reagent	Specific Example	Function in Workflow
CAST Systems	PseCAST (Type I-F) [9]	A lead candidate with demonstrated activity in human cells; used for high-fidelity integration.
	ShCAST (Type V-K) [5]	A compact, well-studied system; used for applications where size is a primary constraint, often requiring engineering to improve fidelity.
Engineering Tools	Cryo-electron Microscopy (cryoEM) [9] [5]	Used to determine the high-resolution structure of CAST complexes, revealing molecular interactions to guide rational engineering (e.g., PAM recognition, complex stability).
	AlphaFold-Multimer [9]	A computational tool used to predict protein-protein interactions within CAST complexes, facilitating the design of chimeric systems.
Delivery Vehicles	Adeno-associated Virus (AAV) [32] [31]	The primary viral vector for in vivo delivery of CAST components and transgene donors in pre-clinical and clinical settings.
	Bacterial Nanosyringes (e.g., SPEAR) [33]	An engineered bacterial contractile injection system that can be loaded with diverse cargos (proteins, RNPs, ssDNA) and retargeted to specific cell types, offering an alternative non-viral delivery method.
Analytical Methods	High-Throughput Sequencing [5]	Essential for genome-wide profiling of integration events, quantifying on-target efficiency, and comprehensively assessing off-target activity.
	Single-Molecule Imaging [5]	Used to visualize the dynamics of single transposase molecules (e.g., TnsC filament formation) to understand the mechanisms of target site selection.
Affinity Purification	Heparin Sepharose, IX-Select Resin [34]	Chromatography resins used for the large-scale purification of FIX protein, relevant for in vitro studies or protein replacement therapy.
Functional Assay	One-Stage APTT Clotting Assay [32] [31]	The standard functional test to measure the biological activity of FIX in plasma samples following transgene integration.

The development of therapeutic workflows for large transgene integration is rapidly advancing with the adoption of CAST systems. Type I-F systems, with their high intrinsic specificity and proven activity in human cells, currently hold an edge for applications where fidelity is paramount, such as ex vivo cell therapy or in vivo gene correction. Type V-K systems, while more compact, require further engineering to mitigate their inherent off-target integration but remain promising for their simplicity. The successful integration of hyperactive FIX transgenes demonstrates the powerful synergy between protein engineering and advanced genome editing tools. As structural insights from cryoEM and functional data from single-molecule studies continue to inform the rational engineering of both system specificity and efficiency [9] [5], CAST systems are poised to become indispensable tools for creating next-generation genetic therapies.

Targeted integration of genetic cargo into specific genomic loci is a cornerstone of modern therapeutic development and functional genomics. This approach allows for the precise insertion of therapeutic genes or genetic circuits into "safe harbor" loci, such as AAVS1 (located within the PPP1R12C gene) and ALB (the albumin locus), or endogenous genes like TRAC (T Cell Receptor Alpha Constant), which is crucial for T-cell therapies [35] [36]. The primary technological arms for achieving this integration encompass a range of systems: from early protein-based editors like Zinc Finger Nucleases (ZFNs) and the adeno-associated virus (AAV) Rep78 protein, to modern RNA-programmable systems such as CRISPR-Cas9, and the more recently developed Prime-Editing-Assisted Site-Specific Integrase Gene Editing (PASSIGE) and Programmable Addition via Site-Specific Targeting Elements (PASTE) [35] [37]. This guide objectively compares the performance of these technologies, with a specific focus on the emerging CRISPR-associated transposase (CAST) systems, framing the discussion within broader research on the editing efficiency and cargo-size capacity of Type I-F versus Type V-K CAST systems.

Technology Performance Comparison

The efficiency, specificity, and cargo-size capacity of genome editing tools are critical for their successful application. The data below quantitatively compares leading technologies.

Table 1: Performance Comparison of Genome Editing Technologies for Targeted Integration

Technology	Average Integration Efficiency	Cargo Size Capacity	Key Advantages	Key Limitations
PASSIGE/eePASSIGE [37]	~23% (single transfection); up to 30-60% with evolved recombinases	>10 kilobases (kb)	High efficiency for large cargo; avoids double-strand breaks (DSBs); RNA-programmable.	Requires pre-installed or PE-installed landing site.
CRISPR-Cas9 (HDR) [38]	Variable; often low (typically <10%)	Limited by HDR efficiency	High programmability; widely adopted.	Prone to indels and off-target effects; requires DSBs.
ZFN [35]	Similar to Rep78	Standard donor vector	Established technology; specific binding.	Cumbersome protein engineering; lower specificity than ZFN.
AAV2 Rep78 Nickase [35]	Similar to ZFN	Standard donor vector	Avoids DSBs (nickase activity).	Lower specificity compared to ZFN.
Type I-F CAST Systems [37]	≤ ~1% in mammalian cells	Programmable	RNA-programmable integration without DSBs.	Currently low efficiency in mammalian cells.
Type V-K CAST Systems [37]	No reported mammalian genomic integration	Programmable	RNA-programmable integration without DSBs.	Not yet demonstrated in mammalian cells.

Table 2: Case Study Summary: Integration Efficiencies at Specific Loci

Genomic Locus	Technology	Model System	Key Outcome/Integration Efficiency
AAVS1 Safe Harbor	PASSIGE with eeBxb1 (eePASSIGE) [37]	Human cell lines	Up to 60% donor integration in cells with pre-installed sites.
AAVS1 Safe Harbor	AAV2 Rep78 [35]	HEK293 cells	Promoted site-specific integration, but with lower specificity than ZFNs.
AAVS1 Safe Harbor	ZFN [35]	HEK293 & human iPSCs	Demonstrated site-specific integration with high specificity.
CCR5 (Therapeutic)	PASSIGE with eeBxb1 (eePASSIGE) [37]	Human cell lines	High integration efficiency demonstrated.
Multiple Loci	PASSIGE with eeBxb1 (eePASSIGE) [37]	Primary Human Fibroblasts	Integration efficiencies up to 30% at therapeutically relevant sites.
TRAC Locus	CRISPR-Cas9 [38] [39]	Human T-cells	Successful integration for CAR-T therapy; basis for FDA-approved therapies.

Detailed Experimental Protocols

Protocol 1: PASSIGE for Site-Specific Integration in Human Cells

The following workflow details the method for achieving high-efficiency, large cargo integration using the PASSIGE system with evolved recombinases [37].

Workflow Description: The process begins with the design of a pegRNA that targets the desired genomic locus (e.g., AAVS1) and contains an RT template encoding the Bxb1 attachment site, attB. A dual-flap Prime Editor (PE) complex, consisting of a nickase Cas9 (nCas9) fused to a reverse transcriptase (RT), is transfected into the cell. The PE complex binds the target DNA, nicks the strand, and reverse transcribes the attB sequence directly into the genome. Following successful installation of the attB landing site, a separately delivered plasmid containing the large gene cargo (e.g., a therapeutic cDNA) flanked by attP sites and an evolved Bxb1 recombinase (evoBxb1 or eeBxb1) catalyzes the recombination between the genomic attB site and the donor attP sites. This results in the precise integration of the large cargo into the genome. Efficiency is typically assessed via flow cytometry or sequencing after puromycin selection of successfully transfected cells [37].

Protocol 2: Assessing CAST System Editing Efficiency

This protocol outlines a general framework for evaluating the nascent Type I-F and Type V-K CRISPR-associated transposase (CAST) systems in mammalian cells, which currently show low but promising integration activity [37].

Workflow Description: The process begins with the identification and cloning of the core CAST system components: the Cas effector (Cas8/11 for Type I-F or Cas12 for Type V-K) and the associated transposase (TniQ). A donor plasmid containing the cargo DNA flanked by the necessary transposon ends is constructed. The CAST ribonucleoprotein (RNP) complex is assembled in vitro by combining the Cas protein, its guide RNA (crRNA), and the transposase. This RNP complex is then delivered into mammalian cells (e.g., HEK293T) via methods like electroporation or lipofection. The complex binds the target DNA via the guide RNA, and the transposase catalyzes the integration of the cargo. Genomic DNA is harvested after a set period, and integration efficiency is quantified using digital PCR (dPCR) or next-generation sequencing (NGS) to detect insertion events. Given the current low efficiencies (<1% for Type I-F), sensitive assays are crucial [37].

The Scientist's Toolkit: Research Reagent Solutions

Successful execution of these integration experiments requires a suite of specific reagents and tools.

Table 3: Essential Reagents for Targeted Integration Research

Reagent/Tool	Function	Examples & Notes
Prime Editor (PE) System	Installs recombinase landing site without DSBs.	PE2 (optimized reverse transcriptase) or PE3 (with additional nicking gRNA) are common choices [40].
Evolved Recombinase	Catalyzes high-efficiency integration of large cargo.	evoBxb1 or eeBxb1 show 3- to 4-fold higher activity than wild-type Bxb1 in PASSIGE [37].
pegRNA / epegRNA	Guides PE to target locus and templates landing site insertion.	epegRNA (engineered pegRNA) with 3' RNA motifs improves stability and editing efficiency by 3-4 fold [40].
Donor Plasmid	Delivers the large genetic cargo for integration.	Must contain appropriate recombinase landing sites (e.g., attP for Bxb1) flanking the cargo [37].
Delivery Vehicle	Transports editing components into cells.	Lentiviral IDLVs (Integrase-Deficient Lentiviral Vectors) or AAVs for in vivo delivery; lipids for in vitro [35] [40].
CAST System Components	For RNA-programmable transposon integration.	Includes Cas effector (I-F or V-K), TniQ transposase, and donor plasmid with transposon ends [37].

Comparative Analysis & Future Directions

The data unequivocally demonstrates that technologies like eePASSIGE, which leverage continuously evolved recombinases, currently set the benchmark for efficiency in integrating large gene-sized cargoes (>10 kb) into mammalian genomes, achieving rates that are therapeutic relevant (exceeding 30% in some cases) [37]. In contrast, while promising for their fully RNA-programmable nature, Type I-F and Type V-K CAST systems are still in their infancy, with notably lower efficiencies in mammalian cells (≤~1% and 0% reported, respectively). This highlights a significant performance gap, framing the current research frontier: the quest to enhance CAST system activity to rival that of recombinase-based methods.

Future directions will likely focus on applying protein engineering and artificial intelligence (AI)-driven design to overcome current limitations. As demonstrated by the development of AI-generated editors like OpenCRISPR-1, computational models can create highly functional genome editors that diverge significantly from natural sequences [41]. This approach could be harnessed to engineer more efficient and specific CAST system components, particularly the transposase. Furthermore, optimizing the delivery and expression of the large, multi-component CAST machinery in human cells remains a critical challenge. The synergy between the programmability of systems like CAST and the high efficiency of evolved recombinases may ultimately yield next-generation editors capable of safe, targeted integration of any cargo, at any genomic location, with unparalleled precision and efficacy.

Overcoming Bottlenecks: Engineering Solutions for Enhanced Efficiency, Specificity, and Delivery

The development of CRISPR-associated transposase (CAST) systems represents a paradigm shift in genome engineering, offering the potential for programmable integration of large DNA cargo without relying on double-strand break (DSB) repair pathways. Among these systems, Type I-F and Type V-K CASTs have emerged as the most prominent platforms, yet both face significant efficiency bottlenecks when deployed in human cells. Understanding whether these limitations stem primarily from inadequate DNA binding to genomic targets or deficiencies in catalytic integration machinery is crucial for guiding future engineering efforts. This analysis examines the distinct molecular architectures of Type I-F and Type V-K systems to identify the primary constraints on their performance in human cells and compares strategic approaches to overcome these barriers.

Molecular Architecture and Bottleneck Analysis

CAST systems are sophisticated multi-protein complexes that couple CRISPR-guided target recognition with transposase-mediated DNA integration. Their operation in the complex environment of human cells presents unique challenges, with Type I-F and Type V-K systems facing distinct bottlenecks.

Table 1: Core Components and Primary Bottlenecks of CAST Systems

System Feature	Type I-F CAST	Type V-K CAST
Targeting Complex	Multi-protein Cascade (Cas6, Cas7, Cas8)	Single effector (Cas12k)
Transposase Components	TnsA, TnsB, TnsC, TniQ	TnsB, TnsC, TniQ
Integration Mechanism	Cut-and-paste	Copy-and-paste (co-integrate formation)
Primary Human Cell Bottleneck	Catalytic Integration	DNA Binding & Fidelity
Evidence	~200-fold improvement via transposase evolution [14]	High RNA-independent off-target integration [5]

Type I-F CAST: Catalytic Integration as the Primary Barrier

Type I-F systems utilize a multi-subunit Cascade complex for target recognition but suffer from inefficient transposition catalysis in human cells. Evidence supporting catalytic integration as the main bottleneck comes from successful protein evolution campaigns. Researchers applied phage-assisted continuous evolution (PACE) to the transposase module (TnsABC) of a Pseudoalteromonas sp. S983 system (PseCAST), performing hundreds of generations of directed evolution [14]. This approach yielded evolved transposase variants with approximately 200-fold improved integration activity in human cells, demonstrating that optimization of the catalytic machinery alone could overcome what was previously a critical limitation [14]. The resulting evolved CAST (evoCAST) system achieved 10-25% integration efficiencies of kilobase-sized DNA cargo across 14 tested genomic loci in HEK293T cells while generating predominantly unidirectional products and undetectable indels [14].

Type V-K CAST: DNA Binding and Fidelity as Primary Constraints

In contrast, Type V-K systems face challenges primarily related to target recognition fidelity. These more compact systems utilize a single Cas12k effector for DNA binding but exhibit significant RNA-independent "untargeted" transposition [5]. Mechanistic studies reveal that Type V-K CASTs maintain parallel integration pathways, with a minimal transpososome (TnsB, TnsC, TniQ) capable of directing integration independently of Cas12k and guide RNA [5]. This pathway preferentially targets AT-rich genomic regions due to TnsC's DNA binding specificity, creating a substantial fidelity challenge [5]. The problem is compounded in human cells where the systems also face obstacles with nuclear localization and proper function of bacterial-derived components in a mammalian environment [4].

Diagram 1: Comparative bottleneck analysis of Type I-F and Type V-K CAST systems, highlighting their distinct primary limitations and engineering solutions.

Performance Comparison in Human Cells

Recent engineering efforts have yielded substantial improvements in the performance of both CAST systems in human cells, though through fundamentally different approaches reflective of their distinct bottlenecks.

Table 2: Performance Metrics of Engineered CAST Systems in Human Cells

Performance Metric	Evolved Type I-F (evoCAST)	Engineered Type V-K (MG64-1)
Integration Efficiency	10-25% [14]	~3% [4]
Cargo Size Demonstrated	Kilobase-scale [14]	3.2-3.6 kb [4]
On-Target Specificity	High (low off-targets) [14]	Moderate (improved with engineering) [4]
Key Engineering Strategy	Transposase evolution via PACE [14]	Metagenomic mining & NLS optimization [4]
Therapeutic Application	Factor IX cDNA, CAR integration [14]	Factor IX at safe-harbor site [4]

The performance differential between these systems reflects their distinct engineering challenges. The Type I-F evoCAST system benefits from hundreds of generations of continuous evolution specifically targeting its catalytic deficiency [14]. Meanwhile, Type V-K systems have been improved through metagenomic mining to identify naturally diverse systems like MG64-1 and through optimization of nuclear localization signals (NLS) to enhance their function in human cells [4]. Notably, the simplicity of the Type V-K system—requiring only a single Cas effector rather than the multi-protein Cascade complex of Type I-F—remains an attractive feature despite current efficiency limitations [4] [42].

Experimental Approaches for Bottleneck Analysis

Phage-Assisted Continuous Evolution (PACE) of Type I-F CAST

The PACE platform for evolving Type I-F CAST systems involved a sophisticated selection circuit that directly linked transposition efficiency to phage propagation [14]. The experimental workflow comprised:

• Selection Phage (SP) Design: Encoding TnsA, TnsB, and TnsC (TnsABC) in place of the essential M13 bacteriophage gene III [14].

• Host Cell Configuration: Containing two complementary plasmids:

  • CP1: Expressing the DNA-targeting QCascade complex
  • CP2: Containing a transposon-encoded promoter sequence upstream of a promoter-less gene III on an accessory plasmid [14]

• Selection Mechanism: Successful transposition placed the promoter upstream of gene III, triggering its expression and enabling phage propagation. Selection stringency was increased throughout evolution by utilizing weaker promoters requiring more integration events [14].

This direct linkage between transposition efficiency and replicative success drove the evolution of transposase variants with dramatically improved catalytic function in human cells, specifically addressing the primary integration bottleneck of Type I-F systems [14].

Mechanism Mapping for Type V-K CAST Fidelity

For Type V-K systems, a combination of biochemical and genomic approaches identified the DNA binding fidelity bottleneck:

• High-Throughput Sequencing: Capturing genome-wide integration events with various genetic perturbations to quantify on-target versus off-target integration [5].

• Component Minimization: Systematically testing transposition with subsets of CAST components, revealing that TnsB, TnsC, and TniQ alone could catalyze RNA-independent transposition [5].

• Single-Molecule Imaging: Visualizing TnsC filament formation on DNA, revealing its intrinsic preference for AT-rich regions independent of Cas12k guidance [5].

• Cryo-EM Structural Analysis: Determining the architecture of the "BCQ transpososome" (TnsB-TnsC-TniQ) responsible for untargeted integration [5].

These approaches collectively demonstrated that cytoplasmic TnsC filaments could initiate transposition independently of the CRISPR targeting system, revealing a fundamental limitation in the DNA binding fidelity of Type V-K CAST systems [5].

Diagram 2: Experimental workflows for identifying and addressing primary bottlenecks in Type I-F and Type V-K CAST systems.

The Scientist's Toolkit: Essential Research Reagents

Table 3: Key Reagents for CAST Research in Human Cells

Reagent Category	Specific Examples	Function & Application
Evolved CAST Systems	evoCAST (Type I-F) [14]	High-efficiency integration in human cells (10-25%)
Metagenomic CAST Variants	MG64-1, MG64-6 (Type V-K) [4]	Diverse Cas12k effectors with optimized PAM preferences
Evolution Platforms	PACE/PANCE systems [14] [37]	Continuous protein evolution for activity enhancement
Specialized Delivery Vectors	Broad-host-range plasmids with NLS [43] [4]	Component expression and nuclear localization in human cells
Donor Template Designs	Transposon with LE/RE sequences [14] [42]	Cargo flanked by necessary transposon ends for integration
Fidelity Assessment Tools	Whole-genome sequencing, Tn-Seq [5] [43]	Comprehensive on-target and off-target integration profiling

The strategic development of CAST systems for therapeutic applications in human cells requires a precise understanding of system-specific bottlenecks. For Type I-F CAST, the primary constraint lies in catalytic integration efficiency, which can be successfully addressed through continuous evolution of the transposase components as demonstrated by the ~200-fold improvement achieved with PACE [14]. In contrast, Type V-K CAST systems face fundamental challenges in DNA binding fidelity due to their inherent RNA-independent transposition pathway, requiring targeted engineering of component interactions and specificity [5] [4]. Future advancements will likely combine these approaches—applying continuous evolution to enhance catalysis while employing mechanistic insights to engineer superior fidelity—ultimately realizing the full potential of CAST systems for therapeutic human genome engineering.

CRISPR-associated transposases (CASTs) represent a revolutionary class of genome-editing tools that combine the precise targeting ability of CRISPR systems with the DNA insertion capabilities of transposases. Unlike conventional CRISPR-Cas systems that create double-strand breaks (DSBs) and rely on host repair mechanisms, CAST systems enable DSB-free integration of large DNA sequences, overcoming a fundamental limitation in therapeutic genome editing [9] [17]. This capability positions CAST technologies as promising platforms for addressing loss-of-function genetic diseases through targeted gene insertion strategies that are mutation-agnostic [14].

The landscape of characterized CAST systems primarily comprises two major categories: multi-subunit Type I-F systems and more compact Type V-K systems. Type I-F CASTs utilize a multi-protein QCascade complex for DNA targeting (containing Cas8, Cas7, Cas6, and TniQ components) coupled with TnsA, TnsB, and TnsC transposition proteins [9] [22]. In contrast, Type V-K CASTs employ a single Cas12k protein for DNA targeting alongside TnsB and TnsC transposition components, lacking the TnsA subunit present in Type I-F systems [44]. This fundamental architectural difference underlies distinct functional characteristics that have emerged from recent structural and engineering studies.

Structural Foundations from Cryo-EM Analysis

Type I-F CAST DNA Recognition Architecture

Recent cryo-electron microscopy (cryo-EM) studies of the Type I-F PseCAST system have revealed intricate details of its DNA recognition mechanism. The QCascade complex comprises a pseudo-helical assembly of six Cas7 subunits that form a backbone for crRNA binding, with Cas8 responsible for PAM recognition at the crRNA 5' end and Cas6 stabilizing the crRNA 3' end hairpin [9] [22]. A key structural finding involves the dynamic behavior of the TniQ dimer, which exhibits remarkable flexibility relative to other complex components [9].

CryoDRGN analysis, a machine-learning approach for cryo-EM data, has revealed that the TniQ dimer populates a wide range of positions pivoting around Cas6 and Cas7.6, adopting both 'open' conformations lacking direct Cas8 interactions and 'closed' conformations that approach the tip of the Cas8 α-helical domain [9] [22]. This structural flexibility appears functionally important, as replacement of the Cas8 α-helical domain with a flexible linker completely abolishes editing activity in human cells [9]. The quality of cryo-EM maps degrades rapidly in the TniQ dimer region compared to the PAM-adjacent region, suggesting inherent dynamics that may be crucial for recruiting transposition components to the target site [22].

Type V-K CAST Transposase Organization

In Type V-K CAST systems, structural insights have primarily focused on the TnsB transposase component. The cryo-EM structure of the Scytonema hofmannii (sh) TnsB transposase bound to strand transfer DNA reveals an intertwined pseudo-symmetrical architecture with four subunits grouped in different conformations [44]. Notably, only two protomers display catalytically competent active sites with properly positioned DDE residues, while the other two serve structural roles with mispositioned catalytic residues [44].

Transposon end recognition is accomplished through NTD1/2 helical domains, with a singular in trans association of NTD1 domains from catalytically competent subunits reinforcing the overall assembly [44]. The DNA in the DDE catalytic pockets exhibits a sharp bend after the strand-transfer reaction, providing mechanistic insights into the integration process. This structural organization suggests that catalysis is coupled to protein-DNA assembly to secure proper DNA integration, with DNA-binding residue mutants showing altered activity profiles [44].

Table 1: Key Structural Features Revealed by Cryo-EM Analysis

Structural Feature	Type I-F CAST	Type V-K CAST
DNA Targeting Complex	Multi-subunit QCascade (Cas8:Cas7:Cas6:TniQ:crRNA)	Cas12k-TniQ complex
PAM Recognition	Cas8 subunit	Cas12k protein
Transposase Organization	TnsA, TnsB, TnsC proteins	TnsB, TnsC proteins (lacks TnsA)
Key Dynamic Element	Flexible TniQ dimer	Bent DNA in TnsB active site
Complex Stoichiometry	1:6:1:2:1 (Cas8:Cas7:Cas6:TniQ:crRNA)	TnsB tetramer with functional asymmetry

Figure 1: Comparative Architecture of Type I-F and Type V-K CAST Systems. Type I-F systems employ multi-subunit complexes for targeting and transposition, while Type V-K systems utilize more compact arrangements with single-protein targeting.

Experimental Approaches for CAST Engineering

Structure-Guided Engineering of PAM Specificity

The combination of cryo-EM structural data with comprehensive target DNA library screens has enabled precise engineering of PAM specificity in CAST systems. For Type I-F PseCAST, structural analysis revealed subtype-specific interactions and RNA-DNA heteroduplex features that informed rational mutagenesis approaches [9]. Researchers combined structural insights with targeted mutations in PAM- and crRNA-interacting regions, successfully generating CAST variants with both increased integration efficiencies and modified PAM stringencies [9] [22].

Experimental protocols for determining PAM specificity involve incubating purified QCascade complexes with double-stranded DNA substrates containing defined target sequences and candidate PAM motifs, followed by binding affinity measurements through electrophoretic mobility shift assays (EMSAs) [9] [22]. For functional validation, engineered CAST variants are tested in human cell lines using reporter assays that quantify integration efficiency at genomic sites with varying PAM sequences [9].

Phage-Assisted Continuous Evolution (PACE) of Transposase Activity

To overcome the limited activity of natural CAST systems in human cells, researchers developed a phage-assisted continuous evolution (PACE) platform that links transposition activity to bacteriophage propagation [14]. This approach involves:

• Selection Phage (SP): Engineered M13 bacteriophage encoding evolving TnsABC genes in place of the essential gIII gene
• Accessory Plasmid (AP): Contains gIII under a promoter-less configuration, requiring transposition of a promoter sequence from a complementary plasmid for expression
• Mutagenesis Plasmid (MP): Provides inducible mutagenesis to generate diversity in the evolving SP population
• Complementary Plasmids (CP1/CP2): Express essential PseCAST targeting components (CP1) and provide promoter-donor sequences (CP2) [14]

Through hundreds of generations of mutation, selection, and replication, PACE identified transposase variants with approximately 200-fold improved integration activity in human cells compared to wild-type PseCAST [14]. This evolved transposase (evoCAST) synergized with structure-guided engineering of the DNA-targeting module to achieve 10-25% integration efficiencies of kilobase-sized DNA cargos across multiple genomic loci in HEK293T cells [14].

Figure 2: Phage-Assisted Continuous Evolution (PACE) Workflow for CAST Engineering. This platform links transposition activity to phage propagation, enabling rapid evolution of improved CAST variants through hundreds of generations of mutation and selection.

Performance Comparison: Type I-F vs. Type V-K CAST Systems

Editing Efficiency and Specificity

Comparative analyses reveal distinct performance characteristics between Type I-F and Type V-K CAST systems. Engineered Type I-F CAST systems, particularly the evolved PseCAST (evoCAST), demonstrate 10-25% integration efficiencies of kilobase-sized DNA cargos across 14 tested human genomic loci in HEK293T cells [14]. These systems generate predominantly unidirectional transposition products without detectable indel formation and maintain low off-target integration rates [14].

In contrast, Type V-K CAST systems exhibit multiple undesirable biochemical properties in heterologous cellular contexts, including reduced specificity, low overall editing efficiencies, and poor product purity [9] [22]. Type I-F CASTs show demonstrably greater efficiencies than Types I-B, I-D, and V-K in bacterial systems, with this advantage extending to engineered variants in human cells [9].

Cargo Size Capacity and Product Purity

Both Type I-F and Type V-K CAST systems support the integration of large DNA sequences, overcoming a critical limitation of conventional genome editing tools. However, Type I-F CASTs exhibit highly specific and homogeneous integration products with strong directionality bias, minimizing byproduct formation [14]. The presence of TnsA in Type I-F systems enables precise cleavage of transposon ends, contributing to cleaner integration events compared to Type V-K systems that lack TnsA [44].

Table 2: Performance Comparison of Engineered CAST Systems

Performance Metric	Type I-F CAST (evoCAST)	Type V-K CAST
Integration Efficiency	10-25% (human cells)	<1% (human cells)
Cargo Size Capacity	Multi-kilobase inserts	Multi-kilobase inserts
Product Purity	High (unidirectional, minimal byproducts)	Moderate (heterogeneous products)
Off-Target Integration	Low levels	Elevated concerns
Indel Formation	Undetectable	Not reported
Therapeutic Validation	Factor IX, CAR, multiple disease genes	Factor IX (preclinical)

Research Reagent Solutions for CAST Engineering

Table 3: Essential Research Reagents for CAST Engineering Studies

Reagent / Material	Function in CAST Engineering	Example Application
Cryo-EM Infrastructure	High-resolution structure determination	Mapping DNA recognition interfaces
PACE Platform	Continuous evolution of transposase activity	Generating hyperactive CAST variants
EMSA Assays	Measuring DNA-binding affinity	Evaluating PAM specificity mutants
Reporter Cell Lines	Quantifying integration efficiency	Testing engineered CAST variants
Library Screening	Profiling PAM preferences	Identifying specificity determinants
AlphaFold-Multimer	Predicting protein-protein interactions	Designing chimeric CAST systems

Structure-guided engineering approaches leveraging cryo-EM insights have dramatically advanced the development of CAST systems for therapeutic genome editing. The integration of structural biology with continuous evolution platforms has transformed Type I-F CAST systems from minimally active curiosities into promising tools capable of efficient, targeted DNA integration in human cells [14]. These engineered systems now achieve integration efficiencies that approach therapeutic relevance while maintaining high specificity and product purity.

The contrasting architectures of Type I-F and Type V-K CAST systems highlight different engineering challenges and opportunities. While Type V-K systems offer advantages in compactness, Type I-F systems have demonstrated superior editing efficiency and product purity in human cells following extensive engineering [9] [14]. The structural insights guiding these improvements—particularly regarding PAM recognition and complex stability—provide a framework for ongoing optimization efforts.

As CAST engineering continues to mature, the translation of these systems to therapeutic applications appears increasingly feasible. Companies like Metagenomi are advancing CAST-based therapeutics toward clinical trials, with first-in-human studies anticipated by 2026 [17]. The unique capability of CAST systems to install large DNA sequences without double-strand breaks positions them as promising platforms for addressing loss-of-function genetic diseases through mutation-agnostic gene insertion strategies [14] [17]. Future developments will likely focus on enhancing delivery efficiency, expanding targetable genomic sites, and further refining integration specificity to realize the full therapeutic potential of CAST genome editing.

The clinical application of any genome-editing technology hinges on its precision. For CRISPR-associated transposases (CASTs), which enable RNA-guided insertion of large DNA cargos without double-strand breaks, minimizing off-target and RNA-independent integration is a critical research and development focus. A direct comparison reveals that Type I-F and Type V-K CAST systems, the two primary classes being engineered for human cell applications, achieve this precision through distinct mechanistic strategies and exhibit different performance trade-offs [6] [17]. This guide objectively compares the engineered efficiency and specificity of these systems, providing a framework for selecting the optimal platform for therapeutic development.

Mechanisms of Integration Fidelity

The foundational difference in fidelity between Type I-F and Type V-K systems stems from their core architectures and the presence of a dedicated proofreading subunit.

• Type I-F CASTs, such as the PseCAST and VchCAST systems, typically feature a multi-subunit Cascade complex for DNA targeting and a heteromeric transposase comprising TnsA, TnsB, TnsC, and TniQ [9] [45]. The presence of TnsA is a key differentiator; it acts as a endonuclease that works with TnsB to cleanly excise the transposon, contributing to a multi-step proofreading process that results in highly specific integration [45]. Structural studies using cryogenic electron microscopy (cryoEM) show that the TniQ dimer bridges the Cascade complex and the TnsC regulator, forming a stable complex that ensures the transposase is recruited only to the intended target site [9].

• Type V-K CASTs, like the ShCAST system from Scytonema hofmannii, are more compact. They utilize a single Cas12k protein for DNA targeting and lack the TnsA subunit [46] [45]. This simplicity is advantageous for delivery but comes with a fidelity cost. The absence of TnsA is correlated with higher rates of off-target integration and the formation of chimeric products when the system is overexpressed [45]. The fidelity relies heavily on the proper assembly of a megadalton-scale "transpososome" complex, where TnsC polymerization on the target DNA is stabilized by interactions with TniQ and Cas12k [46].

The diagram below illustrates the distinct components and integration checkpoints for each system.

Performance Comparison: Efficiency, Specificity, and Cargo

The mechanistic differences between Type I-F and Type V-K systems translate into distinct performance profiles. The following table summarizes key quantitative and qualitative metrics based on recent experimental findings.

Feature	Type I-F CAST	Type V-K CAST
Core Targeting Effector	Multi-subunit Cascade complex [9] [6]	Single protein Cas12k [6] [46]
Transposase Composition	TnsA, TnsB, TnsC, TniQ [9] [45]	TnsB, TnsC, TniQ (lacks TnsA) [46] [45]
Reported Editing Efficiency	Initially low (single-digit %), engineered to >30% with evoCAST [17]	Low initial efficiency, 5x improvement via high-throughput screening [47] [24]
Integration Specificity	Very high (near 100% on-target in bacteria) [45]	Moderate; more prone to off-target integration [6] [45]
Key Fidelity Mechanism	Multi-step proofreading; TnsA/B excision [45]	Transpososome assembly fidelity [46]
Cargo Capacity	Large; demonstrated insertions up to 10 kb [45]	Large; suitable for therapeutic gene insertion [17]
Insertion Orientation	Predominantly unidirectional [6]	Almost unidirectional [6]

Experimental Protocols for Enhancing Fidelity

Researchers employ advanced structural and screening methodologies to understand and improve the precision of CAST systems.

Structure-Guided Engineering of Type I-F CASTs

This protocol focuses on using high-resolution structural data to identify and engineer key protein residues for improved DNA binding and specificity [9].

• Step 1: Complex Purification & Structural Determination. Purify the native QCascade complex (e.g., PseCAST) and determine its structure using single-particle cryoEM. This reveals atomic-level interactions, such as those between the Cas8 protein and the protospacer adjacent motif (PAM), and highlights flexible regions like the TniQ dimer [9].
• Step 2: Identify Engineering Targets. Analyze the cryoEM structure and conformational dynamics (e.g., via cryoDRGN) to pinpoint residues critical for DNA recognition and complex stability. For example, the Cas8 protein's α-helical domain, though flexible, is essential for activity [9].
• Step 3: Generate and Test Variants. Create a library of mutants targeting the identified residues. Screen these variants in human cells (e.g., HEK293T) using a reporter assay that measures the precise integration of a donor plasmid into a defined genomic locus. Mutants with enhanced DNA binding can yield higher integration efficiencies [9].
• Step 4: Assess Specificity. For lead variants, assess off-target integration using unbiased methods like whole-genome sequencing (WGS) to confirm that efficiency gains do not compromise specificity [48].

High-Throughput Mutational Screening for Type V-K CASTs

This protocol describes a scalable method to simultaneously evaluate the activity and specificity of thousands of CAST variants [47] [24].

• Step 1: Create a Saturation Mutagenesis Library. Generate a comprehensive library of a Type V-K CAST system (e.g., Cas12k, TnsB, TnsC, TniQ) where every single amino acid is mutated to all possible alternatives [47] [24].
• Step 2: High-Throughput Delivery & Selection. Deliver the mutant library into a model cell system (e.g., E. coli) using a high-efficiency transformation method. Subject the cells to a selection pressure where only cells with successful CAST-mediated integration of a marker gene survive [47].
• Step 3: Deep Sequencing & Analysis. Extract genomic DNA from the selected cell population and perform deep sequencing of the integrated donor and potential off-target sites. Computational analysis of the sequencing data reveals which mutations increase the on-target integration rate and/or reduce off-target events [47] [24].
• Step 4: Combine Beneficial Mutations. Combine the top-performing mutations from the screen to generate synergistic "hyperactive" variants. Recent studies have reported a fivefold increase in activity without compromising specificity using this approach [47] [24].

The workflow for this high-throughput screening strategy is visualized below.

The Scientist's Toolkit: Essential Research Reagents

The following reagents are fundamental for conducting the experiments cited in this guide and for advancing CAST system research.

Research Reagent	Function in CAST Research
PseCAST QCascade Plasmid	Engineered expression vector for producing the multi-subunit Type I-F targeting complex for structural and functional studies [9].
Cas12k-TniQ-TnsB-TnsC System	The core set of proteins for reconstituting Type V-K CAST activity, often co-expressed from a single plasmid [46] [45].
Strand-Transfer DNA Substrate	A custom-designed double-stranded DNA fragment containing the target sequence and PAM, used for in vitro reconstitution of the transpososome for cryoEM studies [46].
Reporter Cell Line (e.g., HEK293T)	A mammalian cell line engineered with a defined genomic target site, used for quantifying CAST integration efficiency and specificity in a human cell context [9] [17].
dsODN Donor Template	A double-stranded oligodeoxynucleotide or a larger linear DNA fragment containing the cargo to be integrated, flanked by the necessary transposon end sequences [46].

The strategic choice between Type I-F and Type V-K CAST systems involves a direct trade-off between inherent fidelity and engineering simplicity. Type I-F systems offer a structurally robust, high-fidelity foundation due to their multi-component proofreading, making them ideal for applications where specificity is paramount. In contrast, Type V-K systems provide a compact, engineerable chassis that has demonstrated significant improvements in efficiency through high-throughput screening.

The future of CAST precision engineering lies in the convergence of these strategies. Integrating structural insights from Type I-F systems into the more compact Type V-K architecture, combined with powerful directed evolution campaigns, will be pivotal in creating next-generation CAST systems. These systems will need to combine high efficiency, minimal off-target activity, and delivery-friendly packaging to realize their full potential in therapeutic gene insertion.

The emergence of CRISPR-associated transposase (CAST) systems represents a significant leap beyond traditional CRISPR-Cas9 technology, offering the potential for precise integration of large DNA sequences without creating double-strand breaks. These systems combine the programmability of CRISPR with the DNA insertion capabilities of transposases, opening new avenues for therapeutic gene delivery. However, a central challenge persists: efficiently packaging these multi-component systems for in vivo use. This article objectively compares two primary CAST systems—Type I-F and Type V-K—within the context of this delivery challenge, examining their editing efficiency, cargo capacity, and compatibility with current delivery platforms to inform strategic selection for research and therapeutic development.

CAST systems are naturally occurring genetic elements that have been repurposed for precision genome engineering. Their core function involves using a CRISPR-guided complex to direct the integration of a DNA payload into a specific genomic locus, a process that avoids the double-strand breaks associated with conventional CRISPR nucleases.

The following diagram illustrates the distinct component architectures of the two main CAST systems.

The fundamental difference lies in their targeting mechanisms. Type I-F CAST employs a multi-protein Cascade complex (comprising Cas6, Cas7, and Cas8 proteins) for DNA recognition [42]. In contrast, Type V-K CAST utilizes a single Cas12k effector protein for the same purpose, significantly reducing biochemical complexity [17] [4]. This distinction in composition has direct implications for delivery, as the simpler architecture of Type V-K is more amenable to packaging into size-constrained viral vectors.

Performance Comparison: Efficiency, Cargo and Specificity

The architectural differences between Type I-F and Type V-K CAST systems translate directly into distinct performance profiles, particularly in human cells. The table below summarizes key quantitative metrics.

Parameter	Type I-F CAST	Type V-K CAST
Targeting Complex	Multi-subunit Cascade (Cas6, Cas7, Cas8) [42]	Single protein Cas12k [17] [4]
Transposase Core	TnsA, TnsB, TnsC, TniQ [42]	TnsB, TnsC, TniQ [42]
Reported Editing Efficiency in Human Cells	~1% (in HEK293 cells with ~1.3 kb donor) [42]	Up to ~3% (in HEK293 cells with 3.2 kb donor) [4] [42]
Demonstrated Cargo Capacity	Up to ~15.4 kb [42]	Up to 30 kb [42]
Integration Byproduct	Clean, "cut-and-paste" insertion (TnsA present) [42]	Co-integrate product (TnsA absent) [42]
Primary Delivery Challenge	Packaging multiple large proteins	Balancing cargo size with Cas12k delivery

Data synthesized from multiple research studies indicate that while both systems are functional in human cells, Type V-K CAST demonstrates a favorable balance of efficiency and cargo capacity. The recently identified MG64-1 system, a Type V-K variant, achieved approximately 3% integration efficiency of a 3.2 kb donor DNA at the AAVS1 safe harbor locus in HEK293 cells [4] [42]. Furthermore, Type V-K systems have successfully integrated therapeutic genes, such as the full-length Factor IX gene for hemophilia B, showcasing their therapeutic relevance [17] [4].

Notably, the absence of TnsA in Type V-K systems leads to the formation of co-integrate byproducts, where vector backbone sequences may be integrated alongside the intended cargo [42]. In contrast, the presence of TnsA in Type I-F systems enables a cleaner "cut-and-paste" transposition mechanism. This is a critical consideration for therapeutic applications where product purity is paramount.

Delivery Modalities: Navigating the Packaging Challenge

A primary hurdle for the in vivo application of CAST systems is the efficient delivery of their multiple components into target cells. The limited packaging capacity of preferred delivery vectors, particularly adeno-associated viruses (AAVs), creates a significant bottleneck.

The Viral Vector Dilemma

Recombinant Adeno-associated Virus (rAAV) is a leading platform for in vivo gene therapy due to its favorable safety profile and tissue-specific tropism [49] [50]. However, its stringent packaging capacity of less than 5 kb is a major constraint [49] [50]. This limitation directly impacts the choice of CAST system. The simpler architecture of Type V-K CAST, with its single Cas12k effector, presents a more feasible candidate for AAV delivery compared to the multi-protein Cascade complex of Type I-F systems.

Strategies to Overcome Packaging Constraints

Researchers are developing several innovative strategies to circumvent these delivery limitations:

• Use of Compact Effectors: The discovery and engineering of smaller Cas effectors is a primary focus. The single-protein Cas12k of Type V-K systems is inherently more compact than the multi-protein Cascade, offering a strategic advantage [17].
• Dual-Vector and Split-Intein Systems: These approaches split large genetic payloads across two separate AAV vectors. The full-length protein or gene is reconstituted in the target cell, either through co-transduction and recombination or via protein trans-splicing mediated by split inteins [49] [50].
• System Minimization: Ongoing efforts to engineer minimal CAST systems by trimming non-essential regions of proteins and optimizing guide RNA designs are crucial for reducing the total genetic payload [24] [4].

Experimental Workflow for CAST System Evaluation

Evaluating the performance and specificity of engineered CAST systems requires a robust experimental pipeline. The following diagram outlines a high-throughput screening workflow used to profile CAST variants.

This workflow, as employed by researchers at St. Jude Children's Research Hospital, involves creating a comprehensive mutant library and using high-throughput assays to simultaneously measure the activity and specificity of thousands of CAST variants [24]. This method led to the identification of specific mutations that, when combined, boosted CAST activity fivefold without compromising specificity [24]. Such screening platforms are invaluable for engineering next-generation CAST systems with enhanced properties for in vivo applications.

The Research Toolkit: Essential Reagents for CAST System Development

Advancing CAST technology from a bacterial tool to a platform suitable for human therapeutic application requires a specific set of molecular tools and reagents.

Research Tool	Function & Purpose
Metagenomic Datasets	Discovery of novel, diverse CAST systems from uncultivated microbes [4].
Nuclear Localization Signal (NLS)	Peptide tags engineered onto CAST proteins to ensure their import into the nucleus of mammalian cells [4].
Host Factors (e.g., S15, ClpX)	Bacterial proteins (ribosomal protein S15, chaperone ClpX) that are co-expressed to enhance CAST integration efficiency in human cells [4] [42].
Safe Harbor Locus gRNAs	Guide RNAs targeting genetically "safe" genomic regions (e.g., AAVS1, Albumin) to minimize risks in therapeutic integration [17] [4].
Dual AAV Vector System	A delivery strategy where CAST components are split across two AAV vectors to overcome packaging constraints [49] [50].

The journey toward therapeutic in vivo application of CAST systems is fundamentally a delivery problem. The comparative analysis presented here indicates that Type V-K CAST systems, with their simpler single-effector architecture and substantial cargo capacity, currently present a more tractable path forward for overcoming packaging hurdles. However, the cleaner integration mechanism of Type I-F systems remains an attractive feature.

Future progress hinges on the continued engineering of minimal, high-efficiency CAST variants and the parallel development of sophisticated delivery platforms, such as optimized dual-AAV systems and lipid nanoparticles. The convergence of these fields—genome editing and delivery technology—will ultimately determine the success of CAST systems in realizing their potential as transformative therapeutic tools for correcting a wide array of genetic diseases.

Benchmarking Performance: A Data-Driven Comparison of Efficiency, Cargo Size, and Specificity

The programmable integration of large DNA cargo into the human genome represents a central goal in genetic engineering, with profound implications for gene therapy, synthetic biology, and functional genomics. CRISPR-associated transposase (CAST) systems have emerged as leading tools for this purpose, as they facilitate RNA-guided integration without relying on double-strand break (DSB) repair pathways [51]. Among the diverse CAST families, Type I-F and Type V-K systems represent two of the most advanced engineering platforms. This guide provides a direct, data-driven comparison of their editing efficiencies at human genomic loci, synthesizing the most current experimental evidence to inform tool selection for research and therapeutic development.

A critical differentiator between these systems is their molecular complexity. Type V-K systems utilize a single Cas12k effector protein for DNA targeting, making them inherently more compact [4] [52]. In contrast, Type I-F systems rely on a multi-protein Cascade complex (comprising Cas8, Cas7, and Cas6 subunits) and TniQ for target recognition [9]. This architectural difference has significant implications for their deliverability and efficiency in the challenging environment of human cells.

Quantitative Efficiency Comparison at a Glance

The table below summarizes key performance metrics for Type I-F and Type V-K CAST systems, based on the latest peer-reviewed studies in human cells.

Table 1: Head-to-Head Comparison of CAST System Performance in Human Cells

Performance Metric	Type I-F CAST (e.g., PseCAST)	Type V-K CAST (e.g., MG64-1, MG64-6)
Reported Integration Efficiency	~10-25% (with engineered variants) [9]	Demonstrated; specific quantitative rates in human cells not fully detailed in available results [4]
Key System Components	Cas8f, Cas7f, Cas6f, TniQ (dimer), TnsA, TnsB, TnsC [9]	Cas12k, TniQ, TnsB, TnsC, S15 host factor [4]
Protospacer Adjacent Motif (PAM)	5'-CC-3' (for PseCAST) [9]	5'-GTN-3' or 5'-rGTN-3' (for metagenomic systems) [4]
DSB-Free Integration	Yes [9]	Yes [4]
Primary Experimental Evidence	Structure-guided engineering of QCascade complex; efficiency gains from targeted mutations [9]	Engineering for nuclear localization; integration of a therapeutically relevant transgene (Factor IX) at a safe-harbor site [4]

System Architectures and Mechanisms

The divergent efficiencies of Type I-F and Type V-K systems are rooted in their distinct molecular architectures. The following diagram illustrates the core components and their assembly for each system.

Diagram 1: CAST System Architectures. Type V-K uses a simpler, single-effector (Cas12k) targeting complex. Type I-F employs a multi-subunit QCascade complex for targeting, which is a key factor in its typically higher integration efficiency and specificity in human cells [4] [9].

Mechanism of Type V-K CASTs

Type V-K systems utilize a single Cas12k effector, which complexes with TniQ and the bacterial host factor S15 to form the DNA targeting module [4]. This module is guided by a single guide RNA (sgRNA) to the target genomic locus. The targeting module then recruits the transposase machinery (TnsB and TnsC), which catalyzes the excision of the donor DNA from its carrier plasmid and its subsequent integration downstream of the PAM site [4]. A noted characteristic of many native Type V-K systems is the absence of TnsA, which can lead to a higher frequency of co-integrate byproducts (where plasmid backbone is also integrated) rather than simple "cut-and-paste" transposition [4].

Mechanism of Type I-F CASTs

Type I-F systems rely on a more complex QCascade complex for target recognition. This complex includes proteins Cas8, Cas6, and multiple copies of Cas7, which form a filament that binds the crRNA [9]. A stable TniQ dimer is associated with this complex. Upon recognizing the target DNA, the QCascade complex, via TniQ, recruits and activates the TnsABC transposase [9]. The presence of TnsA in these systems enables a clean "cut-and-paste" integration, typically resulting in a single copy of the donor cargo being inserted without accompanying plasmid backbone [9], which is a significant advantage for therapeutic applications requiring high product purity.

Experimental Protocols for Efficiency Assessment

To ensure the reproducibility of the efficiency data presented, this section outlines the core methodologies common to studies quantifying CAST activity in human cells.

Protocol for Measuring Genomic Integration Efficiency

The following workflow is standard for determining the integration rates reported in comparative studies.

Diagram 2: Integration Efficiency Workflow. The percentage of alleles containing the desired integration is calculated from qPCR data or by dividing the number of sequencing reads containing the insert by the total number of reads [4] [9]. NGS: Next-Generation Sequencing.

Key Experimental Details:

• Cell Culture: Experiments are typically performed in standard human cell lines (e.g., HEK293T) maintained under recommended conditions.
• Delivery Method: CAST components are often delivered via plasmid transfection (e.g., using lipofection or electroporation). The systems are typically broken into multiple plasmids to reduce size and improve deliverability (e.g., one for proteins, one for sgRNA, and one for the donor DNA) [4].
• Donor Design: The donor plasmid contains the therapeutic or reporter cargo (e.g., Factor IX) flanked by the necessary terminal inverted repeats (TIRs) recognized by the transposase (TnsB) [4]. For Type I-F systems, the donor is typically designed to minimize co-integrate formation.
• Controls: Essential controls include "no sgRNA" and "catalytically dead transposase" conditions to establish baseline activity and confirm RNA-guided integration.

The Scientist's Toolkit: Key Reagents and Materials

Successful implementation of CAST technology requires a specific set of molecular tools. The table below lists essential reagents for researchers aiming to replicate these studies or develop new applications.

Table 2: Essential Research Reagents for CAST Genome Engineering

Reagent / Solution	Function in the Experiment	Specific Examples & Notes
CAST Expression Plasmids	To express the core protein components (Cas/Cascade, Tns) in human cells.	Codon-optimized for human cells; often split across multiple plasmids. Requires NLS tags (Nuclear Localization Signals) for nuclear import [4].
Guide RNA Expression Vector	To express the sgRNA (V-K) or crRNA (I-F) that programs target specificity.	For V-K, the sgRNA design includes conserved tracrRNA and crRNA elements [4].
Donor Template Plasmid	Carries the DNA cargo to be integrated into the genome.	Must contain Terminal Inverted Repeats (TIRs) recognized by TnsB. Cargo size can range from fluorescent reporters to full therapeutic genes (>10 kb) [4] [37].
Human Cell Lines	The cellular environment for editing.	Commonly used lines: HEK293T, HeLa, and primary human fibroblasts [4] [37].
Transfection Reagent	To deliver nucleic acids (plasmids) into human cells.	Lipofection or electroporation kits suitable for the specific cell type.
Host Factor Supplements	To enhance integration efficiency in a heterologous human environment.	Type V-K systems can require the bacterial S15 ribosomal protein [4]. Type I-F systems can benefit from the bacterial chaperone ClpX [4] [9].

The direct quantitative comparison reveals that while both Type I-F and Type V-K CAST systems represent monumental advances in DSB-free genome engineering, they currently occupy different maturity levels for applications in human cells. Engineered Type I-F systems have demonstrated robust integration efficiencies in the ~10-25% range, a significant benchmark for therapeutic relevance [9]. Their more complex targeting apparatus appears to pay dividends in efficiency and product purity. In contrast, the primary advantage of Type V-K systems is their simplified, compact architecture centered on a single Cas12k effector, which is highly advantageous for delivery in vivo, though this currently comes with trade-offs in efficiency and byproduct profile that require further optimization [4].

The future of CAST technology lies in continuous protein engineering. As demonstrated with Type I-F systems, structure-guided engineering and directed evolution are powerful strategies for overcoming natural bottlenecks in DNA binding and integration [9]. The development of even more efficient CAST systems, potentially through the creation of chimeric proteins that combine optimal modules from different systems, is a highly active area of research. For researchers choosing a system today, the decision hinges on the priority: Type I-F for higher demonstrated efficiency in human cells, or Type V-K for its compactness and potential for future viral delivery.

The evolution of gene therapy and functional genomics has created a pressing demand for technologies capable of inserting large DNA sequences into the genome. While CRISPR-Cas9 has revolutionized precision genome editing, its reliance on DNA double-strand breaks (DSBs) and host repair mechanisms presents significant limitations for kilobase-scale insertions. The repair processes often result in unpredictable outcomes, including indels, partial integrations, and chromosomal rearrangements, making precise large insertions challenging [53]. Furthermore, adeno-associated virus (AAV) vectors, commonly used for therapeutic gene delivery, have a stringent packaging limit of approximately 4.7–5.0 kb, which constrains the size of deliverable genetic material [54]. CRISPR-associated transposase (CAST) systems have emerged as promising solutions, combining the programmability of CRISPR with the efficient insertion machinery of transposons to enable DSB-free integration of large DNA cargo. This analysis provides a comparative assessment of two major CAST systems—Type I-F and Type V-K—focusing on their cargo capacity, editing efficiency, and practical implementation for therapeutic gene insertion.

CAST systems represent a paradigm shift in genome engineering by enabling targeted DNA integration without creating double-strand breaks. These systems naturally consist of two primary modules: a CRISPR-guided targeting complex that identifies specific genomic loci, and a transposase enzyme complex that catalyzes the integration of donor DNA [51] [17]. Unlike traditional CRISPR-Cas systems that induce DSBs and rely on endogenous repair mechanisms, CAST systems directly insert DNA fragments through a cut-and-paste or paste-only mechanism, significantly reducing unintended mutations and enabling more predictable editing outcomes [4].

Table 1: Core Components of Major CAST Systems

Component	Type I-F CAST	Type V-K CAST
Targeting Module	Multi-subunit Cascade complex (Cas8, Cas7, Cas6, TniQ)	Single effector Cas12k with TniQ
Transposase Module	TnsA, TnsB, TnsC	TnsB, TnsC
Guide RNA	crRNA with separate tracrRNA	Single guide RNA (sgRNA)
PAM Recognition	Cas8 subunit	Cas12k protein
Key Structural Feature	TniQ homodimer stably associated with Cascade	More compact architecture; simpler composition

Type I-F CAST: Multi-Subunit Precision

Type I-F CAST systems employ a sophisticated multi-protein CRISPR complex for DNA targeting. The core targeting module comprises Cas8, which recognizes the protospacer adjacent motif (PAM) and facilitates target DNA binding; Cas7 subunits that form a helical backbone stabilizing the crRNA-DNA heteroduplex; and Cas6, which processes crRNA and anchors the TniQ homodimer [9]. The transposase machinery includes TnsB, the catalytic subunit responsible for DNA strand transfer; TnsC, an ATPase that regulates transposase activity; and TnsA, which cleaves the donor DNA, enabling a precise "cut-and-paste" integration mechanism [4]. Recent structural insights into the PseCAST system, a Type I-F CAST, reveal that the TniQ dimer exhibits considerable conformational flexibility relative to the Cascade complex, populating both "open" and "closed" states that may regulate integration efficiency [9].

Type V-K CAST: Compact Simplicity

Type V-K CAST systems offer a markedly simpler architecture, utilizing a single Cas12k protein for DNA targeting instead of the multi-subunit Cascade complex [4] [17]. This compact system retains TniQ, which associates with Cas12k, along with the transposase components TnsB and TnsC. Notably, most natural Type V-K systems lack TnsA, resulting in a "paste-only" mechanism where only the transferred strand is cleaved, potentially leading to co-integration events where plasmid backbone sequences are inserted alongside the desired cargo [4]. Engineering efforts have optimized the sgRNA architecture and nuclear localization signals to enhance function in human cells, with systems like MG64-1 and MG64-6 demonstrating programmable integration with 5'-GTN PAM preferences [4].

Table 2: Cargo Capacity and Performance Metrics of Genome Editing Systems

Editing System	Mechanism	Maximum Cargo Capacity	Therapeutic Applicability	Key Advantages	Key Limitations
AAV Vectors	Viral transduction	~5.0 kb	Limited by packaging capacity	Established delivery platform	Rapid decline in full-length genomes beyond 4.9 kb [54]
CRISPR-Cas9 HDR	DSB-dependent repair	Limited only by delivery	Constrained by low efficiency in non-dividing cells	High precision with donor template	Unpredictable indels; low HDR efficiency [53]
Prime Editing	Reverse transcription without DSBs	<50 bp	Single-base changes to small insertions	No DSBs; high precision	Limited cargo capacity [17]
Type I-F CAST	RNA-guided transposition	Multi-kilobase (theoretically large)	Demonstrated in human cells	Highly specific, homogeneous products	Complex multi-component system [9]
Type V-K CAST	RNA-guided transposition	Multi-kilobase (therapeutically relevant)	Full-length Factor IX integration shown	Compact, single-protein targeting	Co-integration events without TnsA [4]

Cargo Capacity Assessment: Quantitative Analysis

The capacity to deliver large genetic payloads is a critical advantage of CAST systems over other genome editing technologies. While AAV vectors, a common therapeutic delivery vehicle, exhibit significantly reduced proportions of full-length genomes at sizes approaching their 5.0 kb limit [54], CAST systems can theoretically accommodate much larger insertions. Research indicates that the genomic integrity of AAV vectors begins to decline rapidly between 4.9 and 5.0 kb, with an 86.3% reduction in full-length genomes observed when comparing 4.7 kb versus 5.0 kb vectors [54]. This limitation severely constrains AAV-based gene therapy approaches for larger genes.

CAST systems fundamentally overcome this size restriction. Type I-F CAST systems have demonstrated the capability for "multi-kilobase" insertions, though their multicomponent nature presents delivery challenges [9]. Type V-K CAST systems have shown particular promise, with engineered systems successfully integrating full-length therapeutic genes, such as Factor IX (relevant for hemophilia B), into safe harbor loci in human cells [4]. The cargo flexibility of CAST systems represents a significant advancement for therapeutic applications requiring the insertion of complete gene sequences with regulatory elements.

Editing Efficiency and Specificity: Performance Benchmarks

Editing efficiency and specificity are paramount considerations for therapeutic applications. Type I-F CAST systems, such as PseCAST, have demonstrated highly specific and homogeneous integration products but initially showed limited efficiency in human cells [9]. Engineering efforts focusing on the DNA binding domain of PseCAST have yielded variants with improved integration efficiencies, addressing this initial limitation [9].

Type V-K CAST systems have shown promising efficiency profiles in both bacterial and human cells. In E. coli, systems like MG64-1 achieved integration efficiencies up to 80% at engineered loci and 50% at endogenous intergenic regions, with the capability for simultaneous multi-locus targeting [4]. Importantly, off-target integration events were relatively infrequent, occurring at rates below 7% in comprehensively sequenced genomes [4]. In human cells, initial Type V-K CAST activity was limited but was significantly enhanced through engineering of nuclear localization signals and optimization of component ratios, ultimately enabling therapeutic transgene integration across multiple human cell types [4].

CAST Experimental Workflow

Experimental Protocols for CAST Evaluation

In Vitro Integration Assay

The in vitro integration assay provides a controlled system for initial CAST functionality assessment. The protocol involves incubating purified CAST proteins (expressed and purified from E. coli) with in vitro transcribed guide RNA, a linear donor DNA fragment containing terminal inverted repeats, and a target plasmid library encompassing diverse PAM sequences [4]. Integration products are detected through PCR amplification of donor-target junctions using orientation-specific primers, followed by next-generation sequencing to determine integration precision, PAM preferences, and product purity. This assay confirmed that 90% of integration events for MG64-1 and MG64-6 Type V-K CAST systems occurred between 57-67 base pairs from the PAM sequence [4].

Genomic Integration in E. coli

For prokaryotic validation, CAST components are delivered via multiple plasmids (encoding proteins, guide RNA, and donor DNA) into engineered E. coli strains under antibiotic selection [4]. The resulting colonies are pooled and analyzed through probe-based qPCR and whole genome sequencing to quantify on-target efficiency, assess off-target events, and characterize integration structures (single integration versus co-integration). This approach demonstrated that Type V-K CAST systems can achieve up to 80% integration efficiency at engineered loci with minimal off-target effects (<7%) [4].

Mammalian Cell Integration

Mammalian cell integration requires additional engineering considerations. CAST components must be optimized with nuclear localization signals and codon-optimized for eukaryotic expression [4]. Delivery can be achieved through plasmid transfection, in vitro transcribed mRNA, or ribonucleoprotein (RNP) complexes. Integration efficiency is typically assessed using reporter systems (e.g., EGFP) or therapeutic genes targeted to safe harbor loci (e.g., AAVS1), with outcomes measured by flow cytometry, droplet digital PCR, and next-generation sequencing to quantify precise integration and detect potential off-target events [4].

Research Reagent Solutions: Essential Materials for CAST Engineering

Table 3: Key Research Reagents for CAST System Experiments

Reagent Category	Specific Examples	Function and Application
CAST Expression Plasmids	PseCAST (Type I-F), MG64-1 (Type V-K)	Source of codon-optimized CAST proteins for mammalian expression
Guide RNA Scaffolds	Optimized sgRNA for Type V-K	Programmable targeting; can be truncated to 80% of native length without losing activity [4]
Donor Templates	Linear fragments with TIRs; plasmid donors	Cargo for integration; terminal inverted repeats (TIRs) can be reduced by 50% while maintaining function [4]
Delivery Vehicles	Lipid nanoparticles; AAV vectors; electroporation	Introduction of CAST components into cells; format-dependent efficiency optimization
Host Factors	ClpX (for Type I-F); S15 (for Type V-K)	Enhance integration efficiency in non-native environments [9] [4]
Analytical Tools	NGS platforms; ddPCR; flow cytometry	Quantification of integration efficiency, specificity, and cargo integrity

CAST systems represent a transformative approach to kilobase-scale genome editing, offering distinct advantages for therapeutic gene insertion compared to traditional technologies. Type I-F systems provide highly specific integration with a cut-and-paste mechanism but require complex multi-component delivery. Type V-K systems offer a compact architecture with single-protein targeting but may produce co-integration events without TnsA engineering. Current research focuses on enhancing editing efficiency through structural guidance and directed evolution, refining specificity to minimize off-target integration, and optimizing delivery strategies for therapeutic applications. As these engineering challenges are addressed, CAST systems are poised to expand the therapeutic landscape for genetic disorders requiring the insertion of large DNA sequences, potentially enabling treatments for conditions that have remained intractable to previous generations of genome editing technologies.

CRISPR-associated transposases (CASTs) represent a powerful new class of genome-editing tools that combine the programmability of CRISPR systems with the DNA integration capability of transposases. Unlike conventional CRISPR-Cas systems that rely on DNA double-strand breaks (DSBs) and cellular repair mechanisms, CASTs enable DSB-free integration of large DNA cargoes, potentially exceeding 10 kilobases in size [9] [17]. Among the diverse CAST systems identified, Type I-F and Type V-K have emerged as the most promising for biotechnological applications, yet they differ fundamentally in their integration mechanisms and resulting editing outcomes.

The critical distinction between these systems lies in their molecular composition: Type I-F systems contain the transposase proteins TnsA and TnsB, enabling true "cut-and-paste" transposition, while Type V-K systems lack TnsA, resulting in different integration byproducts [4] [8]. This structural difference has direct implications for product purity, with Type I-F systems typically producing homogeneous, unidirectional integrations, whereas Type V-K systems often generate co-integration byproducts where vector backbone sequences are incorporated alongside the intended cargo [4]. Understanding these mechanistic differences is essential for researchers selecting the appropriate CAST system for specific genome engineering applications.

Comparative Integration Mechanisms and Outcomes

Structural Basis for Divergent Integration Strategies

The fundamental difference in integration outcomes between Type I-F and Type V-K CAST systems originates from their distinct molecular architectures. Type I-F systems utilize a multi-subunit Cascade complex (comprising Cas8, Cas7, and Cas6 proteins) for DNA targeting, which associates with a TniQ dimer to recruit downstream transposition components [9] [8]. This Cascade complex recognizes the protospacer adjacent motif (PAM) and facilitates R-loop formation, subsequently recruiting the AAA+ ATPase TnsC, which acts as a bridge to the transposase machinery [8].

Most importantly, Type I-F systems encode both TnsA and TnsB transposases, which function together in a concerted mechanism. TnsB catalyzes cleavage at the 3' ends of the transposon, while TnsA cleaves the 5' ends, enabling complete excision of the donor DNA and its clean integration into the target site [8]. This complete excision prevents the incorporation of non-cargo sequences and ensures high product purity.

In contrast, Type V-K systems employ a significantly simpler architecture, utilizing a single Cas12k effector for DNA targeting along with TniQ [4] [8]. While this compact organization offers advantages for delivery, Type V-K systems notably lack the TnsA protein. Without TnsA, these systems cannot cleave both strands of the donor DNA, leading to incomplete excision from the donor plasmid and resulting in co-integration events where vector backbone sequences are incorporated alongside the intended cargo [4].

Quantitative Comparison of Integration Outcomes

The table below summarizes key differences in integration outcomes between Type I-F and Type V-K CAST systems based on experimental data from multiple studies:

Parameter	Type I-F CAST	Type V-K CAST
Integration Mechanism	Cut-and-paste	Copy-in (without TnsA)
Primary Product	Unidirectional, precise integration	Mixed outcomes: simple insertion & co-integration
Co-integration Frequency	Minimal (system-dependent)	70-80% (with circular plasmid donor) [4]
Simple Insertion Frequency	High (system-dependent)	20-30% (with circular plasmid donor) [4]
Product Purity	Highly homogeneous mixtures [9]	Heterogeneous mixtures
Directionality	Strong bias for one orientation [14]	Variable orientation
Byproduct Formation	Minimal	Significant (plasmid backbone insertion)
Off-target Integration	Low levels in evolved systems [14]	~7% in optimized systems [4]

Table 1: Comparative analysis of integration outcomes between Type I-F and Type V-K CAST systems

Experimental Evidence and Validation

Studies with the PseCAST (Type I-F) system demonstrated its ability to perform highly specific, DSB-free DNA integration in human cells with minimal byproduct formation [9]. When researchers applied phage-assisted continuous evolution (PACE) to this system, they developed evoCAST, which achieved 10-25% integration efficiencies of kilobase-sized DNA cargoes across 14 tested genomic loci in HEK293T cells while generating "predominately unidirectional cut-and-paste transposition products" and no detected indels [14].

For Type V-K systems, characterization of the MG64-1 and MG64-6 systems revealed that both single (20-30%) and co-integration (70-80%) events occur when using a circular plasmid donor [4]. This co-integration byproduct occurs because "the absence of TnsA for second-strand donor cleavage" prevents complete transposon excision, leading to incorporation of plasmid backbone sequences alongside the intended cargo [4].

Experimental Approaches for Characterization

In Vitro Integration Assays

The integration specificity and byproduct profiles of CAST systems are typically first characterized using in vitro integration assays. These assays involve incubating purified CAST proteins (expressed and purified from E. coli) with guide RNA, a linear donor fragment containing the transposon cargo, and a target plasmid library containing diverse PAM sequences [4].

Key steps in this protocol include:

• Donor-Target Junction PCR: Following integration reactions, PCR amplification is performed targeting each donor-target junction for both possible integration orientations [4].
• Next-Generation Sequencing (NGS): PCR products from successful integration events are sequenced via NGS to determine PAM preferences and integration precision [4].
• Byproduct Analysis: Co-integration events are identified through sequencing by detecting vector backbone sequences flanking the integrated cargo.

This approach enabled researchers to determine that 90% of integration events for MG64-1 and MG64-6 Type V-K systems occurred between 57 and 67 base pairs away from the PAM [4].

Genomic Integration Efficiency Assessments

To evaluate CAST performance in biological systems, researchers employ bacterial and human cell assays:

E. coli Genomic Integration Protocol:

• Plasmid Transformation: Three separate plasmids containing protein-coding components, the single guide RNA, and donor DNA are transformed into an engineered E. coli strain [4].
• Selection and Pooling: Transformants are maintained on triple antibiotic selection plates, then pooled for sequencing analysis [4].
• Efficiency Quantification: Probe-based qPCR and unbiased whole genome sequencing are used to analyze the population of genomes for on- and off-target integration frequencies [4].

Human Cell Integration Protocol:

• Component Delivery: CAST components are delivered to HEK293T cells via transfection, typically using an all-in-one mRNA format for better delivery efficiency [17].
• Efficiency Measurement: Integration efficiency is quantified using targeted NGS of the insertion site, with evoCAST reporting efficiencies of 10-25% across multiple genomic loci [14].
• Byproduct Analysis: PCR assays specifically designed to detect co-integration events or misoriented insertions are employed to quantify product purity.

Diagram 1: CAST integration mechanisms and outcomes

The Scientist's Toolkit: Essential Research Reagents

The table below outlines key reagents and methodologies employed in CAST system engineering and characterization:

Reagent/Method	Function in CAST Research	Application Examples
Phage-Assisted Continuous Evolution (PACE)	Accelerated evolution of transposase efficiency	Developed evoCAST with ~200-fold improved activity [14]
High-Throughput Mutational Screening	Simultaneous quantification of CAST variant activity and specificity	Identified mutations improving V-K CAST activity 5-fold [47]
Nuclear Localization Signal (NLS) Tags	Directs bacterial-derived proteins to mammalian nucleus	Essential for CAST function in human cells [4]
Single-Guide RNA (sgRNA) Designs	Programmable targeting of CAST integration	Enables site-specific integration in human genomes [4]
Cryo-Electron Microscopy (Cryo-EM)	Structural determination of CAST complexes	Revealed PAM recognition and TnsC recruitment mechanisms [9]
Whole Genome Sequencing	Unbiased detection of on- and off-target integration	Quantified ~7% off-target rates in optimized V-K systems [4]

Table 2: Essential research reagents and methods for CAST system engineering

The choice between Type I-F and Type V-K CAST systems involves significant trade-offs between product purity and practical implementation. Type I-F systems, particularly evolved variants like evoCAST, offer superior product purity with predominantly unidirectional integration and minimal byproducts, making them ideal for therapeutic applications where homogeneous editing outcomes are critical [14]. However, their multi-component nature presents delivery challenges that must be addressed for clinical translation.

Type V-K systems provide a compact architecture with single-protein targeting through Cas12k, offering advantages for viral packaging and delivery [4]. Nevertheless, their tendency for co-integration byproducts represents a significant limitation for applications requiring precise gene integration. Recent engineering efforts have improved their specificity, with some optimized systems showing off-target rates below 7% [4].

Future directions in CAST engineering will likely focus on combining the favorable attributes of both systems—developing compact architectures that maintain high-fidelity integration—while addressing delivery challenges through continued protein engineering and evolution. As these technologies mature, CAST systems promise to expand the therapeutic landscape for genetic diseases requiring large gene insertion, potentially offering one-time, mutation-agnostic treatments for diverse loss-of-function disorders [14] [17].

CRISPR-associated transposases (CASTs) represent a significant advancement in genome editing technology by enabling RNA-guided, site-specific integration of large DNA sequences without relying on double-strand break (DSB) formation. Unlike conventional CRISPR-Cas systems that create DSBs and depend on endogenous cellular repair mechanisms, CAST systems combine nuclease-deficient CRISPR effectors with transposase enzymes to catalyze precise "cut-and-paste" or "copy-and-paste" integration of genetic cargo. This capability positions CAST systems as particularly promising tools for therapeutic applications requiring the insertion of full gene sequences, such as the treatment of monogenic diseases. The two most extensively characterized CAST families—type I-F and type V-K—diverge significantly in their molecular architecture, DNA targeting mechanisms, and editing outcomes, leading to distinct specificity profiles that warrant systematic comparison for research and therapeutic development [17].

The burgeoning interest in CAST systems stems from their potential to overcome fundamental limitations of current genome editing technologies. While CRISPR-Cas9, base editing, and prime editing have revolutionized genetic manipulation, they face challenges in efficiently inserting large DNA sequences (>1 kb) with high precision and minimal byproducts. CAST systems address this unmet need by providing a single-step integration mechanism that operates independently of host repair pathways, thus bypassing the inefficiencies of homology-directed repair (HDR) and the unpredictability of non-homologous end joining (NHEJ) [14]. As these systems transition from bacterial contexts to human cell applications, understanding their specificity profiles—encompassing both on-target fidelity and genome-wide off-target activity—becomes paramount for assessing their therapeutic potential and guiding further engineering efforts.

Molecular Architecture and Targeting Mechanisms

Type I-F CAST Systems

Type I-F CAST systems employ a multi-subunit Cascade (CRISPR-associated complex for antiviral defense) complex for DNA recognition and targeting. This complex comprises several Cas proteins arranged in a specific stoichiometry: one Cas8 subunit, six Cas7 subunits, one Cas6 subunit, and a dimer of TniQ adaptor proteins. The Cas8 subunit recognizes the protospacer adjacent motif (PAM) at the target DNA site, while the Cas7 subunits form a helical backbone that stabilizes the crRNA-DNA heteroduplex. The Cas6 protein processes the crRNA and anchors the TniQ dimer, which serves as a bridge between the DNA recognition complex and the transposase machinery [9].

Recent structural insights into the PseCAST system (a type I-F CAST from Pseudoalteromonas sp.) revealed unexpected dynamics within the QCascade complex. Cryo-EM analyses demonstrated that the TniQ dimer populates a range of positions relative to other complex components, pivoting around Cas6 and Cas7.6 in both "open" and "closed" conformations. This structural flexibility may influence target site selection and integration efficiency. The Cas8 subunit features two domains: a bulky domain that interacts with Cas7.1 and binds the crRNA 5' end and PAM sequence, and a second α-helical domain that exhibits dynamic behavior and appears essential for RNA-guided DNA integration activity [9].

Type V-K CAST Systems

In contrast to the multi-protein Cascade complex of type I-F systems, type V-K CAST systems utilize a single Cas12k effector protein for DNA targeting, resulting in a substantially more compact molecular architecture. Cas12k recognizes the target DNA sequence through guide RNA complementarity and PAM interaction, while simultaneously associating with TniQ and the transposase components. This simplified targeting mechanism offers practical advantages for therapeutic delivery, as the coding sequence for Cas12k is significantly smaller than the multi-gene cascade complex of type I-F systems [4] [17].

Structural studies of the S. hofmannii CAST (ShCAST) system have revealed that the Cas12k-TniQ complex recruits TnsC, which forms helical filaments on double-stranded DNA. These filaments serve as platforms for recruiting the TnsB transposase, which catalyzes the integration of the donor DNA. Unlike type I-F CASTs, type V-K systems lack the TnsA subunit and therefore mobilize DNA through a copy-and-paste mechanism that produces cointegrate products rather than simple insertions [55].

Table 1: Comparative Molecular Architectures of CAST Systems

Characteristic	Type I-F CAST	Type V-K CAST
Targeting Complex	Multi-subunit Cascade (Cas6/7/8, TniQ dimer)	Single Cas12k effector with TniQ
PAM Preference	5'-CC-3' (PseCAST) [9]	5'-GTN-3' (MG64-1) or 5'-rGTN-3' (MG64-6) [4]
Transposase Components	TnsA, TnsB, TnsC	TnsB, TnsC (lacks TnsA)
Integration Mechanism	Cut-and-paste [14]	Copy-and-paste [55]
Integration Directionality	Unidirectional [14]	Unidirectional [55]
Coding Sequence Size	~8 kb [9]	~5 kb [9]

DNA Recognition and Integration Workflow

The following diagram illustrates the core mechanisms and structural differences between type I-F and type V-K CAST systems:

On-Target Integration Fidelity and Efficiency

Integration Efficiency in Human Cells

A critical metric for evaluating CAST system performance is their efficiency in achieving targeted integration in human cells. Early wild-type CAST systems demonstrated minimal activity in human cells, limiting their therapeutic utility. However, recent protein engineering efforts have yielded substantial improvements. The wild-type PseCAST (type I-F) system initially showed <0.1% integration efficiency in human cells, which could be modestly improved to approximately 1% with the addition of the bacterial unfoldase ClpX, though with associated cytotoxicity [14].

Through phage-assisted continuous evolution (PACE), researchers developed an evolved CAST (evoCAST) system with dramatically enhanced performance. The evolved transposase variants achieved an average 200-fold improvement in integration activity compared to wild-type PseCAST, culminating in 10-25% integration efficiencies of kilobase-sized DNA cargos across 14 tested genomic loci in HEK293T cells. This enhanced efficiency occurred without requiring ClpX, reducing cellular toxicity [14]. Similarly, engineered type V-K CAST systems have demonstrated the capability to integrate therapeutically relevant transgenes, such as the full-length Factor IX gene (relevant for hemophilia B), into safe harbor loci like AAVS1 in multiple human cell types [4] [17].

Product Purity and Byproduct Formation

The presence of undesirable editing byproducts represents a significant concern for therapeutic genome editing applications. Type I-F CAST systems typically exhibit high product purity due to their cut-and-paste transposition mechanism mediated by TnsA and TnsB. The evoCAST system, for instance, generates predominantly unidirectional transposition products without detected indel formation at the target site [14]. This high-fidelity integration profile stems from the coordinated activity of TnsA, which cleaves the 5' strands, and TnsB, which cleaves the 3' strands of the transposon ends, resulting in clean excision and insertion events.

In contrast, type V-K CAST systems lack TnsA and consequently produce a mixture of integration outcomes. When delivered as circular plasmid donors, these systems yield both simple insertions (20-30%) and co-integration events (70-80%), where the entire donor plasmid integrates into the target site alongside the transposon cargo [4]. This heterogeneity complicates their therapeutic application, as co-integrate byproducts may include antibiotic resistance genes and plasmid backbone sequences that could potentially disrupt transgene expression or regulatory elements.

Table 2: Quantitative Comparison of CAST System Performance

Performance Metric	Type I-F CAST	Type V-K CAST
Theoretical Cargo Capacity	Multi-kilobase [14]	Multi-kilobase [17]
Reported Efficiency in Human Cells	10-25% (evoCAST) [14]	Single-digit percentages (engineered systems) [4]
Product Purity	High (>90% precise products) [14]	Moderate (20-30% simple insertions) [4]
On-target Integration Specificity	>99% of total editing events (HiFi variant) [56]	12-76% (varies by guide RNA) [55]
Indel Formation at Target Site	Undetectable [14]	Not comprehensively reported

Genome-Wide Off-Target Analysis

Off-Target Profiles and Molecular Mechanisms

Comprehensive genome-wide analyses reveal fundamental differences in the off-target integration behaviors between type I-F and type V-K CAST systems. Type I-F systems demonstrate exceptional target specificity in cellular contexts. A landmark study performing whole-genome sequencing of single-cell-derived human hematopoietic stem and progenitor cell (HSPC) clones edited with Cas9 ribonucleoprotein complexes found that the collective somatic mutational burden in edited clones was indistinguishable from naturally occurring background genetic heterogeneity [57]. Statistical analysis revealed no significant difference in the number of novel non-targeted indels between Cas9-treated and control samples, and no evidence of Cas9-mediated indel formation at 623 predicted off-target sites [57].

Type V-K CAST systems, however, demonstrate a more complex off-target profile due to their capacity for dual-pathway transposition. Research on the ShCAST system revealed that these systems undergo both RNA-dependent targeted transposition and RNA-independent untargeted transposition [55]. The RNA-independent pathway requires only TnsB, TnsC, and TniQ (the "BCQ" pathway) and exhibits remarkable bias for AT-rich genomic regions. Cryo-EM structural analysis of this untargeted transpososome revealed a TnsB-TnsC-TniQ complex that encompasses two turns of a TnsC filament and otherwise resembles major architectural aspects of the Cas12k-containing targeted transpososome [55].

Methodologies for Off-Target Assessment

Robust assessment of CAST system specificity requires specialized methodologies capable of detecting both targeted and untargeted integration events across the genome. The following diagram illustrates a representative workflow for genome-wide off-target analysis:

Established methods for off-target assessment include both biochemical approaches (e.g., CIRCLE-seq, CHANGE-seq) that use purified genomic DNA and engineered nucleases to map potential cleavage sites in vitro, and cellular methods (e.g., GUIDE-seq, DISCOVER-seq) that assess nuclease activity directly in living or fixed cells to capture biological context [58]. For comprehensive evaluation of CAST systems, whole-genome sequencing of single-cell-derived clones provides the most unambiguous assessment of off-target integration events, as demonstrated in HSPC studies [57].

The selection of appropriate off-target assessment methods should consider their respective strengths and limitations. Biochemical methods offer ultra-sensitive detection of potential cleavage sites but may overestimate biologically relevant off-target editing due to the absence of chromatin structure and cellular repair mechanisms. Cellular methods provide superior biological relevance by capturing the influence of nuclear context but require efficient delivery of both nuclease and detection reagents [58].

Experimental Protocols for Specificity Assessment

Genome-Wide Specificity Profiling Using Whole-Genome Sequencing

The most definitive method for assessing CAST system specificity involves whole-genome sequencing (WGS) of single-cell-derived clones, which provides an unbiased assessment of off-target integration events and other unintended mutations. The following protocol has been successfully applied to human hematopoietic stem and progenitor cells (HSPCs) [57]:

• Cell Preparation and CAST Delivery: Electroporate HSPCs with CAST ribonucleoprotein (RNP) complexes targeted to relevant genomic loci (e.g., CXCR4 on chromosome 2 or AAVS1 on chromosome 19).

• Single-Cell Cloning: Following editing, isolate and expand single-cell-derived clones to establish pure populations for analysis. Include non-CAST-treated control clones to establish baseline mutation rates.

• Library Preparation and Sequencing: Extract high-molecular-weight genomic DNA and prepare sequencing libraries using standardized WGS protocols. Sequence to sufficient coverage (typically 30-50x) to confidently detect somatic variants.

• Bioinformatic Analysis:

  • Identify somatic variants (indels, single nucleotide variants, and structural variants) by comparing CAST-treated clones to control clones.
  • Perform statistical analysis to determine whether any increase in variant number reaches significance compared to baseline.
  • Examine predicted off-target sites for CAST-mediated indel formation.
  • Assess structural variants for any causal connection to CAST editing procedures.

This approach typically identifies >20,000 total somatic variants distributed among CAST-treated and control clones, enabling robust statistical comparison [57].

In Vitro Specificity Assessment Using Biochemical Methods

Biochemical methods such as CHANGE-seq (Circularization for High-throughput Analysis of Nuclease Genome-wide Effects by Sequencing) provide sensitive, genome-wide off-target profiling without cellular influences [58]:

• Genomic DNA Preparation: Extract and purify genomic DNA from relevant cell types (microgram quantities required).

• CAST Complex Assembly: Incube purified genomic DNA with assembled CAST complexes under optimized reaction conditions.

• Library Construction:

  • Digest DNA with CAST complexes and blunt-end the resulting fragments.
  • Add adapters with T7 promoters to ends of DNA fragments.
  • Circularize DNA fragments and digest linear DNA with exonucleases.
  • Perform in vitro transcription and reverse transcription to generate sequencing libraries.

• Sequencing and Analysis: Sequence libraries and map reads to the reference genome to identify cleavage sites. CHANGE-seq can detect rare off-targets with reduced false negatives compared to earlier methods [58].

Research Reagent Solutions for CAST Studies

Table 3: Essential Research Reagents for CAST System Evaluation

Reagent Category	Specific Examples	Research Application
High-Fidelity CAST Variants	Alt-R HiFi Cas9 Nuclease V3 [56], evoCAST [14]	Enhanced specificity with reduced off-target effects
Off-Target Detection Kits	GUIDE-seq, CHANGE-seq, DISCOVER-seq kits [58]	Genome-wide identification of off-target integration events
Control Templates	Validated positive control gRNAs, Synthetic target DNA with known off-target sites	Assay standardization and cross-experiment comparison
Sequence Analysis Tools	Cas-OFFinder, CRISPOR, CCTop [58]	In silico prediction of potential off-target sites during guide design
Delivery Reagents	RNP electroporation kits [57], Lipid nanoparticles [17]	Efficient intracellular delivery of CAST components

Discussion and Future Perspectives

The comparative analysis of type I-F and type V-K CAST systems reveals a fundamental trade-off between editing efficiency and target specificity. While engineered type I-F systems such as evoCAST achieve therapeutically relevant integration efficiencies (10-25%) with minimal off-target activity, type V-K systems offer advantages in delivery feasibility due to their more compact coding size but require further optimization to improve their specificity profiles. The discovery of RNA-independent transposition pathways in type V-K systems represents a particular challenge for therapeutic applications, though modulation of TnsC availability has shown promise in suppressing these untargeted events [55].

Future directions for CAST system development will likely focus on enhancing both efficiency and specificity through continued protein engineering, optimization of delivery modalities, and refinement of off-target assessment methodologies. The implementation of enrichment strategies—such as selective markers, phenotypic screening, or physical separation methods—may further improve the recovery of correctly edited cells, particularly for applications where native editing efficiencies remain limiting [59]. As the field progresses toward clinical translation, standardized approaches for genome-wide off-target assessment will be essential for rigorous evaluation of CAST system safety and specificity [58].

The promising specificity profiles of evolved CAST systems, particularly their capacity for DSB-free integration of large DNA cargoes with minimal indel formation, position them as compelling tools for next-generation therapeutic genome editing. With continued refinement, CAST systems may overcome longstanding challenges in gene therapy, enabling precise gene insertion strategies for treating diverse genetic disorders while mitigating the safety concerns associated with conventional CRISPR-Cas nucleases.

Conclusion

The comparative analysis reveals that Type I-F and Type V-K CAST systems offer distinct and complementary profiles for therapeutic genome engineering. Type I-F systems, particularly evolved variants like evoCAST, currently lead in achieving high-efficiency (10-25%), precise, unidirectional integration of large DNA cargos with minimal byproducts, making them ideal for applications demanding high product purity. In contrast, the compact, single-effector architecture of Type V-K systems offers a significant advantage for delivery, though it may require further engineering to match the efficiency and specificity of optimized Type I-F systems. Future directions will involve refining delivery strategies, such as lipid nanoparticles or novel viral vectors, to accommodate these large molecular machines for in vivo therapy. Furthermore, continuous protein evolution and deeper structural insights will unlock the next generation of CAST systems, solidifying their role as indispensable tools for one-time, mutation-agnostic treatments of loss-of-function genetic diseases and advancing the frontiers of synthetic biology.

References

• 1. Structure-guided engineering of type I-F CASTs for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12375015/]
• 2. Structural basis of target DNA recognition by CRISPR ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8571069/]
• 3. Structural basis for the assembly of the type V CRISPR- ... [https://www.sciencedirect.com/science/article/pii/S0092867422014210]
• 4. Integration of therapeutic cargo into the human genome ... [https://www.nature.com/articles/s41467-025-57416-2]
• 5. Mechanism of target site selection by type V-K CRISPR- ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10771339/]
• 6. CRISPR Plasmids - CRISPR Transposases (CAST) [https://www.addgene.org/crispr/CAST/]
• 7. Distinct horizontal transfer mechanisms for type I and ... [https://www.nature.com/articles/s41467-024-50816-w]
• 8. Unlocking the potential of CRISPR-associated transposons [https://www.sciencedirect.com/science/article/abs/pii/S0168952525000800]
• 9. Structure-guided engineering of type I-F CASTs for ... [https://www.nature.com/articles/s41467-025-63164-0]
• 10. Structural basis for CRISPR RNA-guided DNA recognition ... [https://www.nature.com/articles/nsmb.2019]
• 11. R-loop formation and conformational activation ... [https://www.nature.com/articles/s41586-022-05114-0]
• 12. Article Conformational landscape of the type VK CRISPR ... [https://www.sciencedirect.com/science/article/pii/S1097276524003988]
• 13. R-loop generation during transcription: formation ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6340742/]
• 14. Continuous evolution of CRISPR-associated transposases ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12326709/]
• 15. DNA on the move: mechanisms, functions and applications ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10761935/]
• 16. Unlocking the potential of CRISPR-associated transposons [https://pubmed.ncbi.nlm.nih.gov/40393858/]
• 17. Beyond CRISPR: What CAST could bring to genome editing [https://www.labiotech.eu/trends-news/cast-genome-editing/]
• 18. Structure of TnsABCD transpososome reveals mechanisms ... [https://www.sciencedirect.com/science/article/abs/pii/S0092867424010717]
• 19. Structure of the TnsB transposase-DNA complex of type ... [https://www.nature.com/articles/s41467-022-33504-5]
• 20. The Tn7 transposition regulator TnsC interacts with the ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4104917/]
• 21. Transposase-assisted target-site integration for efficient ... [https://www.nature.com/articles/s41586-024-07613-8]
• 22. Structure-guided engineering of type I-F CASTs for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11429998/]
• 23. CASTing Off for New Shores in Human Genome Editing [https://blog.addgene.org/casting-off-for-new-shores-in-human-genome-editing]
• 24. New screening method boosts efficiency and accuracy of ... [https://www.news-medical.net/news/20250923/New-screening-method-boosts-efficiency-and-accuracy-of-CAST-genome-editing.aspx]
• 25. CRISPR Delivery Methods: Cargo, Vehicles, and Challenges [https://www.synthego.com/blog/delivery-crispr-cas9/]
• 26. Hairpin Internal Nuclear Localization Signals in CRISPR- ... [https://pubmed.ncbi.nlm.nih.gov/40163415/]
• 27. Evaluation of the correlation between nuclear localization ... [https://www.sciencedirect.com/science/article/pii/S0022354924004799]
• 28. - Phage - Wikipedia assisted continuous evolution [https://en.wikipedia.org/wiki/Phage-assisted_continuous_evolution]
• 29. - Phage and non- assisted continuous continuous evolution [https://www.nature.com/articles/s41596-020-00410-3?error=cookies_not_supported&code=0eeec181-81c5-42e6-8e86-403b49555297]
• 30. Directed evolution of hyperactive integrases for site specific insertion... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11317131/]
• 31. Enhanced Factor IX Activity following Administration of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6834974/]
• 32. Generation of a novel factor IX with augmented clotting ... [https://www.sciencedirect.com/science/article/pii/S1538783622068076]
• 33. Targeted delivery of diverse biomolecules with engineered ... [https://www.nature.com/articles/s41587-025-02774-x]
• 34. Large-Scale Purification of Factor-IX: Comparing Two ... [https://www.bioprocessintl.com/chromatography/large-scale-purification-of-factor-ix-comparing-two-affinity-chromatography-resins]
• 35. Efficiency and Specificity of Targeted Integration Mediated ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6016731/]
• 36. Gene regulation technologies for gene and cell therapy [https://www.sciencedirect.com/science/article/pii/S1525001625002783]
• 37. Efficient site-specific integration of large genes in ... [https://www.nature.com/articles/s41551-024-01227-1]
• 38. A Visual Guide to Genome Editors [https://press.asimov.com/articles/crispr-guide]
• 39. Mapping the therapeutic landscape of CRISPR-Cas9 for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12006052/]
• 40. Emerging trends in prime editing for precision genome ... [https://www.nature.com/articles/s12276-025-01463-8]
• 41. Design of highly functional genome editors by modelling ... [https://www.nature.com/articles/s41586-025-09298-z]
• 42. Advances in large-scale DNA engineering with the ... [https://www.nature.com/articles/s12276-025-01530-0]
• 43. CRISPR-Associated Transposase for Targeted ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10367135/]
• 44. Structure of the TnsB transposase-DNA complex of type V-K ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9527255/]
• 45. CRISPR-associated transposons - Wikipedia [https://en.wikipedia.org/wiki/CRISPR-associated_transposons]
• 46. Structures of the holo CRISPR RNA-guided transposon ... [https://www.nature.com/articles/s41586-022-05573-5]
• 47. Screening approach enhances CRISPR genome-editing ... [https://www.stjude.org/media-resources/news-releases/2025-medicine-science-news/screening-approach-enhances-crispr-genome-editing-efficiency.html]
• 48. Off-target effects in CRISPR/Cas9 gene editing - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC10034092/]
• 49. Viral Vectors for the in Vivo Delivery of CRISPR Components [https://pmc.ncbi.nlm.nih.gov/articles/PMC9133430/]
• 50. innovations and challenges in rAAV vector-based CRISPR ... [https://www.nature.com/articles/s41434-025-00573-2]
• 51. Advances in large-scale DNA engineering with the ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12508223/]
• 52. Genome editing using CRISPR, CAST, and Fanzor systems [https://pmc.ncbi.nlm.nih.gov/articles/PMC11278801/]
• 53. CRISPR Gene Therapy: Applications, Limitations, and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7427626/]
• 54. Evaluation of the loading capacity and patterns ... [https://www.sciencedirect.com/science/article/pii/S2329050125000695]
• 55. Mechanism of target site selection by type V-K CRISPR ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10370016/]
• 56. High-fidelity Cas9 for efficient genome editing [https://www.idtdna.com/page/support-and-education/decoded-plus/high-fidelity-cas9-provides-highly-efficient-genome-editing-with-reduced-off-target-effects-even-in-primary-cells/]
• 57. Genome-Wide Analysis of Off-Target CRISPR/Cas9 Activity in ... [https://pubmed.ncbi.nlm.nih.gov/33322084/]
• 58. BLOG: Selecting the Right Gene Editing Off-Target Assay [https://seqwell.com/guide-to-selecting-right-gene-editing-off-target-assay/]
• 59. Enrichment strategies to enhance genome editing | Journal of Biomedical Science | Full Text [https://jbiomedsci.biomedcentral.com/articles/10.1186/s12929-023-00943-1]