A Programmable Dual-RNA–Guided DNA Endonuclease in Adaptive Bacterial Immunity: A Contemporary Review

1. Department of Cardiology, All India Institute of Medical Sciences (AIIMS), New Delhi, India.
2. Department of Internal Medicine, King George’s Medical University, Lucknow, India.

* Correspondence: amaaiims@gmail.com

Keywords: CRISPR-Cas9, genome editing, base editing, prime editing, off-target, clinical trials, ethics

1. INTRODUCTION

The report by [1] that a bacterial CRISPR-associated protein (Cas9) could be guided by a dual-RNA (tracrRNA: crRNA) and engineered single-guide RNA (sgRNA) to catalyze site-specific double-strand DNA cleavage represented a pivotal advance in molecular genetics. Prior genome-editing nucleases—zinc-finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALENs)—required laborious protein engineering to retarget DNA cleavage. The programmable RNA-guided mechanism of Cas9 simplified target design to an easily synthesized RNA sequence and enabled rapid adoption across model systems and species [2]. In the decade since that publication, the CRISPR-Cas9 platform evolved from a laboratory tool to a clinical technology. Innovations have expanded function beyond cutting—enabling single-base changes without double-strand breaks (base editing), precise sequence replacement (prime editing), transcriptional regulation, and epigenetic modulation—while the community developed methods to detect and mitigate off-target effects and to deliver editors to clinically relevant tissues. These advances spurred human clinical trials for in-body and ex-vivo editing approaches, with early results demonstrating both promise and the need for careful assessment of efficacy, durability, and safety. This review has three aims: (1) to summarize the mechanistic and experimental contributions of [3] that enabled programmable editing; (2) to synthesize subsequent technological advances (detection of off-targets, base/prime editing, delivery systems) and their translational implications; and (3) to evaluate current clinical applications, regulatory/ethical challenges, and future directions. Key papers and clinical results are cited to provide an evidence-based perspective for researchers and clinicians.

2. THE JINEK CONTRIBUTION: MECHANISM AND DEMONSTRATION

CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) loci, together with CRISPR-associated (Cas) genes, form adaptive immune systems in bacteria and archaea that capture short fragments of foreign nucleic acids and use them to guide nucleases to matching sequences on subsequent invasions [4]. Jinek and colleagues purified Streptococcus pyogenes Cas9 and demonstrated that Cas9 requires two RNAs—a CRISPR RNA (crRNA) containing the target-complementary guide sequence and a trans-activating crRNA (tracrRNA)—to direct double-strand cleavage at sites adjacent to a short protospacer adjacent motif (PAM). They also showed that a synthetic fusion of tracrRNA and crRNA into a single-guide RNA (sgRNA) retained activity, simplifying the system and making it readily programmable for diverse target sites. Jinek characterized domain activities within Cas9—showing that distinct nuclease domains (HNH and RuvC-like) cleave complementary and non-complementary strands respectively—and mapped the PAM dependency that enforces target specificity. These mechanistic insights explained how a single protein could catalyze precise double-strand breaks at user-defined sites simply by changing the short guide RNA sequence. The engineered sgRNA made the system readily adoptable by laboratories without specialized protein-engineering expertise, a practical pivot that accelerated its dissemination.

3. RAPID ADOPTION AND EXPANSION: EARLY TECHNICAL AND CONCEPTUAL PROGRESS

Within the year after 2012, multiple groups adapted Cas9 for genome modification in bacteria, yeast, plants, and mammalian cells, showing that programmable nuclease activity could induce mutations via endogenous DNA repair pathways—classically non-homologous end joining (NHEJ) for indels and homology-directed repair (HDR) for templated changes [5]. The ease of reprogramming target specificity made CRISPR-Cas9 a universal gene-perturbation tool, enabling rapid genetic screens and functional genomics in cell lines and organisms previously difficult to manipulate. As use expanded, investigators observed that Cas9 can tolerate mismatches between guide RNA and genomic DNA at certain positions, producing off-target cleavage. This motivated systematic development of unbiased, genome-wide detection methods such as GUIDE-seq [6], which captures double-strand break sites via integration of a short oligonucleotide and maps them by sequencing, and other assays (Digenome-seq, CIRCLE-seq, SITE-Seq), enabling more comprehensive profiling of off-target activity and informing design rules to minimize unwanted edits [7]. These detection methods became essential for preclinical evaluation of therapeutic candidates.

4. ENGINEERING GREATER PRECISION: HIGH-FIDELITY NUCLEASES AND NEW EDITING MODALITIES

Protein engineering produced Cas9 variants with reduced off-target cleavage while maintaining on-target efficiency (for example, “enhanced specificity” and “high-fidelity” SpCas9 variants). Rational and directed-evolution approaches altered DNA-binding energetics and guide interactions to increase discrimination, which improved safety margins for therapeutic development [8]. Combining improved guide design algorithms with engineered nucleases markedly reduced off-target burdens measured by GUIDE-seq and related methods. Base editors, developed beginning with cytosine base editors (CBEs) and later adenine base editors (ABEs), fuse a catalytically impaired Cas9 (nickase or dead Cas9) to a nucleotide deaminase to perform direct single-base conversions. These tools convert C→T (or G→A) or A→G (or T→C) within a small editing window without generating double-strand breaks, reducing indels and improving product purity for many clinically relevant single-base corrections [9].

5. PRIME EDITING: EXPANDED PROGRAMMABILITY AND PRECISION

Prime editing, described by [10], couples a Cas9 nickase to an engineered reverse transcriptase and a prime editing guide RNA (pegRNA) that both specifies the target and encodes the desired edit. Prime editors can install small insertions, deletions, and all 12 possible base substitutions without double-strand breaks or donor templates, expanding the repertoire of corrections possible in therapeutic contexts where HDR is inefficient. Therapeutic editing strategies fall into two broad categories. Ex vivo approaches modify cells outside the body (e.g., hematopoietic stem cells, T cells) that are then returned to patients; delivery is typically achieved by electroporation of ribonucleoprotein complexes (RNPs) or viral vectors and offers precise dosing and robust quality control. In vivo approaches aim to deliver editors directly to tissues (liver, muscle, eye) using lipid nanoparticles (LNPs), adeno-associated virus (AAV) vectors, or other modalities. Each approach faces distinct tradeoffs: ex vivo methods permit extensive pre-infusion testing but are limited to cell types amenable to harvesting and reinfusion; in vivo methods can target tissues directly but must address biodistribution, immune responses, and long-term persistence.

6. LIPID NANOPARTICLES AND FIRST IN-HUMAN IN VIVO EDITING

Lipid nanoparticle (LNP) formulations that deliver Cas9 mRNA and guide RNAs have proven effective for liver delivery. The first published in-human study of in vivo CRISPR editing, NTLA-2001 for transthyretin (ATTR) amyloidosis, used LNPs to deliver a Cas9 system targeting the TTR gene and showed dose-dependent knockdown of serum TTR with manageable safety signals in early cohorts [11]. This result provided a proof-of-principle for transient, in vivo editing of a liver-expressed gene using nonviral delivery. AAV vectors have high transduction efficiency in many tissues and a favorable clinical track record but are constrained by packaging size—making delivery of large Cas9 orthologs or base/prime editors challenging. Strategies include split-Cas systems, smaller Cas orthologs (e.g., SaCas9), or dual-AAV designs. Long-term expression from AAV raises additional safety and immunogenicity questions for nuclease activity, prompting careful selection of promoters and regulatory designs.

7. THERAPEUTIC APPLICATIONS AND CLINICAL PROGRESS

One of the most advanced clinical programs using CRISPR-mediated editing is CTX001 (exa-cel), an ex vivo edited autologous hematopoietic stem and progenitor cell product developed by CRISPR Therapeutics and Vertex that disrupts BCL11A erythroid enhancer to induce fetal hemoglobin expression for sickle cell disease (SCD) and transfusion-dependent β-thalassemia (TDT). Clinical reports demonstrated transfusion independence in TDT and resolution of Vaso-occlusive crises in SCD for treated patients, prompting regulatory filings and pivotal trials (CRISPR Therapeutics & Vertex, corporate disclosures and meeting reports). This ex vivo approaches benefit from controlled manufacturing and the ability to perform deep quality testing prior to patient infusion.  As noted above, NTLA-2001 (Intellia) is the leading example of in vivo CRISPR editing delivered via LNPs to the liver to knock down transthyretin production, reducing pathogenic protein levels in ATTR amyloidosis. Published NEJM data reported meaningful reductions in TTR concentrations after single doses, with initial safety signals supporting continued development; follow-up and larger cohorts will be required to demonstrate durable clinical benefit and long-term safety [12].

8. ONCOLOGY AND CELL THERAPIES

CRISPR has accelerated engineering of T cells—for example, to knock out immune checkpoint genes, disrupt endogenous T-cell receptors, or insert synthetic antigen receptors—leading to multiplexed cell therapies under early clinical evaluation. This ex vivo edited cell products leverage CRISPR’s multiplexing capacity to program complex, multi-gene changes that are difficult with older technologies. Early phase trials are exploring safety, persistence, and antitumor activity in hematologic and solid malignancies [13]. Methods such as GUIDE-seq, Di genome-seq, CIRCLE-seq, and SITE-Seq enable comprehensive discovery of potential off-target sites and inform guide design and nuclease engineering [14]. Preclinical pipelines for therapeutic candidates typically combine in vitro unbiased assays, cellular GUIDE-seq, and in vivo sequencing to characterize off-target landscapes and to estimate risk of deleterious edits in clinically relevant tissues. Regulators and developers now expect rigorous off-target evaluation before human dosing.

9. STRATEGIES TO REDUCE OFF-TARGET EVENTS

Multiple complementary strategies are used to reduce off-target editing: (1) using high-fidelity Cas9 variants; (2) optimizing guide RNAs with computational tools that penalize potential off-target matches; (3) delivering RNPs for transient nuclease activity rather than persistent expression; (4) using base or prime editors when appropriate to obviate double-strand breaks; and (5) limiting exposure by tissue-restricted delivery (e.g., LNPs targeting liver). Combining strategies typically yields the best safety profile in preclinical models. Immune responses to bacterial Cas proteins or viral vectors can limit efficacy and cause adverse events. Preexisting antibodies and cellular immunity in humans to commonly used Cas orthologs (e.g., SpCas9 from S. pyogenes) have been described, and clinical protocols often screen for sensitizing antibodies or use immunosuppression transiently. Immunogenicity remains an active area o study as in-vivo systemic delivery becomes more common [15].

10. ETHICAL, REGULATORY, AND SOCIETAL CONSIDERATIONS

The promise of germline genome modification—permanent heritable edits—triggered intense debate. The 2018 case in which a researcher reported births of genome-edited infants in China drew widespread condemnation and highlighted the profound ethical, social, and safety concerns attendant to germline modification (news coverage and analyses). The episode catalyzed policy responses, calls for moratoria on clinical germline editing, and renewed emphasis on international governance and responsible research practices. The scientific community reiterated broad support for continuing somatic therapeutic research while opposing clinical germline interventions outside an agreed global framework [16]. Regulators are adapting to a new class of therapeutics that edit genomes. Requirements emphasize characterization of the editing reagents, demonstration of minimal off-target activity, robust biodistribution data, and long-term safety follow-up. Because genome editing can produce permanent changes, regulatory agencies typically require extended post-trial monitoring and sometimes new registries to capture long-term outcomes. International harmonization of standards remains incomplete, raising challenges for multinational trials [17].

11. EQUITY, ACCESS, AND PUBLIC ENGAGEMENT

High costs and specialized infrastructure threaten to concentrate benefits of gene-editing therapies in high-income settings. Ethical development includes attention to equitable access, meaningful engagement with affected communities, and transparent benefit-sharing. Public trust depends on responsible research practices, clear communication of benefits and risks, and regulatory transparency [18]. Although short-term signals from early trials (e.g., exa-cel, NTLA-2001) are encouraging, long-term durability, late toxicities, clonal selection, and immunogenic consequences require continued surveillance. Because editing is permanent in the targeted cells, unanticipated biology could emerge over years or decades; thus, systems for long-term follow-up and registries are essential.

12. OFF-TARGET STRUCTURAL VARIANTS AND COMPLEX REARRANGEMENTS

While much attention focuses on off-target single-site cleavage, complex outcomes including large deletions, chromosomal rearrangements, and on-target insertional mutagenesis have been reported in some contexts. Sensitive assays and long-read sequencing can reveal such events and should be incorporated into preclinical safety packages when risk is suspected. Many diseases would benefit from editing in tissues that are challenging to reach efficiently (brain, lung, muscle). Engineering delivery vehicles with appropriate tropism, minimizing immunogenicity, and achieving cell-type specificity remain major engineering hurdles. Advances in targeted nanoparticles, improved viral capsids, and cell-type specific promoters are active areas of research.

13. FUTURE DIRECTIONS

Ongoing engineering aims to reduce collateral damage and expand the set of correctable variants. Next-generation base editors with minimized byproducts, prime editors with increased efficiency and reduced indels, and novel enzymes with alternative PAM compatibilities will broaden the disease targets amenable to safe editing [19]. Integration of protein engineering and machine learning promises rational design of higher-precision editors. Community development of consensus pipelines for off-target discovery and reporting standards will help regulators compare candidate therapies. Combining short- and long-read sequencing, unbiased DSB capture methods, and sensitive functional assays will produce a richer safety profile for candidate editors. Public datasets of observed off-targets and harmonized reporting metrics would further accelerate safe translation. Efforts to build capacity in low- and middle-income countries for genetic diagnostics, manufacturing, and clinical trials will be crucial to avoid exacerbating global health inequities. Parallel investments in governance—multi-stakeholder frameworks that include researchers, patients, ethicists, and regulators—are needed to guide responsible deployment. The He Jiankui case remains a cautionary tale underscoring the need for global norms and strong local oversight.

14. CONCLUSION

The study succinctly revealed the programmable nature of Cas9 and set in motion a technological revolution in genetics and medicine. In a relatively brief time, the field has progressed from bench demonstrations of targeted cleavage to sophisticated editing platforms (base and prime editors), improved safety profiling (GUIDE-seq and related assays), delivery solutions for both ex vivo and in vivo therapy, and early human clinical trials that provide cautious optimism about therapeutic potential. Yet significant scientific, ethical, and social challenges remain—chiefly the need to minimize unintended edits, understand long-term effects, ensure equitable access, and prevent irresponsible germline editing. Careful, transparent, and collaborative development, guided by rigorous science and inclusive governance, will determine whether the transformative promise of programmable RNA-guided nucleases is realized safely and broadly.

REFERENCES

1. Anzalone, A. V., Randolph, P. B., Davis, J. R., Sousa, A. A., Koblan, L. W., Levy, J. M., … & Liu, D. R. (2019). Search-and-replace genome editing without double-strand breaks or donor DNA. Nature, 576(7785), 149–157.
2. Doudna, J. A., & Charpentier, E. (2014). Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science, 346(6213), 1258096.
3. Gaudelli, N. M., Komor, A. C., Rees, H. A., Packer, M. S., Badran, A. H., Bryson, D. I., & Liu, D. R. (2017). Programmable base editing of A•T to G•C in genomic DNA without DNA cleavage. Nature, 551(7681), 464–471.
4. Gillmore, J. D., Gane, E., Taubel, J., Kao, J., Fontana, M., Maitland, M. L., … & Khatchadourian, R. (2021). CRISPR–Cas9 in vivo gene editing for transthyretin amyloidosis. New England Journal of Medicine, 385(6), 493–502.
5. Jinek, M., Chylinski, K., Fonfara, I., Hauer, M., Doudna, J. A., & Charpentier, E. (2012). A programmable dual-RNA–guided DNA endonuclease in adaptive bacterial immunity. Science, 337(6096), 816–821.
6. Komor, A. C., Kim, Y. B., Packer, M. S., Zuris, J. A., & Liu, D. R. (2016). Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage. Nature, 533(7603), 420–424.
7. Tsai, S. Q., Zheng, Z., Nguyen, N. T., Liebers, M., Topkar, V. V., Thapar, V., … & Joung, J. K. (2015). GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR–Cas nucleases. Nature Biotechnology, 33(2), 187–197.
8. Vertex Pharmaceuticals / CRISPR Therapeutics. (2023). Vertex and CRISPR Therapeutics present new data in 22 patients (press release).
9. Public and journal commentary on the He Jiankui germline editing case and its consequences for policy and governance (Nature, New Yorker, TIME) are summarized in news and analysis articles (e.g., Nature editorial coverage, 2019; news analyses linked above).
10. Liu, Z., & Song, S. (2026). Reprogramming innate immunity through viral interference: A double-edged strategy for enhancing and containing gene therapies. Molecular Therapy Nucleic Acids, 37(2).
11. Faure, G., Saito, M., Wilkinson, M. E., Quinones-Olvera, N., Xu, P., Flam-Shepherd, D., … & Zhang, F. (2025). TIGR-Tas: A family of modular RNA-guided DNA-targeting systems in prokaryotes and their viruses. Science, 388(6746), eadv9789.
12. Brück, M., & Randau, L. (2025). TIGR on the loose: A dual-guide RNA system for DNA targeting. Molecular Cell, 85(9), 1712-1713.
13. Ishihara, K., Matsumoto, S., Gerle, C., Gopalasingam, C. C., Shigematsu, H., Shirai, T., & Numata, T. (2026). Sequential structural rearrangements at the PAM-distal site of a type I-F3 CRISPR-Cas effector enabling RNA-guided DNA transposition. Nucleic Acids Research, 54(1), gkaf1415.
14. Xiang, W., Lin, X., Yang, Y., Huang, L., Chen, Y., Chen, J., & Liu, L. (2025). Cas12h is a crRNA-guided DNA nickase that can be utilized for precise gene editing. Cell Reports, 44(5).
15. Ruden, D. M. (2025). TIGR-Tas and the Expanding Universe of RNA-Guided Genome Editing Systems: A New Era Beyond CRISPR-Cas. Genes, 16(8), 896.
16. Schuler, G., Hu, C., & Ke, A. (2022). Structural basis for RNA-guided DNA cleavage by IscB-ωRNA and mechanistic comparison with Cas9. Science, 376(6600), 1476-1481.
17. Fernández, J. H. (2022). CRISPR: 10 años de su descubrimiento, un-Premio Nobel y en proceso, el aprendizaje de edición de genes. Revista Mutis, 12(2).
18. Chang, H. Y., & Qi, L. S. (2023). Reversing the Central Dogma: RNA-guided control of DNA in epigenetics and genome editing. Molecular cell, 83(3), 442-451.
19. Liu, Y., Wang, L., Zhang, Q., Fu, P., Zhang, L., Yu, Y., … & Zhu, H. (2025). Structural basis for RNA-guided DNA degradation by Cas5-HNH/Cascade complex. Nature communications, 16(1), 1335.