Addressing Off-Target Effects in CRISPR Technologies

The first FDA-approved CRISPR-based therapy marked the transition of therapeutic gene editing from a dream to reality¹. CRISPR technologies have streamlined gene editing, and CRISPR screens have become an important approach for identifying genes involved in disease processes². This technique introduces targeted mutations across numerous genes, enabling large-scale identification of gene functions, interactions, and pathways³. Identifying the full range of biological changes, particularly any undesirable off-target effects, is an essential part of this process.

Technologies and Developments
Off-target alterations in gene editing have been a long-standing concern for drug developers and regulatory agencies. Over the years, numerous techniques have been developed to identify these changes. Among the earliest and most widely used techniques is GUIDE-seq (genome-wide, unbiased identification of DSBs enabled by sequencing)^4,5, which integrates double-stranded oligodeoxynucleotides at CRISPR-induced breaks to map off-target sites across the genome. CIRCLE-seq (circularization for in vitro reporting of cleavage effects by sequencing)⁶ is a cell-free methodology that uses circularized DNA as a substrate for off-target detection. DISCOVER-seq (discovery of in situ Cas off-targets and verification by sequencing)⁷ offers another layer of insight by capturing DNA-protein interactions that occur during CRISPR-induced breaks.

Other popular methods include Digenome-seq⁸, BLISS (breaks labeling in situ and sequencing)⁹, and rhAmpSeq¹⁰, with newer methods like Tracking-seq¹¹ and AID-seq (adaptor-mediated off-target identification by sequencing)¹² offering unique advantages in sensitivity and specificity. Whole genome sequencing (WGS) is a still beneficial approach for comprehensive detection, although it is significantly more resource-intensive and lacks the sensitivity to detect low frequency editing events¹³.

Each of these methods has its own set of strengths and limitations. For example, GUIDE-seq provides a cell-based measurement of off-targets, but it has limitations in oligonucleotide integration efficiency and sensitivity, leading to poor performance in certain important cell types¹⁴. CIRCLE-seq, while streamlined and cell-free, demonstrates a high false positive rate for off-target nomination which requires downstream validation in cells⁶. DISCOVER-seq’s high precision and ability to provide real-time insights into DNA-protein interactions make it invaluable for understanding gene editing, but its lower sensitivity led to the development of the updated DISCOVER-seq+¹⁵.

In Silico Approaches
In addition to experimental techniques, computational methods play some role in identifying potential off-target sites. These in silico approaches allow researchers to predict where off-target effects might occur, providing a pre-screening step before experimental validation. One well-known tool is Cas-OFFinder¹⁶, which examines potential off-target sites of Cas9 RNA-guided endonucleases by scanning the entire genome for sequence similarities. CRISPOR¹⁷, a web-based tool, was designed to find gRNAs in an input sequence and rank them based on different scores that evaluate potential off-target effects in the genome of interest, as well as predict on-target activity.

A recently developed tool named CRISOT (CRISPR Off-Target)¹⁸ characterizes molecular dynamics and uses machine learning to predict off-target effects. Similarly, CRISPR-M¹⁹ uses a multi-view deep learning approach and was designed to handle target sites with indels and mismatches. Other notable methods include CRISPRMatch²⁰, DeepCRISPR²¹, and older tools like MIT²² and CROP-IT²³.

While these computational methods are beneficial, they have well-known limitations. For instance, biases inherent in algorithm design can lead to false positives or negatives²⁴. Additionally, these methods often do not account for the complex intranuclear environment, where numerous factors can influence off-target effects. In silico predictions remain a valuable tool, but their accuracy must be confirmed through experimental validation to account for the complexities of the biological environment.

INDUCE-seq^® Technology
Felix Dobbs, Ph.D., Co-Founder and CEO of Broken String Biosciences, stated that understanding and addressing the challenges of CRISPR off-target effects was the primary reason for founding the company. Originating from academic research, Broken String’s work led to the development of INDUCE-seq^®, an NGS-based platform that enables high-precision measurement of CRISPR off-target effects²⁵. “We realized that a key challenge in the space was not being able to measure [off-target edits] accurately," stated Dobbs. INDUCE-seq addresses this challenge by providing clean, quantitative data on genomic break sites, which is essential for pre-clinical assessments and therapeutic development.

The technology works by taking a sample of whole cells and directly labeling breaks formed in the genomes of those cells in situ, explained Dobbs. During this process, DSBs are labeled with a first sequencing adapter, followed by a novel PCR-free library preparation with a second adapter that prepares these fragments for sequencing. The result is a direct readout of break sites throughout the genome using NGS. Every read generated corresponds to a single break from the starting cell population, and the technology offers a clear snapshot of where breaks occur, whether naturally or due to gene editing. This allows researchers to pinpoint on- and off-target effects with high precision.

Dobbs noted that INDUCE-seq offers several advantages over alternative methods for detecting CRISPR off-target effects. One key benefit is its cell-based approach, which measures off-targets in physiologically or clinically relevant cell samples. "Off-targets form in a very cell-type specific way," Dobbs stated, "and there's a real need to measure these sites in your intended tissue target or actual cell therapy products." Another significant advantage is the technology's PCR-free nature, which eliminates biases commonly associated with PCR amplification. Furthermore, INDUCE-seq is designed to be highly scalable, making it suitable for a wide range of applications including discovery and therapeutic development.

While the primary focus of Broken String Biosciences is on gene editing and therapeutic applications, INDUCE-seq has potential uses beyond its original purpose. "The platform as a whole is very broad," Dobbs said. "We're focusing today on gene editing for therapeutic programs, but in the future we can envision it also being used for screening many hundreds of gRNAs earlier in the program and for ongoing monitoring in patient-focused phases." INDUCE-seq's versatility also extends to different types of nucleases and cell types, making it an invaluable tool for various research and clinical applications. Dobbs also mentioned its potential future uses in oncology, where the technology could identify break signatures in different cancers, aiding in disease characterization and potentially in diagnostics.

Broken String Biosciences is actively working on commercializing INDUCE-seq, Dobbs announced, with plans to make it widely available. Their efforts involve creating a deployable product offering that includes laboratory components and bioinformatics analysis, making it accessible and user-friendly for researchers and clinical partners. “We didn't want this to just remain an academic technology," Dobbs reflected. "We needed to take this and build it into a product that can become the gold standard in the market." Towards this goal, Broken String is proud to be working within the HESI CT-TRACS and NIST Genome Editing Consortium working groups, focused on standardizing technologies for assessing off-targets induced by gene editing.

Final Thoughts
These developments are essential for ensuring the precision and safety of gene editing technologies like CRISPR. As the field continues to develop with novel editors and editing modalities like base and prime editing, it will remain essential to profile off-target DSBs in the genome in products for therapeutic use. With innovative methods like INDUCE-seq^®, researchers can more effectively identify unintended off-target effects and pave the way for safer and more reliable gene therapies.

References

1. Sheridan C. The world’s first CRISPR therapy is approved: who will receive it? Nature Biotechnology. Published November 21, 2023.
2. Kim HS, Kweon J, Kim Y. Recent advances in CRISPR-based functional genomics for the study of disease-associated genetic variants. Experimental & Molecular Medicine. 2024;56(4):861-869. doi:https://doi.org/10.1038/s12276-024-01212-3
3. Bock C, Datlinger P, Chardon F, et al. High-content CRISPR screening. Nature Reviews Methods Primers. 2022;2(1):8. doi:https://doi.org/10.1038/s43586-021-00093-4
4. Tsai SQ, Zheng Z, Nguyen NT, et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nature Biotechnology. 2015;33(2):187-197. doi:https://doi.org/10.1038/nbt.3117
5. Malinin, Nikolay L, Lee G, Lazzarotto CR, et al. Defining genome-wide CRISPR–Cas genome-editing nuclease activity with GUIDE-seq. Nature Protocols. 2021;16(12):5592-5615. doi:https://doi.org/10.1038/s41596-021-00626-x
6. Tsai SQ, Nguyen NT, MalagonLopez J, Topkar, Ved V, Aryee MJ, Keith JJ. CIRCLE-seq: a highly sensitive in vitro screen for genome-wide CRISPR–Cas9 nuclease off-targets. Nature Methods. 2017;14(6):607-614. doi:https://doi.org/10.1038/nmeth.4278
7. Wienert B, Wyman SK, Yeh CD, Conklin BR, Corn JE. CRISPR off-target detection with DISCOVER-seq. Nature Protocols. 2020;15(5):1775-1799. doi:https://doi.org/10.1038/s41596-020-0309-5
8. Kim D, Bae S, Park J, et al. Digenome-seq: genome-wide profiling of CRISPR-Cas9 off-target effects in human cells. Nature Methods. 2015;12(3):237-243. doi:https://doi.org/10.1038/nmeth.3284
9. Yan WX, Mirzazadeh R, Garnerone S, et al. BLISS is a versatile and quantitative method for genome-wide profiling of DNA double-strand breaks. Nature Communications. 2017;8(1):15058. doi:https://doi.org/10.1038/ncomms15058
10. Turk R. CRISPR Applications in Medicine Depend on Minimizing Off-Target Editing. Genetic Engineering & Biotechnology News. 2021;41(9):64-66. doi:https://doi.org/10.1089/gen.41.09.24
11. Zhu M, Xu R, Yuan J, et al. Tracking-seq reveals the heterogeneity of off-target effects in CRISPR–Cas9-mediated genome editing. Nature Biotechnology. Published online 2024. doi:https://doi.org/10.1038/s41587-024-02307-y
12. Tian R, Cao C, He D, et al. Massively parallel CRISPR off-target detection enables rapid off-target prediction model building. Med. 2023;4(7):478-492.e6. doi:https://doi.org/10.1016/j.medj.2023.05.005
13. Atkins A, Chung C, Allen AG, et al. Off-Target Analysis in Gene Editing and Applications for Clinical Translation of CRISPR/Cas9 in HIV-1 Therapy. Frontiers in Genome Editing. 2023;3, 673022. doi:https://doi.org/10.3389/fgeed.2021.673022
14. Yang Z, Deng D, Gao Z, et al. OliTag-seq enhances in cellulo detection of CRISPR-Cas9 off-targets. Communications Biology. 2024;7(1):696. doi:https://doi.org/10.1038/s42003-024-06360-w
15. Zou RS, Liu Y, Reyes OE, et al. Improving the sensitivity of in vivo CRISPR off-target detection with DISCOVER-Seq+. Nature Methods. 2023;20(5):706-713. doi:https://doi.org/10.1038/s41592-023-01840-z
16. Bae S, Park J, Kim J. Cas-OFFinder: a fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases. Bioinformatics. 2014;30(10):1473-1475. doi:https://doi.org/10.1093/bioinformatics/btu048
17. Concordet J, Haeussler M. CRISPOR: intuitive guide selection for CRISPR/Cas9 genome editing experiments and screens. Nucleic Acids Res. 2018;46(W1):W242-W245. doi:https://doi.org/10.1093/nar/gky354
18. Chen Q, Chuai G, Zhang H, et al. Genome-wide CRISPR off-target prediction and optimization using RNA-DNA interaction fingerprints. Nature Communications. 2023;14(1):7521. doi:https://doi.org/10.1038/s41467-023-42695-4
19. Sun J, Guo J, Liu J. CRISPR-M: Predicting sgRNA off-target effect using a multi-view deep learning network. PLOS Computational Biology. 2024;20(3):e1011972-. doi:https://doi.org/10.1371/journal.pcbi.1011972
20. You Q, Zhong Z, Ren Q, Hassan F, Zhang Y, Zhang T. CRISPRMatch: An Automatic Calculation and Visualization Tool for High-throughput CRISPR Genome-editing Data Analysis. Int J Biol Sci. 2018;14(8):858-862. doi:https://doi.org/10.7150/ijbs.24581
21. Chuai G, Ma H, Yan J, et al. DeepCRISPR: optimized CRISPR guide RNA design by deep learning. Genome Biology. 2018;19(1):80. doi:https://doi.org/10.1186/s13059-018-1459-4
22. Hsu PD, Scott DA, Weinstein JA, et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nature Biotechnology. 2013;31(9):827-832. doi:https://doi.org/10.1038/nbt.2647
23. Singh R, Kuscu C, Quinlan A, Qi Y, Adli M. Cas9-chromatin binding information enables more accurate CRISPR off-target prediction. Nucleic Acids Res. 2015;43(18):e118-e118. doi:https://doi.org/10.1093/nar/gkv575
24. Guo C, Ma X, Gao F, Guo Y. Off-target effects in CRISPR/Cas9 gene editing. Frontiers in Bioengineering and Biotechnology, 2023; 11, 1143157. doi:https://doi.org/10.3389/fbioe.2023.1143157
25. Dobbs FM, Eijk van, Fellows MD, Loiacono L, Nitsch R, Reed SH. Precision digital mapping of endogenous and induced genomic DNA breaks by INDUCE-seq. Nature Communications. 2022;13(1):3989. doi:https://doi.org/10.1038/s41467-022-31702-9