Non-Homologous End Joining in CRISPR: Mechanisms and Applications
By Dr. Zubair Khalid, DVM, MS, PhD ·

Introduction to Non-Homologous End Joining in CRISPR
What is NHEJ?
Non-homologous end joining (NHEJ) is the predominant DNA double-strand break (DSB) repair pathway in mammalian cells. Unlike homologous recombination, which requires a sister chromatid as a template, NHEJ directly ligates the two broken DNA ends with minimal or no sequence homology. This repair mechanism is inherently error-prone because it often introduces small insertions or deletions (indels) at the break site. In the context of CRISPR-Cas9 genome editing, NHEJ is the default pathway that cells use to repair the site-specific DSB created by the Cas9 nuclease. The Non Homologous End Joining Repair pathway operates throughout the cell cycle, making it the most accessible repair option for CRISPR-induced breaks.
The biological rationale for NHEJ is straightforward: it is fast and does not depend on the availability of a homologous template. In a typical human cell, NHEJ can repair a DSB within minutes to hours, whereas homologous recombination takes several hours and is restricted to the S and G2 phases of the cell cycle. For CRISPR applications, this means that the majority of editing events will be NHEJ-mediated indels rather than precise homology-directed repair (HDR) events.
CRISPR-Cas9 and Double-Strand Breaks
The CRISPR-Cas9 system, derived from the bacterial adaptive immune system, consists of two core components: the Cas9 endonuclease and a single-guide RNA (sgRNA). The sgRNA contains a 20-nucleotide spacer sequence that is complementary to the target genomic locus, followed by a scaffold sequence that binds Cas9. When the sgRNA guides Cas9 to the target site, the nuclease generates a blunt-ended DSB exactly 3 base pairs upstream of the protospacer adjacent motif (PAM), which is the sequence NGG for Streptococcus pyogenes Cas9.
This DSB triggers the cellular DNA damage response. The MRE11-RAD50-NBS1 (MRN) complex and the Ku70/Ku80 heterodimer compete for binding to the broken ends. In most cell types and cell cycle phases, Ku70/Ku80 wins this competition, initiating the Non Homologous End Joining Pathway. The outcome is an indel at the target site, which can disrupt gene function. This is the basis of CRISPR Knockout technology. Understanding the molecular choreography of NHEJ is essential for predicting editing outcomes and designing experiments that either exploit or suppress this pathway.
The NHEJ Repair Pathway: Step-by-Step
Ku70/Ku80 Heterodimer
The first responder to a DSB is the Ku70/Ku80 heterodimer, a ring-shaped protein complex that binds with high affinity to free DNA ends. Ku70 (70 kDa) and Ku80 (80 kDa) form a basket-like structure that threads onto the DNA double helix, encircling it without making sequence-specific contacts. This binding is extremely stable, with a dissociation constant in the picomolar range, and it protects the DNA ends from exonucleolytic degradation.
Once bound, Ku serves as a loading platform for downstream factors. The Ku-DNA complex recruits the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs), a 469 kDa serine/threonine kinase. The assembly of Ku, DNA-PKcs, and DNA forms the active DNA-PK holoenzyme. DNA-PKcs undergoes autophosphorylation at multiple sites, including the ABCDE cluster (Serine 2056, Threonine 2609, and others), which induces a conformational change that allows end processing enzymes access to the DNA termini. This step is critical because the ends produced by Cas9 are not always directly ligatable; they may contain damaged bases, abasic sites, or other chemical modifications that must be removed.
DNA-PKcs and End Processing
The ends of a CRISPR-induced DSB are blunt and contain 5' phosphate and 3' hydroxyl groups, which are chemically compatible with ligation. However, the cellular environment often modifies these ends before ligation occurs. Nucleases such as Artemis, which has 5'→3' exonuclease and endonuclease activity, can trim overhangs or open hairpin structures. The MRN complex and CtBP-interacting protein (CtIP) can also resect ends, although limited resection favors NHEJ while extensive resection commits the cell to homologous recombination.
Polymerases play a crucial role in end processing. DNA polymerase mu (Pol μ) and DNA polymerase lambda (Pol λ) are template-independent and template-dependent polymerases, respectively, that can fill in gaps or add nucleotides to create microhomologies. Pol μ is particularly adept at adding nucleotides across a break without a template, which explains why NHEJ frequently introduces small insertions. The balance between nucleolytic trimming and polymerase activity determines the final sequence at the repair junction. This is why CRISPR-induced indels are not random; they often show a bias toward 1 bp insertions or deletions at homopolymeric stretches.
Ligation by XRCC4/Ligase IV
The final step of NHEJ is ligation, catalyzed by DNA ligase IV in complex with XRCC4. XRCC4 (X-ray repair cross-complementing protein 4) stabilizes ligase IV and recruits it to the DNA-PK complex. The ligase IV-XRCC4 complex also interacts with XLF (XRCC4-like factor, also called Cernunnos), which stimulates ligase activity and helps bridge the two DNA ends. Additional factors such as PAXX (paralog of XRCC4 and XLF) and the recently identified CYREN (cell cycle regulator of NHEJ) modulate the efficiency of this final step.
The ligation reaction requires ATP and Mg²⁺, and it proceeds through a covalent DNA-adenylate intermediate. Ligase IV is uniquely suited for NHEJ because it can ligate ends that are not perfectly complementary, including ends with short overhangs or mismatched bases. The result is a repaired DSB that contains a scar—the indel—which is the hallmark of NHEJ. The Non Homologous End Joining Proteins involved in this pathway are conserved from yeast to humans, underscoring the fundamental importance of this repair mechanism.
Why CRISPR Preferentially Uses NHEJ
Cell Cycle Dependence
The choice between NHEJ and homologous recombination is governed primarily by cell cycle phase and the availability of a homologous template. NHEJ is active in all phases of the cell cycle, including G1, when no sister chromatid is available. In contrast, homologous recombination requires a sister chromatid and is therefore restricted to S and G2 phases. In a typical asynchronous cell population, the majority of cells are in G1, which means that NHEJ will dominate the repair outcome.
The cell cycle regulation is mediated by cyclin-dependent kinases (CDKs). CDK activity in S and G2 phases promotes the phosphorylation of CtIP, which stimulates DNA end resection—the first committed step toward homologous recombination. In G1, CDK activity is low, CtIP is not phosphorylated, and resection is inhibited. This favors Ku binding and NHEJ. Additionally, the protein CYREN directly inhibits NHEJ during S and G2 phases, further biasing repair toward homologous recombination when a template is available.
Competition with Homology-Directed Repair
When a CRISPR-Cas9 DSB is generated, the repair pathway choice is a kinetic competition. Ku binds to DNA ends within seconds, whereas the resection machinery that initiates homologous recombination takes minutes to engage. If Ku wins, NHEJ proceeds rapidly. If resection occurs first, the ends are committed to homologous recombination or, if resection is extensive, to single-strand annealing or microhomology-mediated end joining (MMEJ).
For CRISPR applications, this competition has practical consequences. The efficiency of HDR is typically 1–10% of total editing events in mammalian cells, while NHEJ accounts for the remaining 90–99%. This is why CRISPR Knock (knock-in) experiments, which require HDR, are substantially more challenging than CRISPR Knockout experiments. The cell's default preference for NHEJ is an evolutionary adaptation that prioritizes genome stability over repair fidelity—a broken chromosome is more dangerous than a mutated one.
Outcomes of NHEJ: Indels and Gene Knockout
Frameshift Mutations
The most common outcome of NHEJ-mediated repair of a CRISPR-induced DSB is a small insertion or deletion of 1–10 base pairs. When the indel size is not a multiple of three, it causes a frameshift mutation. The reading frame of the gene is shifted, so that every codon downstream of the mutation is altered. This almost invariably leads to the production of a non-functional protein.
For example, consider a gene with the coding sequence ATG GGA CCC AAA (Met-Gly-Pro-Lys). A 1 bp deletion of the first G in GGA would shift the reading frame to ATG GAC CCA AA... (Met-Asp-Pro-Lys...), completely changing the amino acid sequence from that point onward. The probability of a frameshift is approximately 67% for a random indel, since only indels that are multiples of three preserve the reading frame.
Premature Stop Codons
Frameshift mutations frequently generate premature stop codons. A stop codon (TAA, TAG, or TGA) appearing in the new reading frame terminates translation prematurely, producing a truncated protein. If the stop codon appears early in the coding sequence, the truncated protein is typically non-functional and often degraded by the nonsense-mediated decay (NMD) pathway. NMD recognizes mRNAs with premature termination codons located more than 50–55 nucleotides upstream of the last exon-exon junction and targets them for degradation.
The combination of frameshift mutations and premature stop codons makes NHEJ an extremely effective tool for gene knockout. However, it is important to note that not all indels cause gene disruption. An in-frame deletion of 3, 6, or 9 base pairs will remove one or more amino acids but may leave the protein partially or fully functional. This is why validating knockout efficiency requires protein-level analysis (e.g., western blot) rather than relying solely on DNA sequencing.
Methods to Detect and Analyze NHEJ Events
T7 Endonuclease I Assay
The T7 Endonuclease I (T7E1) assay is a simple and cost-effective method to detect NHEJ-induced mutations. The principle is based on the ability of T7E1 to recognize and cleave mismatched heteroduplex DNA. Genomic DNA is extracted from edited cells, and the target region is amplified by PCR. The PCR products are then denatured and re-annealed. If NHEJ has introduced indels, some of the re-annealed duplexes will contain mismatches where a wild-type strand pairs with a mutant strand. T7E1 cleaves at these mismatched sites, producing smaller DNA fragments that can be visualized by agarose gel electrophoresis.
The assay protocol typically involves: PCR amplification using primers flanking the target site (amplicon size 400–800 bp), denaturation at 95°C for 5 minutes, re-annealing by slow cooling (e.g., −2°C/second to 85°C, then −0.1°C/second to 25°C), digestion with 5–10 units of T7E1 at 37°C for 15–60 minutes, and analysis on a 2% agarose gel. The editing efficiency is estimated from the band intensities using the formula: % editing = 100 × (1 − √(1 − fraction of cleaved DNA)). The T7E1 assay is semi-quantitative and can detect editing efficiencies as low as 1–5%, but it cannot determine the exact nature of the indels.
Sanger Sequencing and TIDE
For a more detailed analysis, Sanger sequencing of the target region followed by computational analysis with TIDE (Tracking of Indels by DEcomposition) is recommended. TIDE uses the raw Sanger sequencing trace files (AB1 format) from both edited and control samples. The software decomposes the mixed sequencing signal and quantifies the frequency and types of indels present.
The TIDE algorithm works by comparing the sequencing chromatogram of the edited sample to that of the control. It identifies the position of the breakpoint and fits the data to a model that includes wild-type sequence plus a set of common indel sequences. The output includes the total editing efficiency, the spectrum of indels (e.g., 1 bp deletions, 2 bp insertions), and the statistical significance of each indel. TIDE can detect indels with frequencies as low as 1% and is accurate for indels up to 50 bp. This method is widely used because it is inexpensive, requires only standard Sanger sequencing, and provides quantitative information about the mutation spectrum.
Next-Generation Sequencing
For comprehensive analysis, including the detection of large deletions, off-target effects, and complex rearrangements, next-generation sequencing (NGS) is the gold standard. Targeted amplicon sequencing involves PCR amplification of the target locus, library preparation, and deep sequencing (typically 10,000–100,000 reads per sample). Bioinformatics tools such as CRISPResso2 or Cas-Analyzer align the reads to the reference sequence and quantify the frequency of each indel type.
NGS offers several advantages: it can detect indels of any size, identify the exact breakpoint sequences, and distinguish between NHEJ and MMEJ outcomes based on the presence of microhomologies at the junction. The main disadvantages are cost, turnaround time, and the need for bioinformatics expertise. For most undergraduate laboratory courses, T7E1 and TIDE provide sufficient information to assess NHEJ efficiency.
Enhancing or Suppressing NHEJ in CRISPR Experiments
Chemical Inhibitors
For experiments that require HDR rather than NHEJ, suppressing NHEJ can increase HDR efficiency. Several small-molecule inhibitors target key NHEJ proteins. SCR7, a DNA ligase IV inhibitor, is the most commonly used. SCR7 binds to the DNA-binding domain of ligase IV and prevents the final ligation step. In cultured cells, treatment with 1–10 μM SCR7 for 24–48 hours after transfection can increase HDR efficiency by 2–5-fold, although the effect varies by cell type.
Other inhibitors include NU7441 and KU-57788, which inhibit DNA-PKcs. These compounds block the kinase activity of DNA-PKcs, preventing end processing and downstream signaling. Wortmannin, a broad-spectrum PI3K inhibitor, also inhibits DNA-PKcs at micromolar concentrations but has significant cytotoxicity. The combination of NHEJ inhibition with HDR template delivery (e.g., single-stranded oligodeoxynucleotide, ssODN) is a common strategy to improve knock-in efficiency.
Cell Cycle Synchronization
Since NHEJ is active throughout the cell cycle but HDR is restricted to S/G2, synchronizing cells to the S/G2 phase can enhance HDR. This can be achieved by treating cells with aphidicolin, which inhibits DNA polymerase α and arrests cells at the G1/S boundary. After release from the arrest, cells enter S phase synchronously, and CRISPR delivery at this time maximizes HDR.
Alternatively, nocodazole arrests cells in M phase, and after release, cells progress through G1 into S phase. The timing of CRISPR delivery must be optimized empirically for each cell line. A simpler approach is to use cell cycle inhibitors such as lovastatin (arrests in G1) or RO-3306 (CDK1 inhibitor, arrests in G2). These synchronization strategies are labor-intensive but can increase HDR efficiency by 3–10-fold in some cell types.
CRISPR-Mediated NHEJ Knockout
A genetic approach to suppress NHEJ is to knock out or knock down essential NHEJ genes such as XRCC4, LIG4, or KU80. For example, a cell line with a stable knockout of LIG4 will have severely impaired NHEJ, forcing repair through homologous recombination. However, this approach has significant drawbacks: NHEJ-deficient cells are sensitive to DNA-damaging agents, grow slowly, and may accumulate chromosomal abnormalities. This strategy is therefore limited to specific experimental contexts where the benefits outweigh the costs.
Conversely, enhancing NHEJ is rarely needed for CRISPR experiments, since NHEJ is already the dominant pathway. However, in certain cell types with high HDR activity (e.g., mouse embryonic stem cells), suppressing homologous recombination by inhibiting RAD51 or CtIP can shift the balance toward NHEJ. This is occasionally useful for generating knockout cell lines in cell types where HDR is unexpectedly efficient.
NHEJ in CRISPR Applications: Beyond Gene Knockout
Functional Genomics Screens
NHEJ-mediated gene knockout is the foundation of CRISPR loss-of-function screens. In a typical screen, a pooled library of sgRNAs targeting thousands of genes is delivered to a cell population. Each cell receives one sgRNA, and NHEJ introduces indels at the target locus. Cells that lose a gene essential for a particular phenotype (e.g., resistance to a drug, survival under stress) are depleted or enriched, and the sgRNA sequences in surviving cells are identified by deep sequencing.
The power of this approach lies in its scalability. Genome-wide libraries containing 70,000–100,000 sgRNAs can be screened in a single experiment. The CRISPR Cas 9 system's specificity and the efficiency of NHEJ make it possible to achieve near-complete gene knockout in a population. This has revolutionized functional genomics, enabling the identification of genes involved in drug resistance, viral infection, cancer cell viability, and many other biological processes.
Therapeutic Gene Disruption
NHEJ is not only a research tool but also a therapeutic strategy. The most advanced application is in the treatment of genetic diseases where disrupting a specific gene is beneficial. For example, in β-thalassemia and sickle cell disease, reactivating fetal hemoglobin (HbF) can ameliorate symptoms. The BCL11A gene encodes a transcription factor that represses γ-globin expression. CRISPR-Cas9-mediated NHEJ knockout of BCL11A in hematopoietic stem cells leads to sustained HbF expression. Clinical trials using this approach (e.g., CTX001) have shown promising results.
Another therapeutic application is in chimeric antigen receptor (CAR) T-cell therapy. NHEJ can be used to knock out the endogenous T-cell receptor (TCR) or major histocompatibility complex (MHC) genes to create universal donor cells that are not rejected by the recipient's immune system. The CRISPR in Medicine field is rapidly expanding, and NHEJ-based gene disruption is a key enabling technology.
Base Editing Context
Base editors, which are fusions of a catalytically dead Cas9 (dCas9) or nickase Cas9 (nCas9) with a deaminase enzyme, do not create DSBs and therefore do not activate NHEJ. Instead, they convert one DNA base to another (e.g., C→T or A→G) through chemical deamination. This is advantageous because it avoids the stochastic indels produced by NHEJ. However, base editors can still cause bystander edits and off-target deamination, and they are limited to specific base transitions.
In contrast, prime editing uses a nCas9 fused to a reverse transcriptase and a prime editing guide RNA (pegRNA) to introduce precise insertions, deletions, or base substitutions without generating DSBs. Both base editing and prime editing were developed in part to overcome the limitations of NHEJ-mediated editing, particularly the inability to achieve precise, predictable outcomes. Nevertheless, NHEJ remains the workhorse of CRISPR editing for applications where gene disruption is the goal.
Common Pitfalls and Troubleshooting in NHEJ Studies
Misinterpreting Indel Patterns
A common mistake is assuming that all indels are equivalent. In reality, the indel spectrum is influenced by the local DNA sequence context. Homopolymeric runs (e.g., poly-A or poly-T tracts) are hotspots for 1 bp insertions and deletions because the repair polymerases can slip during synthesis. The presence of microhomologies (2–6 bp direct repeats) flanking the break site can shift repair toward MMEJ, which produces larger deletions. When analyzing NHEJ outcomes, it is essential to consider the sequence context and to validate the functional consequences of the specific indels obtained.
Another frequent error is failing to account for the possibility of in-frame indels. A 3 bp deletion or a 6 bp insertion will not cause a frameshift and may produce a protein that is partially functional. If the goal is complete gene knockout, screening multiple clones and verifying protein loss by western blot is essential. Relying solely on DNA sequencing can lead to false conclusions about knockout efficiency.
Ignoring Off-Target Effects
CRISPR-Cas9 can cleave at off-target sites that share sequence homology with the sgRNA. The tolerance for mismatches is highest in the 5' region of the sgRNA (the seed region is the 12 bp adjacent to the PAM). Off-target NHEJ events can cause mutations in genes other than the intended target, leading to confounding phenotypes. To minimize off-target effects, use sgRNAs with high specificity scores (e.g., from the Zhang lab or Benchling algorithms), use truncated sgRNAs (17–18 nucleotides), or use high-fidelity Cas9 variants such as SpCas9-HF1 or eSpCas9.
Detection of off-target effects requires either computational prediction followed by targeted sequencing, or unbiased methods such as GUIDE-seq (genome-wide unbiased identification of DSBs evaluated by sequencing) or CIRCLE-seq. For undergraduate projects, computational prediction using tools like Cas-OFFinder is a reasonable starting point, but it should be recognized that prediction algorithms have both false positives and false negatives.
Confusing NHEJ with MMEJ
Microhomology-mediated end joining (MMEJ) is a distinct repair pathway that also produces indels but has different mechanistic requirements. MMEJ requires 5–25 bp of microhomology flanking the break site, uses the MRN complex and CtIP for end resection, and employs DNA polymerase theta (Pol θ) for repair synthesis. The hallmark of MMEJ is deletions that remove one copy of the microhomology repeat. In contrast, canonical NHEJ does not require microhomology and typically produces smaller indels.
Distinguishing between NHEJ and MMEJ is important because the two pathways have different genetic requirements and can be differentially modulated. For example, inhibiting Pol θ (e.g., with the small molecule ART558) suppresses MMEJ but not NHEJ. When analyzing CRISPR editing outcomes, examining the junction sequences for microhomologies can help determine which pathway was used. This distinction is particularly relevant in BRCA1- or BRCA2-deficient cells, where homologous recombination is impaired and MMEJ becomes more prominent.
Summary and Key Takeaways
NHEJ is the dominant DNA repair pathway that acts on CRISPR-Cas9-induced double-strand breaks. The pathway involves Ku70/Ku80 binding, DNA-PKcs activation, end processing by nucleases and polymerases, and ligation by XRCC4/Ligase IV. The error-prone nature of NHEJ produces indels that frequently cause frameshift mutations and premature stop codons, making it the preferred method for gene knockout. The choice between NHEJ and HDR is governed by cell cycle phase and the competition between Ku binding and end resection. Experimental modulation of NHEJ can be achieved through chemical inhibitors, cell cycle synchronization, or genetic ablation of NHEJ components. Beyond simple knockout, NHEJ is used in functional genomics screens and therapeutic gene disruption. Accurate analysis of NHEJ outcomes requires appropriate detection methods and careful interpretation of indel patterns.
Frequently Asked Questions
What is non-homologous end joining in CRISPR?
Non-homologous end joining (NHEJ) is the cellular DNA repair pathway that ligates the two broken ends of a DNA double-strand break without requiring a homologous template. In CRISPR-Cas9 genome editing, NHEJ repairs the site-specific DSB created by Cas9, typically introducing small insertions or deletions (indels) at the cut site. These indels can disrupt gene function, making NHEJ the basis of CRISPR knockout technology.
Why does CRISPR use NHEJ instead of homologous recombination?
CRISPR-Cas9 does not "choose" NHEJ; rather, the cell's repair machinery does. NHEJ is active in all cell cycle phases and is kinetically faster than homologous recombination. The Ku70/Ku80 heterodimer binds to DNA ends within seconds, whereas homologous recombination requires DNA end resection and a homologous template, which is only available in S/G2 phases. In most cells, NHEJ outcompetes homologous recombination, resulting in 90–99% of editing events being NHEJ-mediated indels.
How does NHEJ lead to gene knockout?
NHEJ introduces indels at the CRISPR cut site. If the indel causes a frameshift mutation (an insertion or deletion that is not a multiple of three), the reading frame of the gene is shifted, leading to a completely different amino acid sequence downstream. This often generates a premature stop codon, which truncates the protein. The truncated protein is typically non-functional and may be degraded by nonsense-mediated decay.
What is the difference between NHEJ and HDR?
NHEJ is an error-prone repair pathway that directly ligates DNA ends without a template, producing indels. Homology-directed repair (HDR) uses a homologous DNA template (e.g., a sister chromatid or an exogenous donor) to repair the break precisely, without introducing mutations. NHEJ is active throughout the cell cycle and is the default pathway; HDR is restricted to S/G2 phases and requires a template. For CRISPR applications, NHEJ is used for gene knockout, while HDR is used for precise gene knock-in or base substitution.
How can I detect NHEJ-induced mutations?
Common methods include the T7 Endonuclease I (T7E1) assay, which detects mismatched heteroduplex DNA after PCR amplification and re-annealing; Sanger sequencing followed by TIDE analysis, which quantifies the frequency and types of indels; and next-generation sequencing (NGS), which provides comprehensive information about the mutation spectrum. Western blotting is used to confirm loss of protein expression.
Can NHEJ be inhibited to favor HDR?
Yes. Chemical inhibitors such as SCR7 (DNA ligase IV inhibitor), NU7441 (DNA-PKcs inhibitor), and wortmannin can suppress NHEJ. Cell cycle synchronization to enrich for S/G2 phase cells can also favor HDR. Genetic approaches, such as knocking out LIG4 or KU80, are effective but have significant side effects. These strategies typically increase HDR efficiency by 2–10-fold, but HDR remains less efficient than NHEJ in most cell types.
What are common mistakes when analyzing NHEJ results?
Common mistakes include: (1) assuming all indels cause gene knockout without verifying protein loss; (2) ignoring the possibility of in-frame indels that preserve protein function; (3) failing to account for off-target mutations; (4) confusing NHEJ with MMEJ, which has different mechanistic requirements and produces different indel patterns; and (5) using detection methods with insufficient sensitivity or resolution to accurately quantify editing efficiency.
Key Takeaways
- NHEJ is the dominant repair pathway for CRISPR-Cas9-induced double-strand breaks, producing indels that enable gene knockout.
- The pathway involves Ku70/Ku80 binding, DNA-PKcs activation, end processing, and ligation by XRCC4/Ligase IV.
- NHEJ is preferred over HDR because it is active throughout the cell cycle and does not require a homologous template.
- Indels that are not multiples of three cause frameshift mutations and premature stop codons, leading to non-functional proteins.
- NHEJ can be suppressed with chemical inhibitors (SCR7, NU7441) or cell cycle synchronization to favor HDR.
- Detection of NHEJ events requires appropriate methods: T7E1 assay for quick screening, TIDE for quantitative analysis, and NGS for comprehensive characterization.
- NHEJ is used not only for gene knockout but also in functional genomics screens and therapeutic gene disruption, such as BCL11A knockout for hemoglobinopathies.
Further Reading
- Ruis BL, Bielinsky AK, Hendrickson EA. Gene editing and CRISPR-dependent homology-mediated end joining. Experimental & molecular medicine. 2025. PubMed 40745005
- DeCarlo A et al. Targeting synthetic lethality between non-homologous end joining and radiation in very-high-risk medulloblastoma. Cell reports. Medicine. 2025. PubMed 40562042
- Hu Q et al. Non-homologous end joining shapes the genomic rearrangement landscape of chromothripsis from mitotic errors. Nature communications. 2024. PubMed 38965240
- Dev H et al. Shieldin complex promotes DNA end-joining and counters homologous recombination in BRCA1-null cells. Nature cell biology. 2018. PubMed 30022119
- Huang Y et al. Mycobacterial non-homologous end joining is required for antiphage defense. Nucleic acids research. 2026. PubMed 41603729
- Ferreira da Silva J et al. Genome-scale CRISPR screens are efficient in non-homologous end-joining deficient cells. Scientific reports. 2019. PubMed 31673055