# Base Editing: A Precise CRISPR Tool for Single-Letter DNA Changes The genetic code of every living organism is written in just four letters—A, T, G, and C—arranged in a sequence of roughly three billion [Nucleotide Base](/knowledge/molecular-biology/nucleotide-base) pairs in the human genome. A single change in that sequence, such as a C replaced by a T, can cause disease or alter a trait. For decades, scientists could read these letters with ease but struggled to write them with precision. Traditional CRISPR-Cas9 technology offered a way to edit DNA, but it worked by cutting both strands of the double helix, which is a blunt and sometimes destructive approach. Base editing is a newer, more refined CRISPR-based technique that changes a single nucleotide to another without ever cutting both DNA strands. It is, in effect, a molecular pencil that corrects a typo in the genome without tearing out the page. ## What Is Base Editing? Base editing is a genome-editing technology that uses a modified CRISPR system to convert one [Base Pairing](/knowledge/molecular-biology/base-pairing) into a different base pair at a precise location in the DNA. The conversion is chemical, not mechanical: an enzyme attached to a disabled Cas9 protein chemically alters a nucleotide base so that it pairs with a different partner during DNA replication or repair. The result is a permanent, heritable change in the DNA sequence—a single-letter substitution—without introducing a double-strand break (DSB) in the DNA backbone. ### From CRISPR-Cas9 to Base Editing To understand base editing, you first need to understand its predecessor. The classic CRISPR-Cas9 system consists of two key components: a guide RNA (gRNA) that carries a 20-[nucleotide sequence](/knowledge/molecular-biology/nucleotide-sequence) complementary to a target DNA region, and the Cas9 protein, an endonuclease that cuts DNA. When the gRNA finds its matching sequence in the genome, Cas9 binds and introduces a blunt double-strand break about three nucleotides upstream of a sequence motif called the protospacer adjacent motif (PAM), typically NGG for *Streptococcus pyogenes* Cas9. The cell then repairs this break, usually through one of two pathways: non-homologous end joining (NHEJ), which often introduces small insertions or deletions (indels) that disrupt the gene, or homology-directed repair (HDR), which can incorporate a repair template if one is provided. CRISPR-Cas9 is powerful, but it has limitations. The double-strand break is a traumatic event for the cell. NHEJ is error-prone and frequently produces a mixture of different indels, making it difficult to achieve a precise single-nucleotide change. HDR, while capable of precise edits, is inefficient in most cell types and nearly absent in non-dividing cells. Moreover, the presence of a DSB can trigger cellular stress responses, including p53 activation and apoptosis. Base editing was developed to overcome these problems. Instead of cutting both strands, a base editor uses a Cas9 variant that has been mutated to cut only one strand—or no strand at all—and fuses it to a deaminase enzyme that chemically modifies a nucleotide base. The first base editors were described in 2016 by David Liu's laboratory at the Broad Institute. They demonstrated that a cytidine deaminase fused to a nickase Cas9 could convert a cytosine (C) to a uracil (U), which then pairs with adenine (A) during replication, ultimately resulting in a C•G to T•A conversion. ### Why Precision Matters Precision matters because many genetic diseases are caused by point mutations—single-nucleotide substitutions. Sickle cell anemia, for example, results from a single A to T change in the sixth codon of the β-globin gene (*HBB*), converting glutamic acid to valine. Cystic fibrosis can be caused by any of hundreds of different point mutations in the *CFTR* gene. To correct these mutations, you need to change one letter without disturbing the rest of the gene. Base editing is uniquely suited to this task because it makes a targeted chemical change at a single base, leaving the surrounding sequence untouched. ## How Does Base Editing Work? Base editing is a multi-component system that performs a precise chemical reaction at a defined genomic locus. The process can be broken down into the components of the complex and the step-by-step mechanism by which the edit is made. ### The Base Editor Complex A base editor is a fusion protein composed of three main parts: 1. **A catalytically impaired Cas9 (nickase or dead Cas9):** The Cas9 protein is mutated to reduce or eliminate its endonuclease activity. A "dead" Cas9 (dCas9) has both nuclease domains inactivated (D10A and H840A mutations in *S. pyogenes* Cas9), meaning it can bind DNA but cannot cut it. A "nickase" Cas9 (nCas9) has one of the two nuclease domains inactivated, typically the HNH domain (H840A), leaving it able to cut only the non-target strand. Most base editors use nCas9, because nicking the non-edited strand stimulates the DNA repair machinery to replace that strand, which helps lock in the desired edit. 2. **A deaminase enzyme:** This is the catalytic engine of the base editor. Cytidine deaminases (e.g., APOBEC1 from rats, or human AID) convert cytosine to uracil. Adenine deaminases (e.g., an engineered variant of *E. coli* TadA) convert adenine to inosine, which is read as guanine during replication. 3. **A guide RNA (gRNA):** A single-guide RNA (sgRNA) of approximately 100 nucleotides, containing a 20-nucleotide spacer that is complementary to the target DNA sequence. The gRNA directs the base editor to the correct genomic location through Watson-Crick base pairing. Some base editors also include additional components, such as a uracil DNA glycosylase inhibitor (UGI), which prevents the cell's [base excision repair](/knowledge/molecular-biology/base-excision-repair) machinery from removing the uracil created by cytidine deamination, thereby increasing editing efficiency. ### Step-by-Step Mechanism The base editing process proceeds through a series of ordered steps: 1. **Guide RNA binding:** The gRNA forms a complex with the base editor protein. The 20-nucleotide spacer region of the gRNA scans the genome for a complementary DNA sequence. When it finds a match, it forms an RNA-DNA hybrid, creating an R-loop in which the non-target DNA strand is displaced as a single-stranded loop. 2. **Cas9 binding and PAM recognition:** The Cas9 portion of the base editor recognizes a PAM sequence adjacent to the target site. For *S. pyogenes* nCas9, this is typically NGG. The PAM is essential for Cas9 to unwind the DNA and is located on the non-target strand, downstream of the target site. 3. **Deamination of the target base:** The single-stranded non-target DNA in the R-loop is accessible to the deaminase enzyme. For a cytosine base editor (CBE), the cytidine deaminase converts the target cytosine (C) into uracil (U). For an adenine base editor (ABE), the adenine deaminase converts the target adenine (A) into inosine (I). This deamination reaction occurs within a specific "activity window"—typically nucleotides 4–8 of the spacer sequence, counting from the PAM-distal end. 4. **Nick of the non-edited strand:** The nCas9 introduces a single-strand nick on the non-edited (target) strand, opposite the deaminated base. This nick signals the cell that DNA damage has occurred and triggers DNA repair pathways. 5. **DNA repair and base fixation:** The cell's mismatch repair (MMR) machinery recognizes the U•G or I•T mismatch. In the case of a CBE, the uracil is read as thymine during replication, so the U•G pair becomes T•A after one round of replication. The nick on the non-edited strand biases the repair to use the edited strand as the template, so the G on the non-edited strand is replaced with an A. The result is a permanent C•G to T•A conversion. For an ABE, inosine is read as guanine, so the I•T pair becomes G•C after repair, yielding an A•T to G•C conversion. 6. **UGI protection (for CBEs):** In cytosine base editors, the UGI protein inhibits uracil DNA glycosylase, an enzyme that would otherwise remove the uracil and initiate [base excision repair](/knowledge/molecular-biology/base-excision-repair), which could revert the edit back to C. By blocking this pathway, UGI increases the likelihood that the uracil persists and is copied as thymine. ## Types of Base Editors There are two major classes of base editors, each catalyzing a specific type of nucleotide conversion. They are distinguished by the deaminase enzyme they employ and the base pair they convert. ### Cytosine Base Editors (CBEs) Cytosine base editors convert a C•G base pair to a T•A base pair. They achieve this by deaminating cytosine to uracil. The first-generation CBE (BE1) used a rat APOBEC1 cytidine deaminase fused to dCas9. BE2 added a UGI to improve efficiency. BE3, the most widely used version, switched to nCas9 (D10A mutation, which inactivates the RuvC nuclease domain, leaving only the HNH domain active) and retained UGI, achieving editing efficiencies of 20–40% in mammalian cells. The deaminase domain in CBEs acts on single-stranded DNA, which is why the R-loop formed by Cas9 binding is essential. The activity window of APOBEC1-based CBEs is typically positions 4–8 of the spacer, with position 5 being the most efficiently edited. Newer CBEs, such as BE4 and BE4max, include two copies of UGI and optimized nuclear localization signals to increase efficiency. Other cytidine deaminases, such as human AID and *Petromyzon marinus* CDA1 (pmCDA1), have also been used. ### Adenine Base Editors (ABEs) Adenine base editors convert an A•T base pair to a G•C base pair. This is more challenging than C-to-T conversion because there is no natural adenine deaminase that acts on DNA. The adenine deaminase TadA from *E. coli* normally acts on tRNA, not DNA. David Liu's group engineered TadA through [directed evolution](/knowledge/molecular-biology/directed-evolution) to accept DNA as a substrate, creating TadA\* (TadA7.10). ABE7.10, the first functional ABE, fused TadA\* to nCas9 (with the H840A mutation, which inactivates the HNH domain, leaving the RuvC domain active). ABE7.10 achieves A•T to G•C conversion with efficiencies of 20–50% in mammalian cells. The mechanism is analogous to CBEs: the adenine deaminase converts adenine to inosine, which base pairs with cytosine during replication. The nick on the non-edited strand biases repair to replace the original thymine with cytosine, yielding a G•C pair. ABEs do not require UGI because inosine is not recognized by uracil DNA glycosylase. The following table summarizes the key differences between the two types of base editors: | Feature | Cytosine Base Editor (CBE) | Adenine Base Editor (ABE) | |---|---|---| | Target conversion | C•G → T•A | A•T → G•C | | Deaminase enzyme | APOBEC1, AID, pmCDA1 | Engineered TadA (TadA\*) | | Intermediate base | Uracil (U) | Inosine (I) | | Cas9 variant | nCas9 (D10A) | nCas9 (H840A) | | UGI required | Yes (to prevent uracil removal) | No | | Activity window | Positions 4–8 of spacer | Positions 4–7 of spacer | | Typical efficiency | 20–40% | 20–50% | ## Advantages Over Traditional CRISPR Base editing offers several significant advantages over traditional CRISPR-Cas9, particularly for applications that require precise point mutations. ### Fewer Unintended Mutations Traditional CRISPR-Cas9 introduces a double-strand break, which the cell repairs primarily through NHEJ. This pathway is error-prone and frequently produces indels of varying sizes, creating a heterogeneous population of edited cells. Even when HDR is used to introduce a specific mutation, the process is inefficient and often accompanied by unintended indels at the cut site. Base editing, by contrast, does not create a DSB, so it does not trigger NHEJ. The result is a much cleaner edit: the desired base pair change is made without the collateral damage of random insertions or deletions. Studies have shown that base editors produce indels at frequencies below 0.1%, compared to the substantial indel rates seen with Cas9 nuclease. ### Editing Without Cutting Because base editing does not require a double-strand break, it can be used in non-dividing cells, such as neurons, hepatocytes, and muscle cells. HDR, the pathway that allows precise edits with traditional CRISPR, is largely inactive in non-dividing cells because it requires the homologous chromosome as a template during the S/G2 phases of the cell cycle. Base editing, however, relies on the cell's base excision repair and mismatch repair pathways, which are active in all cell types. This makes base editing the only CRISPR-based method that can achieve precise point mutations in post-mitotic cells without relying on viral delivery of a repair template. Additionally, the absence of a DSB reduces the activation of cellular stress responses. Double-strand breaks can trigger p53-mediated cell cycle arrest or apoptosis, which can select against edited cells and reduce overall editing efficiency. By avoiding DSBs, base editing is less toxic to cells and allows for higher survival rates of edited cells. ## Applications of Base Editing Base editing has found applications across [biomedical research](/blog/news/biomedical-research), medicine, agriculture, and basic science. Its ability to make precise, efficient point mutations has opened new avenues for studying gene function and treating genetic disease. ### Correcting Disease-Causing Mutations The most compelling application of base editing is the correction of pathogenic point mutations. Sickle cell disease, for example, is caused by a single A•T to T•A transversion in the *HBB* gene. An adenine base editor can convert the mutant T back to A, restoring the normal β-globin sequence. In 2021, researchers demonstrated that ABE could correct the sickle cell mutation in patient-derived hematopoietic stem cells with efficiencies exceeding 80%, and the corrected cells produced functional hemoglobin when transplanted into mice. Similarly, base editing has been used to correct mutations associated with cystic fibrosis (*CFTR*), Duchenne muscular dystrophy (*DMD*), and familial hypercholesterolemia (*PCSK9*). In each case, the approach is the same: design a gRNA that targets the mutated base, deliver the base editor, and verify that the correction has occurred without introducing off-target edits. Base editing can also be used to introduce protective mutations. For example, a specific point mutation in the *PCSK9* gene is associated with lower LDL cholesterol levels and reduced risk of cardiovascular disease. By introducing this mutation into liver cells, researchers have achieved durable reductions in cholesterol levels in animal models. ### Engineering Crops and Livestock In agriculture, base editing offers a way to introduce beneficial traits into crops and livestock without the regulatory burden associated with transgenic organisms. Because base editing makes small, precise changes to the organism's own DNA, the resulting organisms are often classified as non-transgenic in some regulatory frameworks. In wheat, for example, base editing has been used to introduce mutations in the *MLO* gene that confer resistance to powdery mildew. In rice, base editing has been used to create herbicide-resistant varieties by introducing point mutations in the *ALS* gene. In livestock, base editing has been used to introduce the "double-muscling" mutation in the *MSTN* gene of pigs and cattle, which increases muscle mass and meat yield. Base editing is also used to create disease models in animals. By introducing specific point mutations known to cause human diseases, researchers can create mouse, rat, or pig models that faithfully recapitulate the human condition, enabling the study of disease mechanisms and the testing of new therapies. ## How Is Base Editing Studied? Studying base editing involves designing the components, delivering them into cells, and analyzing the outcomes. Each step requires careful optimization and validation. ### Designing Guide RNAs The first step is to design a gRNA that targets the base of interest. The gRNA must be complementary to a 20-[nucleotide sequence](/knowledge/molecular-biology/nucleotide-sequence) adjacent to a PAM site (NGG for *S. pyogenes* Cas9). The target base must fall within the activity window of the base editor—typically positions 4–8 of the spacer, counting from the PAM-distal end. Several online tools, such as BE-Designer and CRISPResso, can help identify suitable target sites and predict off-target effects. It is important to note that the PAM sequence constrains the targetable space. For a CBE, the target C must be on the non-target strand (the strand that is not complementary to the gRNA), and it must be within the activity window. For an ABE, the target A must be on the non-target strand as well. If the target base is not in an optimal position, alternative Cas9 variants with different PAM specificities (e.g., SpCas9-NG, which recognizes NG PAMs) can be used. ### Delivery Methods Base editors are large proteins—typically 150–200 kDa—which makes delivery challenging. The most common delivery methods are: 1. **Plasmid transfection:** The base editor and gRNA are encoded on a plasmid and transfected into cells using lipid-based reagents or electroporation. This is straightforward for immortalized cell lines but can be inefficient for primary cells. 2. **mRNA and gRNA co-delivery:** The base editor is delivered as in vitro-transcribed mRNA, and the gRNA is delivered separately as a chemically synthesized RNA. This avoids the risk of plasmid DNA integrating into the genome and reduces the duration of base editor expression, which can reduce off-target effects. 3. **Ribonucleoprotein (RNP) delivery:** The base editor protein is purified and complexed with the gRNA in vitro, then delivered into cells by electroporation. This is the most transient delivery method and minimizes off-target edits, but it requires protein purification and is limited by the stability of the RNP complex. 4. **Viral vectors:** Adeno-associated virus (AAV) vectors are commonly used for in vivo delivery because they are non-pathogenic and can transduce a wide range of cell types. However, the packaging capacity of AAV is approximately 4.7 kb, which is too small for most base editors (which are typically 5–6 kb). To overcome this, base editors have been split into two halves that are delivered by separate AAV vectors and reassemble in the cell, or smaller Cas9 orthologs such as *Staphylococcus aureus* Cas9 (SaCas9) have been used. ### Assessing Editing Efficiency After delivery, the editing outcomes must be analyzed. The standard method is PCR amplification of the target region, followed by Sanger sequencing or next-generation sequencing (NGS). The editing efficiency is calculated as the percentage of reads that contain the desired base change. For a typical experiment, the PCR is performed with primers flanking the target site, using a high-fidelity polymerase such as Q5 or Phusion, with an annealing temperature of 60–65°C and 30–35 cycles. The PCR product is then purified and sequenced. For a more detailed analysis, NGS can be used to assess not only the intended edit but also the frequency of bystander edits (unintended base changes within the activity window) and indels. The data are typically analyzed using software such as CRISPResso2, which aligns the sequencing reads to the reference sequence and quantifies the different editing outcomes. ## Limitations and Challenges Despite its advantages, base editing has several limitations that must be considered when designing experiments or therapies. ### Off-Target Effects Base editors can introduce edits at unintended genomic sites that share sequence similarity with the target site. The gRNA can bind to off-target sequences with up to several mismatches, and the deaminase can then act on bases within those sequences. Off-target editing is a particular concern for therapeutic applications, as it could introduce harmful mutations. Several strategies have been developed to reduce off-target effects, including the use of high-fidelity Cas9 variants (e.g., SpCas9-HF1, eSpCas9), the optimization of gRNA length, and the use of transient delivery methods that limit the duration of base editor expression. ### Bystander Edits Within the activity window of a base editor, there may be multiple cytosines or adenines. The deaminase can act on any of them, leading to bystander edits—unintended base changes at positions adjacent to the target base. For example, if a CBE targets a C at position 5 of the spacer, but there is another C at position 6, the editor may convert both to T, resulting in two mutations instead of one. Bystander edits can be problematic if they alter the [amino acid sequence](/blog/guides/amino-acid-sequence) in an unintended way. The use of engineered deaminases with narrower activity windows, such as the YE1 and YE2 mutants of APOBEC1, can reduce bystander editing. ### Delivery Barriers The large size of base editors limits the choice of delivery vehicles, particularly for in vivo applications. AAV vectors, which are the most commonly used for in vivo gene therapy, have a limited packaging capacity. While split-intein approaches have been developed to deliver base editors via dual AAV vectors, the efficiency of this approach is lower than that of a single-vector delivery. Lipid nanoparticles (LNPs) are an alternative delivery method that can deliver mRNA encoding the base editor, but they are primarily taken up by the liver, limiting their use for other tissues. ### Limited Target Windows The requirement for a PAM sequence and the fixed activity window of the deaminase restrict the number of targetable sites. Not every base in the genome can be edited with a given base editor. The development of Cas9 variants with relaxed PAM requirements (e.g., SpCas9-NG, SpRY) and deaminases with shifted activity windows has expanded the targetable space, but some mutations remain inaccessible. ## Common Pitfalls and Misconceptions Several misconceptions about base editing are common among newcomers to the field. Understanding these pitfalls can help you design better experiments and interpret results correctly. ### Base Editing vs. Prime Editing Base editing is often confused with [Prime Editing](/knowledge/molecular-biology/prime-editing), but the two technologies are fundamentally different. Base editing makes a single-nucleotide substitution by chemically modifying a base, and it can only perform C•G to T•A or A•T to G•C conversions. Prime editing, by contrast, uses a Cas9 nickase fused to a reverse transcriptase and a prime editing guide RNA (pegRNA) that carries both the target sequence and the desired edit. Prime editing can make all four types of base substitutions, as well as small insertions and deletions, without requiring a double-strand break or a donor template. However, prime editing is generally less efficient than base editing for simple point mutations. ### Not All Mutations Are Editable Base editing can only make transitions (purine-to-purine or pyrimidine-to-pyrimidine changes): C to T, G to A, A to G, and T to C. It cannot make transversions (purine-to-pyrimidine or pyrimidine-to-purine changes), such as C to A or A to T. If a disease-causing mutation is a transversion, base editing cannot correct it directly. In some cases, it may be possible to edit the complementary strand to achieve the desired result, but this is not always feasible. ### Base Editing Does Not Create Double-Strand Breaks A common misconception is that base editing, like traditional CRISPR, cuts the DNA. It does not. The nCas9 component introduces a single-strand nick, not a double-strand break. This distinction is important because it explains why base editing produces fewer indels and is less toxic to cells. However, it also means that base editing cannot be used to knock out genes by introducing frameshift mutations, which require the insertion or deletion of nucleotides. ### Base Editing Cannot Insert or Delete Large Sequences Base editing is limited to single-nucleotide substitutions. It cannot insert a gene, delete a large genomic region, or introduce a multi-nucleotide sequence. For these applications, traditional CRISPR-Cas9 with a donor template, or prime editing, would be more appropriate. ## Summary and Future Directions Base editing represents a major advance in genome engineering, offering a way to make precise single-nucleotide changes without the collateral damage of double-strand breaks. It has already been used to correct disease-causing mutations in human cells, engineer crops with desirable traits, and create animal models of human disease. However, it is not a universal tool: it is limited to transition mutations, constrained by PAM requirements and activity windows, and subject to off-target and bystander effects. ### Key Takeaways - Base editing is a CRISPR-based technique that converts one base pair to another (C•G to T•A or A•T to G•C) without cutting both DNA strands. - A base editor consists of a catalytically impaired Cas9 (nickase), a deaminase enzyme, and a guide RNA. - Cytosine base editors (CBEs) use cytidine deaminases to convert C to U, while adenine base editors (ABEs) use engineered adenine deaminases to convert A to I. - Base editing produces fewer unintended mutations than traditional CRISPR-Cas9 and can edit non-dividing cells. - Applications include correcting genetic diseases, engineering crops, and creating disease models. - Limitations include off-target edits, bystander edits, limited target windows, and delivery challenges. - Base editing is distinct from prime editing, which can make all types of base substitutions and small indels. ### What's Next for Base Editing? The field is evolving rapidly. Newer base editors with higher fidelity, such as the "SECURE" (S. aureus Cas9-based engineered) variants, show reduced off-target activity. The development of base editors with expanded PAM compatibility, such as SpRY-CBE and SpRY-ABE, has made nearly any genomic site targetable. Researchers are also exploring the use of base editing in combination with other tools, such as [Base Excision Repair](/knowledge/molecular-biology/base-excision-repair) inhibitors, to further improve efficiency. In the clinic, base editing is being tested in early-phase clinical trials for conditions such as sickle cell disease and β-thalassemia. The first in vivo base editing therapy, VERVE-101, which targets *PCSK9* to lower LDL cholesterol, entered clinical trials in 2022. These trials will provide critical data on the safety and efficacy of base editing in humans. The future of base editing will likely involve the development of editors with even narrower activity windows, reduced off-target effects, and the ability to perform transversion mutations. The integration of base editing with [Organism That RNA Editing Involves Enzymes](/knowledge/molecular-biology/organism-that-rna-editing-involves-enzymes) and [Biologist Offering RNA Editing Tools](/knowledge/molecular-biology/biologist-offering-rna-editing-tools) may also expand the toolkit for correcting genetic information at both the DNA and RNA levels. As the technology matures, base editing is poised to become a standard tool in molecular biology, with applications ranging from basic research to gene therapy. ## Frequently Asked Questions ### What is base editing? Base editing is a genome-editing technology that uses a modified CRISPR system to convert one nucleotide base pair to another at a precise location in the DNA. It makes single-letter changes—such as C•G to T•A or A•T to G•C—without cutting both strands of the DNA double helix. ### How does base editing work? A base editor consists of a catalytically impaired Cas9 protein (nickase), a deaminase enzyme, and a guide RNA. The guide RNA directs the complex to the target DNA sequence. The deaminase chemically modifies a specific nucleotide base—converting cytosine to uracil or adenine to inosine—and the cell's DNA repair machinery then fixes the opposite strand, resulting in a permanent base pair change. ### What is the difference between base editing and CRISPR-Cas9? Traditional CRISPR-Cas9 introduces a double-strand break in the DNA, which is repaired by error-prone NHEJ or inefficient HDR. Base editing does not cut both strands; it makes a chemical change to a single base and uses the cell's repair machinery to lock in the edit. This results in fewer unintended mutations and allows editing of non-dividing cells. ### What are the types of base editors? There are two main types: cytosine base editors (CBEs), which convert C•G to T•A using cytidine deaminases such as APOBEC1, and adenine base editors (ABEs), which convert A•T to G•C using an engineered adenine deaminase called TadA\*. ### What are base editors used for? Base editors are used to correct disease-causing point mutations in human cells, introduce beneficial traits in crops and livestock, create animal models of human disease, and study the effects of specific nucleotide changes on gene function. ### What are the limitations of base editing? Base editing can only make transition mutations (C to T, G to A, A to G, T to C), not transversions. It is also limited by PAM sequence requirements, a fixed activity window that can cause bystander edits, potential off-target effects, and challenges in delivering the large base editor protein into cells. ### Is base editing safe? Base editing is generally safer than traditional CRISPR-Cas9 because it avoids double-strand breaks and produces fewer indels. However, it can still cause off-target edits and bystander edits, which could be harmful. The safety of base editing is being evaluated in clinical trials, and ongoing research aims to improve its specificity and reduce unintended effects. ## Further Reading - Kantor A, McClements ME, MacLaren RE. *CRISPR-Cas9 DNA Base-Editing and Prime-Editing*. International journal of molecular sciences. 2020. [PubMed 32872311](https://doi.org/10.3390/ijms21176240) - Porto EM et al. *Base editing: advances and therapeutic opportunities*. Nature reviews. Drug discovery. 2020. [PubMed 33077937](https://doi.org/10.1038/s41573-020-0084-6) - Coelho MA et al. *Base editing screens map mutations affecting interferon-γ signaling in cancer*. Cancer cell. 2023. [PubMed 36669486](https://doi.org/10.1016/j.ccell.2022.12.009) - Martin-Rufino JD et al. *Massively parallel base editing to map variant effects in human hematopoiesis*. Cell. 2023. [PubMed 37137305](https://doi.org/10.1016/j.cell.2023.03.035) - Cuella-Martin R et al. *Functional interrogation of [DNA damage response](/knowledge/molecular-biology/dna-damage-response) variants with base editing screens*. Cell. 2021. [PubMed 33606978](https://doi.org/10.1016/j.cell.2021.01.041) - Hooper AJ, Tang XL, Burnett JR. *VERVE-101, a CRISPR base-editing therapy designed to permanently inactivate hepatic PCSK9 and reduce LDL-cholesterol*. Expert opinion on investigational drugs. 2024. [PubMed 38878270](https://doi.org/10.1080/13543784.2024.2369747)
## Related Clinical & Scientific Guides * [MAPK Pathway: Mechanism, Function, and Clinical Relevance](/knowledge/molecular-biology/mapk-pathway) * [Mammalian Cell Culture Bioreactors: A Practical Guide](/knowledge/molecular-biology/mammalian-cell-culture-bioreactor) * [Nucleotide Formation: Biosynthesis and Assembly of DNA/RNA Building Blocks](/knowledge/molecular-biology/nucleotide-formation)