BMB Reports 2024; 57(1): 19-29  https://doi.org/10.5483/BMBRep.2023-0224
Mitochondrial genome editing: strategies, challenges, and applications
Kayeong Lim1,2,*
1Brain Science Institute, Korea Institute of Science and Technology (KIST), Seoul 02792, 2Division of Bio-Medical Science & Technology, KIST School, Korea University of Science and Technology, Seoul 02792, Korea
Correspondence to: Tel: +82-2-958-7234; Fax: +82-2-958-7034; E-mail: kylim@kist.re.kr
Received: November 10, 2023; Revised: December 12, 2023; Accepted: December 21, 2023; Published online: January 4, 2024.
© Korean Society for Biochemistry and Molecular Biology. All rights reserved.

cc This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Mitochondrial DNA (mtDNA), a multicopy genome found in mitochondria, is crucial for oxidative phosphorylation. Mutations in mtDNA can lead to severe mitochondrial dysfunction in tissues and organs with high energy demand. MtDNA mutations are closely associated with mitochondrial and age-related disease. To better understand the functional role of mtDNA and work toward developing therapeutics, it is essential to advance technology that is capable of manipulating the mitochondrial genome. This review discusses ongoing efforts in mitochondrial genome editing with mtDNA nucleases and base editors, including the tools, delivery strategies, and applications. Future advances in mitochondrial genome editing to address challenges regarding their efficiency and specificity can achieve the promise of therapeutic genome editing.
Keywords: Mitochondrial DNA mutations, Mitochondrial genome editing, mtDNA base editor, mtDNA nuclease, Protein-based mtDNA editing tools
INTRODUCTION

Mitochondria are essential organelles that play a crucial role in energy production and managing the cellular stress response. These organelles consist of their own genome, known as mitochondrial DNA (mtDNA), which is located within the mitochondrial matrix bounded by the double-membrane system (1). A complex interplay of genetic factors governs mitochondrial function with control by both nuclear DNA and mtDNA (2, 3). Human mtDNA is a 16,569 base-pair (bp) double-stranded circular DNA that contains 37 genes that are essential for oxidative phosphorylation (OXPHOS) (1, 4). These genes include 13 subunits of protein complexes of the OXPHOS system, two ribosomal RNAs, and 22 transfer RNAs that are required for mitochondrial translation. Unlike nuclear DNA, which is typically present in two copies per cell, mtDNA copy numbers vary widely depending on cell type, with up to 100,000 copies per cell (5).

Mutations in mtDNA, whether inherited or acquired somatically, can lead to mitochondrial dysfunction, resulting in respiratory chain deficiency. Such mutations have been linked to mitochondrial disease, aging, and cancer (1, 6, 7). Approximately 100 diverse mtDNA mutations have been confirmed to be clinically pathogenic, causing diseases such as Leigh syndrome, Mitochondrial encephalopathy, lactic acidosis and stroke-like episodes (MELAS) syndrome, Leber hereditary optic neuropathy (LHON), and Myoclonic epilepsy and ragged-red fiber disease (MERRF) (8). These pathogenic mtDNA mutations are primarily comprised of point mutations (95/100), in addition to deletions (2/100), insertions (2/100), and inversions (1/100). The prevalence rate for mtDNA mutations is estimated to be 1 in 5,000 adults (3), making them relatively common; however currently, there is no direct treatment.

The mtDNA mutations can result in two states: heteroplasmy, where the cell harbors both mutant and wild-type mtDNA, or homoplasmy, where all mtDNA carries the same mutation (9). Heteroplasmy levels may fluctuate during the cell division (mitotic or meiotic). The presence of several copies of mtDNA in cells provides a mechanism to tolerate abnormal mtDNA by balancing with normal wild-type mtDNA. Only when the level of mutant mtDNA exceeds a specific threshold, known as the mitochondrial threshold effect, do the phenotypic manifestations of mtDNA mutations emerge (10). Lowering the level of mutant mtDNA below this threshold can alleviate the onset of mitochondrial disease (Fig. 1, middle).

CRISPR genome editing technologies, including CRISPR-nucleases (11), base editors (12, 13), and prime editors (14), can efficiently correct or install different variations in the nuclear genome. These CRISPR systems make use of guide RNA to pinpoint specific loci in the genome. Unfortunately, when it comes to mitochondrial genome editing, this process is hindered by challenges in delivering RNA into the mitochondria (15). Successful manipulation of the mitochondrial genome has been achieved exclusively using RNA-free, protein-based tools, including programmable nucleases and base editors.

This mini review provides an overview of the mitochondrial genome editing tools developed to date, as well as outlining the unique mitochondrial environment that enables these tools to function effectively. Moreover, the delivery strategies for accessing the mitochondrial genome are examined, and their practical applications discussed. Finally, the challenges remaining in the field of mitochondrial genome editing, which must be addressed to enhance its therapeutic potential, are explored.

DEVELOPMENT OF MITOCHONDRIAL GENOME EDITING TECHNIQUES

Mitochondrial DNA repair

Genome editing tools directly modify DNA, and the outcomes of these modifications depend on the cellular DNA repair pathways and replication processes. The mtDNA repair mechanisms either degrade or repair damaged mtDNA, thereby maintaining the mitochondrial function (16). While mtDNA repair mechanisms are less characterized than nuclear DNA repair, mtDNA repair enzymes are encoded by nuclear genes, and transported to the mitochondria after translation. The enzymes related to the DNA repair mechanisms, which include base excision repair (BER) and mismatch repair (MMR) among others, are found within the mitochondria, providing evidence of the existence of the mitochondrial DNA repair system (16, 17).

When double-strand breaks (DSBs) occur in mtDNA, mitochondria tend to eliminate the damaged mitochondrial genome, due to the inefficiency of DSB repair (16, 18). This depletion of mtDNA then triggers the replication of the remaining mtDNA to preserve the copy number and mtDNA pool. Conversely, in cases of damaged base caused by oxidation, deamination, alkylation, or methylation, the damaged mitochondrial genome is repaired through the BER pathway, which is the primary repair mechanism within the mitochondrion (19). In the BER pathway, a DNA glycosylase recognizes and removes the damaged base, leading to the formation of an abasic (apurinic/apyrimidinic, AP) site. Subsequently, the AP site is processed through the short-patch or long-patch mechanisms.

Strategies for editing the mitochondrial genome

To effectively manipulate the mitochondrial genome, it is essential to introduce stable mutations that can overcome the natural mechanisms responsible for mitochondrial genome maintenance and mtDNA repair. In addition, the mitochondrial genome is located within the double-membraned structure of the mitochondria, necessitating a unique approach to render mitochondrial editing tools that are accessible to the mitochondrial genome. The mitochondrial targeting sequence (MTS) is a short signal peptide at the N’-terminus of proteins that guides the transport of a protein to the mitochondria (20). Fusing the MTS to the terminus of the genome editing protein can successfully deliver it into the interior of the mitochondria. Therefore, a common feature of mitochondrial genome editing tools is their use of RNA-free, protein-based systems with MTS fusion. Mitochondrial genome editing tools are typically categorized into two types: nucleases, and base editors.

Nucleases for mitochondrial genome editing

MtDNA nuclease proteins consist of DNA-binding proteins (e.g., zinc finger arrays or TAL-effector [TALE] arrays), which recognize specific DNA sequences, with DNA cleavage domains (e.g., FokI endonucleases). Nucleases are designed to create double strand breaks (DSBs) in mtDNA, leading to mtDNA elimination, rather than repair (18). Pathogenic mutant sequence-specific nucleases can selectively remove only the mutant mtDNA, resulting in a shift in the mutant mtDNA levels within heteroplasmic populations through mtDNA copy number recovery. Consequently, when the level of mutant mtDNA drops below the pathogenic threshold, cells can exhibit a normal phenotype (Fig. 1, left).

Various nuclease-based tools have been developed for mitochondrial genome editing, including zinc finger nucleases (ZFNs) (21-23), TALE nucleases (TALENs) (24-29), mitochondria-targeted restriction endonucleases (24, 30-34), mitoTev-TALE (35), and mitoARCUS (36) (Fig. 2A-F). ZFNs and TALENs operate by fusing customizable DNA-binding domains, such as Zinc finger arrays and TALE arrays, with the endonuclease domain from the FokI enzyme (Fig. 2D, E). To cleave the target DNA, obligatory heterodimeric FokI needs to dimerize on the DNA substrate. ZFNs (21-23) and TALENs (24-29) are often used as dimers by connecting each FokI endonuclease to a pair of zinc finger arrays or TALE repeats. Alternatively, they can function as monomers using a single-chain version of FokI (Fig. 2F) (37). In contrast, mitochondria-targeted restriction endonucleases (24, 30-34), mitoTev-TALE (35), and mitoARCUS (36) utilize restriction endonucleases or homing endonucleases (Fig. 2A-C). As these tools require a fixed DNA sequence for their functioning, they can only be designed for specific cases.

The design of mutant sequence-specific nucleases is the most crucial factor in determining the success of mitochondrial genome editing. The lower the sequence similarity between wild-type and mutant mtDNA, the more likely it is to work successfully. Moreover, nucleases are not the optimal choice for homoplasmic mutations; even after depleting the mutant mtDNA, only the mutant mtDNA remains, making it impossible to induce changes in the mtDNA population. Additionally, generating new mtDNA mutations to model mitochondrial disease is challenging, as selective removal of mtDNA is a major outcome.

Base editors for mitochondrial genome editing

Base editing technology for nuclear DNA was initially developed before the emergence of programmable base editors for mitochondrial DNA (12, 13). Nuclear DNA base editors utilize CRISPR nickases connected with cytosine or adenine deaminases acting on single-stranded DNA (ssDNA). When the guide RNA of CRISPR nickase binds to the DNA target loci, it unwinds the DNA to form an RNA−DNA hybrid, and exposes the displaced ssDNA without generating a DSB. Subsequently, deaminases convert cytosine or adenine in the exposed ssDNA to uracil or hypoxanthine, which is further transformed into thymine or guanine, respectively, during mtDNA replication. The CRISPR-guide RNA is essential for the functioning of ssDNA deaminase in base editing; however, due to the inefficiency of RNA delivery into mitochondria, CRISPR-based editors are unattainable for mtDNA base editing (15).

The discovery of DddAtox, a novel cytidine deaminase that operates on double-stranded DNA (dsDNA), has instigated new possibilities for mitochondrial base editing by developing the DddA-derived cytosine base editors (DdCBEs) (Fig. 1, right, and Fig. 2G) (38). DddAtox is the deaminase domain of the interbacterial toxin DddA originating from Burkholderia cenocepacia (Ddd_Bc), which DddA when delivered intracellularly, causes cellular toxicity. To reduce this toxicity and make it suitable for utilization, DddAtox is split into two segments, and engineered to function in hetero-dimer manner. The DdCBE pair is created by fusing each of the split-DddAtox halves (with the split site at either G1333 or G1397) with TALE arrays, a uracil glycosylase inhibitor (UGI), and a mitochondrial targeting sequence (MTS). Activation of split-DddAtox occurs when their halves are brought together by adjacent TALE arrays and induce C-to-U deamination on both DNA strands containing the TC sequence motif. Damaged bases caused by deamination are typically removed by DNA glycosylase within the BER pathway. However, the UGI in DdCBE inhibits the function of uracil DNA glycosylases, preserving the uracil, which eventually transforms into thymine during mtDNA replication (Fig. 1, right). Comprising an optimal combination (with MTS, mtDNA targeting TALE array, split DddAtox half, and UGI), DdCBE induces C-to-T conversion in mtDNA with very high efficiency.

Another advance is the development of transcription-activator-like effector (TALE)-linked deaminases (TALED), which allow for A-to-G conversion in mtDNA (Fig. 2H) (39). The TadA8e deaminase is an engineered protein derived from tRNA-specific adenosine deaminase (TadA) (40), originally used in CRISPR-adenine base editors, where it functions on ssDNA generated by CRISPR guide RNA. To enable TadA8e to deaminate in mtDNA, ssDNA must be created through the RNA-free system. Surprisingly, when DddAtox binds to the dsDNA substrate, it transiently generates the ssDNA (41), providing a substrate for TadA8e. The TALED system was constructed from DdCBE by adding the adenine deaminase TadA8e and removing UGI, which is not involved in adenine deamination. As a result, the TALED system, consisting of TALE arrays, split-DddAtox halves, and TadA8e, is an efficient adenine base editor for mitochondrial genome editing. Depending on the combination of components in TALED systems, different versions (split TALED, sTALED; dual TALED, dTALED; and monomeric TALED, mTALED) can be designed (Fig. 2H).

CHALLENGES AND IMPROVEMENTS IN mtDNA BASE EDITING

Sequence preference and base editing efficiency

The DddAtox cytosine deaminase displays a sequence preference for the TC motif, making DdCBE a valuable tool for editing TC sequences within the mtDNA target (38). However, the original DdCBE shows reduced efficacy toward AC or CC sequences and has little effect on GC sequences, limiting its overall utility. To overcome these limitations, evolved DddAtox variants were developed through protein engineering using rapid phage-assisted continuous evolution (PACE) and related phage-assisted non-continuous evolution (PANCE) methods (42). These variants include DddA6, which has shown better performance in editing TC targets, and DddA11, which exhibits enhanced performance in editing mtDNA sequences with HC (H = T, A, or C) targets (Fig. 2G).

An alternative approach is to use a different type of dsDNA deaminase. This approach involves replacing Burkholderia cenocepacia DddAtox (Ddd_Bc) with homologous and orthologous proteins identified based on the amino acid sequences of DddA (Fig. 2G) (43-46). Specifically, DdCBE_Ss (=FZY2−DdCBE), Q2L7−DdCBE, and RsDdCBE use a DddA homolog from Simiaoa sunii (Ddd_Ss = FZY2) (43, 44), Streptomyces sp-BK438 (Q2L7) (44), and Ruminococcus sp. (RsDddA) (45), respectively; while mitoCBE2.0 uses a DddA ortholog from Roseburia intestinalis (Ddd_Ri) (46). These base editors can even target GC sequences that were previously inaccessible. Q2L7−DdCBE demonstrates a strong preference for GC targets, while DdCBE_Ss function on DC (D = T, A, or G). Moreover, RsDdCBE and mitoCBE2.0 act on NC (N = T, A, C, or G) targets, thereby enhancing sequence compatibility. In addition, the DdCBE_Bc (E1370N) variant demonstrates higher editing efficiency by incorporating variations from Ddd_Ss into Ddd_Bc, compared to the original DdCBE.

Using AI-assisted structural prediction, ssDNA deaminases (Sdds) and dsDNA deaminases (Ddds) are newly discovered through protein structure-based clustering (47). The DdCBE variants containing selected Ddds have similar or higher editing efficiencies than those using the original Ddd_Bc. Ddd1 and Ddd9 show a preference for GC targets, while Ddd8 targets WC (W = A or T) sequences, and Ddd7 exhibits a sequence preference for TC targets (Fig. 2G).

Thus, various versions of improved DdCBEs have been developed, expanding the sequence preference of DddAtox, and improving their base editing efficiency.

Additional accessories for improving base editing efficiency

The overall efficiency of base editing can be improved by combining additional elements with the mtDNA editing tools. To deliver the mitochondrial base editor into the mitochondria more effectively, DdCBE-NES improves efficiency by incorporating a nuclear export signal (NES) with the DdCBE (48). Additionally, the efficiency of DdCBE can be further enhanced by co-injecting mtDNA-targeted TALEN (mitoTALENs) to eliminate unedited mtDNA (48).

Another method involves applying a transactivation domain (such as VP64, P65, or Rta) to mtDNA base editors, which by regulating chromatin accessibility, is known to boost the editing efficiency of CRISPR-based editors (46). MitoCBE2.1, which results from fusing the transactivator Rta to the tail of mitoCBE2.0 (DdCBE_Ri), demonstrates moderate improvements, averaging a 1.7-fold increase in mtDNA base editing efficiency. Further combinations with additional factors influencing mtDNA base editing will likely expand the scope of base editing within the mitochondrial genome in the future.

Minimizing the size of the mtDNA base editor

MtDNA base editors, like DdCBEs and TALEDs, rely on TALE arrays as DNA sequence-specific binding proteins. These TALE arrays consist of repeating units of about 34 amino acids, with each unit targeting a single DNA nucleotide. To target a 12−20 nucleotide sequence of DNA, 12−20 TALE repeats must be linked together, resulting in large overall size (38, 39). Additionally, when split-DddAtox halves are used, mtDNA base editors must be used in pairs, further increasing the size of the constructs required for base correction. Unfortunately, the substantial size of these base editors poses a challenge when using them in delivery systems with size restrictions, such as AAV.

To address these size limitations, researchers developed Zinc finger deaminase (ZFD) (49) and ZF−DdCBE (50), which use zinc finger arrays as an alternative DNA-binding protein (Fig. 2I, J). Zinc finger arrays are compact proteins that recognize three DNA nucleotides per ZF module, with 3−6 fingers connected to recognize 9−18 nucleotides of the target DNA. The ZF array is significantly smaller than half of the TALE array. Additionally, ZF proteins from mammalian sources exhibit low immunogenicity. Compact ZF arrays can be engineered friendly, allowing for the fusion of split-DddAtox halves at either the N’− or C’− terminus of the ZF array (49). ZFD and ZF−DdCBE, created by connecting split DddAtox halves and UGI to the zinc finger array, effectively induce mtDNA base editing. ZFD can be used in a hybrid pair form with DdCBE, and ZFD pairs or ZFD/DdCBE hybrid pairs can generate unique mutation patterns that are unattainable with DdCBEs alone (49). ZF−DdCBE, constructed in parallel to ZFD, further enhances its operational efficiency through architectural optimization and improvements to ZF scaffolds (50). Moreover, the compact size of ZF−DdCBEs enables their successful delivery via a single AAV9 system into the heart, liver, and skeletal muscle of postnatal mice for the introduction of mtDNA mutations associated with disease (50).

The development of a full-length DddAtox by reducing the toxicity of DddAtox allows for the use of a single construct, reducing the size of the mtDNA base editor. The GSVG variant, obtained through random mutagenesis of DddAtox, functions as a monomeric deaminase, reducing intracellular toxicity, while maintaining efficient base editing (51). Unlike DdCBE with Split-DddAtox, which operates as a heterodimer in a paired system, DdCBE with the GSVG variant can be used in a monomeric form (mDdCBE) (Fig. 2K). Consequently, mDdCBE can be packaged into AAV, leading to successful base editing in human cell lines using AAV-mDdCBE.

Off-target base editing

Deaminases, originally toxin proteins that trigger DNA deamination, can lead to unintended base editing in both mtDNA and nuclear DNA. The in vivo genome-wide off-target analysis method, known as genome-wide off-target analysis by two-cell embryo injection (GOTI), has confirmed that DdCBE induces genome-wide off-target base editing within the nuclear genome. Considering that DdCBE utilizes the mitochondrial targeting signal (MTS) at the N’-terminus for mitochondrial localization, it raises the concern that the MTS may not effectively prevent DdCBE from entering the nucleus in mammalian cells (52).

Moreover, using the Detect-seq method for in vitro off-target assessment by capturing dU in the extracted DNA from DdCBE-treated cells, it has been revealed that DdCBE induces hundreds of off-target sites, including TALE array sequence (TAS)-dependent and TAS-independent (53). A one-sided TALE array is sufficient to guide the spontaneously assembled DddAtox, which causes TAS-dependent off-target editing. TAS-independent off-target sites are frequently found near CTCF-binding sites. Engineered DdCBEs, such as UGI−NES−DdCBEs with NES fused to the C’-terminus or DddIA−DdCBE with separately co-expressed DddIA (an inhibitor of the DddAtox) fused to nuclear localization signals (NLSs), effectively reduce off-target base editing, while enhancing the specificity of DdCBE (Fig. 2L). Similarly, the homolog pair of DddA−DddIA, DddIA(FZY2) of the FZY2 deaminase also exhibits the inhibitory effects, while co-expression of NLS−DddIA(FZY2) minimizes the nuclear DNA off-targeting of FZY2−DdCBE (44).

Furthermore, the spontaneous assembly of split DddAtox has the potential to cause off-target base editing, so DdCBE is engineered by introducing mutations in amino acid residues at the split dimer interface to prevent TALE-independent assembly. HiFi−DdCBEs, DdCBE variants with K1389A, T1391A, or V1411A mutations in DddAtox, depend on DNA-binding by TALEs, and reduce mitochondrial genome-wide off-target base editing (Fig. 2M) (54).

Efforts to enhance the specificity of the tools have also been made in zinc finger-based base editors. ZFD with the QQ variant, achieved by incorporating R(−5)Q mutations in each zinc finger in the ZFDs, inhibits non-specific DNA contact in ZF, thus reducing off-target base editing and improving specificity (Fig. 2I) (49). To enhance the specificity of ZF−DdCBE, high-specificity (HS) ZF−DdCBE variants have been developed using strategies such as truncating split-DddAtox halves, introducing point mutations in DddAtox-C, increasing electrostatic repulsion, and fusing a catalytically inactivated DddAtox-N to hinder target-independent reassembly (Fig. 2J) (50).

Ensuring precise genome editing of the mitochondrial genome, especially when utilizing mtDNA editing tools to specifically remove or replace mutant mtDNA sequences, is crucial to prevent severe and hazardous consequences from off-target editing. Mitochondrial genome editing tools, which are generated by fusing functional domains, such as nucleases or deaminases, to DNA-binding proteins, can further enhance specificity by refining the operation of functional domains and DNA-binding proteins, making the process more precise.

Base editing window of the mtDNA base editor

The precision of genetic modifications within the target mtDNA is defined by the base editing window of the mtDNA base editor. This base editing window refers to the region where the mtDNA base editor can effectively induce nucleotide changes. The editing window is influenced by the molecular characteristics of the mtDNA base editor, including its catalytic deaminase domain and the DNA-binding protein for targeting. Base editing with a pair of mtDNA editors generally occurs in the spacer region, which is located between the binding regions of two DNA-binding proteins (38, 39, 42, 49). When used in monomeric form, base editing occurs in one direction of the binding region of the DNA-binding protein (39, 51). The presence of multiple editable nucleotides within the base editing window introduces the possibility of unwanted base changes occurring at sites other than the primary target nucleotide, known as bystander editing (55, 56). In therapeutic applications, the ultimate goal is to harness the precision of base editing to correct disease-causing genetic mutations with minimal bystander editing, ensuring that only the intended genetic changes occur. The choice of the appropriate mtDNA base editor is important depending on the purpose of use, as it determines the position of base editing within the editing window.

The strategic positioning of DNA-binding proteins, such as TALED-binding sites (39), ZFD-binding sites, and hybrid ZFD/DdCBE-binding sites (49), is a decisive factor in determining the position of base editing. By carefully manipulating DNA-binding sites, researchers can effectively modulate the editing window, demonstrating sophisticated control over the molecular tools used in mtDNA base editing. Additionally, DddA11, a variant created through protein engineering, exhibits a broader base editing window compared to the original DddAtox due to catalytic changes in the deaminase domain (42). Conversely, DddAtox variants with K1389A or V1411A mutations show a narrowed editing window (54). To refine mitochondrial genome editing technologies, controlling the base editing window and minimizing bystander effects remain key objectives for the successful application of mtDNA engineering approaches.

Strand-selective mtDNA base editing

To date, mtDNA base editors have relied on double-stranded DNA deaminase, and have worked on both the top and bottom strands of dsDNA. However, recent developments have introduced strand-selective mtDNA base editing using mitoBEs (57) and CyDENT (58). These advances use ssDNA deaminase and DNA nickase, allowing for more precise editing. By applying nickase to create a nick in only one strand near the target loci, transient ssDNA is generated, resulting in deamination of the exposed adenine or cytosine in that ssDNA. During the repair of the nicked strand and replication of the mtDNA, the deaminated adenine or cytosine remains in the unnicked strand, resulting in A-to-G or C-to-T editing on only one strand.

MitoBEs (57) are created by combining the nickases MutH, MutH*, or Nt.BspD6I(C) with ssDNA-specific deaminases like TadA8e adenine deaminase or UGI-linked rAPOBEC1 cytosine deaminase. This combination produces mitoABEMutH and mitoABENt.BspD6I(C) for A-to-G editing, and mitoCBEMutH and mitoCBENt.BspD6I(C) for C-to-T editing in mtDNA (Fig. 2N, O). The effectiveness of strand-selective base editing of mtDNA depends on several factors, such as the sequence preference of the nickase, the distance between TALE-binding and the nick motif, and the architecture of the nickase and TALE. Nickase MutH operates on the GATC sequence motif, MutH* on the GAT sequence motif, while Nt.BspD6I(C) has no specific sequence preference. These mitoBEs can be configured as dimeric or monomeric units. Monomeric mitoBEs provide greater ease of delivery, especially when using adeno-associated virus (AAVs) as a vector. An additional system, cytidine deaminase-exonuclease-nickase-TALE (CyDENT) (58), utilizes TALE for DNA-specific binding, FokI nickase for one strand nicking, exonuclease for recognizing and digesting the nicked DNA strand, and ssDNA cytosine deaminase with UGI for cytosine deamination (Fig. 2P). These components can be used separately or together, enabling precise mtDNA base editing on a particular DNA strand, and facilitating cytosine or adenine changes.

DELIVERY OF MITOCHONDRIAL GENOME EDITING AND APPLICATIONS

Delivery platforms for mitochondrial genome editing

MtDNA nucleases and base editors, developed for mitochondrial genome editing, are protein-based tools that need to be efficiently delivered into the mitochondrial matrix to gain direct access to the mitochondrial genome. The effectiveness of these tools depends on the delivery system used, with different preferred delivery forms and methods for various cells, tissues, and organisms. DNA or RNA, encoding the genome-editing protein, is delivered into the cell, transcribed in the nucleus, translated in the cytoplasm, and then directed to the mitochondrial matrix via a mitochondrial targeting sequence that is located at the protein’s terminus. This can be achieved using non-viral delivery vehicles, such as liposomes, lipid nanoparticles, or polyplexes; physical delivery systems, like electroporation or microinjection; or viral vectors, like AAV (Fig. 3).

Non-viral delivery carriers are the most used systems for transfecting cultured cells. These systems include lipofectamine and polyethylenimine (PEI), among others (59). Using the cationic properties of vehicles, they form condensed complexes with negatively charged nucleic acids through electrostatic interactions. These transfection complexes enter the cell through endocytosis, and are released via endosomal escape (60). They can easily be utilized by simply mixing them with nucleic acids to form complexes, and directly delivering them to cell culture media. As a result, they are often employed to develop mitochondrial genome editing tools and assess their efficiency on a large scale in cells. Plasmid DNA, mRNA, and circular RNA (57), encoding mtDNA editing proteins, have been successfully delivered to human cells (21, 25, 38, 39, 41-47, 50, 51, 54, 57, 58), including HEK293T, HCT116, and HT1080 cells, as well as mouse cells (50, 61), including NIH3T3 and C2C12 cells, demonstrating efficient mitochondrial genome editing.

The physical delivery system imports nucleic acids by creating a temporary pore in the cell membrane; for example, by electroporation, which uses electrical pulses, or microinjection, which employs a fine needle. Electroporation is commonly used for transfecting various cell types, including human, mouse, rat, and patient-derived cells. In addition, the delivery of genome engineering tools in the form of mRNA into embryonic cells is a widely practiced method (24, 46, 48, 55, 61-67). This allows for mitochondrial genome editing in the early stages of embryonic development, and the generation of new models for genetic disease. Through microinjection of embryos, mtDNA-associated disease models have been created using mitochondrial genome editing in mouse (46, 48, 55, 61, 62), zebrafish (63, 64), and rat (65-67). Moreover, the efficiency and accuracy of mitochondrial genome editing have been evaluated by applying the microinjection method to clinically discarded human embryos with three pronuclei (3PN) or two pronuclei (2PN) (56, 68).

Viral delivery methods often utilize adeno-associated viruses (AAV), adenovirus, and lentivirus, which offer several advantages, such as the ability to target specific tissues through various serotypes and promoters (69). AAVs facilitate the transfer of genome editing tools to specific tissues in neonatal or adult mice, thereby validating the potential of genome editing therapeutics (23, 27, 31, 32, 36, 50, 51, 70). Due to the size limitation of AAV, which is approximately 4.8 kilobases, researchers employ either single-AAV or dual-AAV delivery, depending on the specific genome editing tool. Importantly, AAV is considered a safe option, in comparison to other viral systems that can integrate into the host genome. However, emerging safety concerns have been raised regarding the use of high-dose AAV vectors (71).

Applications in mitochondrial genome editing

The applications of mtDNA nucleases are focused on the elimination of pathogenic mutation and improving the symptoms by reducing heteroplasmy levels below the threshold. mtDNA nucleases can selectively target and deplete the specific mtDNA mutations with substitution and deletion in cultured cybrid cells (21, 22, 24-26, 30, 33-35, 37), patient-derived cells (25, 28, 29), NZB/BALB heteroplasmic mouse (24, 31, 32), and the m.5024C>T mouse model (23, 27, 36). In particular, the AAV-delivery method is utilized to deliver the mtDNA nucleases to mouse in vivo (23, 27, 31, 32, 36) to shift the heteroplasmy, which helps to alleviate pathological outcomes.

On the other hand, mtDNA base editors can be used to both introduce pathogenic point mutations to generate disease models, and correct pathogenic point mutations. In the cultured cell lines, utilizing various delivery strategies as previously mentioned, mtDNA base editors can target virtually all genes within the mtDNA, encompassing human, mouse, and rat cells. Furthermore, in vivo mtDNA base editing has been confirmed in various organisms, including mouse (46, 48, 50, 55, 61, 62, 70), zebrafish (63, 64), rat (65-67), and even plant (72). However, the effectiveness of target achievement and efficiency may vary depending on the design rules of individual mtDNA base editors. Additionally, since mtDNA base editors operate dependently on DNA deaminase, they can induce changes limited to C-to-T or A-to-G substitutions.

Given that mtDNA nucleases and base editors serve different purposes, they can be selected and used according to the specific goals. The advancements of these two tools are expected to make significant contributions to the understanding of mitochondrial function and the treatment of related diseases in the future.

CONCLUSION

The mitochondrial genome, governed by its multicopy, circular DNA known as mtDNA, plays a pivotal role in cellular energy production and stress response. Mutations in mtDNA can lead to mitochondrial dysfunction, impacting tissues and organs with high energy demand, and are closely associated with a spectrum of disease, including mitochondrial disorders and age-related conditions. This review has shed light on the ongoing efforts to manipulate the mitochondrial genome using mtDNA nucleases and base editors, highlighting the tools, delivery strategies, and potential applications. As the field of mitochondrial genome editing continues to evolve, the promise of therapeutic genome editing is becoming more tangible. However, for mitochondrial genome editing to reach its full therapeutic potential, challenges related to efficiency and specificity still need to be addressed. In addition to the advancements discussed in this review, it is worth noting that adapting the CRISPR system for mitochondrial genome editing represents a promising avenue for future research. While this integration is not currently feasible due to challenges in delivering RNA into the mitochondria, achieving success in this area could instigate entirely new possibilities, similar to the revolutionary impact CRISPR has had on nuclear DNA editing, including base editing and prime editing. Understanding the complexities of the mitochondrial genome and advancing the technologies capable of manipulating it are essential steps toward a future where mitochondrial disease and age-related conditions may be treated with targeted precision.

ACKNOWLEDGEMENTS

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2023-00210965) and the Korea Institute of Science and Technology (KIST) Institutional Program (2E32161). Figures were created with BioRender.com.

CONFLICTS OF INTEREST

The author has no conflicting interests.

FIGURES
Fig. 1. mtDNA editing approaches. (Middle) Mitochondrial heteroplasmy and threshold effect of disease onset. The concept of mitochondrial heteroplasmy and its relationship to the threshold effect of disease onset is illustrated. Varying levels of mutant and wild-type mitochondrial DNA (mtDNA) within a mitochondrion are shown. The dashed line represents the threshold at which the accumulation of mutant mtDNA leads to the onset of mitochondrial disease. (Left) mtDNA elimination strategy using mtDNA nucleases. Mutant mtDNA-specific nucleases (shown in green) can selectively target and remove only the mutant mtDNA. Selectively elimination of mutant mtDNA results in a shift in the mutant mtDNA levels within heteroplasmic populations as mtDNA copy numbers are restored. As the concentration of mutant mtDNA drops below the pathogenic threshold, cells can regain a normal phenotype. (Right) mtDNA base editing strategy using mtDNA base editors. mtDNA base editors (shown in green) bind specifically the target DNA and deaminate cytosine (C) or adenine (A) to uracil (U) or hypoxanthine (I), which is further transformed into thymine (T) or guanine (G) during mtDNA replication. MtDNA base editors can be used to create disease models by introducing pathogenic point mutations. Additionally, they can be employed to correct pre-existing pathogenic point mutations.
Fig. 2. Mitochondrial genome editing technologies, including mtDNA nucleases and mtDNA base editors. The components of each mtDNA editing tool and their architectures on the target DNA are depicted. To facilitate efficient delivery into mitochondria, mtDNA editing tools are fused with a mitochondrial targeting sequence (MTS, shown in blue) or additionally combined with a nuclear export signal (NES, shown in sky blue) at either the N-terminus or C-terminus of the protein. Programmable DNA-binding proteins, such as ZF arrays or TALE arrays, are highlighted in green, and the DNA target sequences they recognize are also shown in green. Furthermore, protein domains that introduce direct changes in DNA are indicated in pink for nucleases, red or purple for cytosine or adenine deaminases, and yellow for nickases.
Fig. 3. Delivery strategies for mitochondrial genome editing. This figure illustrates various methods for delivering genome editing tools to the mitochondria, including: (1) non-viral delivery vehicles, (2) physical delivery systems such as electroporation and microinjection, and (3) viral vectors like Adeno-Associated Virus (AAV). In each of the depicted method, DNA or RNA encoding mtDNA editing tools is introduced into the cell. Subsequently, these genetic materials are transcribed and translated to produce mitochondrial editing proteins. These proteins are tagged with mitochondrial targeting sequences that enable their transport into the mitochondrial matrix, where the mitochondrial genome editing process takes place.
REFERENCES
  1. Rackham O and Filipovska A (2022) Organization and expression of the mammalian mitochondrial genome. Nat Rev Genet 23, 606-623.
    Pubmed CrossRef
  2. Gupta R, Kanai M and Durham TJ et al (2023) Nuclear genetic control of mtDNA copy number and heteroplasmy in humans. Nature 620, 839-848.
    Pubmed KoreaMed CrossRef
  3. Gorman GS, Schaefer AM and Ng Y et al (2015) Prevalence of nuclear and mitochondrial DNA mutations related to adult mitochondrial disease. Ann Neurol 77, 753-759.
    Pubmed KoreaMed CrossRef
  4. Anderson S, Bankier AT and Barrell BG et al (1981) Sequence and organization of the human mitochondrial genome. Nature 290, 457-465.
    Pubmed CrossRef
  5. Chen X, Prosser R, Simonetti S, Sadlock J, Jagiello G and Schon EA (1995) Rearranged mitochondrial genomes are present in human oocytes. Am J Hum Genet 57, 239-247.
    Pubmed KoreaMed
  6. Taylor RW and Turnbull DM (2005) Mitochondrial DNA mutations in human disease. Nat Rev Genet 6, 389-402.
    Pubmed KoreaMed CrossRef
  7. Kopinski PK, Singh LN, Zhang S, Lott MT and Wallace DC (2021) Mitochondrial DNA variation and cancer. Nat Rev Cancer 21, 431-445.
    Pubmed CrossRef
  8. Kogelnik AM, Lott MT, Brown MD, Navathe SB and Wallace DC (1996) MITOMAP: a human mitochondrial genome database. Nucleic Acids Res 24, 177-179.
    Pubmed KoreaMed CrossRef
  9. He Y, Wu J and Dressman DC et al (2010) Heteroplasmic mitochondrial DNA mutations in normal and tumour cells. Nature 464, 610-614.
    Pubmed KoreaMed CrossRef
  10. Rossignol R, Faustin B, Rocher C, Malgat M, Mazat JP and Letellier T (2003) Mitochondrial threshold effects. Biochem J 370, 751-762.
    Pubmed KoreaMed CrossRef
  11. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA and Charpentier E (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816-821.
    Pubmed KoreaMed CrossRef
  12. Komor AC, Kim YB, Packer MS, Zuris JA and Liu DR (2016) Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage. Nature 533, 420-424.
    Pubmed KoreaMed CrossRef
  13. Gaudelli NM, Komor AC and Rees HA et al (2017) Programmable base editing of A*T to G*C in genomic DNA without DNA cleavage. Nature 551, 464-471.
    Pubmed KoreaMed CrossRef
  14. Anzalone AV, Randolph PB and Davis JR et al (2019) Search-and-replace genome editing without double-strand breaks or donor DNA. Nature 576, 149-157.
    Pubmed KoreaMed CrossRef
  15. Gammage PA, Moraes CT and Minczuk M (2018) Mitochondrial genome engineering: the revolution may not be CRISPR-Ized. Trends Genet 34, 101-110.
    Pubmed KoreaMed CrossRef
  16. Fontana GA and Gahlon HL (2020) Mechanisms of replication and repair in mitochondrial DNA deletion formation. Nucleic Acids Res 48, 11244-11258.
    Pubmed KoreaMed CrossRef
  17. Silva-Pinheiro P and Minczuk M (2022) The potential of mitochondrial genome engineering. Nat Rev Genet 23, 199-214.
    Pubmed CrossRef
  18. Peeva V, Blei D and Trombly G et al (2018) Linear mitochondrial DNA is rapidly degraded by components of the replication machinery. Nat Commun 9, 1727.
    Pubmed KoreaMed CrossRef
  19. Krokan HE and Bjoras M (2013) Base excision repair. Cold Spring Harb Perspect Biol 5, a012583.
    Pubmed KoreaMed CrossRef
  20. Omura T (1998) Mitochondria-targeting sequence, a multi-role sorting sequence recognized at all steps of protein import into mitochondria. J Biochem 123, 1010-1016.
    Pubmed CrossRef
  21. Gammage PA, Rorbach J, Vincent AI, Rebar EJ and Minczuk M (2014) Mitochondrially targeted ZFNs for selective degradation of pathogenic mitochondrial genomes bearing large-scale deletions or point mutations. EMBO Mol Med 6, 458-466.
    Pubmed KoreaMed CrossRef
  22. Gammage PA, Gaude E and Van Haute L et al (2016) Near-complete elimination of mutant mtDNA by iterative or dynamic dose-controlled treatment with mtZFNs. Nucleic Acids Res 44, 7804-7816.
    Pubmed KoreaMed CrossRef
  23. Gammage PA, Viscomi C and Simard ML et al (2018) Genome editing in mitochondria corrects a pathogenic mtDNA mutation in vivo. Nat Med 24, 1691-1695.
    Pubmed KoreaMed CrossRef
  24. Reddy P, Ocampo A and Suzuki K et al (2015) Selective elimination of mitochondrial mutations in the germline by genome editing. Cell 161, 459-469.
    Pubmed KoreaMed CrossRef
  25. Bacman SR, Williams SL, Pinto M, Peralta S and Moraes CT (2013) Specific elimination of mutant mitochondrial genomes in patient-derived cells by mitoTALENs. Nat Med 19, 1111-1113.
    Pubmed KoreaMed CrossRef
  26. Hashimoto M, Bacman SR and Peralta S et al (2015) MitoTALEN: a general approach to reduce mutant mtDNA loads and restore oxidative phosphorylation function in mitochondrial diseases. Mol Ther 23, 1592-1599.
    Pubmed KoreaMed CrossRef
  27. Bacman SR, Kauppila JHK and Pereira CV et al (2018) MitoTALEN reduces mutant mtDNA load and restores tRNA(Ala) levels in a mouse model of heteroplasmic mtDNA mutation. Nat Med 24, 1696-1700.
    Pubmed KoreaMed CrossRef
  28. Yang Y, Wu H and Kang X et al (2018) Targeted elimination of mutant mitochondrial DNA in MELAS-iPSCs by mitoTALENs. Protein Cell 9, 283-297.
    Pubmed KoreaMed CrossRef
  29. Yahata N, Boda H and Hata R (2021) Elimination of mutant mtDNA by an optimized mpTALEN restores differentiation capacities of heteroplasmic MELAS-iPSCs. Mol Ther Methods Clin Dev 20, 54-68.
    Pubmed KoreaMed CrossRef
  30. Srivastava S and Moraes CT (2001) Manipulating mitochondrial DNA heteroplasmy by a mitochondrially targeted restriction endonuclease. Hum Mol Genet 10, 3093-3099.
    Pubmed CrossRef
  31. Bayona-Bafaluy MP, Blits B, Battersby BJ, Shoubridge EA and Moraes CT (2005) Rapid directional shift of mitochondrial DNA heteroplasmy in animal tissues by a mitochondrially targeted restriction endonuclease. Proc Natl Acad Sci U S A 102, 14392-14397.
    Pubmed KoreaMed CrossRef
  32. Bacman SR, Williams SL, Duan D and Moraes CT (2012) Manipulation of mtDNA heteroplasmy in all striated muscles of newborn mice by AAV9-mediated delivery of a mitochondria-targeted restriction endonuclease. Gene Ther 19, 1101-1106.
    Pubmed KoreaMed CrossRef
  33. Tanaka M, Borgeld HJ and Zhang J et al (2002) Gene therapy for mitochondrial disease by delivering restriction endonuclease SmaI into mitochondria. J Biomed Sci 9, 534-541.
    Pubmed CrossRef
  34. Alexeyev MF, Venediktova N, Pastukh V, Shokolenko I, Bonilla G and Wilson GL (2008) Selective elimination of mutant mitochondrial genomes as therapeutic strategy for the treatment of NARP and MILS syndromes. Gene Ther 15, 516-523.
    Pubmed KoreaMed CrossRef
  35. Pereira CV, Bacman SR and Arguello T et al (2018) mitoTev-TALE: a monomeric DNA editing enzyme to reduce mu-tant mitochondrial DNA levels. EMBO Mol Med 10, e8084.
    Pubmed KoreaMed CrossRef
  36. Zekonyte U, Bacman SR and Smith J et al (2021) Mitochondrial targeted meganuclease as a platform to eliminate mutant mtDNA in vivo. Nat Commun 12, 3210.
    Pubmed KoreaMed CrossRef
  37. Minczuk M, Papworth MA, Miller JC, Murphy MP and Klug A (2008) Development of a single-chain, quasi-dimeric zinc-finger nuclease for the selective degradation of mutated human mitochondrial DNA. Nucleic Acids Res 36, 3926-3938.
    Pubmed KoreaMed CrossRef
  38. Mok BY, de Moraes MH and Zeng J et al (2020) A bacterial cytidine deaminase toxin enables CRISPR-free mitochondrial base editing. Nature 583, 631-637.
    Pubmed KoreaMed CrossRef
  39. Cho SI, Lee S and Mok YG et al (2022) Targeted A-to-G base editing in human mitochondrial DNA with programmable deaminases. Cell 185, 1764-1776.
    Pubmed CrossRef
  40. Richter MF, Zhao KT and Eton E et al (2020) Phage-assisted evolution of an adenine base editor with improved Cas domain compatibility and activity. Nat Biotechnol 38, 883-891.
    Pubmed KoreaMed CrossRef
  41. Yin L, Shi K and Aihara H (2023) Structural basis of sequence-specific cytosine deamination by double-stranded DNA deaminase toxin DddA. Nat Struct Mol Biol 30, 1153-1159.
    Pubmed KoreaMed CrossRef
  42. Mok BY, Kotrys AV, Raguram A, Huang TP, Mootha VK and Liu DR (2022) CRISPR-free base editors with enhanced activity and expanded targeting scope in mitochondrial and nuclear DNA. Nat Biotechnol 40, 1378-1387.
    Pubmed KoreaMed CrossRef
  43. Mi L, Shi M and Li YX et al (2023) DddA homolog search and engineering expand sequence compatibility of mitochondrial base editing. Nat Commun 14, 874.
    Pubmed KoreaMed CrossRef
  44. Sun H, Wang Z and Shen L et al (2023) Developing mitochondrial base editors with diverse context compatibility and high fidelity via saturated spacer library. Nat Commun 14, 6625.
    Pubmed KoreaMed CrossRef
  45. Cheng K, Li C and Jin J et al (2023) Engineering RsDddA as mitochondrial base editor with wide target compatibility and enhanced activity. Mol Ther Nucleic Acids 34, 102028.
    Pubmed KoreaMed CrossRef
  46. Guo J, Yu W and Li M et al (2023) A DddA ortholog-based and transactivator-assisted nuclear and mitochondrial cytosine base editors with expanded target compatibility. Mol Cell 83, 1710-1724.
    Pubmed CrossRef
  47. Huang J, Lin Q and Fei H et al (2023) Discovery of deaminase functions by structure-based protein clustering. Cell 186, 3182-3195.
    Pubmed CrossRef
  48. Lee S, Lee H and Baek G et al (2022) Enhanced mitochondrial DNA editing in mice using nuclear-exported TALE-linked deaminases and nucleases. Genome Biol 23, 211.
    Pubmed KoreaMed CrossRef
  49. Lim K, Cho SI and Kim JS (2022) Nuclear and mitochondrial DNA editing in human cells with zinc finger deaminases. Nat Commun 13, 366.
    Pubmed KoreaMed CrossRef
  50. Willis JCW, Silva-Pinheiro P, Widdup L, Minczuk M and Liu DR (2022) Compact zinc finger base editors that edit mitochondrial or nuclear DNA in vitro and in vivo. Nat Commun 13, 7204.
    Pubmed KoreaMed CrossRef
  51. Mok YG, Lee JM and Chung E et al (2022) Base editing in human cells with monomeric DddA-TALE fusion deaminases. Nat Commun 13, 4038.
    Pubmed KoreaMed CrossRef
  52. Wei Y, Li Z and Xu K et al (2022) Mitochondrial base editor DdCBE causes substantial DNA off-target editing in nuclear genome of embryos. Cell Discov 8, 27.
    Pubmed KoreaMed CrossRef
  53. Lei Z, Meng H and Liu L et al (2022) Mitochondrial base editor induces substantial nuclear off-target mutations. Nature 606, 804-811.
    Pubmed CrossRef
  54. Lee S, Lee H, Baek G and Kim JS (2023) Precision mitochondrial DNA editing with high-fidelity DddA-derived base editors. Nat Biotechnol 41, 378-386.
    Pubmed KoreaMed CrossRef
  55. Guo J, Chen X and Liu Z et al (2022) DdCBE mediates efficient and inheritable modifications in mouse mitochondrial genome. Mol Ther Nucleic Acids 27, 73-80.
    Pubmed KoreaMed CrossRef
  56. Chen X, Liang D and Guo J et al (2022) DdCBE-mediated mitochondrial base editing in human 3PN embryos. Cell Discov 8, 8.
    Pubmed KoreaMed CrossRef
  57. Yi Z, Zhang X and Tang W et al (2023) Strand-preferred base editing of organellar and nuclear genomes using CyDENT. Nat Biotechnol, 1-12. https://doi.org/10.1038/s41587-023-01791-y
    CrossRef
  58. Hu J, Sun Y and Li B et al (2023) Strand-preferred base editing of organellar and nuclear genomes using CyDENT. Nat Biotechnol, 1-10. https://doi.org/10.1038/s41587-023-01910-9
    CrossRef
  59. Paunovska K, Loughrey D and Dahlman JE (2022) Drug delivery systems for RNA therapeutics. Nat Rev Genet 23, 265-280.
    Pubmed KoreaMed CrossRef
  60. Degors IMS, Wang C, Rehman ZU and Zuhorn IS (2019) Carriers break barriers in drug delivery: endocytosis and endosomal escape of gene delivery vectors. Acc Chem Res 52, 1750-1760.
    Pubmed KoreaMed CrossRef
  61. Lee H, Lee S and Baek G et al (2021) Mitochondrial DNA editing in mice with DddA-TALE fusion deaminases. Nat Commun 12, 1190.
    Pubmed KoreaMed CrossRef
  62. Silva-Pinheiro P, Mutti CD and Van Haute L et al (2023) A library of base editors for the precise ablation of all protein-coding genes in the mouse mitochondrial genome. Nat Biomed Eng 7, 692-703.
    Pubmed KoreaMed CrossRef
  63. Guo J, Zhang X and Chen X et al (2021) Precision modeling of mitochondrial diseases in zebrafish via DdCBE-mediated mtDNA base editing. Cell Discov 7, 78.
    Pubmed KoreaMed CrossRef
  64. Sabharwal A, Kar B and Restrepo-Castillo S et al (2021) The FusX TALE Base Editor (FusXTBE) for rapid mitochondrial DNA programming of human cells in vitro and zebrafish disease models in vivo. CRISPR J 4, 799-821.
    Pubmed KoreaMed CrossRef
  65. Qi X, Chen X and Guo J et al (2021) Precision modeling of mitochondrial disease in rats via DdCBE-mediated mtDNA editing. Cell Discov 7, 95.
    Pubmed KoreaMed CrossRef
  66. Qi X, Tan L and Zhang X et al (2023) Expanding DdCBE-mediated targeting scope to aC motif preference in rat. Mol Ther Nucleic Acids 32, 1-12.
    Pubmed KoreaMed CrossRef
  67. Tan L, Qi X and Kong W et al (2023) A conditional knockout rat resource of mitochondrial protein-coding genes via a DdCBE-induced premature stop codon. Sci Adv 9, eadf2695.
    Pubmed KoreaMed CrossRef
  68. Wei Y, Xu C and Feng H et al (2022) Human cleaving embryos enable efficient mitochondrial base-editing with DdCBE. Cell Discov 8, 7.
    Pubmed KoreaMed CrossRef
  69. Wang D, Tai PWL and Gao G (2019) Adeno-associated virus vector as a platform for gene therapy delivery. Nat Rev Drug Discov 18, 358-378.
    Pubmed KoreaMed CrossRef
  70. Silva-Pinheiro P, Nash PA, Van Haute L, Mutti CD, Turner K and Minczuk M (2022) In vivo mitochondrial base editing via adeno-associated viral delivery to mouse post-mitotic tissue. Nat Commun 13, 750.
    Pubmed KoreaMed CrossRef
  71. Lek A, Wong B and Keeler A et al (2023) Death after High-Dose rAAV9 gene therapy in a patient with Duchenne's Muscular Dystrophy. N Engl J Med 389, 1203-1210.
    Pubmed CrossRef
  72. Kang BC, Bae SJ and Lee S et al (2021) Chloroplast and mitochondrial DNA editing in plants. Nat Plants 7, 899-905.
    Pubmed KoreaMed CrossRef


This Article


Cited By Articles
  • CrossRef (0)

Author ORCID Information

Funding Information

Collections

Services
Social Network Service

e-submission

Archives