MNs also referred to as homing endonucleases recognize a large DNA stretches (12–40 base pairs) to facilitate cleavage in the most genomes (Rouet et al., 1994b). MNs are generally encoded by introns or inteins to promote homing of their respective genetic elements into intron or intein-free homologous allelic sites (Epinat et al., 2003). One of the distinctive features of these MNs is their high specificity, due to the tight coupling of their binding site. This tight coupling recognizes a single locus within the yeast nuclear or mitochondrial genome (Paques and Duchateau, 2007). The LAGLIDADG proteins are the most well-studied homing endonuclease. They interact with their targets by nonspecific interactions between the ß strands and the backbone of the target DNA through the recognition of a sequence of 2–4 bp region. Consequently, several new engineered endonuclease variations derived from the following homing endonuclease: (i) I-CreI which was discovered in the chloroplast genome of Chlamydomonas reinhardtii, and (ii) I-SceI which is present in the mitochondria of Saccharomyces cerevisiae (Prieto et al., 2007; Gao et al., 2010; Zekonyte et al., 2021; Lee et al., 2022). These engineered endonucleases enable in vivo and in vitro genetic modifications. Due to the tiny molecular weight of modified MNs makes the in vivo delivery possible. However, the editing efficiency of this strategy is low compared to later developed nuclease-mediated technologies (Aubert et al., 2016; Zhu et al., 2017). Furthermore, reengineering MNs to expand the spectrum of DNA target sequences is complex and laborious and therefore vastly limit its application (Gouble et al., 2006). Recently, Zekonyte et al. (2021) used the engineered I-CreI meganuclease administered via intravenous (IV) injection of AAV9 into mice for the correction of m.5024C>T mutation in the mt-tRNA^Ala gene, as a curative method for disorders caused by heteroplasmic mitochondrial DNA mutation. This resulted to the elimination of mutant mitochondrial DNA followed by the restoration of mt-tRNA^Ala level. Moreover, this approach has also been used in pig model of autosomal dominant Retinitis Pigmentosa (adRP) for the correction of P23H mutation in the rhodopsin (RHO) gene (Jalligampala et al., 2021) and in non-human primates for the modification of the proprotein convertase subtilisin/kexin type 9 (PCSK9) gene responsible for hypercholesterolemia (Wang et al., 2018a; Wang et al., 2021).

Zinc Finger Proteins (ZFPs) are artificially synthetic engineered hybrid heterodimeric proteins for site specific genome editing. ZFPs include a sequence of 3–6 peptides (called Zinc Finger (ZF) domains), each binding to a specific sequence of 3–6 base pairs for a specific attachment to a gene sequence (Durai et al., 2005). Two ZFPs are required to fuse with Flavobacterium okeanokoites endonuclease I (FokI) to induce a DSB at a specific genomic site. ZFNs are frequently used for gene silencing and knockout (Santiago et al., 2008; Gutschner et al., 2011; Gaj et al., 2012; Sun et al., 2018). Therefore, ZFNs have emerged as a versatile tool for gene targeting in various mammalian cells and organisms for the treatment of hereditary diseases (Almeida and Matos, 2019) and creation of animal models for diseases (Petersen and Niemann, 2015a).

ZFNs represent the first gene editing method applied for clinical treatment of diseases. ZFNs have been used to modify autologous CD4⁺ T-cells to inhibit the function of the human chemokine (C-C motif) receptor 5 gene (CCR5) receptor and reduce the infection of these cells by HIV (Tebas et al., 2014). The results showed that infusion of genetically modified CD4⁺ T-cells was well tolerated, and the HIV viral load has been decreased in blood level of most patients. Another experiment also showed that the HIV-specific CD8⁺ T-cell responses are substantially restored (Tebas et al., 2021). The ZFN approach was also successfully used for the treatment of ß-hemoglobinopathies (Hoban et al., 2015; Chang et al., 2017; Psatha et al., 2018; Smith et al., 2019). Indeed, the phase I/II clinical trial (NCT03432364) sponsored by Sangamo Therapeutics Inc., aimed to assess the safety, tolerability and efficacy of ST-400 for the treatment of transfusion-dependent beta-thalassemia. The ST-400 are patient’s hematopoietic stem cells genetically modified by ZFNs to disrupt a specific and precise sequence of the enhancer of the BCL11A gene in order to boost the expression of fetal hemoglobin (HbF) (Bauer and Orkin, 2015; Masuda et al., 2016). A cohort of six participants will be completed by November 2022. In other studies involving two patients manifesting different genotypic profiles showed a prompt hematopoietic reconstitution with long term increased HbF levels; however, serious adverse events (e.g., Hypersensitivity) have been recorded with one patient as a result of reengineered ST-400 (Smith et al., 2019). Scientists at Bioverative Inc., a Sanofi company, are conducting a phase I/II clinical trials (NCT03653247) in a cohort of eight patients to evaluate the safety, tolerability and efficacy of autologous hematopoietic stem cell transplantation using BIVV003 for the treatment of severe Sickle Cell Disease (SCD) in adults. This trial was supported by encouraging preclinical results that showed a robust long-term engraftment of ex-vivo modified hematopoietic stem and progenitor cells (HSPC) from patients. In other phase I/II clinical trials conducted by Sangamo Therapeutics Inc., the UCSF Benioff Children’s Hospital delivered SB-318 (NCT02702115) and SB-913 (NCT03041324) into participants to insert the corrected copy of α-L-iduronidase (IDUA) and iduronate-2-sulfatase (IDS) transgenes respectively into the albumin locus to provide permanent liver specific expression of iduronidase. Preliminary results showed the evidence of albumin-IDS mRNA transcripts in liver and the hepatocytes are able to generate active IDS enzyme (Muenzer et al., 2019). The summary of ongoing clinical trials is presented in Table 2.

A class of naturally occurring DNA binding proteins called the Transcription Activator-Like Effector (TALE) has been identified in the plant pathogen Xanthomonas. These TALEs regulate the transcription of several host target genes (Sanjana et al., 2012). TALENs are artificial engineered proteins combining the DNA-binding properties of a TALE protein and the DNA cleavage of the FokI endonuclease (Moscou and Bogdanove, 2009). The central region of TALEs is composed of 34 amino acid repeats amongst which 32 are constant and 2 are variable and recognized as repeat variable diresidues (RVDs) (Scholze and Boch, 2011). RVDs are involved in the DNA target recognition (Boch et al., 2009). Two TALENs target binding sequences are required to form a FokI dimer that induces a DSB (Feng et al., 2014).

TALENs have successfully been used for the modification of T-cell receptors for the treatment of leukemia (Qasim et al., 2017; Benjamin et al., 2020). UCART19, a CAR-T-cell product engineered with TALENs, was tested in children and adults in phase I clinical trials (NCT02808442 and NCT02746952) to cure advanced lymphoid malignancies and refractory B-cell acute lymphoblastic leukemia (B-ALL). These trials demonstrated the potential of UCART19 in patients with aggressive leukemia, but significant adverse events such as cytokine release syndrome, acute graft-versus-host of the skin and infectious complications have also been observed (Benjamin et al., 2020). There are many phase I trials sponsored by Cellectis Inc., using programmed allogenic engineered T-cells expressing different CARs such as UCART123 (NCT04106076, NCT03203369), UCARTCS1A (NCT04142619) and UCART22 (NCT04150497) respectively to treat acute myeloid leukemia, blastic plasmacytoid dendritic cell neoplasm (BPDCN), multiple myeloma and CD22⁺ B cell acute lymphoblastic leukemia. Ongoing clinical trials using this approach are summarized in Table 3.

CRISPR-Cas9 system is a sophisticated gene editing tool that revolutionized the genome engineering field and generated excitement for the potential of novel therapeutic approaches to treat human diseases. The CRISPR-Cas system is divided into class I and class II (Makarova et al., 2015). The class I uses multi-protein complexes for nucleic acid cleavage and is subdivided into CRISPR-Cas types I, III, and IV. The class II uses a single protein effector domain for the cleavage and is subdivided in CRISPR-Cas type II, V, and VI. The CRISPR-Cas9 belongs to type II system, which is simple to use and thus become the most widely utilized tool for biological research and translational applications (Tang and Fu, 2018).

Over the past decade, CRISPR-Cas9 system has been modified and adapted to become a versatile tool for genome editing in eukaryotes (Ran et al., 2013a; Perez-Pinera et al., 2013; Shalem et al., 2014a; Gao et al., 2019; Casas-Mollano et al., 2020). The system requires a Cas9 nuclease and a single guide RNA (sgRNA) adapted from CRISPR RNA (crRNA), which specifies the target site (the spacer sequence), fused with a trans-activating RNA (tracrRNA), which forms a complex with the crRNA (the scaffold sequence) (Jinek et al., 2012a). The sgRNA forms a stable ribonucleoprotein complex with Cas9 nuclease which initially attaches to a protospacer adjacent motif (PAM) to initiate the first conformational changes of the protein. The targeting activity is driven by 20 nucleotides of RNA-DNA base-pairing between the target DNA strand protospacer and the complementary RNA strand and through interactions between the non-target DNA strand PAM. Subsequently, mediates the second conformational change of the protein which then becomes active. Once activated, the HNH domain of Cas9 cleaves the DNA strand to which the sgRNA is attached and the RuvC domain cleaves at the PAM strand (Jinek et al., 2012b). One of the advantages of CRISPR/Cas9 is that it only requires a sgRNA to specify the DNA sequence where a DSB needs to be generated.

There are many clinical trials based on CRISPR-Cas9 and half of these trials (Phase I and Phase II) have already been successfully completed, the ongoing clinical trials are summarized in Table 4. In a phase I trial (NCT03399448), Pennsylvania University used a multiplex CRISPR-Cas9 to knockout the TCRα, TCRβ, PD-1 genes to treat various malignancies (Stadtmauer et al., 2020). The results showed that the modified T cells engrafted in patients at stable levels for at least 9 months and were barely immunogenic, indicating the feasibility of CRISPR-Cas9 gene editing for cancer immunotherapy (Stadtmauer et al., 2020). In addition to that, some clinical trials moved the technology from ZFNs or TALENs to CRISPR-Cas9 (Yi and Li, 2016; Lu et al., 2020). This was done for example to knockout PD-1 and CD52 for the different types of cancer by electroporating Cas9 and a sgRNA to edit the cells ex vivo. The results indicated that the clinical applications of CRISPR-Cas9 gene-edited T-cells are generally safe and feasible (Lu et al., 2020; Frangoul et al., 2021; Modarai et al., 2021). The disruption of erythroid enhancer of the BCL11A gene by CRISPR-Cas9 for the treatment of ß-thalassemia has been observed with serious adverse events. The reported serious adverse events are sepsis and pneumonia in the presence of neutropenia, vaso-occlusive liver disease, abdominal pain and cholelithiasis after CTX001 (autologous CRISPR-Cas9–edited CD34⁺ HSPCs) intravenous (IV) administration, although the substantially raised of hemoglobin levels in fetal blood cells (Frangoul et al., 2021). Two clinical trials are currently conducted to evaluate the safety and efficacy of CRISPR-Cas9 strategies either to restore ß-globin expression in ß-thalassemic patients harboring the IVS-2-654C>T mutation (NCT04205435) or to de-repress γ-globin in transfusion dependent thalassemia (TDT) patients (NCT04211480), but no results are yet available. Recently, another study prepared NTLA-2001, an in vivo gene-editing therapeutic agent, made by lipid nanoparticles encapsulating messenger RNA of Cas9 protein and single guide RNA targeting misfolding transthyretin (TTR) responsible of transthyretin amyloidosis. The result showed that the TTR is durably knockout after a single dose of NTLA-2001 (Gillmore et al., 2021; Philippidis, 2021).

Base editors are mainly used to target point mutations that may result in an altered DNA sequence with novel or enhanced functions and gene inactivation (Gaudelli, 2017; Gapinske et al., 2018; Zhang et al., 2019). Briefly DNA base editing requires two main components: a Cas9n fused with a deaminase and a sgRNA which binds to a specific DNA sequence (Komor et al., 2016). Three (3) types of base editors have been created including the Cytidine Base Editor (CBE) (Komor et al., 2016), Adenosine Base Editor (ABE) (Gaudelli, 2017) and Glycosylase Base Editors (GBE) (Zhao et al., 2021a) (Figure 2A).

Theoretically, base editors can function in both dividing and non-dividing cells to address several of single-nucleotide polymorphisms (SNPs) associated with human diseases.

Base editing is being used to study and treat hereditary diseases in a variety of cell types (Zeng et al., 2020; Tekel et al., 2021) and organisms (Kim et al., 2021c), including animal models of human hereditary diseases (Liu et al., 2018; Caso and Davies, 2022). Alternatively, some preclinical experiments demonstrated the prospects of using the base editing technologies to treat diseases. Due to the base editor, hA3A-BE3 creates C-to-T conversion at NGN PAM sites efficiently. Wang et al. (2020) showed that the binding site of BCL11A (TGACCA: −114 to −119) represents an ideal target site for binding with hA3A-BE3 to induce a mutation and raise the level of fetal γ-globin expression to ameliorate the ß-hemoglobinopathies. Furthermore, a transformer BE (tBE) system to eliminate unintended mutations was developed and delivered by AAV into mice creating a premature stop codon in proprotein convertase subtilisin/kexin type 9 (PCSK9) gene which significantly reduced the serum PCSK9 and cholesterol (Wang et al., 2021a). Newby et al. (2021) used mRNA encoding the Base editor to treat hematopoietic stem and progenitor cells (HSPCs) from patient with SCD generating high percentage conversion of the SCD allele (HBB^S) into Makassar ß-globin (HBB^G), which is a non-pathogenic variant. It has been shown that base edited CAR T-cells for combinational therapy against T-cell malignancies permitted to enhance molecular remission prior to allo-HSCT for T-cell malignancies (Georgiadis et al., 2021).

Genome editing with base editors effectively induced C→T, G→A, A→G, T→C, C→G, and C→A base substitutions without inducing DSB (Gaudelli, 2017; Zhao et al., 2021b; Kurt et al., 2021). However, they are unable to correct variants beyond these six transition mutations, or other modifications like insertions and deletions of DNA fragments which are successfully achieved by Prime editing (PE). PE uses an engineered Cas9 nickase fused to a reverse transcriptase (RT) enzyme and a modified sgRNA known as prime editing guide RNA (pegRNA) (Figure 2B) (Anzalone et al., 2019). Recent efforts stepwise improved the efficiency of PE system to PEmax system (Chen et al., 2021) and engineer pegRNA known as epegRNA (Nelson et al., 2022b). PE makes possible the accurate insertion up to 1 kb (Wang et al., 2022a) and the deletion of up to 10 kb (Choi et al., 2022) DNA fragment.

As base editing, prime editing has not yet entered clinical trials due to its immature development. However, the potential of PE has so far been demonstrated during the past 2 years in vitro (Surun et al., 2020; Habib et al., 2022; Happi Mbakam et al., 2022; Petri et al., 2022; Tremblay et al., 2022) and in animal models (Liu et al., 2021b; Jang et al., 2022; Zheng et al., 2022; Zhi et al., 2022). Unfortunately, the gene modification efficiency is very low in some models. This may be due to the use of the split viruses to overcome the size of PE, which cannot be packed into a single viral delivery vector. Prime editing has the potential to increase the safety and expand the scope of genome-editing in T-cells showing that method is adaptable to enhance the efficiency of CAR T-cell therapy by concurrently introducing additional complex gene edits into T-cells (Petri et al., 2022).