The ability to precisely modify human genes has been made possible by the development of tools such as mega nucleases, zinc finger nucleases, TALENs, and CRISPR/Cas. These now make it possible to generate targeted deletions, insertions, gene knock outs, and point variants; to modulate gene expression by targeting transcription factors or epigenetic machineries to DNA; or to target and modify RNA. Endogenous repair mechanisms are used to make the modifications required in DNA; they include non-homologous end joining, homology-directed repair, homology-independent targeted integration, microhomology-mediated end joining, base-excision repair, and mismatch repair. Off-target effects can be monitored using in silico prediction and sequencing and minimized using Cas proteins with higher accuracy, such as high-fidelity Cas9, enhanced-specificity Cas9, and hype accurate Cas9. Alternatives to Cas9 have been identified, including Cpf1, Cas12a, Cas12b, and smaller Cas9 orthologs such as CjCas9. Delivery of gene-editing components is performed ex vivo using standard techniques or in vivo using AAV, lipid nanoparticles, or cell-penetrating peptides. Clinical development of gene-editing technology is progressing in several fields, including immunotherapy in cancer treatment, antiviral therapy for HIV infection, and treatment of genetic disorders such as β-thalassemia, sickle cell disease, lysosomal storage disorders, and retinal dystrophy. Here we review these technological advances and the challenges to their clinical implementation.
Graphical Abstract
Introduction
In recent years, various platforms for genetically engineering somatic and pluripotent stem cells have been developed. They include zinc finger nucleases (ZFN), transcription activator-like effector nucleases (TALENs), mega nucleases (MNs), and clustered regularly interspaced short palindromic repeats (CRISPR) in combination with CRISPR-associated protein (CRISPR/Cas). From agriculture to biomedical science, these platforms are being explored in various fields and, more recently, in the first clinical trials (Barrangou and Doudna, 2016, Naldini, 2015, Rodriguez-Rodriguez et al., 2019, Zhan et al., 2019).Several developments are taking place in parallel. First, technological improvements and variations on gene-editing strategies are being reported with unsurpassed speed. Second, many possible applications are being developed to address a wide variety of biomedical questions. Third, the first platforms for gene editing are entering the clinical testing stage. Here, we follow these themes to present an overview of these recent developments, focusing on the likely clinical implementation of gene-editing strategies as well as discussing recent technological advances, and aspects such as safety, efficacy, and delivery that are relevant to clinical implementation. We provide a short overview of the progress gene editing is making toward applications in the clinic.
Technological Advances
Basics of CRISPR/Cas
Genome editing depends on the ability to generate specific pre-designed alterations in the genome. Inducing double-strand breaks (DSBs), single-strand breaks (SSBs) (also termed “nicks”), or specific base changes result in the activation of endogenous repair mechanisms that can be used to alter the genome. MNs, ZFNs, and TALENs were the first tools for genome editing in mammalian cells. MNs are naturally occurring endonucleases and can be re-targeted to new sites. Gene editing with ZFNs and TALENS results from the fusion of the FokI nuclease domain to the DNA-binding modules of zinc finger proteins (in the case of ZFNs) or transcription activator-like effector proteins (TALEs) (in the case of TALENS). However, the design and construction of these gene-editing tools can be labor-intensive, and their efficiencies for performing gene-editing varies. The discovery of CRISPR/Cas as a new gene-editing platform introduced a fast, cheap, and relatively efficient genome-editing method that revolutionized genome engineering. The mechanism of CRISPR/Cas is based on its role in adaptive immunity in prokaryotes, in which short stretches of invading foreign nucleic acids, so-called protospacers, are incorporated into the CRISPR locus of the bacterial or archaeal genome. After acquisition, CRISPR RNA (crRNA) is generated from the protospacers at the CRISPR locus. This crRNA can bind to complementary foreign nucleic acid and directs the Cas protein to recognize invading sequences. A second RNA known as the trans-activating CRISPR RNA (tracrRNA) is transcribed from a genomic locus upstream of the CRISPR locus and forms a complex with the crRNA. The crRNA:tracrRNA complex associates with a Cas protein (the nuclease) and creates an active ribonucleoprotein (RNP) complex that targets foreign nucleic acids for degradation (Mojica and Montoliu, 2016, Maeder and Gersbach, 2016).For genome editing, the tracrRNA and crRNA are fused into a single guide RNA (sgRNA), which binds complementarily to a DNA target and guides the Cas protein to the desired target site, creating a DSB (Jinek et al., 2012). The target sequence is based on the presence of a protospacer adjacent motif (PAM), which is an absolute prerequisite for Cas protein to induce a DSB. The first publications on CRISPR arrays date back as early as 1987 (Ishino et al., 1987), Cas genes were discovered in 2002 (Jansen et al., 2002), and CRISPR/Cas was shown to cleave bacteriophage and plasmid DNA in vivo at specific sites in 2010 (Garneau et al., 2010). However, it was not until 2012 that two groups (Gasiunas et al., 2012, Jinek et al., 2012) adapted this system into a gene-editing tool. A review by Fernández and colleagues provides an extensive history of genome editing tools (Fernandez et al., 2017).Its accessibility and relatively low costs have brought CRISPR/Cas a large number of applications in research worldwide and have catalyzed further research on understanding its mechanism of action, improving its functional capacities, and extending its biomedical applications. The CRISPR/Cas9 system originally applied has downsides that include a lower specificity than other gene-editing tools and a relatively large cargo size that hampers delivery to cells via vectors with limited cargo-size capacity (Fernandez et al., 2017, Guha and Edgell, 2017, Guha et al., 2017, Gupta and Musunuru, 2014, Zych et al., 2018). However, as we discuss below, innovative research has produced many adaptations to the original system that enhance its versatility and improve properties such as specificity and efficacy (Zhang et al., 2016, Wu et al., 2018).
Versatility of CRISPR/Cas-Mediated Gene Editing
As nucleases merely induce breaks in the DNA, the introduction of specific alterations in the genetic code requires the exploitation of distinct DNA-repair mechanisms. Intelligent engineering of the original genome-editing tools has broadened the scope of this toolkit, making it possible to achieve a great number of applications. As a result, CRISPR/Cas can be applied to interfere at multiple steps of gene-expression processes and can target processes at genomic and transcriptomic levels. Approaches include gene knock-out, precise correction of disease-associated variants, insertion of a cDNA in a safe harbor (a location in the genome—such as the AAVS1 locus—where there is no risk of insertional mutagenesis) (van der Wal et al., 2018, Sadelain et al., 2011); and manipulation of gene-expression regulatory elements such as promoter activity (Baliou et al., 2018) or splicing (Bergsma et al., 2018, Smith et al., 2018).
Targeted Deletion and Gene Knock-Out
The DNA repair pathway that is most commonly used to create deletions and specific gene knock-outs is non-homologous end joining (NHEJ). NHEJ repairs blunt or incompatible double strands, either by direct ligation or through mediation by microhomology of 5-25 nucleotides that flank the DSB to facilitate end joining (Chang et al., 2017a). NHEJ represents the major double-strand break repair system in mammalian cells and is active throughout the cell cycle (Chang et al., 2017a, Ranjha et al., 2018). Repair can be precise, leaving the target site intact for recleaving by the Cas nuclease. However, NHEJ-mediated DNA repair is rather error prone, as it can introduce random insertions or deletions of base pairs (indels) that will destroy the target site. These errors can result in a frameshift that inactivates gene products through mRNA decay (Ranjha et al., 2018). This effect is exploited in gene editing to create targeted gene knock-outs. Examples of clinical applications include knocking out disease-promoting genes such as oncogenes or restoring a reading frame by interfering with splice sites such as in Duchenne muscular dystrophy (Amoasii et al., 2018, Tabebordbar et al., 2016, Nelson et al., 2016).
Targeted Gene Editing and Knock-In
Traditionally, the homology-directed repair (HDR) pathway is used to achieve a precise knock-in. HDR recognizes DSBs and utilizes a homologous template—which, under physiological conditions, is the sister chromatid—to repair the defect (Ranjha et al., 2018). For precision gene editing, this pathway is exploited by introducing a donor template that is used in the repair of the DSB rather than the sister chromatid. This donor template has the desired alterations, which are flanked by 3′ and 5′ homology arms on both sites of the DSB. In this manner, it is possible to accurately correct a point variant (as in cystic fibrosis) (Schwank et al., 2013); a tandem repeat (as in Fragile X syndrome) (Xie et al., 2016); large inserts or deletions (as in Duchenne dystrophy) (Li et al., 2015); or other genetic defects.Due to the use of a donor template, HDR is a very precise repair pathway that is less error-prone than NHEJ. However, HDR is less active in cells and therefore much less efficient in genome editing than NHEJ, which limits its clinical potential. To favor its activation over NHEJ induction, its efficiency would have to be increased. This has cast new light on research into the regulation of DNA repair pathways (Mateos-Gomez et al., 2017, Schimmel et al., 2017, Zelensky et al., 2017). Several strategies using small molecules to inhibit NHEJ have been followed with varying success (Chu et al., 2015, Maruyama et al., 2015, Song et al., 2016, Yu et al., 2015, Pinder et al., 2015). Recently, cold shock was found to increase HDR in cells in vitro (Guo et al., 2018a).
Leave a Reply