Exploiting DNA Endonucleases to Advance Mechanisms of DNA Repair

Abstract

The earliest methods of genome editing, such as zinc-finger nucleases (ZFN) and transcription activator-like effector nucleases (TALENs), utilize customizable DNA-binding motifs to target the genome at specific loci. While these approaches provided sequence-specific gene-editing capacity, the laborious process of designing and synthesizing recombinant nucleases to recognize a specific target sequence, combined with limited target choices and poor editing efficiency, ultimately minimized the broad utility of these systems. The discovery of clustered regularly interspaced short palindromic repeat sequences (CRISPR) in Escherichia coli dates to 1987, yet it was another 20 years before CRISPR and the CRISPR-associated (Cas) proteins were identified as part of the microbial adaptive immune system, by targeting phage DNA, to fight bacteriophage reinfection. By 2013, CRISPR/Cas9 systems had been engineered to allow gene editing in mammalian cells. The ease of design, low cytotoxicity, and increased efficiency have made CRISPR/Cas9 and its related systems the designer nucleases of choice for many. In this review, we discuss the various CRISPR systems and their broad utility in genome manipulation. We will explore how CRISPR-controlled modifications have advanced our understanding of the mechanisms of genome stability, using the modulation of DNA repair genes as examples.

Keywords: CRISPR; base excision repair; gene editing; homologous recombination; microhomology-mediated end-joining; mismatch repair; non-homologous end-joining.

Figure 2

Timeline of the major CRISPR-associated protein structures. Structural biology has played a major role in understanding the organization of CRISPR proteins and their interactions with the target DNA, crRNA, and other associated proteins. The majority of CRISPR systems fall into Class I (rainbow-colored structures). To differentiate Cas1 and Cas2 within the Cas1-Cas2 acquisition complex, they are represented in solid colors that are maintained in their independent structures (see legend). The multi-subunit Class I effector, Cascade, is colored to differentiate each of the five proteins that make up the assembly: the large subunit (orange), small subunit (navy), Cas5 (blue), Cas6 (red), and Cas7 (pink). Starting in 2014 with the structure of Cas9, there is a shift in the proportion of solved structures belonging to Class II systems. Represented Class II structures are colored to illustrate the endonuclease domain organization. Additionally, in highlighting the nuclease domains, we can appreciate the bilobed (NUC and REC lobes) organization of Class II effectors. Protein data bank (PDB) IDs are included for each structure [18,30,31,35,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73].

Figure 1

Comparison between Class I and Class II CRISPR immunity pathways. CRISPR immunity takes place in three general steps: acquisition, biogenesis, and interference. Type I-A, II-A, V-B, V-A, and VI-A are examples to illustrate some of the differences between each CRISPR type. In the majority of CRISPR systems, the acquisition is carried out by the Cas1-Cas2 complex, which locates a protospacer within the foreign DNA, excises it, and incorporates the new spacer into the CRISPR array. Biogenesis occurs upon recognition of a previous invader, producing the CRISPR-associated proteins, a long unprocessed CRISPR RNA (crRNA), and any additional RNAs, such as the trans-activating CRISPR RNA (tracrRNA). Processing may occur by the effector protein alone (Cas12a/13a), the effector protein with additional factors (Cas9/12b), or by a subunit of the effector complex (Cas6). Interference begins with the fully assembled effector complex scanning for a target sequence complementary to the spacer of the crRNA and protospacer adjacent motif (PAM) (yellow). Relative to the protospacer on the noncomplementary strand, Cascade and Cas12a/b look for a 5′ PAM, whereas Cas9 looks for a 3′ PAM. Cascade recruits Cas3 and cleaves the exposed non-complementary strand, manipulating the target strand into an ssDNA molecule. Similarly, Cas9 and Cas12a/b locate a target sequence followed by the invasion of the DNA double helix by the crRNA. Cas9 forms a blunt double-stranded break (DSB) within the complementary region, where the HNH domain cleaves the target strand and the RuvC-like domain cleaves the non-target strand. Cas12a and Cas12b form staggered DSBs, cleaving within and 5′ of the complementary region on the non-target and target strands, respectively. In the case of RNA targeting Cas13a, the cleavage location is determined by the nucleotide 3′ of the protospacer, known as the protospacer flanking sequence (PFS) (red). The Cas13a HEPN domain cleaves the target 5′ of the complementary region. CRISPR interference triggers the degradation of foreign nucleic acids.

Figure 3

Pathways that repair SpCas9-directed DNA breaks. The pathway employed to repair Cas9 generated double-strand breaks (DSBs) is dependent on the repair factors present, level of resection, extent of homology, and cell cycle phase. If the ends of a DSB are immediately capped by Ku proteins, preventing resection, the nonhomologous end-joining pathway (NHEJ) initiates to ligate the DNA ends back together. NHEJ leads to uncontrolled insertions and deletions (indels), varying in size. Specific insertions can be incorporated through NHEJ via homologous independent targeted insertions (HITI) by supplying a double-stranded (ds) DNA donor flanked by Cas9 cleavage sites. During the M and S phase, when low levels of resection occur in the presence of 2–25 bp of homology, microhomology-mediated end-joining (MMEJ) takes place that results in small indels. During S and G2 phases, homology-directed repair (HDR) can achieve precise deletions, insertions, and edits, utilizing ss or ds donors.

Figure 4

Mechanism of DNA base and prime editors using Cas9 nickase. (A) The illustrated generation 4 base editor (BE4) contains a nCas9 fused with two inhibitors uracil glycosylase inhibitors (UGI) and the cytosine deaminase, APOBEC1, capable of performing C to U transitions. APOBEC1 deaminates the target cytosine on the non-complementary strand, creating a uracil. The UGIs inhibit the repair machinery activated by the presence of the uracil base. nCas9 nicks the complementary strand and persuades the repair machinery to correct the non-deaminated strand, thereby promoting the incorporation of the edit. The final result is a C:G to T:A transition. (B). Prime editors contain a nCas9-reverse transcriptase (RT) fusion guided by an extended crRNA known as a primer editing guide RNA (pegRNA). The 5′ end of the pegRNA resembles a crRNA with a spacer region and a stem-loop. The 3′ end is extended to include a template containing the desired edit for the reverse transcriptase, and a 3′ primer binding site. The pegRNA guides nCas9 to the target DNA and forms Watson and Crick base-pairing at both the 5′ and 3′ ends. nCas9 nicks the PAM strand between the primer binding site and RT template. Using the primer binding site, the RT extends the PAM strand using the template containing the desired edit. Base pairing of the 3′ flap containing the edit leads to the removal of the 5′ flap. Through replication and DNA repair, the edit is incorporated into both strands.