Marker-free quantification of repair pathway utilization at Cas9-induced double-strand breaks

Abstract

Genome integrity and genome engineering require efficient repair of DNA double-strand breaks (DSBs) by non-homologous end joining (NHEJ), homologous recombination (HR), or alternative end-joining pathways. Here we describe two complementary methods for marker-free quantification of DSB repair pathway utilization at Cas9-targeted chromosomal DSBs in mammalian cells. The first assay features the analysis of amplicon next-generation sequencing data using ScarMapper, an iterative break-associated alignment algorithm to classify individual repair products based on deletion size, microhomology usage, and insertions. The second assay uses repair pathway-specific droplet digital PCR assays ('PathSig-dPCR') for absolute quantification of signature DSB repair outcomes. We show that ScarMapper and PathSig-dPCR enable comprehensive assessment of repair pathway utilization in different cell models, after a variety of experimental perturbations. We use these assays to measure the differential impact of DNA end resection on NHEJ, HR and polymerase theta-mediated end joining (TMEJ) repair. These approaches are adaptable to any cellular model system and genomic locus where Cas9-mediated targeting is feasible. Thus, ScarMapper and PathSig-dPCR allow for systematic fate mapping of a targeted DSB with facile and accurate quantification of DSB repair pathway choice at endogenous chromosomal loci.

Figure 1.

ScarMapper analysis pipeline. (A) The target region of a Cas9 generated DSB (scissors) is amplified and sequenced. Blue and red lines denote sequence upstream and downstream of the DSB site and is represented as a dashed line in the consensus read when deleted, relative to the reference. Orange lines represent inserted sequence. Junctions at sites where one copy of identical sequence is deleted (microhomology; MH) have the microhomology represented in purple. Step 1: ScarMapper generates a dictionary of 10 mers consistent with increasing deletion in 1 nt increments of upstream sequence flanking the cas9 target site. The upstream flank (UF) is then located in each consensus read by the identification of a matching 10mer from this dictionary that has the least amount of upstream flank deletion. In step 2, the same process is used to identify the downstream flank (DF). In step 3, each read is classified as having deletions, deletions with microhomology, or insertions (with or without deletion), then further assigned to repair pathways using pathway definitions as noted in (B).

Figure 2.

ScarMapper Analysis of the Murine Rosa26 Locus. (A) ScarMapper graphical output of Rosa26 locus paired-end NGS data for WT (left panel), Polq^–/– (middle panel), and Ku70^–/– (right panel) MEFs. The x-axis is the number of nucleotides deleted and/or inserted to the left or right of the Cas9 cut site. The values in each scar type are ranked according to the allelic frequency. The height of the bars is proportional to the frequency. The plots show frequencies ≥0.00025. (B) Pol θ dependency of scar patterns. The allelic frequency ratio (WT/Polq^−/−) for the 21 most prevalent repair products observed in WT cells. Red bars indicate repair products that are Pol θ-dependent (ratio ≥2). (C) Histogram of scar types in the different cell lines. Mean ± SEM are shown. Significance by two-tailed unpaired Student's t-test. *P <0.05; **P <0.02; ***P <0.002; ****P <0.0001.

Figure 3.

PathSig-dPCR assay designs to monitor NHEJ, TMEJ and HR repair at the murine Rosa26 locus. (A) Schematic diagram of PathSig-dPCR assays applied to the murine Rosa26 locus. Cells are transfected with Flag-Cas9 and sgRosa26 by Neon electroporation. 24–48 hours post transfection, genomic DNA is assayed by dPCR for signature products reflecting repair by NHEJ, TMEJ and HR. (B) NHEJ repair - one ‘T’ insertion repair product was validated in WT, Ku70^−/− and Polq^−/− cells by dPCR. The NHEJ forward primer includes a ‘T’ to ‘A’ mismatch to destabilize the primer against amplification of the wild-type sequence. (C-E) Three TMEJ repair products were validated by dPCR. (C) Signature of TMEJ-del 23 bp was validated in WT, Polq^−/− and Polq^−/− + hPOLQ cells. TMEJ-del 39 bp (D) and TMEJ-del 95 bp (E) were validated in WT, Polq^−/−, Polq^−/− + hPOLQ and Ku70^−/− cells. (F) Scatter plots of droplets showing HR events induced by Rosa26 cut site. (G) HR repair products were validated by adding 200 bp HR donor (HRD-200) or 500 bp HR donor (HRD-500) in WT and Brca2^−/− mutant cells. Mean ± SEM are shown. Statistical significance was assessed by two-tailed t-tests. *P <0.05, ** P <0.01 and **** P <0.0001.

Figure 4.

Mre11 hypomorphism alters DSB repair pathway utilization. (A) Schema depicting the role of Mre11-mediated end resection in the regulation of DSB repair pathway choice. (B) Mre11^ATLD1/ATLD1 MEFs exhibit reduced expression of the MRN complex by immunoblotting. (C–F) DSB repair products were identified in WT, Mre11^ATLD1-1 and Mre11^ATLD1-2 cells. (C) TMEJ-del 23 bp. (D) TMEJ-del 39 bp. (E) HR repair (HRD-500 donor) and (F) NHEJ-ins +1 bp. Mean ± SEM are shown. Statistical significance was assessed by two-tailed t-tests. * P < 0.05, ** P < 0.01, *** P < 0.001 and **** P < 0.0001.

Figure 5.

Inhibition DNA-PK alters DSB repair pathway utilization. (A-C) Evaluation of (A) NHEJ-Ins +1bp, (B) HR (HRD-500), and (C) TMEJ-del 95bp signature repair products after increasing doses of DNA-PKi in WT MEF cells. (D–F) ES cells (TC1) were transfected with Cas9-RNP targeting the Rosa26 locus and a homologous donor (HRD-500), followed by PathSig-dPCR assays for (D) NHEJ-Ins +1 bp, (E) HR repair (HRD-500), and (F) TMEJ-del 23 bp. Mean ± SEM are shown. Statistical significance was assessed by two-tailed t-tests. * P < 0.05, ** P < 0.01, *** P < 0.001 and **** P < 0.0001.

Figure 6.

Kinetics of DSB repair in WT, Ku70^−/− and Polq^−/− cells. WT, Ku70^−/− and Polq^−/- MEFs were transfected with Cas9 ribonucleoprotein (Cas9-RNP) targeting the Rosa26 locus and a homologous donor (HRD-500). Cells were collected and extracted for genomic DNA at 1, 2, 4, 6, 8, 12, 24 and 48 h, and PathSig-dPCR assays were performed to detect NHEJ-ins +1 bp (A), HR repair (B), and TMEJ-del 23 bp (C). Error bars indicate ± SEM.