Chromosomal translocations in human cells are generated by canonical nonhomologous end-joining

Abstract

Breakpoint junctions of the chromosomal translocations that occur in human cancers display hallmarks of nonhomologous end-joining (NHEJ). In mouse cells, translocations are suppressed by canonical NHEJ (c-NHEJ) components, which include DNA ligase IV (LIG4), and instead arise from alternative NHEJ (alt-NHEJ). Here we used designer nucleases (ZFNs, TALENs, and CRISPR/Cas9) to introduce DSBs on two chromosomes to study translocation joining mechanisms in human cells. Remarkably, translocations were altered in cells deficient for LIG4 or its interacting protein XRCC4. Translocation junctions had significantly longer deletions and more microhomology, indicative of alt-NHEJ. Thus, unlike mouse cells, translocations in human cells are generated by c-NHEJ. Human cancer translocations induced by paired Cas9 nicks also showed a dependence on c-NHEJ, despite having distinct joining characteristics. These results demonstrate an unexpected and striking species-specific difference for common genomic rearrangements associated with tumorigenesis.

Figure 1. Intrachromosomal DSB repair in c-NHEJ deficient human cells is inefficient and shows a shift towards longer deletions and microhomology

A. Absence of LIG4 in both L4^−/− and X4^−/− HCT116 cells. B. The ZFN^EWS cleavage site overlaps an AseI restriction site. C. After ZFN^EWS expression, AseI-resistant PCR products (*: 724bp) were purified and re-amplified (examples shown for wt, L4^+/−, and L4^−/− cells) for sequencing. D. AseI-resistant NHEJ junctions at the ZFN^EWS site from L4^−/− and X4^−/− cells demonstrate a shift towards longer deletions and increased microhomology (wt, n = 4; L4, n = 2; X4, n = 3). Junctions from L4 and X4 heterozygous or deficient cell lines are combined here and below, although results from individual cell lines were similar. The median deletion length is indicated, and each value represents the combined deletion from both ends of an individual junction. In the microhomology analysis, junctions that contain insertions were not included. Deletion and microhomology distributions here and below were compared by Mann-Whitney analysis. For random microhomology distribution, the probability that a junction will have microhomology by chance assumes an unbiased base composition (Roth et al. 1985). ***, p < 0.0001. E. Two DSBs 3.2 kb apart were introduced by ZFN^FLI-A and ZFN^FLI-B in a FLI1 gene intron. To detect the 3.2-kb deletion, an ~890 bp fragment was PCR amplified using primers flanking the two DSBs. F. Limiting dilution of genomic DNA to estimate the frequency of deletions after expression of ZFN^FLI-A and ZFN^FLI-B. PCR amplification of the 3.2-kb deletion product was performed with serial dilutions of genomic DNA (100, 50, 25, 12.5, 6.25, 3.125, and 1.56 ng). The number of times a PCR fragment was detected from 3 independent amplifications is indicated below the gel. G. Junctions from joining two DSBs 3.2 kb apart at the FLI1 locus from L4^−/− and X4^−/− cells demonstrate a shift towards longer deletions and increased microhomology. See also Figure S1.

Figure 2. c-NHEJ generates chromosomal translocations in HCT116 human cells

A. Induction of chromosomal translocations with sequence-specific nucleases. Derivative chromosomes. Der19 and Der22 were detected by PCR using primers that flank the cleavage sites after ZFN^p84 and ZFN^EWS expression. B. Translocation frequency is reduced in L4^−/− and X4^−/− HCT116 cells. Translocations were quantified as follows: ZFN^EWS and ZFN^p84, Der19 and Der22 (wt, n = 7; L4, n = 3; X4, n = 4); ZFN^EWS and TAL^p84, Der22 (n = 4); TAL^LAM and TAL^p84, Der1 (n = 4). Der19 and Der22 were assessed in the same experiment and the frequencies averaged. Error bars, +/− SEM. Nuclease expression was detected 48 h after transfection. C. – E. Translocation junction analysis from X4^−/− cells demonstrates a shift towards longer deletions and an increased presence of microhomology. Der19 and Der22 junctions derived from ZFN expression were pooled. C, Deletion lengths from individual Der19 and Der22 junctions are indicated by the triangle and circle, respectively. D, Junctions are grouped according to whether the deletions were restricted to the overhang or were short (≤ 30 bp) or long (> 30 bp) deletions extending outside of the overhang. E, Microhomology distribution. See also Figure S2.

Figure 3. Chromosomal translocation formation is impaired in human LIG4 mutant cells

A. Locations of LIG4 mutations in N114 pre-B cells and 411BR fibroblasts. B. Western blotting for LIG4 and ZFNs. C. LIG4-null N114 cells have a significantly reduced ZFN-induced translocation frequency (n = 3), while hypomorphic 411BR cells have only a mild reduction (n = 4). Der19 and Der22 formation is pooled. Error bars, +/− SEM. D. High insertion frequency in Der19 and Der22 junctions from NALM6. E. – G. Translocation junction analysis from LIG4 mutant cells. E, F, Junctions show longer deletions which always (N114) or mostly (411BR) extend beyond the ZFN overhang. G, Microhomology distribution at translocation junctions. Microhomology can only be determined at junctions without insertions (only 33% of junctions from wild-type NALM6 pre-B cells). Microhomology distributions from both pre-B cell lines was different from that expected by chance. By contrast, neither wild-type HDFa nor hypomorphic 411BR fibroblasts had microhomology distributions that differed from chance. See also Figure S3.

Figure 4. Cancer translocation induced by paired nicks and DSBs

A. Induction of NPM-ALK cancer translocations with sequence specific nucleases. Der5 encodes the NPM-ALK fusion. B. Der5 is detected only when DSBs or paired nicks are induced on both chromosomes. Wild-type Cas9 with gRNAs for NPM1 and ALK1 gives rise to translocations as does nCas9 with gRNAs NPM1+NPM2 and ALK1+ALK2. Relative positions of cleavage sites are indicated. C. Indel formation at the ALK locus, as monitored by the T7-endonuclease assay. See also Figure S4.

Figure 5. Cancer translocations induced by paired nicks and DSBs

A. – E. Der5 translocations induced by expression of the indicated nucleases in wild-type and mutant HCT116 cells. Nucleases: TAL, TAL^ALK+TAL^NPM (n = 4); Cas9, Cas9+gRNAs (ALK1+NPM1) (n = 4); nCas9, nCas9+gRNAs (ALK1+ALK2 and NPM1+NPM2) (n = 3). A, Translocation frequency. Error bars, +/− SEM. B. The NPM-ALK fusion protein was detected 48 h after expression of the indicated nucleases. C-E, Translocation junction analysis. Junctions from L4^−/− cells exhibited longer deletions and more microhomology. C, Deletion lengths from individual Der5 junctions. For wild-type Cas9, a large fraction of junctions from control L4^+/− cells were blunt end ligations of the two ends but these were absent from L4^−/− cells. For nCas9, the deletions are quite long, reflecting the long overhangs; however, in L4^−/− cells, unlike L4^+/− cells, most of the deletions extended well past the overhangs (see last two columns with asterisks for deletion lengths beyond the overhang). D, Junctions are grouped according to whether the deletions extend beyond the overhang (≤ 30 bp or > 30 bp) or are restricted to the overhang (or, in the case of wild-type Cas9, are 0 bp). E, Microhomology distribution. Microhomology in junctions from control L4^+/− cells is distributed almost identically to that expected by chance with wild-type Cas9 but is greater with nCas9 which leaves long overhangs. See also Figure S5.

Figure 6. LIG3 deficiency does not affect translocations in human cells

A. – C. Neither translocation frequency nor junction characteristics are altered by LIG3 loss. Translocations are induced by expression of the indicated nucleases: ZFN, ZFN^p84+ZFN^EWS, generating Der22 (n = 3); Cas9, Cas9+gRNAs (ALK1+NPM1) or nCas9, nCas9+gRNAs (ALK1+ALK2 and NPM1+NPM2), generating Der5 (n = 2). B, Translocation frequency. Error bars, +/− SEM. C, Deletion lengths from individual Der22 (ZFNs) and Der5 (Cas9, nCas9) junctions. D, Microhomology distribution. Microhomology in junctions from control and LIG3-deficient cells was similar. See also Figure S6.

Figure 7. CtIP affects translocations in XRCC4-deficient but not wild-type cells

A. Translocation frequency is reduced with CtIP knockdown in X4^−/− but not wild-type human cells (n = 3). Error bars, +/− SEM. FLAG-ZFN levels at the time of transfection (t0) and 48 h later (t2) is shown in the inset. B. Deletion lengths from individual Der22 junctions are reduced in X4^−/− but not wild-type cells with CtIP knockdown. C. Fewer deletions extend ≤ 30 bp beyond the overhang in X4^−/− cells with CtIP knockdown. D. Microhomology in junctions from control and X4^−/− cells is not affected by CtIP. See also Figure S7.