Homology and enzymatic requirements of microhomology-dependent alternative end joining

Abstract

Nonhomologous DNA end joining (NHEJ) is one of the major double-strand break (DSB) repair pathways in higher eukaryotes. Recently, it has been shown that alternative NHEJ (A-NHEJ) occurs in the absence of classical NHEJ and is implicated in chromosomal translocations leading to cancer. In the present study, we have developed a novel biochemical assay system utilizing DSBs flanked by varying lengths of microhomology to study microhomology-mediated alternative end joining (MMEJ). We show that MMEJ can operate in normal cells, when microhomology is present, irrespective of occurrence of robust classical NHEJ. Length of the microhomology determines the efficiency of MMEJ, 5 nt being obligatory. Using this biochemical approach, we show that products obtained are due to MMEJ, which is dependent on MRE11, NBS1, LIGASE III, XRCC1, FEN1 and PARP1. Thus, we define the enzymatic machinery and microhomology requirements of alternative NHEJ using a well-defined biochemical system.

Figure 1

Schematic presentation of cell-free repair assay to evaluate microhomology-mediated alternative DNA end joining. (a) Outline of experimental strategy used for detection of MMEJ. Two double-stranded oligomers possessing 10 nt microhomology were incubated with buffer and cell-free extracts. After heat inactivaction, end-joined products were subjected to radioactive PCR and the products were resolved on 8% denaturing polyacrylamide gel. (b and c) Schematic showing strategy used for detection of alternative NHEJ products using radiolabelled oligomers. Examples of 10 nt (b) and 22 nt (c) microhomology-bearing DNA substrates are depicted to show strategy employed for detection of MMEJ. Dark rectangles indicate positions of microhomology. PCR primer positions are also indicated

Figure 2

Evaluation of different experimental conditions for the development of a cell-free repair system to assess microhomology-mediated alternative DNA end joining. (a) Evaluation of MMEJ in presence of increasing concentrations of cell-free extracts. Testicular extracts (0, 0.025, 0.050, 0.075, 0.1, 0.2, 0.5, 0.75, 1.0, 1.5, 2.0, 2.5 and 3.0 μg) were incubated with DNA substrates (4 nM) 10 nt microhomology for 2 h at 30 °C. Lane 1 indicates no protein control. (b) Time kinetics of MMEJ on a 10-nt microhomology-containing DNA substrates. Rat testicular extracts (1 μg) were incubated with DNA substrates for 0, 5, 15 and 30 min, and 1, 2, 4, 6, 8, 10 and 12 h, and products were analyzed on 8% denaturing PAGE. (c) MMEJ assay at increasing incubation temperatures. A unit of 1.0 μg of extract was incubated with 10 nt microhomology-containing DNA substrates in rat testicular extracts for 2 h at 4, 16, 25, 30, 37 and 40 °C. Lane 1 is no protein control. (d) Effect of MgCl₂ on MMEJ catalyzed by cell-free extracts. Rat testicular extracts (1 μg) were incubated with microhomology substrates and increasing concentrations of MgCl₂ at 30 °C. Lane 1 is no protein control. Lanes 2–7 indicate MMEJ in the presence of 0, 1, 2, 5, 10 and 20 mM of MgCl₂, respectively. (e) Effect of ATP on MMEJ catalyzed by cell-free extracts. Rat testicular extracts (1 μg) was incubated with microhomology substrates and increasing concentrations of ATP for 2 h. Lanes 2–8 indicate MMEJ in the presence of 0, 0.1, 0.25, 0.5, 1, 2 and 4 mM of ATP, respectively. In panels a–e, bar diagram showing quantification based on at least three independent experiments are provided. MMEJ products are indicated by an arrow, while C-NHEJ products are bracketed. In each case, lower panel serves as the loading control for equal DNA, indicated as ‘input DNA'. M' and M indicate 60 nt marker and 50 nt ladder, respectively. PSLU in y axis of bar diagram indicates photostimulated luminescence units.*P<0.05; **P<0.01; ***P<0.001. Error bars represent S.E.M.

Figure 3

Comparison of frequency and mechanism of MMEJ between normal and cancer cells. (a) SDS-PAGE profile of cell-free extracts prepared from testis, thymus, K562 and Reh. (b) Bar diagram showing densitometry analysis for proteins in PAGE shown in panel a. (c) Comparison of MMEJ and C-NHEJ efficiency catalyzed by testis, thymus, K562 and Reh cell-free extracts. Cell-free extracts (1.0 μg) were incubated with oligomeric DNA possessing 10 nt microhomology in a buffer containing 10 mM Tris⋅HCl (pH 8.0), 20 mM MgCl₂, 1 mM ATP, 10% PEG 8000 and 1 mM DTT for 2 h at 30 °C. The products were resolved on a PAGE and visualized. For other details refer Figure 2 legend. M is 50 nt ladder. M is marker for 60 nt position. MMEJ and NHEJ products are indicated. (d) Bar diagram showing comparison of MMEJ and C-NHEJ catalyzed by testis, thymus, K562 and Reh, relative to input. *P<0.05. Error bars represent S.E.M. (e) Comparison of different modes of NHEJ among normal tissues and cancer cell lines. The total end-joining junctions from testis, thymus, K562 and Reh cells were PCR amplified, cloned and sequenced. Each sequence shown is derived from an independent clone. Cases where microhomology is used are indicated in the column. The ‘joined products' indicate the usage of substrates. Red color indicates sequences that are deleted, while blue indicates insertions. Green indicates mutations in the sequence. Microhomology region is underlined and the sequence is indicated in bold. (f) Western blots showing expression profile of canonical and noncanonical NHEJ proteins in testes, thymus and leukemic cell lines. Bands marked with an asterisk could be an isoform. Both isoforms of Ligase III, α and β, are indicated. PCNA and actin were used as loading control

Figure 4

Determination of length of microhomology requirement during MMEJ. (a) Depiction of sequence and probable MMEJ product following joining of ds oligomeric substrates possessing 3, 5, 8, 10, 13, 16, 19 and 22 nt of microhomology, which are indicated in red. Restriction enzyme sites generated due to microhomology-mediated joining are also indicated. (b) Comparison of MMEJ catalyzed by testicular extracts when different microhomology regions were used. Rat testicular extracts were incubated with oligomeric DNA substrates harboring 3, 5, 8, 10, 13, 16, 19 and 22 nt microhomology for 2 h at 25 ºC. End-joined products were detected by radioactive PCR. In case of every substrate, reactions are shown with either of the substrate or both. MMEJ products are indicated by arrow, while NHEJ products are bracketed. M and M' are molecular weight markers. (c) Comparison of MMEJ of oligomeric DNA containing DSBs when flanked with different length microhomology catalyzed by Reh cell-free extract. M and M' are markers as indicated

Figure 5

Sequence analysis of MMEJ junctions, when DSBs with different microhomology regions were used. Products due to possible MMEJ (3, 5, 8, 10, 13 and 16 nt microhomology) catalyzed by testicular and thymic extracts were gel purified, cloned and sequenced. The sequences denoted in red color are the deleted nucleotides. (a) MMEJ junctions from rat testicular extracts. (b) MMEJ junctions derived from rat thymic extracts. For other details, refer Figures 1 and 4 legends

Figure 6

Evaluation of the effect of C-NHEJ proteins on joining of DNA substrates containing DSBs flanked with 3, 10 and 19 nt microhomology regions. (a) Western blots showing immunodepletion of KU70, KU80 and LIGASE IV proteins from rat testicular cell extracts. ‘Control' is whole cell extract and ‘IP' is immunodepleted extract. GAPDH was used as an internal loading control. (b) Efficiency of MMEJ and NHEJ following immunodepletion of KU70, KU80 and LIGASE IV on DNA substrates containing 3, 10 and 19 nt microhomology. (c) Comparison of MMEJ and C-NHEJ in NALM6 and Ligase IV genetic knockout cell lines. Cell-free extracts were incubated with 10 nt microhomology substrates as described, products from multiple experiments were quantified and presented. *P<0.05; **P<0.01. Error bars represent S.E.M. (d) Western blots showing immunodepletion of polymerase μ and λ and evaluation of MMEJ efficiency using the immunodepleted extracts. (e) Effect of wortmannin, a DNA-PKcs inhibitor on MMEJ. Wortmanin (1, 10 and 100 μM) was incubated with cell-free extracts of testis (2 μg) and oligomeric DNA containing 10 nt microhomology and analyzed. In panels b–e, MMEJ products are indicated by arrow, while NHEJ products are bracketed. (f) siRNA-mediated knockdown of classical NHEJ proteins in Reh cells. Reh cells were transfected with siRNA against KU70, KU80, XRCC4 and LIGASE IV and harvested after 48 h. Cell-free extracts were prepared and efficiency of knockdown was evaluated by western blotting and was quantified (shown as bar diagram). Scrambled siRNA was used as control. (g) Efficiency of MMEJ and NHEJ following siRNA-mediated knockdown of expression of KU70, KU80, XRCC4 and LIGASE IV. MMEJ products are boxed, while NHEJ products are bracketed. M is marker

Figure 7

Determination of proteins responsible for microhomology-mediated alternative end joining. (a) Bar diagram showing immunodepletion of XRCC1, PARP1 and LIGASE I proteins from rat testicular extracts. Lane 1, control IP. Lanes 2 and 3 are immunodepleted extract following incubation with two different concentrations of antibody (0.2 and 0.3 μg/50 μl). Following quantitation data is normalized against respective loading controls and presented. (b) Efficiency of MMEJ and C-NHEJ following immunodepletion of XRCC1 and PARP1 and LIGASE I on DNA substrates containing 10 nt microhomology. Joining assay and its quantification (%) are shown. The highest concentration (1 mM) was not considered for quantification, as it inhibited the joining in a nonspecific manner. (c) Effect of increasing concentrations of mirin, a MRN complex inhibitor, on MMEJ. Mirin (100, 200, 400, 600, 800 μM and 1 mM) was incubated with cell-free extracts of testis and 3, 10 and 19 nt microhomology-containing substrates and analyzed. Joining assay and its quantitation are presented. (d) siRNA-mediated knockdown of proteins in Reh cells. Reh was transfected with siRNA against LIGASE III, PARP1, MRE11 and RAD50, and harvested after 48 h. Cell-free extracts were prepared and efficiency of knockdown was evaluated by western blotting and quantified (shown as bar diagram). Scrambled siRNA was used as control. (e) Efficiency of MMEJ and NHEJ following siRNA-mediated knockdown of expression of LIGASE III, PARP1, MRE11, NBS1 and RAD50. MMEJ products are indicated by arrow, while NHEJ products are bracketed. M and M' are markers

Figure 8

Mechanism of microhomology-mediated alternative NHEJ. Schematic presentation shows different modes of DSB repair via NHEJ and MMEJ. When a DSB is generated flanking a microhomology region, there are two outcomes. If KU proteins bind to the ends and recruit C-NHEJ proteins, it can result in C-NHEJ. Alternatively, if the microhomology is recognized by MRN complex and PARP1 or a yet unidentified protein, it can follow MMEJ. This is followed by recruitment of FEN1/unknown endonucleases, which can remove the flap. Further, recruitment of XRCC1–LIGASE III at the site helps in ligating the DNA ends leading to an intact DNA