Essential Roles for Polymerase θ-Mediated End Joining in the Repair of Chromosome Breaks

Abstract

DNA polymerase theta (Pol θ)-mediated end joining (TMEJ) has been implicated in the repair of chromosome breaks, but its cellular mechanism and role relative to canonical repair pathways are poorly understood. We show that it accounts for most repairs associated with microhomologies and is made efficient by coupling a microhomology search to removal of non-homologous tails and microhomology-primed synthesis across broken ends. In contrast to non-homologous end joining (NHEJ), TMEJ efficiently repairs end structures expected after aborted homology-directed repair (5' to 3' resected ends) or replication fork collapse. It typically does not compete with canonical repair pathways but, in NHEJ-deficient cells, is engaged more frequently and protects against translocation. Cell viability is also severely impaired upon combined deficiency in Pol θ and a factor that antagonizes end resection (Ku or 53BP1). TMEJ thus helps to sustain cell viability and genome stability by rescuing chromosome break repair when resection is misregulated or NHEJ is compromised.

Figure 1. Effect of end-joining deficiencies on cell proliferation

(A) The noted cell types were seeded at 5×10³ cells/well in 24 well dishes, then counted 12 hours after plating and every 24 hours afterwards. Data shown are the mean +/- SEM, n=3. (B) Ku70^-/- cells were infected with lentivirus containing Cas9 and either control guides or guides targeting sites in regions coding for mouse Pol θ, seeded into 10 cm plates, and stained after 10 days (right panel). Colonies of more than 100 cells were counted from triplicate experiments (left panel); Data shown are the mean +/- SEM, n=3. Statistical significance was assessed by ANOVA with Bonferoni correction for multiple comparisons. ***p<0.001. See also Figure S1.

Figure 2. Effect of end-joining deficiencies on repair of pre-resected ends

(A) Linear substrates with varied end structures were introduced into noted cell types and end-joining products characterized by quantitative polymerase chain reaction (top panels) or electrophoresis of amplified products (bottom panels); Data shown below are the mean +/- SEM, n=3. Statistical significance was assessed by one-way ANOVA with Bonferoni correction of p values to account for multiple comparisons. NS not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. (B) A substrate with symmetrical 45 nt 3′ single stranded DNA overhangs was introduced into the noted cell types, and the mean efficiency of end-joining determined by qPCR and expressed as a fraction of that observed in WT cells. Amplified products generated in a parallel fixed-cycle number PCR were also characterized by electrophoresis; products of size consistent with joining after overhang clipping vs. overhang retention are noted. (C) A substrate with 70 nt 3′ overhangs was introduced into the noted cell lines and joining characterized as described above. (D) A substrate as in panel B, except with a 20 nt non homologous tail ending in a 5′ terminal biotin-streptavidin group, was introduced into the noted cell lines and joining characterized as described above. See also Figure S2.

Figure 3. Mechanism of Pol θ mediated end-joining

(A) The 45 nt 3′ overhang substrate described in Fig. 2B was introduced into cells that were WT, Polq^-/-, and Polq^-/- engineered to express wild type human Polq or variants deficient in helicase-like domain ATPase activity (K121M), or polymerase activity (D2330A+Y2331A), and joining efficiency assessed by qPCR using primers that require retention of at least 10 nt of each 3′ overhang sequence in junctions. Joining efficiency was also characterized in Poll^-/-Polm^-/- MEFs, relative to matched WT control MEFs. Data shown are the mean +/- SEM, n=3. Statistical significance was assessed by one-way ANOVA with Bonferoni correction. NS =not significant, *p<0.05, **p<0.01, ***p<0.001. (B) Observed frequency of different joining products of the 70 nt overhang substrate (Fig. 2C) recovered from wild type cells, including the 4 most common products (above dashed line) and 4 other representative examples (see also Table S1). The first 3 columns summarize the structures of the inferred intermediates. (C) For experiments described in Fig. 3A (using the 45 nt overhang) we report the fraction of products directed by the terminal 4 bp microhomology (shown in panel B), as determined by sequencing. Data shown are the mean +/- SEM, n=3. Statistical significance was assessed by two-way ANOVA (p values as above). (D) The extent each sized microhomology is enriched (bars right of Y axis) or depleted (bars left of Y axis) in recovered junctions, relative to their representation in the set of all possible deletion products. See also Figure S3, Table S1.

Figure 4. Effect of end-joining deficiencies on repair of Cas9-induced chromosome breaks

(A) Cas9 and guide RNA specific to a chromosome 6 target were expressed in the noted cell types, and products of end-joining characterized by target site amplification and sequencing. (B) Bars denote extent of deleted DNA for each end-joining product relative to the Cas9 target site, and are shaded according to the extent of microhomology as noted in the legend. The height of each bar on the Y axis defines the proportion of each product in a set of 20,000 estimated input molecules, and is averaged from 3 experiments for each of WT (top left), Polq^-/-(top right), Ku70^-/-, (bottom left), and Polq^-/-Ku70^-/- cells (bottom right). (C) The percentage of all end-joining products with microhomology >2 bp and deletion (Δ) within the noted size ranges. Data shown are the mean +/- SEM, n=3. Statistical significance was assessed by two-way ANOVA with Bonferoni correction (results for the 5-50 bp category shown). *p<0.05, ****p<0.0001. (D) The percentage of all end-joining products with insertion >4 bp recovered from MEF lines described above. Data shown are the mean +/- SEM, n=3. Statistical significance was assessed by one-way ANOVA with Bonferoni correction (p values as above). (E) Structure of selected Pol θ mediated insertions. Successive segments denote template switching, with opposing arrows identifying use of opposite strands. The lengths of microhomologies in flanks of successive segments are in parentheses. Black segments, from chromosome 6; red segments, from other chromosomes. See also Figure S4, Table S2.

Figure 5. Effect of aberrant end resection on repair and viability

(A) Template DNA from the experiment described in Fig. 4 (left panel), or from Ku70^-/- cells infected with lentivirus containing Cas9 and either a pool of control guides or guides targeting Mre11 at amino acids 572 or 603 respectively (right panel), was amplified and characterized by electrophoresis. (B) Polq^-/- cells were infected with lentivirus containing Cas9 and a pool of either control guides or guides targeting 53BP1. Colony forming ability was assessed as in Fig. 1B. Data shown are the mean +/- SEM, n=3. Statistical significance was assessed by unpaired T-test. ***p<0.001. See also Figure S5.

Figure 6. Effect of end-joining deficiencies on repair of chromosomal breaks by homologous recombination and translocation

(A, B) Homologous recombination was assessed by introduction of Cas9, guide RNA, and a plasmid homology donor that introduces a ScaI recognition site at the site of breakage in chromosome 6 into noted cell types. DNA was harvested 48 hours later, amplified, and digested with ScaI followed by electrophoresis. (C) The mean fraction of the chromosome 6 target site that acquired sensitivity to ScaI. Data shown are the mean +/- SEM, n=3. Statistical significance assessed by one-way ANOVA with Bonferoni correction. NS =not significant, ***p<0.001, ****p<0.0001. (D) Cas9 and guide RNAs targeting chromosome 6 and chromosome 11 were expressed in the noted cell types. Genomic DNA was isolated after 48 hours and used as template for parallel emulsion PCRs (ddPCR) specific for the t(6,11) der11 translocation product and an input genome control. (E) Mean frequency of der11 translocations is determined by the number of translocations over the number of input genomes; Data shown are the mean +/- SEM, n=3, with >4×10⁴ genomes assessed/experiment. Statistical significance was assessed by one-way ANOVA with Bonferoni correction (p values as above). See also Figure S6, Table S3.

Figure 7. Mechanism and role of Pol θ mediated end-joining

Suggested contexts for engagement of Pol θ mediated end-joining, and proposed mechanism including a i) search for microhomology, ii) removal of non-homologous tails, and iii) synthesis.