Fig. 3: Strand-specific repair of ends requiring processing before ligation.

a. Repair by NHEJ of substrates with non-complementary 3’ overhangs generates different products if repaired by model 1 or model 2 paths to both strands joined. b The substrate in panel a (3’-GAG/GAG) was introduced into wild-type MEFs (WT), MEFs deficient in only Pol μ (Polm^-/-), MEFs deficient in Pol μ, Pol λ, and Pol β (PolX^-/-), and PolX^-/- MEFs complemented by the introduction of wild-type Pol μ (1xWT μ, 1x = 100 ng), the same amount of Pol μ F46A (1xF46A μ), or five times the amount of Pol μ F46A (5xF46A μ). Vertical dashed line separates independent experiments. Data are the mean amounts of +C NHEJ product, as shown in panel a, ±s.d. from n = 3 independent biological replicates, relative to WT cells. Means were analyzed by unpaired t-test (left of dashed line) or one-way ANOVA with Dunnett’s test for multiple comparisons (right of dashed line). ****p < 0.0001, ns = not significant, n.d. = not detected. c Alphafold models⁴² of the Pol λ (cyan, left inset; PDB ID code 2JW5)⁸ or Pol μ (magenta, right inset; PDB ID code 2HTF)⁷ interaction with Ku (rose; based on PDB ID code 7LSY)¹², with the hydrophobic residues required for this interaction in red stick representation. d The substrate in panel a was introduced into MEFs. Relative strand-specific repair plotted as in Fig. 2b except using the noted substrate and plotting top strand Pol μ-dependent repair ( + C product, defined in panel a, green), bottom strand repair using this top strand product as template (blue), and the fraction of both products sensitive to alkali treatment (shaded purple) compared to total (mock-treated) product. Data are the mean ± s.d. of n = 3 independent biological replicates, with the exception of 30 min top strand fraction RNA containing 2 replicates, with all product amounts expressed relative to the total (mock treated, purely DNA and RNA-embedded products both) top strand Pol μ-dependent product after 30 min in cells, as determined by digital PCR.