Fig 2: SHLD1 is required for class switch recombination.a Concentration of indicated isotypes in the serum of 6- to 8-week-old sex- and aged-matched unimmunized WT, Shld1-/-, Xlf-/-, and Shld1-/-Xlf-/- mice. Horizontal lines show mean, n = 11 independent mice (except Shld1-/- Xlf-/- for which n = 9), two-sided Wilcoxon–Mann–Whitney test, p values are indicated in Source Data file. b Representative dot plots showing IgM and IgG1 expression in WT, Shld1-/-, 53bp1-/-, Xlf-/-, and Shld1-/-Xlf-/- stimulated B cells. c Percentages of WT, Shld1-/-, 53bp1-/-, Xlf-/-, and Shld1-/-Xlf-/- B cells expressing either IgG1 or IgG2b isotype after 4 days of in vitro stimulation with anti-IgD dextran, LPS and IL-4 (IgG1) or LPS (IgG2b). Bars represent mean ± s.e.m., WT (n = 11 for IgG1 and n = 7 for IgG2b), Shld1-/- (n = 11 for IgG1 and n = 8 for IgG2b), 53bp1-/- (n = 4 for IgG1 and n = 2 for IgG2b), Xlf-/- (n = 8 for IgG1 and n = 7 for IgGb2), Shld1-/- Xlf-/- (n = 9 for IgG1 and n = 6 for IgG2b), two-sided Wilcoxon–Mann–Whitney test, p values are indicated in Source Data file. d Percentages of IgM-IgG1+ and IgMlow IgG1- cells. Bars represent mean ± s.e.m., n = 8 (except 53bp1-/- for which n = 4, and Xlf-/- and Shld1-/- Xlf-/- for which n = 6), two-sided Wilcoxon–Mann–Whitney test, p values are indicated in Source Data file. e CTV dilution assay in purified B cells cultured in the presence of anti-IgD dextran, LPS, and IL-4 for 96 h. Representative data, n > 2 mice. f SHLD1 predicted domains and variants used. Mutant 1: L95A, R96A, S98A, L99A, F102A, Y103A; Mutant 2: L125A, I129A, L132A; Mutant 3: Y140A, M147A, V150A, I151A, D155A, F169A; Mutant 4: P186A, G187A, L188A, S189A, D191A, I192A, F195A, L196A, L197A. g CSR assay in Shld1-/- primary B cells complemented with SHLD1 variants. Bars represent the average ± s.e.m. relative to Shld1-/- primary B cells complemented with SHLD1 WT, n = 9 independent experiments (non-complemented cells and ?1–70), n = 8 (EV), n = 10 (SHLD1 WT), n = 4 (?1–60 and Mutant 4), n = 6 (? 71–205), n = 2 (?18–21), n = 3 (Mutant 1; Mutant 2; Mutant 3; Y103F; K203A), two-sided Wilcoxon–Mann–Whitney test (****p < 0.0001 (vs EV; vs ?1–70); ***p = 0.001 (vs ?1–60; vs Mutant 4); ***p = 0.0001 (vs ?71–205); **p = 0.0035 (vs Mutant 1; vs Mutant 2; vs Mutant 3; vs Y103F)). h Percentages of WT, Shld1-/-, Xrcc4-/- and Shld1-/-Xrcc4-/- B cells expressing either IgG1 or IgG2b isotype after 4 days of in vitro stimulation with anti-IgD dextran, LPS and IL-4 (IgG1) or LPS (IgG2b). Bars represent mean ± s.e.m., WT (n = 6), Shld1-/- (n = 4), Xrcc4-/- (n = 5 for IgG1 and n = 4 for IgG2b) and Shld1-/-Xrcc4-/-(n = 5), two-sided Wilcoxon–Mann–Whitney test, p values are indicated in Source Data file. i Percentages of IgM-IgG1+ versus IgMlow IgG1- cells. Bars represent mean ± s.e.m., n = 6 (WT), n = 4 (Shld1-/-), n = 5 (Xrcc4-/-; Shld1-/-Xrcc4-/-), two-sided Wilcoxon–Mann–Whitney test, p values are indicated in Source Data file. j Percentages of IgM- IgA+ versus IgMlow IgA- CH12 cells after in vitro stimulation with anti-CD40 antibody, IL-4, and TGF-ß for the indicated genotypes. Bars represent mean ± s.e.m., n = 4 independent experiments with two clones per genotype, two-sided Wilcoxon–Mann–Whitney test, p values are indicated in Source Data file. n.s. non-significant (p = 0.05), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.

Fig 3: SHLD1 promotes functional CSR in XLF-proficient and XLF-deficient cells by limiting DNA end resection.a Representative images of Igh breaks in aberrant metaphases, as quantified in b. Scale bars, 3 µm. b Quantification of Igh breaks (clear bar) and translocations (filled bar) in metaphases of WT, Shld1-/-, Xlf-/-, and Shld1-/-Xlf-/- cells. Bars represent means. n = 1005 (WT), n = 1075 (Shld1-/-), n = 1201 (Xlf-/-), n = 1000 (Shld1-/- Xlf-/-), two-sided Fisher’s exact test from four independent experiments (****p < 0.0001). See also Supplementary Table 4. c Schematic representation of the murine Igh locus. Arrows indicate primers used for the long-range PCR. d Long-range PCR analysis of Sµ to S?1 rearrangements in WT, Xlf-/-, Shld1-/-, and Shld1-/-Xlf-/- stimulated B cells. Il2 gene PCR was used as a loading control. e Normalized coverage of the mapped PacBio reads in WT, Shld1-/-, Xlf-/-, and Shld1-/- Xlf-/- stimulated B cells. Vertical dashed lines indicate the Sµ switch region and vertical lines indicate the S?1 region. f Distribution of read length (in bp) in WT, Shld1-/-, Xlf-/-, and Shld1-/- Xlf-/- cells. n = 3 independent samples. For all boxplots: minima is minimum value, maxima is maximum value, center is median and quartiles shown by box and whiskers, and cross is mean. n.s. non-significant (p = 0.05), ****p < 0.0001. Source data are provided as a Source Data file.

Fig 4: SHLD1 deficiency and long DNA end resection leads to unbiased directional joining during CSR.a Illustration of activation of CSR in normal B cells stimulated with anti-IgD dextran/LPS/IL-4 which induces AID and activates transcription of I?1. Deletional CSR events are indicated with a blue arrow and inversional recombination events are indicated with a red arrow. Intra-switch recombination events are indicated with a dotted line. b CSR-HTGTS-seq analysis of break joining between 5'Sµ and downstream acceptor S-regions in WT, Shld1-/-, Xlf-/-, and Shld1-/-Xlf-/- splenic B cells stimulated with anti-IgD dextran/LPS/IL-4. The blue line indicates deletional joining, and the red line indicates inversional joining. c Zoom-in view of CSR-HTGTS-Seq junctions located in the AID-targeted ectopic S?1 region from WT, Shld1-/-, Xlf-/-, and Shld1-/-Xlf-/- cells. Junctions are plotted at 200 bp bin size. d Bar graph showing percentages of junctions located in Sµ and S?1 regions from WT, Shld1-/-, Xlf-/-, and Shld1-/-Xlf-/- splenic B cells stimulated with anti-IgD dextran/LPS/IL-4. Data represent mean ± s.e.m. from three independent repeats. e Microhomologies (MH) usage of Sµ-S?1 junctions, plotted as a percentage of total junctions. n = 3 independent samples. f, g Percentages of Sµ-S?1 direct joins in f and Sµ-S?1 joins with 4-bp or longer MH in g in the different genetic backgrounds are compared. Bars represent mean ± s.e.m. n = 3 independent samples, unpaired two-tailed t-test, p values are indicated in Source Data file. h Percentage of resection at the S?1 region in the different genetic backgrounds as indicated. Bars represent mean ± s.e.m., n = 3 independent samples, unpaired two-tailed t-test, p values are indicated in Source Data file. i Inversional/deletional joining ratio at the S?1 region in the different genetic background as indicated. Bars represent mean ± s.e.m., n = 3 independent samples, unpaired two-tailed t-test, p values are indicated in Source Data file. n.s. non-significant (p = 0.05), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file. j Working model for the specific roles of 53BP1 and components of the SHLD complex in antigen receptor gene diversification. The 53BP1-dependent chromatin response acts independently of SHLD to promote chromosome structural changes that are essential for long-range V(D)J recombination and CSR as well as for RAG-DSB repair in the context of an unstable post-cleavage complex (i.e., XLF-deficient cells) (left panel), while the 53BP1-SHLD axis limits resection at AID-induced DSB ends, providing an end-protection mechanism that permits productive CSR by NHEJ and alt-NHEJ (right panel).

Fig 5: 53BP1 displays SHLD1-independent functions during V(D)J recombination.a Schematic representation of the murine Tcrß locus. Variable (V), diversity (D), joining (J), and constant (C) segments are shown. The star marks a pseudogene (Jß2.6). b PCR analysis of Dß1 to Jß1 rearrangements in Rag2-/-, WT, Shld1-/-, 53bp1-/-, Xlf-/-, and Shld1-/-Xlf-/- thymocytes. The arrows indicate the region in germline configuration. The bands marked by arrowheads represent rearrangements of Dß to one of the Jß segments. Il2 gene PCR was used as a loading control. c PCR analysis of Dß2 to Jß2 rearrangements. d Schematic representation of the mouse Tcra/d locus with the four different Tcrd rearrangements depicted individually below. e Relative frequency of Tcrd locus rearrangements in total thymocytes from WT, Shld1-/- and 53bp1-/- mice. Quantitative assessment of genomic DNA rearrangements of Dd1 to Dd2, Dd2 to Jd1, and Vd2 and Vd5 to (D)Jd1 genes were performed by qPCR and normalized to the signal of the non-rearranging DNA 3' of Jd2. Histograms represent the average relative to WT mice for each rearrangement ± s.e.m., n = 5 independent samples (except 53bp1-/- for which n = 4), two-sided Wilcoxon–Mann–Whitney test (*p = 0.0179). f Schematic of inversional Igk6-23 –J1 rearrangement (position of primers (arrows) used to assay coding joint are shown). g Semi-quantitative nested PCR analysis of Igk6-23 –J1 coding joint (CJ) in v-Abl pro-B cell lines treated for 72 h with ABLki. Il2 gene PCR was used as a loading control. (WT: clone #12095; 53bp1-/-: clone #9999; Shld1-/-: clone #O37; Xlf-/-: clone #16488; Shld1-/-Xlf-/-: clone #XO2-8; Xlf-/-53bp1-/-: clone #9X2). CJ/Il2 ratio of band intensity is indicated below the gel. h Schematic of pMX-INV recombination substrate. i Bar plot showing V(D)J recombination efficiency of pMX-INV recombination substrate in v-Abl pro-B cell lines. v-Abl pro-B cell lines of each genotype were treated with ABLki for 72 h and GFP+ cells were analyzed by flow cytometry. Bars represent mean ± s.e.m., n = 4 independent clones (WT; 53bp1-/-), n = 3 (Shld1-/-), n = 2 (Xlf-/-; Shld1-/- Xlf-/-; 53bp1-/- Xlf-/-), two-sided Wilcoxon–Mann–Whitney test (****p < 0.0001). WT: clones #O38 (n = 4 independent experiments), #GBB (n = 4), #12096 (n = 4) and #12095 (n = 11); 53bp1-/-: clones #9999 (n = 7), #1110 (n = 3), #BP95-2 (n = 3) and #BP95-5 (n = 3); Shld1-/-: clones #O32 (n = 3), #O37 (n = 6) and #O44 (n = 3); Xlf-/-: clones #16488 (n = 10) and #X95-3 (n = 2); Shld1-/-Xlf-/-: clones #XO2-8 (n = 4) and #XO2-24 (n = 6); Xlf-/-53bp1-/-: clones #9X1 (n = 8) and #9X2 (n = 6). n.s. non-significant (p = 0.05), *p < 0.05, ****p < 0.0001. Source data are provided as a Source Data file.