Activation-induced cytidine deaminase (AID) catalyses the deamination of deoxycytidines to deoxyuracils within immunoglobulin genes to induce somatic hypermutation and class-switch recombination1,2. AID-generated deoxyuracils are recognized and processed by subverted base-excision and mismatch repair pathways that ensure a mutagenic outcome in B cells3,4,5,6. However, why these DNA repair pathways do not accurately repair AID-induced lesions remains unknown. Here, using a genome-wide CRISPR screen, we show that FAM72A is a major determinant for the error-prone processing of deoxyuracils. Fam72a-deficient CH12F3-2 B cells and primary B cells from Fam72a−/− mice exhibit reduced class-switch recombination and somatic hypermutation frequencies at immunoglobulin and Bcl6 genes, and reduced genome-wide deoxyuracils. The somatic hypermutation spectrum in B cells from Fam72a−/− mice is opposite to that observed in mice deficient in uracil DNA glycosylase 2 (UNG2)7, which suggests that UNG2 is hyperactive in FAM72A-deficient cells. Indeed, FAM72A binds to UNG2, resulting in reduced levels of UNG2 protein in the G1 phase of the cell cycle, coinciding with peak AID activity. FAM72A therefore causes U·G mispairs to persist into S phase, leading to error-prone processing by mismatch repair. By disabling the DNA repair pathways that normally efficiently remove deoxyuracils from DNA, FAM72A enables AID to exert its full effects on antibody maturation. This work has implications in cancer, as the overexpression of FAM72A that is observed in many cancers8 could promote mutagenesis.
This is a preview of subscription content
Subscription info for Chinese customers
We have a dedicated website for our Chinese customers. Please go to naturechina.com to subscribe to this journal.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Muramatsu, M. et al. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 102, 553–563 (2000).
Feng, Y., Seija, N., JM, D. I. N. & Martin, A. AID in antibody diversification: there and back again. Trends Immunol. 41, 586–600 (2020).
Cascalho, M., Wong, J., Steinberg, C. & Wabl, M. Mismatch repair co-opted by hypermutation. Science 279, 1207–1210 (1998).
Di Noia, J. & Neuberger, M. S. Altering the pathway of immunoglobulin hypermutation by inhibiting uracil-DNA glycosylase. Nature 419, 43–48 (2002).
Wiesendanger, M., Kneitz, B., Edelmann, W. & Scharff, M. D. Somatic hypermutation in MutS homologue (MSH)3-, MSH6-, and MSH3/MSH6-deficient mice reveals a role for the MSH2-MSH6 heterodimer in modulating the base substitution pattern. J. Exp. Med. 191, 579–584 (2000).
Rada, C., Di Noia, J. M. & Neuberger, M. S. Mismatch recognition and uracil excision provide complementary paths to both Ig switching and the A/T-focused phase of somatic mutation. Mol. Cell 16, 163–171 (2004).
Rada, C. et al. Immunoglobulin isotype switching is inhibited and somatic hypermutation perturbed in UNG-deficient mice. Curr. Biol. 12, 1748–1755 (2002).
Guo, C. et al. Ugene, a newly identified protein that is commonly overexpressed in cancer and binds uracil DNA glycosylase. Cancer Res. 68, 6118–6126 (2008).
Muramatsu, M. et al. Specific expression of activation-induced cytidine deaminase (AID), a novel member of the RNA-editing deaminase family in germinal center B cells. J. Biol. Chem. 274, 18470–18476 (1999).
Lawson, K. A. et al. Functional genomic landscape of cancer-intrinsic evasion of killing by T cells. Nature 586, 120–126 (2020).
Pan-Hammarstrom, Q. et al. Impact of DNA ligase IV on nonhomologous end joining pathways during class switch recombination in human cells. J. Exp. Med. 201, 189–194 (2005).
Ward, I. M. et al. 53BP1 is required for class switch recombination. J. Cell Biol. 165, 459–464 (2004).
Yan, C. T. et al. IgH class switching and translocations use a robust non-classical end-joining pathway. Nature 449, 478–482 (2007).
Perez-Duran, P. et al. UNG shapes the specificity of AID-induced somatic hypermutation. J. Exp. Med. 209, 1379–1389 (2012).
Frieder, D., Larijani, M., Collins, C., Shulman, M. & Martin, A. The concerted action of Msh2 and UNG stimulates somatic hypermutation at A.T base pairs. Mol. Cell. Biol. 29, 5148–5157 (2009).
Thientosapol, E. S. et al. Proximity to AGCT sequences dictates MMR-independent versus MMR-dependent mechanisms for AID-induced mutation via UNG2. Nucleic Acids Res. 45, 3146–3157 (2017).
Martin, A. et al. Msh2 ATPase activity is essential for somatic hypermutation at a-T basepairs and for efficient class switch recombination. J. Exp. Med. 198, 1171–1178 (2003).
Phung, Q. H. et al. Increased hypermutation at G and C nucleotides in immunoglobulin variable genes from mice deficient in the MSH2 mismatch repair protein. J. Exp. Med. 187, 1745–1751 (1998).
Delbos, F., Aoufouchi, S., Faili, A., Weill, J. C. & Reynaud, C. A. DNA polymerase eta is the sole contributor of A/T modifications during immunoglobulin gene hypermutation in the mouse. J. Exp. Med. 204, 17–23 (2007).
Arakawa, H., Saribasak, H. & Buerstedde, J. M. Activation-induced cytidine deaminase initiates immunoglobulin gene conversion and hypermutation by a common intermediate. PLoS Biol. 2, E179 (2004).
Di Noia, J. M. & Neuberger, M. S. Immunoglobulin gene conversion in chicken DT40 cells largely proceeds through an abasic site intermediate generated by excision of the uracil produced by AID-mediated deoxycytidine deamination. Eur. J. Immunol. 34, 504–508 (2004).
Campo, V. A. et al. MSH6- or PMS2-deficiency causes re-replication in DT40 B cells, but it has little effect on immunoglobulin gene conversion or on repair of AID-generated uracils. Nucleic Acids Res. 41, 3032–3046 (2013).
Di Noia, J. M., Rada, C. & Neuberger, M. S. SMUG1 is able to excise uracil from immunoglobulin genes: insight into mutation versus repair. EMBO J. 25, 585–595 (2006).
Wang, Q. et al. The cell cycle restricts activation-induced cytidine deaminase activity to early G1. J. Exp. Med. 214, 49–58 (2017).
Cappelli, E. et al. Rates of base excision repair are not solely dependent on levels of initiating enzymes. Carcinogenesis 22, 387–393 (2001).
Krusong, K., Carpenter, E. P., Bellamy, S. R., Savva, R. & Baldwin, G. S. A comparative study of uracil-DNA glycosylases from human and herpes simplex virus type 1. J. Biol. Chem. 281, 4983–4992 (2006).
Krokan, H. E. et al. Error-free versus mutagenic processing of genomic uracil–relevance to cancer. DNA Repair 19, 38–47 (2014).
Zeng, X. et al. DNA polymerase eta is an A–T mutator in somatic hypermutation of immunoglobulin variable genes. Nat. Immunol. 2, 537–541 (2001).
Belcheva, A. et al. Gut microbial metabolism drives transformation of MSH2-deficient colon epithelial cells. Cell 158, 288–299 (2014).
Ramachandran, S. et al. The SAGA deubiquitination module promotes DNA repair and class switch recombination through ATM and DNAPK-mediated γH2AX formation. Cell Rep. 15, 1554–1565 (2016).
Aregger, M., Chandrashekhar, M., Tong, A. H. Y., Chan, K. & Moffat, J. Pooled lentiviral CRISPR–Cas9 screens for functional genomics in mammalian cells. Methods Mol. Biol. 1869, 169–188 (2019).
Sarno, A. et al. Uracil–DNA glycosylase UNG1 isoform variant supports class switch recombination and repairs nuclear genomic uracil. Nucleic Acids Res. 47, 4569–4585 (2019).
Li, C. et al. The H2B deubiquitinase Usp22 promotes antibody class switch recombination by facilitating non-homologous end joining. Nat. Commun. 9, 1006 (2018).
Li, C. et al. Early-life programming of mesenteric lymph node stromal cell identity by the lymphotoxin pathway regulates adult mucosal immunity. Sci. Immunol. 4, aax1027 (2019).
Liu, M. et al. Two levels of protection for the B cell genome during somatic hypermutation. Nature 451, 841–845 (2008).
Siriwardena, S. U., Perera, M. L. W., Senevirathne, V., Stewart, J. & Bhagwat, A. S. A tumor-promoting phorbol ester causes a large increase in APOBEC3A expression and a moderate increase in APOBEC3B expression in a normal human keratinocyte cell line without increasing genomic uracils. Mol. Cell. Biol. 39, e00238-18 (2019).
So, C. C., Ramachandran, S. & Martin, A. E3 ubiquitin ligases RNF20 and RNF40 are required for double-stranded break (DSB) repair: evidence for monoubiquitination of histone H2B lysine 120 as a novel axis of DSB signaling and repair. Mol. Cell. Biol. 39, e00488-18 (2019).
Boulianne, B. et al. AID-expressing germinal center B cells cluster normally within lymph node follicles in the absence of FDC−M1+ CD35+ follicular dendritic cells but dissipate prematurely. J. Immunol. 191, 4521–4530 (2013).
We thank P. Poussier and M. Shulman for critical review of the manuscript. Y.F., C.L. and A.A.-Q. are recipients of the Canadian Institutes of Health Research Postdoctoral Fellowship. N.S.D. is the recipient of a doctoral award from the FRQ-S (Fonds de recherche Santé Québec). J.A.S. was supported by a competitive graduate research assistantship from the Wayne State University and R.M.-R. was supported by a Thomas C. Rumble fellowship. J.M.D.N. is a Merit scholar from the Fonds de recherche de Quebec–Santé and is supported by CIHR (PJT-155944). J.M. is supported from a CIHR project grant (CBT-438323) and holds a Canada Research Chair in Functional Genomics. D.D. is a Canada Research Chair (Tier I) and work in the D.D. laboratory was supported by a grant from the CIHR (FDN143343). A.S.B. was supported by a National Institutes of Health grant (1R21AI144708) and Bridge Funding grant from Wayne State University. A.M. is supported by grants from the CIHR (PJT-153307 and PJT-156330).
The authors declare no competing interests.
Peer review information Nature thanks Uttiya Basu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
(a) Western blot analysis to identify Cas9-expressing CH12 cell subclones after transduction with a lentiviral vector expressing Cas9. Experiment was repeated 2 times independently with similar results. (b) Schematic of the CRISPR/Cas9 sorting screen in CH12 cells. CH12 cells stably expressing Cas9 were transduced with a mouse gRNA library that contains ~5 gRNA per gene. Transduced populations were treated with CIT cocktail (CD40 ligand, IL4, TGFβ) to induce CSR from IgM to IgA, followed FACS sorting to separate the switched (IgA+) versus the unswitched (IgA-) population. Genomic DNA from the initially transduced cells (T0), and from unsorted, IgA-, IgA+ population was isolated, and sequenced. (c) gRNA was ranked using a NormZ plot (standard deviations from the mean) comparing IgA- cells to the unsorted population. (d) Guide RNA (gRNA) targeting strategy against mouse Aicda and Fam72a genes to validate the role of FAM72A in CSR in bulk CH12 cells expressing Cas9. (e) Quantification of Fam72a mRNA relative to HPRT in wild-type (WT) and Fam72a-/- CH12 clones by qPCR. (f) Sequenced Fam72a, Msh2, Ung, and ligase4 alleles in Fam72a-/-, Ung-/-, Fam72a-/-Ung-/-, Msh2-/-, Fam72a-/-Msh2-/-, and Ligase 4-/- CH12 clones generated using CRISPR/Cas9. Underlined sequence denotes gRNA target sequence, with the wildtype amino acid sequence indicated at the bottom. (g) Assessing the role of FAM72A in DNA double-strand break repair pathways. WT, Fam72a-/- and Ligase 4-/- CH12 clones stably expressing EJ5-GPF, EJ2-GFP, and DR-GFP substrates that measure non-homologous end joining (NHEJ), alternative end joining (A-EJ), and homologous recombination (HR), respectively. Cells were mock transfected or transfected with yeast endonuclease I-SceI expressing vector, pCBA-SceI. GFP expression was monitored by flow cytometry three days post-transfection. Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test. Left panel: p=0.028 for the comparison between WT clone 1 and Lig4-/- clone; middle panel: p=0.014 for the comparison between WT clone I and Fam72a-/- clone 1; p=0.0003 for the comparison between WT clone I and Lig4-/- clone; ns: not significant. Data are representative of 3 independent experiments.
(a) Schematic representation of Fam72a gene disruption strategy. The whole coding sequence of Fam72a was replaced with a LacZ/neo cassette by homologous recombination using a bacterial artificial chromosome (BAC)-based targeting vector. The floxed neo cassette was removed by further breeding to a ubiquitous Cre mouse strain. Strain development was done at MMRRC (UC, Davis). PCR genotyping primer sequence can be found in Supplementary Table 2. (b) qPCR analysis of Fam72a mRNA from resting spleen B cells or spleen B cells that were ex vivo stimulated with LPS for 3 days (n= 4 mice per group). Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test (p= 0.0006 for the comparison in the left panel). Data are representative of 2 independent experiments.
Extended Data Fig. 3 Comparable B cell profiles in the bone marrow of Fam72a-/- and Fam72a+/+ littermate mice.
(a) Representative FACS plots of B cells derived from the bone marrow of Fam72a-/- or Fam72a+/+ littermate mice. (b) Full minus one (FMO)-derived background staining for B220, CD43, BP-1, CD24 and IgM, respectively. (c) The frequencies of indicated B cell fractions in the bone marrow of Fam72a-/- or Fam72a+/+ littermates (n= 4 mice per group). Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test (ns: not significant). Data are representative of 2 independent experiments. (d) Same as c except that total cell numbers were reported.
Extended Data Fig. 4 Comparable B cell profiles in the spleen of Fam72a-/- and Fam72a+/+ littermates.
(a) Representative FACS plots of B cells in the spleen of Fam72a-/- or Fam72a+/+ littermate mice. (b) FMO-derived background staining for B220, CD93, CD23 and IgM, respectively. (c) The frequencies of indicated B cell subsets in the spleen of Fam72a-/- or Fam72a+/+ littermates (n= 4 mice per group). Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test (ns: not significant). Data are representative of 2 independent experiments. MZ: marginal zone B cells; FO: follicular B cells. (d) Same as c except that total cell numbers were reported.
Extended Data Fig. 5 CSR and germinal center formation analysis from Fam72a-/- Fam72a+/- and WT mice and CH12 clones, as well as AID and germline transcripts, cell proliferation and cell cycle profile of splenic B cells.
(a) Analysis of ex vivo CSR in splenic B cells from Fam72a+/+, Fam72a+/-, and Fam72a-/- mice (n=2 mice per group; duplicate assays per mouse). For IgG1 panel, p= 0.0014 for the comparison between Fam72a+/+ and Fam72a-/- group; for IgG2b panel, p= 0.0005 for the comparison between Fam72a+/+ and Fam72a-/- group; for IgG3 panel, p= 0.0143 for the comparison between Fam72a+/+ and Fam72a+/- group; for IgA panel, p= 0.0002 for the comparison between Fam72a+/+ and Fam72a-/- group, and p= 0.0074 for the comparison between Fam72a+/+ and Fam72a+/- group. (b) CH12 clones of indicated genotype were treated with CIT for 2 days and analyzed for CSR to IgA. Data are representative of 3 independent experiments. (c) Characterization of the humoral response in NP-CGG immunized WT and Fam72a-/- mice. Splenic germinal center (GC) B cell analysis in WT and Fam72a-/- mice after NP-CGG immunization for 8 days. Y-axis in the left panel shows the frequency of GC (GL-7+Fas+) among live CD45+CD4-CD8-CD11c-F4/80-CD19+IgD- splenic cells(n= 3-5 mice per group). For the left panel, p= 0.0059 for the comparison between naïve and d8-immunized Fam72a+/+ group, and p= 0.0308 for the comparison between d8-immunized Fam72a+/+ and Fam72a-/- group; for the right panel, p= 0.0176 for the comparison between naïve and d8-immunized Fam72a+/+ group. (d) Same as (c), except that follicular helper T (Tfh) cells were examined. Y-axis in the left panel shows the frequency of CXCR5+PD-1+ among live CD4+B220- splenic cells (n= 3-5 mice per group). For the left panel, p= 0.0123 for the comparison between naïve and d8-immunized Fam72a+/+ group; for the right panel, p= 0.0254 for the comparison between naïve and d8-immunized Fam72a+/+ group. (e) Representative immunofluorescence analysis of splenic GCs post NP immunization in WT and Fam72a-/- mice. Magnification 20x. Spleens were collected day 8 post immunization. Cryosection were stained with PNA-FITC and IgD-PE. Right panel: The average GC size for each mouse is reported, in which ~2-6 individual GCs were analyzed per mouse (n=4 mice per group). (f) qPCR analysis of AID mRNA, Iμ-Cμ and Iγ3-Cγ3 germline transcripts of d3-LPS stimulated splenic B cells, as well as Iα-Cα germline transcripts of splenic B cells stimulated to switch to IgA from Fam72a-/- or Fam72a+/+ littermate mice (n= 3-4 mice per group). Data are representative of 2 independent experiments. (g) Evaluation of cell proliferation of LPS-stimulated splenic B cells from Fam72a-/- or Fam72a+/+ littermate mice (n=4 mice per group), and data was tested by two-way ANOVA and representative of 2 independent experiments (left panel). Proliferation was also assessed by CFSE dilution (Middle panel), and apoptosis by Annexin V-staining (right panel; n=3-4 mice per group; p= 0.0341). (h) Gating strategy for cell cycle analysis. (i) The compiled cell cycle analysis of Fam72a-/- or Fam72a+/+ splenic B cells that were stimulated by LPS for 3 days (n= 4 mice per group). Data are representative of 2 independent experiments. Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test (ns: not significant).
(a) Gating strategy used to sort germinal center B cells from Peyer’s patches for JH4 region and Bcl6 gene sequencing. (b) The frequencies of germinal center B cells in Peyer’s patches from naïve Fam72a+/+, Fam72a-/- or Aicda-/- mice (n= 3 mice per group). Data are representative of 2 independent experiments. Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test (ns: not significant). (c) The compiled spectrum of unique top-strand (coding-strand) mutations shown as mutation frequencies (top panel) and percentage of total mutations (bottom panel). These two panels represented the pooled data from 4 Fam72a+/+ and 4 Fam72a-/- mice, respectively. Data shown in the upper panels showing mutation frequencies for each individual mutation type were analyzed using a Mann-Whitney test, and all types of mutations were significantly reduced in the Fam72a-/- mice, except for C to A mutations. (d) Analysis of SHM in the JH4 region of germinal center B cells for each Fam72a+/+ or Fam72a-/- mouse. (e) Size of insertion in the JH4 region is shown (n= 4 mice per group). In addition, there was also no statistical significance in the frequency of insertions between Fam72a+/+ and Fam72a-/- mice (data not shown). (f) Size of deletions in the JH4 region is shown (n= 4 mice per group). In addition, there was also no statistical significance in the frequency of deletions between Fam72a+/+ and Fam72a-/- mice (data not shown).
(a) Mutation characteristics at the 5′ Sμ region was analyzed in WT, Fam72a-/-, Ung-/-, and Ung-/- Fam72a-/- CH12 clones after 5 days of CIT treatment. Mutations at WRC and GYW motifs are considered AID hotspot mutation, where W=A/T, R=A/G, and Y=C/T. Mutation frequency was calculated from total mutations pooled from three independent experiment divided by total number of nucleotide sequenced. Data is summarized in Fig. 3a. (b) Distribution of mutations at the 5’μ switch region in CH12 cells of the indicated genotype after 5 days of CIT stimulation. (c) AID expression levels of WT, Fam72a-/-, Ung-/-, and Fam72a-/-Ung-/- CH12 cells as assessed by Western blot analysis. β-ACTIN was used as loading control. Experiment was repeated 3 times independently with similar results. (d) Sμ germline transcripts (μGLT) and Sα germline transcript (αGLT) of the indicated CH12 clones before and after stimulation with CIT for 2 days. Data are representative of 3 independent experiments. (e) Analysis of ex vivo CSR to IgG2b, IgG3, IgA and IgE using splenic B cells from WT, Fam72a-/-, Msh2-/-, or Fam72a-/-Msh2-/- littermate mice (n=2 mice per group; duplicate assays per mouse). Data are representative of 2 independent experiments. For IgG3 panel, p= 0.0004 for the comparison between WT and Fam72a-/- group, p= 0.011 for the comparison between Fam72a-/- and Msh2-/- group, and p= 0.005 for the comparison between Fam72a-/- and Fam72a-/-Msh2-/- group; for IgG2b panel, p= 0.0003 for the comparison between WT and Fam72a-/- group, p= 0.0024 for the comparison between Fam72a-/- and Msh2-/- group, and p= 0.0059 for the comparison between Fam72a-/- and Fam72a-/-Msh2-/- group; for IgE panel, p= 0.0184 for the comparison between WT and Fam72a-/- group, p= 0.0326 for the comparison between Fam72a-/- and Msh2-/- group, and p= 0.0005 for the comparison between Fam72a-/- and Fam72a-/-Msh2-/- group. (f) The compiled spectrum of unique top-strand (coding-strand) mutations shown as mutation frequencies per 10, 000bp. The data were pooled from 4 WT mice, 3 Fam72a-/-, 2 Msh2-/-, or 3 Fam72a-/-Msh2-/- littermate mice, respectively. (g) The frequency of A or T mutations in the JH4 region of WT, Fam72a-/-, Msh2-/-, or Fam72a-/-Msh2-/- littermate mice (n=2-4 mice per group). (h) Western blots measuring for UNG expression level in ex vivo LPS-activated spleen B cells from WT, Fam72a-/-, Msh2-/-, and Fam72a-/-Msh2-/- mice. Experiment was repeated 2 times independently with similar results.
(a) Fam72a-/- DT40 clones were generated using CRISPR/Cas9. Depletion of Fam72a was confirmed at the genomic DNA and mRNA level in 6 different DT40 clones. All the clones were mostly IgM- as assessed by flow cytometry and only 4 clones (renames K01-04) were picked for fluctuation analysis. Prmt5 was used as a control for mRNA extraction. (b) Growth curve analysis in WT and Fam72a-/- DT40 cells. The average doubling time (DT) of WT and Fam72a-/- clones in cultures for each individual clone is shown in the bottom panel and revealed reduced growth kinetics for all Fam72a-/- clones (c) Fluctuation analysis for gene conversion in WT and Fam72a-/- DT40 cells based on same number of cell divisions. Data are representative of 3 independent experiments. (d) Western blots for UNG were carried out on the indicated clones.
(a) Streptavidin affinity purification from 293T cell lysates expressing N-terminal 3xFLAG-miniTURBO tagged FAM72A upon proximity biotinylation. FLAG-miniTURBO-SV40-NLS (NLS) was used as control. Strepavidin bound proteins and whole-cell extracts (WCEs) were immunoblotted with the indicated antibodies. (b) Purified mFAM72A and UNG2 proteins were electrophoresed on an SDS-PAGE gel and stained with Coomassie Brilliant Blue dye. The 6XHis-mFAM72A (20 kDa), mUNG2-polyGly-FLAG (35kDa), and hUNG2 (35kDa) are seen next to the precision plus protein standard (Bio-Rad). (c) Murine UNG2 and FAM72A proteins were mixed in equal molar amounts and pulled-down using polyHis-tag or FLAG-tag on mFAM72A and mUNG2 proteins, respectively. The input and elutions were analyzed by Western blot using anti-His tag or anti-FLAG tag antibodies. Cross-species pull-down of mFAM72A with hUNG2 is also shown: the hUNG2 protein was detected using anti-UDG antibody. (d) Purified murine UNG2 and Fam72A proteins were pre-incubated at a 1:5 molar ratio, respectively, then reacted with either a single-stranded uracil substrate (ssU) or double-stranded uracil substrates with either U:G or U:A pairs. Reaction products were separated on a denaturing polyacrylamide gel. (e) Quantification of three independent experiments of the type shown in Extended Data Fig. 9d using 0:1, 1:1 or 5:1 molar ratios of mFAM72A to mUNG2. (f) Inhibition of hUNG2, but not E. coli UNG by mFAM72A. Purified E. coli UNG or hUNG2 proteins were pre-incubated with a five-fold molar excess of mFAM72A and then reacted with a double-stranded DNA substrate, with uracils at position 9 and 33 in a 55 bp DNA. Both the ends of the uracil-containing oligomer are labeled with 6-FAM (shown as stars). Consequently, two major excision products (U9 and U33) are observed. Uracil glycosylase inhibitor (Ugi) was used in one of the reactions. (g) The mRNAs for Aicda, Ung2, and Fam72a genes were quantified from WT (clone 1) and Fam72a-/- (clone 1) cell lines that had been stimulated with the CIT cocktail (+CIT) or untreated (-CIT). Mean values and standard deviations of three independent qRT-PCR reactions are shown for each treatment. The expression was normalized to the reference gene TBP. Ung transcript levels were also quantified in resting primary mouse B cells by qRT-PCR (bottom right panel; n=4 per group). Data were presented as mean ± SEM and were analyzed using two-tailed unpaired Student t test (ns: not significant). (h) Western blots for UNG, MSH2, and β-ACTIN (as control) in CIT-stimulated CH12 clones of the indicated genotype. UNG1 and UNG2 are indicated on the gel. Experiment was repeated 3 times independently with similar results. (i) Western blot analysis for UNG in WT CH12 cells expressing the empty vector or overexpressing FAM72A. Experiment was repeated 3 times independently with similar results. (j) WT CH12 clone#1 and Fam72-/- CH12 clone#1 were treated with CHX or CHX plus MG132 for the indicated time, and lysates were probed with anti-UNG or anti-β-actin antibodies. Data are representative of 3 independent experiments. (k) Lysates from G1-synchronized Fam72a-/- CH12 cells were diluted to determine the level of increase of UNG2 protein compared to undiluted WT CH12 cells, then probed for UNG protein by western blot. These blots suggest a 3-4 fold increase in UNG2 protein in Fam72a-/- CH12 cells compared to controls. Data are representative of 3 independent experiments.
FAM72A antagonizes UNG2 in the G1 phase of the cell cycle, leading to reduced processing of the AID-induced dU in Ig genes, which leads to increased G:U mismatches. The increased G:U lesions can either be replicated to produce transition mutations at G:C basepairs, or can engage the mismatch repair system to enhance mutagenesis, or provide DNA lesions, in collaboration with UNG2 that are required for CSR. In the absence of FAM72A, the protein expression and enzymatic activity of UNG2 are enhanced, resulting in increased excision of dU, which favors faithful repair by base excision repair pathway, as evidenced by diminished mutation frequencies in both μ switch region and JH4 region. Furthermore, the accelerated removal of dU in the context of FAM72A deficiency fails to recruit the mismatch repair system in the G1 or S phase, which is required for both CSR and SHM. One question that arises, is why doesn’t CSR increase with increased UNG2 protein/activity in Fam72a-/- B cells? The key is the engagement of the MMR system by FAM72A to induce CSR. First, deleting FAM72A in Msh2-/- CH12 cells and Msh2-/- spleen B cells moderately increases CSR (Fig. 3d, f) consistent with an increased UNG2 level and activity inducing larger number of breaks to facilitate CSR. Importantly, CSR is not increased to WT levels. For SHM, the increased UNG2 activity in Fam72a-/- B cells would cause an increase in ssDNA breaks in the V-region, but ultimately lead to faithful repair of the dUs, thereby erasing many AID-induced uracil lesions. Second, removal of either Ung or Msh2 leads to a defect in CSR that is not additive suggesting an interaction between these two proteins during CSR (Fig. 3c, d). For SHM, UNG and MMR pathways converge to induce error-prone repair. The exact mechanism likely involves the interruption of faithful MMR repair by uracil excision by UNG2. The interaction between these both pathways has not been formally shown in CSR, although there was evidence for this notion in the literature (e.g. see figure 3d in6). As to how MMR and UNG2 work together to induce CSR is not clear, and requires more work, but the mechanism might be similar to that hypothesized to what occurs during SHM. For CSR, we therefore hypothesize that FAM72A antagonizes UNG2, leading to the accumulation of U:G mispairs that can be engaged by the MMR system to induce CSR.
About this article
Cite this article
Feng, Y., Li, C., Stewart, J.A. et al. FAM72A antagonizes UNG2 to promote mutagenic repair during antibody maturation. Nature 600, 324–328 (2021). https://doi.org/10.1038/s41586-021-04144-4