Phenotypic Mutation 'bellezza' (pdf version)
Mutation Type missense
Coordinate122,561,185 bp (GRCm38)
Base Change A ⇒ G (forward strand)
Gene Aicda
Gene Name activation-induced cytidine deaminase
Synonym(s) Aid
Chromosomal Location 122,553,801-122,564,180 bp (+)
MGI Phenotype FUNCTION: [Summary is not available for the mouse gene. This summary is for the human ortholog.] This gene encodes a RNA-editing deaminase that is a member of the cytidine deaminase family. The protein is involved in somatic hypermutation, gene conversion, and class-switch recombination of immunoglobulin genes. Defects in this gene are the cause of autosomal recessive hyper-IgM immunodeficiency syndrome type 2 (HIGM2). [provided by RefSeq, Feb 2009]
PHENOTYPE: Homozygous mutation of this gene results in elevated IgM levels and impairment of B cell class switching. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_009645; MGI:1342279

Mapped Yes 
Amino Acid Change Asparagine changed to Aspartic acid
Institutional SourceBeutler Lab
Gene Model predicted gene model for protein(s): [ENSMUSP00000040524] [ENSMUSP00000125093]
SMART Domains Protein: ENSMUSP00000040524
Gene: ENSMUSG00000040627
AA Change: N101D

Pfam:APOBEC_N 11 178 4.6e-66 PFAM
Pfam:APOBEC_C 120 171 1.8e-23 PFAM
Predicted Effect probably benign

PolyPhen 2 Score 0.029 (Sensitivity: 0.95; Specificity: 0.82)
(Using ENSMUST00000043301)
SMART Domains Protein: ENSMUSP00000125093
Gene: ENSMUSG00000040627

Pfam:APOBEC_C 1 35 6.3e-17 PFAM
Predicted Effect probably benign
Phenotypic Category
Phenotypequestion? Literature verified References
LPS-induced Necroptosis - decreased
Macrophage necroptosis: low
NALP3 inflammasome signaling defect
Nlrc4 inflammasome: high response
NLRP3 inflammasome: high response
OVA-specific IgE - decreased
ratio of OVA-specific IgE over the total IgE - increased
T-dependent humoral response defect- decreased antibody response to OVA+ alum immunization
T-dependent humoral response defect- decreased antibody response to rSFV
TLR signaling defect: hypersensitivity to PAM3CSK4
TLR signaling defect: TNF production by macrophages
Total IgE After 2nd OVA/Alum Challenge (day 7) - decreased
total IgE level - decreased
Alleles Listed at MGI

All Mutations and Alleles(18) : Targeted(16) Transgenic(2)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL01775:Aicda APN 6 122561053 missense probably damaging 1.00
IGL03328:Aicda APN 6 122562437 missense probably benign 0.01
creeper UTSW 6 122561867 missense probably damaging 1.00
R1370:Aicda UTSW 6 122561185 missense probably benign
R2207:Aicda UTSW 6 122561285 missense possibly damaging 0.88
R4012:Aicda UTSW 6 122559490 missense probably benign 0.07
R4177:Aicda UTSW 6 122561084 missense probably benign 0.00
R4698:Aicda UTSW 6 122553888 start gained probably benign
R5000:Aicda UTSW 6 122561867 missense probably damaging 1.00
R5110:Aicda UTSW 6 122561185 missense probably benign 0.03
Mode of Inheritance Autosomal Recessive
Local Stock
Last Updated 2018-07-18 11:23 AM by Anne Murray
Record Created 2017-03-24 11:05 PM by Jin Huk Choi
Record Posted 2018-07-18
Phenotypic Description

Figure 1. Bellezza mice exhibited increased OVA-specific Ige to total IgE levels in response to ovalbumin/alum. IgE levels were determined by ELISA. Normalized data are shown. Abbreviations: WT, wild-type; REF, homozygous reference mice; HET, heterozygous variant mice; VAR, homozygous variant mice. Mean (μ) and standard deviation (σ) are indicated.

Figure 2. Bellezza mice exhibited reduced total IgE levels in the serum. IgE levels were determined by ELISA. Normalized data are shown. Abbreviations: WT, wild-type; REF, homozygous reference mice; HET, heterozygous variant mice; VAR, homozygous variant mice. Mean (μ) and standard deviation (σ) are indicated.

Figure 3. Bellezza mice exhibit diminished T-dependent IgG responses to ovalbumin administered with aluminum hydroxide. IgG levels were determined by ELISA. Normalized data are shown. Abbreviations: WT, wild-type; REF, homozygous reference mice; HET, heterozygous variant mice; VAR, homozygous variant mice. Mean (μ) and standard deviation (σ) are indicated.

Figure 4. Bellezza mice exhibit diminished T-dependent IgG responses to recombinant Semliki Forest virus (rSFV)-encoded β-galactosidase (rSFV-β-gal). IgG levels were determined by ELISA. Normalized data are shown. Abbreviations: WT, wild-type; REF, homozygous reference mice; HET, heterozygous variant mice; VAR, homozygous variant mice. Mean (μ) and standard deviation (σ) are indicated.

The bellezza phenotype was identified among N-ethyl-N-nitrosourea (ENU)-mutagenized G3 mice of the pedigree R5110, some of which showed an increase in the ratio of OVA-specific IgE to total IgE levels (Figure 1) due to decreased amounts of total IgE levels in the serum (Figure 2). The T-dependent antibody responses to ovalbumin administered with aluminum hydroxide (Figure 3) and to recombinant Semliki Forest virus (rSFV)-encoded β-galactosidase (rSFV-β-gal) were also diminished (Figure 4). 
Nature of Mutation

Figure 5. Linkage mapping of the reduced IgE secretion using an additive model of inheritance. Manhattan plot shows -log10 P values (Y-axis) plotted against the chromosome positions of 59 mutations (X-axis) identified in the G1 male of pedigree R5110. Log phenotype data are shown for single locus linkage analysis without consideration of G2 dam identity. Horizontal pink and red lines represent thresholds of P = 0.05, and the threshold for P = 0.05 after applying Bonferroni correction, respectively.

Whole exome HiSeq sequencing of the G1 grandsire identified 59 mutations. All of the above phenotypes were linked by continuous variable mapping to mutations in two genes on chromosome 6: Pparg and Aicda. The mutation in Aicda was presumed causative due to its known effects on immunology and the bellezza phenotypes mimic those found in creeper mice. The Aicda mutation is an A to G transition at base pair 122,561,185 (v38) on chromosome 6, or base pair 7,377 in the GenBank genomic region NC_000072 encoding Aicda. The strongest association was found with an additive model of inheritance to the total IgE phenotype, wherein four variant homozygous mice and 30 heterozygous mice departed phenotypically from 20 homozygous reference mice with a P value of 2.343 x 10-9 (Figure 5). A slight semidominant effect was observed in most of the assays, but the mutation is preponderantly recessive.


The mutation corresponds to residue 394 in the mRNA sequence NM_009645 within exon 3 of 5 total exons.



96  -E--F--L--R--W--N--P--N--L--S--L-


The mutated nucleotide is indicated in red. The mutation results in an asparagine to aspartic acid substitution at residue 101 (N101D) in the AID protein, and is strongly predicted by PolyPhen-2 to be benign (score = 0.029).

Protein Prediction
Figure 6. Domain organization of AID. The location of the bellezza mutation is indicated. Mutations found in AID are noted in red. Click on each mutation to view more information. Abbreviations: NLS, nuclear localization signal; NES, nuclear export signal.
Figure 7. Crystal structure of the MBP-AIDv(Δ15) fusion protein. AIDv(Δ15) is a soluble human AID variant. MBP is gray and AID is blue. Figure generated by UCSF Chimera and is based on PDB:5JJ4.

Activation-induced cytidine deaminase (AID) is a member of the polynucleotide deaminase families, which also includes the APOBEC enzymes. The AID/APOBEC proteins share a characteristic zinc coordination motif at the core of the catalytic site (1).


AID has a bipartite nuclear localization signal (amino acids 1 to 30), a catalytic domain (amino acids 56 to 94), an APOBEC-like domain (amino acids 112 to 184), CMP/dCMP-type deaminase domain (amino acids 23 to 129), and a nuclear export signal (amino acid 183 to 190) [Figure 6; reviewed in (2)]. Glu58 within the catalytic domain serves as a general acid-base catalyst. His56, Cys87, and Cys90 within the catalytic domain bind zinc and are required for catalytic activity. The APOPBEC-like domain binds DNA surrounding cytosines to be deaminated and influences substrate specificity. Amino acids 13 to 26 mediate DNA binding. Amino acids 113 to 123 constitute the hotspot recognition loop (corresponding to loop 7; see below), which dictates substrate specificity (3;4). Several regions of AID are required for protein-protein interactions (Table 1).


Table 1. Select AID-interacting proteins [adapted from (2)]


Interaction region on AID (if known)




Amino acids 2 to 26

Histone chaperone that differentially regulates AID functions in CSR and SHM through H3K4me3 on AID-target loci



Amino acids 39 to 42

Spliceosome-associated factor that promotes AID localization or association with RNA spliceosomes



Amino acids 88 to 116

E3 ligase



Amino acids 180 to 198

Recruits and/or stabilizes AID at target S regions



Amino acids 56 to 198

(see the record for clover)


HSP90, HSP40 DnaJa1, and eEF1A


Stabilizes cytoplasmic AID


HP1 and KAP1


Tethers AID to Sμ region bearing H3K9me3




Transports AID to B-cell nucleus and targets AID to actively transcribed Ig V(D)J regions




Stabilizes nuclear AID



The crystal structure of a soluble human AID variant, AIDv(Δ15), has been solved [Figure 7; PDB:5JJ4; (16)]. The structure of native AID was not solved as it aggregated. AIDv(Δ15) has 12 mutations and three amino acid deletions at the N-terminus in helix 1 and loop 1 and three surface mutations at the end of strand β1. AIDv(Δ15) also has a 15 residue C-terminal truncation. The alterations to AID reduced the charge of AID and enhanced crystallization, but did not alter the biochemical properties of AID. AIDv(Δ15) has a classic APOBEC fold, with six α-helices surrounding a central five-stranded β-sheet. His56, Cys87 and Cys90 are bound to a catalytic zinc ion, and Glu58 putatively functions as a proton shuttle during catalysis (17). The 3′OH of the ribose on Tyr114, Val57, Cys87, and Asn51 are proposed to form a hydrogen bond with Asn43 as well as bridge oxygen in the phosphodiester backbone. Leu113 and Cys116 cap a hydrophobic pocket consisting of residues from β-strand 5 and α-helices 4 and 6. AIDv(Δ15) has a positively charged channel on both sides of the active site, which putatively function as a ssDNA substrate binding surface (16).


AID undergoes several posttranslational modifications. AID is phosphorylated by PKA at Thr27 and Ser38 (6;18;19). Phosphorylation of Thr27 putatively affects AID enzymatic activity (19). Phosphorylation of Ser38 is essential for AID interaction with endonucleases that generate DNA breaks required for class-switch recombination (20). Phosphorylation of Ser38 increases AID function in class-switch recombination, somatic hypermutation, and gene conversion, but it does not alter the catalytic activity of AID on ssDNA (18;21). AID is also phosphorylated at Tyr184, but phosphorylation of Tyr184 does not alter the activity or function of AID (19). PKC-mediated phosphorylation of Ser3 and Thr140 putatively negatively regulates AID activity in B cells and limits off-target activity (22;23). AID is putatively monoubiquitinated on several residues by RNF126 (8).


The bellezza mutation results in an asparagine to aspartic acid substitution at residue 101 (N101D) in the AID protein. Amino acid 101 is not within a defined domain, but is within the region that interacts with RNF126.


AID is expressed in germinal center B cells, but not in naïve B cells, resting memory B cells, or plasma cells. AID is expressed in leukemia and B lymphomas of germinal center origin (24). AID expression is induced by T cell-dependent CD40L (see the record for walla)/CD40 (see the record for bluebonnet) engagement and T cell-independent TLR engagement. Cytokines (IL-4, TGF-β in humans; and IL-4, TGF-β and IFN-γ in mice) also stimulate Aicda expression. AID localizes to both the nucleus and the cytoplasm (25), and can shuttle between the nucleus and cytoplasm (26;27).

Figure 8. AID functions in class switch recombination, somatic hypermutation, and Ig gene conversion. A schematic diagram of the Ig heavy chain locus is shown. Somatic hypermutation causes point mutations in the vicinity of the V exon. Gene conversion involves the transfer of sequence information from a pseudogene (ψV) into the variable region exon. Class switch recombination involves looping out and deletion of DNA between two switch regions, thereby swapping the constant region of the expressed heavy chain.

In germinal centers, B cells proliferate, differentiate, and undergo somatic hypermutation (SHM) and class-switch recombination (CSR) during antibody responses. AID is a single-stranded (ss) DNA-specific cytidine deaminase that functions in CSR, SHM, and gene conversion of immunoglobulin genes in B cells (Figure 8) (28-30). In all processes, AID deaminates cytosines, converting them to uracils. The uracil conversion results in U:G mismatch DNA lesions that are converted into point mutations during SHM and into DNA double-stranded breaks (DSBs) during CSR or aberrant chromosomal translocations. AID exhibits preference for deaminating cytosine within WRC (W is A/T and R is A/G) motifs (31). As a result of this preference, there are SHM hot spots within IgV and S-regions (32).


CSR facilities the production of antibodies of different isotypes in mature B cells during a humoral immune response (33;34). CSR is a recombination reaction that occurs between paired DSBs in immunoglobulin heavy chain (Igh) switch regions (S-regions) that flank Igh constant regions (35). The S-regions contain a repetitive sequence that can serve as a substrate for proximal microhomology-mediated intra-switch repair by C-NHEJ (36;37). During CSR, AID converts cytosines into uracils at the S-region (38). The excision of uracils from both DNA strands results in staggered DNA breaks at donor and acceptor switch regions (38). The IgH locus lesions are detected as DSBs by the Mre11/Nbs1/Rad50 (MRN) complex, which leads to phosphorylation of H2AX, the recruitment of 53BP1 (see the record for lentil) to the IgH locus, and eventual end joining by C- or A-NHEJ (35;39;40).


SHM contributes to Ig diversity and antibody affinity maturation (41). RNA polymerase II exposes the single-stranded DNA template within Ig variable region (V) exons for AID, which subsequently deaminates cytosine to uracil. Replication over the uracil results in C to T or G to A transitions. Processing by uracil DNA glycosylase generates an abasic site that can be replicated over or repaired in an error-prone manner (possibly by Rev1 or other translesion synthesis polymerases) to give rise to transition and tranversion mutations (indicated as ‘N’) at C-G nucleotides. Recognition of the U-G mismatch by Msh2/Msh6 (see the record for medea) followed by the action of Exo1 and Polη spreads mutations to surrounding A-T nucleotides. Ung and Msh2/Msh6 can also act as parts of the normal base excision repair and mismatch repair pathways, respectively, resulting in high-fidelity repair of the uracil and no mutation [reviewed in (42)].


AID also functions in Ig gene conversion. Gene conversion involves the transfer of sequence information from a pseudogene (ψV) into the variable region exon. Gene conversion can be used in addition to or instead of SHM to diversify the IgV.


Mutations in human AICDA are linked to type 2 immunodeficiency with hyper-IgM (HIGM2; OMIM: #605258; (43)), which is characterized by normal or elevated serum IgM levels with absence of IgG, IgA, and IgE. HIGM2 patients exhibit increased susceptibility to bacterial infections. When acting off-target, AID can also generate non-Ig genomic mutations, which cause B-cell lymphoma or leukemia (44;45).

Putative Mechanism
Figure 9.   Overview of the T-dependent humoral immune response

Recruitment of antigen-primed T cells
During infection, initial contact with antigen activates peripheral dendritic cells (DCs) that acquire the antigen. Activated DCs produce inflammatory cytokines that promote the differentiation of monocytes to form additional DCs, which migrate to the T cell zones of secondary lymphoid organs such as the lymph nodes. There, they activate CD4+ T helper (Th) cells by presenting antigen in an MHC class II context. This results in selection and preferential expansion of antigen-specific Th cell clones. Activated Th cells are induced to transiently express CD40L, permitting them to engage CD40 expressed on DCs. CD40 signaling is essential for the maturation and survival of DCs, resulting in secretion of cytokines, and upregulation of costimulatory molecules. Once activated, Th cells migrate to the border between the B cell follicles and the T cell zone where they encounter antigen-activated B cells.

T cell interaction with antigen-primed B cells
Naïve B cells reside in the follicles of lymph nodes and move around the heterogeneous reticular network containing follicular dendritic cells (FDCs), CD169+ macrophages, and marginal reticular cells (not shown). While moving along the stromal processes of these cells, B cells encounter antigen presented on FDCs or CD169+ macrophages. FDCs obtain antigen-antibody complexes via their Fc or complement receptors and retain antigen on the cell surface in a native, unprocessed form for long periods of time (e.g. over a year). CD169+ macrophages present antigen within the context of MHC class II. B cells bind antigen through their membrane Ig, leading to antigen endocytosis, proteolytic processing, and presentation at the cell surface within class II MHC molecules.  Once activated, B cells migrate to the T/B border where they encounter activated, cognate Th cells expressing CD40L. The peptide-MHC II complex expressed on activated B cells binds to the antigen-specific T cell receptors (TCRs) of Th cells. This B cell/T cell interaction leads to rapid clonal expansion of B cells within the T or B zones. B cells located in T cell zones differentiate into short-lived plasma cells (plasmablasts) that secrete germline-encoded IgM and then IgG, but do not undergo somatic hypermutation. Activated B cells that expand within B cell zones seed secondary follicles and enter the germinal center (GC) reaction. CD40/CD40L interactions are essential for GC formation, progression, and maintenance.

Germinal center reaction
B cells that enter the GC reaction rapidly proliferate (as centroblasts) and undergo isotype switching and somatic hypermutation. B cells expressing variant BCR exit the cell cyle (as centrocytes) and move to the light zone where they interact with FDCs and antigen-specific Th cells. Centrocytes selected by their ability to interact with antigen held on FDC and/or Th cells either re-enter the GC cycle or further differentiate into long-lived plasma cells or memory B cells with high affinity B cell receptors.  B cell/FDC interactions through Fc/Ig and CD21/C3d send a strong survival signal to B cells. However, the majority of centrocytes fail to be selected and undergo apoptosis and clearance by tingible body macrophages. The GC reaction is sustained by signals, including IL-21 and CD40L, from T follicular helper cells (Tfh). Tfh cells also depend on CD40L and IL-21 for their generation and function.

The B cell response to T-dependent antigen involves induction of a variety of cell surface molecules within hours of activation. Cell cycle entry follows, with early proliferating B cells located in T cell zones differentiating into short-lived plasma cells (plasmablasts) that secrete germline-encoded IgM and then IgG, but do not undergo somatic hypermutation. Activated B cells that expand within B cell zones seed secondary follicles, rapidly proliferate, and interact with Th cells in the germinal center (GC) reaction where isotype switching and somatic hypermutation occur. Selected GC B cells further differentiate into long-lived plasma cells or memory B cells with high affinity B cell receptors of the switched isotypes. The phenotype observed in creeper indicates loss of AIDcreeper function.


Aicda deficient mice exhibit reduced class switch recombination to IgG1 and IgG3 as well as reduced somatic hypermutation frequency in Peyer’s patch B cells (46). Mice carrying a knock-in point mutation in Aicda had much less SHM but had normal amounts of immunoglobulin in both serum and intestinal secretions (47). In addition, the knock-in mice had absent class switching in B cells as well as defects in IgG1 and IgG3 CSR (48)

Primers PCR Primer

Sequencing Primer
bellezza_seq(F):5'- ATGCAGGTCACGTCACCAGTG -3'
bellezza_seq(R):5'- AGTTTCCCCGCTGACACTCAC -3'
Science Writers Anne Murray
Illustrators Diantha La Vine
AuthorsBeibei Fang, Jin Huk Choi, Bruce Beutler