Phenotypic Mutation 'snowcock' (pdf version)
Mutation Type missense
Coordinate101,630,603 bp (GRCm38)
Base Change T ⇒ G (forward strand)
Gene Rag2
Gene Name recombination activating gene 2
Synonym(s) Rag-2
Chromosomal Location 101,624,718-101,632,529 bp (+)
MGI Phenotype Homozygotes for targeted null mutations exhibit arrested development of T and B cell maturation at the CD4-8- thymocyte or B220+/CD43+pro-B cell stage due to inability to undergo V(D)J recombination.
Accession Number

NCBI RefSeq: NM_009020; MGI:97849

Mapped Yes 
Amino Acid Change Cysteine changed to Tryptophan
Institutional SourceBeutler Lab
Gene Model predicted gene model for protein(s): [ENSMUSP00000106858]
SMART Domains Protein: ENSMUSP00000106858
Gene: ENSMUSG00000032864
AA Change: C419W

Pfam:RAG2 51 389 6.7e-193 PFAM
Pfam:RAG2_PHD 414 491 1.1e-50 PFAM
Predicted Effect probably damaging

PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99)
(Using ENSMUST00000111227)
Phenotypic Category decrease in B cells, decrease in B2 cells, decrease in CD4+ T cells, increase in CD44 MFI in CD4, increase in CD44 MFI in CD8, T-dependent humoral response defect- decreased antibody response to OVA+ alum immunization, T-dependent humoral response defect- decreased antibody response to rSFV, T-independent B cell response defect- decreased TNP-specific IgM to TNP-Ficoll immunization
Alleles Listed at MGI

All alleles(14) : Targeted(14)


Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00647:Rag2 APN 2 101630617 missense probably benign 0.15
IGL01358:Rag2 APN 2 101630020 missense possibly damaging 0.93
IGL01774:Rag2 APN 2 101630047 missense probably damaging 1.00
IGL02267:Rag2 APN 2 101630031 missense probably damaging 1.00
IGL02507:Rag2 APN 2 101630710 missense probably damaging 0.99
IGL02615:Rag2 APN 2 101629568 nonsense probably null 0.00
IGL02690:Rag2 APN 2 101629494 missense probably benign 0.00
IGL03087:Rag2 APN 2 101630214 missense probably benign 0.00
IGL03261:Rag2 APN 2 101630263 missense probably damaging 1.00
R0266:Rag2 UTSW 2 101630603 missense probably damaging 1.00
R0284:Rag2 UTSW 2 101630119 missense probably damaging 0.98
R1250:Rag2 UTSW 2 101630439 missense probably damaging 0.96
R1520:Rag2 UTSW 2 101630131 missense probably damaging 0.99
R1641:Rag2 UTSW 2 101629615 missense probably benign 0.22
R2260:Rag2 UTSW 2 101630238 missense probably benign 0.00
R2571:Rag2 UTSW 2 101629967 missense probably damaging 0.99
R3441:Rag2 UTSW 2 101630300 missense probably damaging 0.99
R3752:Rag2 UTSW 2 101630776 missense probably damaging 0.99
R4894:Rag2 UTSW 2 101629677 missense probably damaging 1.00
R5197:Rag2 UTSW 2 101630740 missense probably damaging 1.00
R5236:Rag2 UTSW 2 101629660 missense probably damaging 1.00
X0027:Rag2 UTSW 2 101630373 missense probably damaging 1.00
Z31818:Rag2 UTSW 2 101630805 missense probably damaging 1.00
Mode of Inheritance Autosomal Recessive
Local Stock Live Mice
MMRRC Submission 037091-MU
Last Updated 05/17/2017 8:05 AM by Anne Murray
Record Created 09/27/2013 10:54 AM by Kuan-Wen Wang
Record Posted 01/03/2014
Phenotypic Description
Figure 1. The snowcock mice (red dots, R0266 pedigree) lack a T-dependent IgG response to OVA-Alum.

Figure 2. The snowcock mice (red dots, R0266 pedigree) lack a T-dependent IgG response to β-galactosidase (β-gal)​.

Figure 3. The snowcock mice (red dots, R0266 pedigree) lack a T-independent IgM response to TNP-Ficoll.
Figure 4. The snowcock mice have reduced numbers of CD4+ T cells (top) and B cells (bottom). Snowcock mice are denoted by the red dots in the R0266 pedigree. Values are denoted as percentage of total peripheral blood lymphocytes.

The snowcock mutation was induced by N-ethyl-N-nitrosourea (ENU)-mutagenesis and discovered in G3 mice screened for T-dependent (T-D) and T-independent (T-I) humoral responses.  The snowcock mice lack a T-dependent IgG response to OVA-Alum (Figure 1), a T-dependent IgG response to rSFV-encoded β-galactosidase (β-gal) (Figure 2), and a T-independent IgM response to TNP-Ficoll (Figure 3). Flow cytometric analysis determined that the snowcock mice have reduced numbers of CD4+ T cells (Figure 4, top) and B cells (Figure 4, bottom).

Nature of Mutation

Whole exome HiSeq sequencing of the G1 grandsire identified 77 mutations. Six G3 mice with the snowcock phenotype were genotyped at all 77 mutation sites and two mutations on chromosome 2 affecting Olfr1131 and Rag2 were homozygous in all six of the snowcock mice; ten unaffected mice were wild-type or heterozygous at both the Olfr1131 (LOD=8.254) and Rag2 (LOD=8.708) loci. Snowcock mice phenocopy mutant models of Rag2, supporting a causal relationship between the mutation in Rag2 and the snowcock phenotype. The Rag2 mutation is a T to G transversion at base pair 101,630,603 (v38) on chromsome 2, or base pair 5,856 in the GenBank genomic region NC_000068 encoding Rag2. The mutation corresponds to residue 1,468 in the NM_009020 mRNA sequence (equivalent to residue 1498 in the ENSMUST00000044031 cDNA sequence) in exon 3 of 3 total exons and residue 1407 in the ENSMUST00000111227 cDNA sequence in exon 2 of 2 total exons.


414  -G--Y--W--I--T--C--C--P--T--C--D-


Genomic numbering is shown corresponding to NC_000068. The mutated nucleotide is in red. The snowcock mutation results in a cysteine (C) to tryptophan (W) substitution at amino acid 419 of RAG2.

Protein Prediction
Figure 5. Domain structure of RAG2. The position of the core domain is shown as well as the acidic region and the plant homology domain (PHD). Phosphorylation of regulates the localization and  degradation of RAG2. The position of the snowcock mutation is indicated.
Figure 6. Representative Kelch-like repeat folded into a 6-bladed β-propeller. The crystalized peptide is from mouse Kelch-like ECH-associated protein 1 (KEAP1). The figure was modified from PDB:3WDZ and was generated using using Chimera software. The image is interactive; click to rotate.

The unusual structure of the recombination activating gene (RAG) locus is present in most vertebrate genomes. Within the locus, the genes encoding RAG1 and RAG2 lie immediately adjacent to each other (separated by only a few kb), are convergently transcribed, and have an exceptionally compact organization with the entire open reading frame of each gene contained in a single exon (exon 2 in Rag2) (1;2).


Rag2 encodes the 527 amino acid RAG2 protein that can be divided into two functional regions, an N-terminal “core” domain (amino acid 1-383) and a C-terminal “non-core” domain (amino acids 384-527) [Figure 5; (3); reviewed in (4)]. The RAG2 core domain is necessary and sufficient for variable (V), diversity (D), joining (J) (V(D)J) recombination in vivo as well as V(D)J cleavage at recombination signal sequences (RSS) in vitro (3;5-7). RSS sequences flank each V, D, and J encoding gene segment and consist of moderately well-conserved heptamer (CACAGTG) and nonamer (ACAAAAACC) sequences separated by 12 or 23 (±1) base pairs of nonconserved spacer DNA (designated a 12- or 23-RSS, respectively).  The RAG2 core domain is predicted to fold into a six-bladed β-propeller structure, with each blade of the propeller composed of a 50 amino acid Kelch motif [Figure 6; PDB:3WDZ; (8-10); reviewed in (11)]. Kelch motifs consist of four antiparallel β-strands and often function as docking platforms for proteins or small signaling molecules [(8); reviewed in (4)]. The sixth Kelch motif of the RAG2 β-propeller (amino acids 314-371) mediates the interaction of RAG2 with RAG1. Mutation of Trp317 (W317Y) within the sixth Kelch motif results in loss of RAG1-RAG2 complex formation and a subsequent detriment in RSS recognition and cleavage (8). The association of RAG2 with RAG1 is believed to induce a conformational change within RAG1 that switches the catalytic center of RAG1 into an active conformation [(8); reviewed in (4)].


The non-core domain of RAG2 is conserved throughout evolution and is dispensable for the catalytic activity of RAG2 [(3;12;13); reviewed in (11)]. However, mice that express only the core domain of RAG2 have impaired B cell development (i.e., reduction in the number of mature, pre-B, and immature B cells as well as an increase in the percentage of pro-B cells) and reduced T cell numbers, indicating that there is an essential physiological function for the non-core domain of RAG2 (12;14;15). Subsequent studies determined that the RAG2 non-core domain restricts RAG1/2-mediated transposition (16-18). Reports have been conflicting as to whether the non-core domain also functions in the suppression of hybrid joint formation, in which a signal end rejoins to a coding end during V(D)J recombination. Some studies indicate that full-length RAG2 supports more hybrid joint formation than the core RAG2 domain alone (16-17), while others demonstrate that full-length RAG2 suppresses hybrid joint formation significantly more than the core RAG2 domain alone (18-19). Reasons for the conflicting results were not indicated and the function of the non-core domain of RAG2 in hybrid joint formation has not been resolved. In addition, the non-core domain of RAG2 regulates the recombinatorial order during V(D)J recombination, inhibiting direct V to D rearrangements prior to D-J rearrangement. Non-core RAG2 also suppresses some forms of inter-chromosomal translocations between TCRβ and TCRδ D gene segments (13). In addition, non-core RAG2 helps to enforce the use of the proper RSSs during recombination (13)


Figure 7. Solution structure of the PHD finger motif of mouse RAG2. Nuclear magnetic resonance (NMR) spectroscopy was used to determine the structure of the RAG2 PHD finger motif in solution. The figure was modified from PDB:2JWO and (3) and was generated using Chimera software. The pink-colored α-helix denotes the L2 segment. The zinc atoms are represented as Zn1 and Zn2. The residues that bind the zinc atoms (Cys419, Cys423, Cys446, His452, His455, Cys458, Cys478, and His481) are labeled and colored yellow. The location of the snowcock mutation (Cys419) is denoted in red.  The image is interactive; click to rotate.

Within the non-core domain, amino acids 417-484 fold into a plant homeodomain (PHD)-type zinc finger that is important for VH-to-DJH rearrangement (3;12;14;20)]. PHD zinc fingers are a member of the treble class of zinc-binding domains that function in PtdInsP-binding, nucleosome interaction, chromatin modification, and E3 ubiquitin ligase activities in different proteins (10;21-25). Treble clef motifs include RING finger domains (often found in E3 ubiquitin ligase enzymes) and FYVE finger domains (found in proteins that bind PtdInsP) domains]. PHD zinc fingers fold into a two-strand anti-parallel β-sheet and a C-terminal α-helix (not present in all PHDs) that is stabilized by two zinc atoms. Using nuclear magnetic resonance (NMR) spectroscopy, Elkin et al. determined that Cys419, Cys423, Cys446, His452, His455, Cys458, Cys478, and His481 within the PHD zinc finger of RAG2 bind the two zinc ions in a characteristic interleaved topology shared by members of the treble class of zinc-binding domains [Figure 7; PDB: 2JWO; (3)]. RAG2 does not display E3 ubiquitin ligase activity, but does bind to PtdInsPs (preferentially to bis-phosphorylated PtdInsPs) (3). The PHD zinc finger of RAG2 plus amino acids 488-527 (i.e., the “basic patch”) were necessary and sufficient for PtdInsP binding; each domain alone did not bind PtdInsPs (3). Arg464 and His468 are proposed to be involved in PtdInsP recognition and Trp453 and Asn474 are predicted to influence the molecular surface area of the α-helix formed by the L2 segment (Figure 7) of the RAG2 zinc finger; mutations in Trp453 and Asn474 (e.g., W453R and N474S, respectively) alter the surface area of the α-helix and, subsequently, the interaction of the PHD domain with PtdInsPs (3). In addition to the association of the PHD zinc finger to PtdInsP, the PHD zinc finger of RAG2 recognizes trimethylated histone H3K4, a modification that occurs mostly within accessible chromatin; mutations that disrupt the RAG2-histone association impair V(D)J recombination in vivo [(26-28); reviewed in (4)].


RAG2 contains an “acidic region” (amino acids 374-414) that acts as a linker between the core domain and the PHD zinc finger (2;5;15;29). The acidic region is required for the interaction of RAG2 with histones as well as for complete recombination of the IgH locus in B cells (2;5;15). Full-length RAG2 binds directly to core histones H2A, H2B, H3, and H4 via amino acids 397-408; mutation of Tyr402, Asn403, Asp406, or Glu407 to alanine in full-length RAG2 diminished the histone interaction (15). West et al. propose that the binding of RAG2 to histones could stabilize RAG1/2 binding to the RSS; or that the RAG2 C-terminus could recognize specific histones that bear unique posttranslational modification patterns bringing the recombinase to, or stabilizing it, at very specific RSSs; and/or the interaction of RAG2 with histones could function in the stabilization or protection of DNA ends during the processing and joining phases of the recombination reaction (15).


Cyclin A/CDK2-mediated phosphorylation of RAG2 at amino acid Thr490 facilitates RAG2 degradation at the G1/S transition; degradation of RAG2 at G1/S prevents the formation of RAG1/2-initiated DNA breaks during replication and restricts V(D)J recombination to the G1 stage of the cell cycle (6;30-32). Upon Thr490 phosphorylation, RAG2 is translocated from the nucleus to the cytoplasm, where it is degraded by the proteasome (31,32). Treatment of RAG2-expressing human embryonic kidney cells (HEK-293) with the cyclin-dependent kinase inhibitor, p27Kip1, inhibited the activity of cyclin A/CDK2 and subsequently increased RAG2 stability (31). In addition, p27Kip1 induced the localization of RAG2 to the nucleus and its subsequent stabilization (32). Mizuta et al. propose that after the completion of V(D)J recombination and the cell cycle progresses to S phase, p27Kip1 is degraded, allowing cyclin A/CDK2 to become active (32). Cyclin A/CDK2 subsequently phosphorylates RAG2 at Thr490, removing the C-terminal regulatory domain function that inhibits the cytoplasmic localization of RAG2; RAG2 is subsequently translocated from the nucleus to the cytoplasm where it is ubiquitinated and degraded by the 26S proteasome (32). The mechanism by which Thr490 phosphorylation promotes the cytoplasmic localization of RAG2 as well as the identity of the ubiquitin ligase are unknown (32).


The snowcock mutation (C419W) is within the PHD domain of RAG2. Cys419 is directly involved in binding one of two zinc atoms associated with RAG2.


RAG2 is expressed specifically in developing B and T cells during V(D)J recombination (2). V(D)J recombination is restricted to the G0/G1 stage of the cell cycle, and the onset of V(D)J recombination correlates with the RAG2 protein expression level (i.e., RAG2 accumulates at G0/G1 and decreases rapidly at the G1/S transition); Rag2 mRNA levels remain constant throughout the entire cell cycle (32;33). RAG2 can shuttle between the nucleus and cytoplasm (see the “Protein Prediction” section) (32). Within the nucleus, RAG2 is distributed throughout, with the exception of the nucleolus (34)


The RAG1 and RAG2 proteins carry out the first enzymatic step of V(D)J recombination, the process by which the variable region of antigen receptor genes is assembled in developing B and T lymphocytes. The complex containing RAG1 and RAG2 recognizes, binds, and catalyzes two double-stranded DNA cleavages between the RSS heptamer and the flanking coding sequence.


Please see maladaptive for more information about the function of RAG2.


Mutations in RAG2 are linked to severe combined immunodeficiencies (SCID), disorders with varying degrees of defective cellular and humoral immune function. These include combined cellular and humoral immune defects with granulomas (CCHIDG; OMIM: #233650), Omenn syndrome (OMIM: #603554), and B cell-negative severe combined immunodeficiency (SCID; OMIM: #601457) (35-37). Null mutations in RAG1 or RAG2 underlie approximately half of the human T cell-negative, B cell-negative SCIDs (35). Affected patients begin to have problems with oral candidiasis, diarrhea, and failure to thrive in the first months of life, and are later identified after several more months of persistent infections by opportunistic organisms. CCHIDG is a less severe form of SCID that is accompanied by noninfectious granuloma formation; residual T and B cell function may exist, conferring some protection against infections. Hypomorphic mutations in RAG2 cause Omenn syndrome, an autosomal recessive SCID characterized by enlarged lymphoid tissue, severe erythroderma, hypereosinophilia, elevated serum IgE, few B cells, and oligoclonal expansion of T cells.

Putative Mechanism

Rag2-deficient mice are viable, but B and T cell development are blocked at an early progenitor stage due to an inability to initiate V(D)J rearrangement (39). The snowcock mutation affects Cys419, a zinc-binding residue in the PHD zinc finger domain of RAG2 (Figure 7; (3)]. The mutation may destabilize the PHD domain, the integrity of which is necessary for RAG1/2 recombinase activity in vivo (3;26-28). The snowcock mutation may also cause disinhibition of RAG1/2-mediated transposition (16-18) and hybrid joint formation (19), and dysregulate recombinatorial order during V(D)J recombination (13). 

Primers PCR Primer

Sequencing Primer
snowcock_seq(F):5'- TTCCTTGGCATACCAGGAGAC -3'

Snowcock genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide transition.

PCR Primers




Sequencing Primer

Snowcock_seq(F): 5’- TTCCTTGGCATACCAGGAGAC -3’


PCR program

1) 94°C             2:00

2) 94°C             0:30

3) 55°C             0:30

4) 72°C             1:00

5) repeat steps (2-4) 40X

6) 72°C             10:00

7) 4°C               ∞


The following sequence of 790 nucleotides (from Genbank genomic region NC_000068 for linear DNA sequence of Rag2) is amplified:


aggaggaatc tctgtctcca gtgcaatcct cactcaaaca aacaatgatg aatttgttat

tgtgggtggt tatcagctgg aaaatcagaa aaggatggtc tgcagccttg tctctctagg

ggacaacacg attgaaatca gtgagatgga gactcctgac tggacctcag atattaagca

tagcaaaata tggtttggaa gcaacatggg aaacgggact attttccttg gcataccagg

agacaataag caggctatgt cagaagcatt ctatttctat actttgagat gctctgaaga

ggatttgagt gaagatcaga aaattgtctc caacagtcag acatcaacag aagatcctgg

ggactccact ccctttgaag actcagagga attttgtttc agtgctgaag caaccagttt

tgatggtgac gatgaatttg acacctacaa tgaagatgat gaagatgacg agtctgtaac

cggctactgg ataacatgtt gccctacttg tgatgttgac atcaatacct gggttccgtt

ctattcaacg gagctcaata aacccgccat gatctattgt tctcatgggg atgggcactg

ggtacatgcc cagtgcatgg atttggaaga acgcacactc atccacttgt cagaaggaag

caacaagtat tattgcaatg aacatgtaca gatagcaaga gcattgcaaa ctcccaaaag

aaaccccccc ttacaaaaac ctccaatgaa atccctccac aaaaaaggct ctgggaaagt



Primer binding sites are underlined and the sequencing primer is highlighted; the mutated nucleotide is shown in red text.

Science Writers Anne Murray
Illustrators Peter Jurek
AuthorsKuan-Wen Wang, Jin Huk Choi, Ming Zeng, Bruce Beutler