Phenotypic Mutation 'Unnatural' (pdf version)
AlleleUnnatural
Mutation Type missense
Chromosome6
Coordinate128,761,174 bp (GRCm39)
Base Change T ⇒ C (forward strand)
Gene Klrb1c
Gene Name killer cell lectin-like receptor subfamily B member 1C
Synonym(s) Ly55c, Nk1.1, Ly59, Nk-1, Nkrp1-c, Nk1, CD161, Nk-1.2, Ly-59, NKR-P1, NK-RP1, NK-1.1, NKR-P1C
Chromosomal Location 128,755,448-128,765,514 bp (-) (GRCm39)
MGI Phenotype PHENOTYPE: This locus controls an antigen on natural killer cells. The a allele determines the Nk1.1 antigen in strains CE, C57BL/6, C57BR/cd, C57L, C58, DBA/1, MA/My, NZB, SJL, SM and B10.D2. The b allele determines the Nk1.2 antigen in strains CBA/J, BALB/c, C3H/He, A/J, DBA/2, LP and 129. [provided by MGI curators]
Accession Number

NCBI RefSeq:  NM_008527; MGI: 107538

MappedYes 
Amino Acid Change Aspartic acid changed to Glycine
Institutional SourceBeutler Lab
Gene Model not available
AlphaFold no structure available at present
SMART Domains Protein: ENSMUSP00000127297
Gene: ENSMUSG00000030325
AA Change: D148G

DomainStartEndE-ValueType
transmembrane domain 87 109 N/A INTRINSIC
CLECT 139 256 1.65e-25 SMART
Predicted Effect probably benign

PolyPhen 2 Score 0.085 (Sensitivity: 0.93; Specificity: 0.85)
(Using ENSMUST00000167691)
SMART Domains Protein: ENSMUSP00000134184
Gene: ENSMUSG00000030325
AA Change: D98G

DomainStartEndE-ValueType
transmembrane domain 41 63 N/A INTRINSIC
CLECT 90 207 1.65e-25 SMART
Predicted Effect
SMART Domains Protein: ENSMUSP00000134504
Gene: ENSMUSG00000030325
AA Change: D151G

DomainStartEndE-ValueType
transmembrane domain 87 109 N/A INTRINSIC
CLECT 142 259 1.65e-25 SMART
Predicted Effect probably benign

PolyPhen 2 Score 0.085 (Sensitivity: 0.93; Specificity: 0.85)
(Using ENSMUST00000174404)
SMART Domains Protein: ENSMUSP00000134055
Gene: ENSMUSG00000030325

DomainStartEndE-ValueType
transmembrane domain 76 98 N/A INTRINSIC
Predicted Effect probably benign
Meta Mutation Damage Score Not available question?
Is this an essential gene? Probably nonessential (E-score: 0.052) question?
Phenotypic Category Autosomal Semidominant
Candidate Explorer Status loading ...
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI

All alleles(3) : Targeted, knock-out(1) Other(2)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL02636:Klrb1c APN 6 128765515 missense probably benign 0.01
Eccentric UTSW 6 128761702 missense possibly damaging 0.88
Freakish UTSW 6 128761148 missense probably benign 0.38
wacky UTSW 6 128757306 missense probably damaging 1.00
Weird UTSW 6 128761220 missense probably benign 0.00
Wild UTSW 6 128762968 missense probably benign 0.09
R0463:Klrb1c UTSW 6 128757366 missense probably benign 0.07
R3157:Klrb1c UTSW 6 128761702 missense possibly damaging 0.88
R3779:Klrb1c UTSW 6 128757306 missense probably damaging 1.00
R5111:Klrb1c UTSW 6 128762968 missense probably benign 0.09
R5149:Klrb1c UTSW 6 128760670 missense probably benign 0.07
R5196:Klrb1c UTSW 6 128757262 missense probably benign 0.00
R5568:Klrb1c UTSW 6 128765877 intron probably benign
R5620:Klrb1c UTSW 6 128761706 missense possibly damaging 0.67
R6000:Klrb1c UTSW 6 128761120 missense probably damaging 1.00
R6483:Klrb1c UTSW 6 128761148 missense probably benign 0.38
R6854:Klrb1c UTSW 6 128765381 missense possibly damaging 0.87
R7283:Klrb1c UTSW 6 128761220 missense probably benign 0.00
R7697:Klrb1c UTSW 6 128757273 missense probably benign 0.02
R7946:Klrb1c UTSW 6 128766072 intron probably benign
R8789:Klrb1c UTSW 6 128761148 missense probably benign 0.38
R9665:Klrb1c UTSW 6 128760625 critical splice donor site probably null
Z1177:Klrb1c UTSW 6 128765410 missense probably damaging 1.00
Mode of Inheritance Autosomal Semidominant
Local Stock Sperm, gDNA
MMRRC Submission 032644-UCD
Last Updated 2016-05-13 3:09 PM by Stephen Lyon
Record Created unknown
Record Posted 2009-08-03
Phenotypic Description

Figure 1. G3 mice screened for the percentage of cells in the blood expressing the indicated markers. A9401, the index Unnatural mutant, failed to express NK1.1 (red circle).


Figure 2. Pedigree of A9401, the index of Unnatural mutant.

The Unnatural phenotype was discovered in a screen of ENU-mutagenized G3 mice (C57BL/6J) for alterations in hematopoietic cell surface markers. The index mouse was found to lack expression of the cell surface protein NK1.1 (also known as NKR-P1C), which is expressed on natural killer (NK) cells, and a subset of T cells (NKT cells) (Figures 1 and 2).  A 50% reduction in the NK1.1 mean fluorescence is observed in Unnatural heterozygotes, indicating that the mutation is semidominant.

Nature of Mutation
Due to the lack of expression of NK1.1, the NK1.1 gene Klrb1c on Chromosome 6 was directly sequenced, and an A to G transition at position 398 of the Klrb1c transcript in exon 4 of 6 total exons was identified.
 
382 TGGCTTTTACACCGAGATAAGTGCTTTCACGTT
95  -W--L--L--H--R--D--K--C--F--H--V-
 
The mutated nucleotide is indicated in red lettering, and causes an aspartic acid to glycine change at residue 100 of the NKR-P1C protein.
Illustration of Mutations in
Gene & Protein
Protein Prediction

Figure 3. The NKC gene complex on mouse Chromosome 6. The Nkrp1 (Klrb1) and Clr genes are color coded to indicate the function of their gene products. Chromosomal gene orientation is not shown. Proteins encoded by Nkrp1b and Clr-b, as well as Nkrp1f and Clr-g, have been demonstrated to interact (see Background). Figure not to scale. Adapted from Carlyle et al. Semin Immunol. 20, 321-330 (2008).

The 39 kDa, 220-amino acid NKR-P1C (natural killer receptor protein 1C) belongs to a family of receptors that are mainly expressed on natural killer (NK) cells (1-3). The genes encoding these receptors are genetically linked on Chromosome 6 in a complex known as the NK gene complex (NKC) (Figure 3) (4). In addition to the Nkrp1 family, this complex includes the Ly49 genes and the C-type lectin-related (Clr) genes (4-7). Members of the latter group have now been shown to encode transmembrane proteins that bind to NKR-P1 receptors (6;7), although the ligand for NKR-P1C has not been identified. Many of the murine NK cell receptors are divergent from NK cell receptors present in humans, but the NKR-P1 family is conserved in rodents and humans (8). Humans possess only a single family member, designated NKR-P1A (also known as KLRB1, killer cell lectin-like receptor subfamily B, member 1; and CD161), which is 48% identical to mouse NKR-P1C. However, while mouse NKR-P1C is an activating receptor, human NKR-P1A is an inhibitory receptor (9).
 
Figure 4. Domain structure of NKR-P1C. The location of the Unnatural mutation is indicated with a red asterisk. LBM, Lck-binding motif; TM, transmembrane domain; CTLD, C-type lectin domain.
Like other NKR-P1/Ly49 receptors, NKR-P1C is classified as a type II membrane C-type lectin [reviewed in (10)]. C-type lectins are calcium-dependent lectins that are carbohydrate-binding proteins of animal origin.  Carbohydrate-binding activity of C-type lectins is based on the function of the carbohydrate recognition domain (CRD) whose structure is highly conserved among this family (11). Members of the C-type lectin-like family of NK receptors (Ly49A-W, NKR-P1A-E, CD94/NKG2A, and CD69) are transmembrane proteins with a single transmembrane domain (amino acids 39-66 for NKR-P1C), extracellular carboxyl termini and cytoplasmic amino termini (Figure 4). The CRD or C-type lectin-like domain (CTLD; amino acids 98-208 for NKR-P1C) is located at the extracellular carboxyl terminus (10). Unlike other C-type lectins, the NK receptors are not carbohydrate-binding proteins and do not appear to be calcium-dependent (12). They bind to other proteins and are involved in signal transduction based on cell-cell recognition (5;10)
 

Figure 5. Crystal structure of human CD69 in noncovalent dimeric form. The crystallized protein (beginning with Gly70) does not include the interchain dimerization cysteine (Cys68), but the overall fold and dimer arrangement are identical to those described previously (see Reference 15). The structure has a classical C-type lectin-like fold composed of two α-helices and three β-sheets. 31% identity exists between mouse NKR-P1C and human CD69 (within aa 91-209 for NKR-P1C and aa 85-196 for CD69). The residue corresponding to Asp100 mutated in Unnatural (Arg94 in CD69), located adjacent to strand β0, is indicated in red. The residue corresponding to Ser191 (Phe175 in CD69), located within strand β3, shown in yellow. UCSF Chimera model is based on PDB 3CCK, Vanek et al., FEBS J. 275, 5589-5606 (2008). Click on the 3D structure to view it rotate.

The three-dimensional structure of the CTLDs for several of the C-type lectin transmembrane NK cell receptor proteins have been elucidated, with or without binding to ligand, including Ly49 family members and CD69 (11-17). However, the structure of the NKR-P1 family has not been studied, and the exact three-dimensional characteristics of these proteins remain unknown. Despite the diversity in ligands that bind to NK cell receptors, the characterized extracellular domains do display a highly conserved core structure, consisting of two antiparallel β-sheets formed by β-strands β0, β1 and β5, and β2, β2’, β3 and β4, and, in most cases, two α-helices (α1 and α2) arranged in a conserved folding pattern (Figure 5, PDB ID 3CCK) [reviewed in (10)]. β-strands 1-5 and α1, in particular, are conserved in structure, while the orientation of α2 is variable.   Considerable variation is observed in the sequence and conformation of the loops connecting the β-strands, with loops 1 and 2 being more similar in structure amongst different proteins, and large differences observed in loops 3 and 5. Loop 3 comprises the ligand-binding region of these proteins (12;17), and the variation in conformation allows ligand-binding specificity amongst the NK cell receptors (10). The CTLDs from these proteins also contain three conserved disulfide bonds between α1 and β5, β3 and the loop following β4, and β0 and β1. NK cell receptors, including NKR-P1 receptors, exist at the cell-surface in homo- or heterodimeric form, which is stabilized by an interchain disulfide bond (18). In characterized NK cell receptors, the subunits associate through the β0 strand and the C-terminal ends of α2 (when present) to form a hydrophobic interface. The β0 strands are linked by two hydrogen bonds and contain a conserved W-X-X-Y/H motif (10).
 
The NK1.1 antibody recognizes distinct NKR-P1 receptors in various mouse strains. For instance, the NK1.1 antibody binds specifically to NKR-P1C in C57BL/6 mice, and to NKR-P1B in SW/J mice, while not binding to any NKR-P1 receptors in BALB/c animals (1;3;19). This specificity appears to be partially due to the presence of a serine instead of a threonine at residue 191 in β3 of the CTLD. Residue 191 is not sufficient to generate the NK1.1 epitope, as rat NKR-P1A and NKR-P1B sequences contain a serine at the correct position, but are not recognized by NK1.1. This suggests that the NK1.1 epitope relies on additional context-dependent amino acid residues, and that denaturation of the three-dimensional structure of the epitope destroys its reactivity (20).
 
NK1.1 (NKR-P1C)-mediated activation occurs through association with the FcRγ adapter molecule (originally called FcεRIγ) (21), an integral membrane protein that associates with and transduces signals from the ligand-binding α chains of FcγRI, FcγRIII and FcεRI (22-25). The FcRs, receptors for the Fc domain of immunoglobulins (Ig), function in antibody transport across epithelial tissues, antibody-dependent cellular cytotoxicity (see Background), phagocytosis, and other processes (26). The FcRγ chain is homolgous to the CD3ζ chain that is part of the TCR complex. The association between NKR-P1C and FcRγ involves charged transmembrane residues present in both NKR-P1C and FcRγ (R63 for NKR-P1C) (19).   NKR-P1 receptors also contain a conserved motif C-X-C-P-R/H in their cytoplasmic domains that binds to the lymphocyte protein tyrosine kinase (Lck) and is necessary for signaling (27;28) (see the record for iconoclast).   
 
The Unnatural mutation substitutes an aspartic acid to glycine at residue 100 of NKR-P1C. This residue is located near N-terminal end of the extracellular CTLD domain in the loop between β0 and β1. The effect of the mutation on expression and localization of the NKR-P1C protein is not known.
Expression/Localization
NKR-P1 family members are mainly expressed on the surface of NK cells, although a subset of mature T cells also express the NK1.1 epitope (NKT cells) (29). In C57BL/6J, the NK1.1 antigen (therefore NKR-P1C) is expressed on virtually all developing and mature NK cells (2;3;30). During development, a transient population of NK1.1+ cells in fetal thymus is observed on day 14, but not on day 16 of gestation.  On day 16 of gestation, NK1.1+ cells are detected only in liver and spleen (30). Later studies determined that NKR-P1C is first expressed in the fetal thymus in progenitors committed to T and NK cell fates (31). Interestingly, the Nkrp1b gene (and protein) is expressed in these progenitors prior to entry into the thymus (19).
 
Transcription of the mouse Nkrp1c gene is controlled by three functional promoters that regulate expression of Nkrp1c transcripts at various stages. Promoter 1 is located immediately upstream of the first coding exon and is active at all stage of NK cell development (32). Promoter 2 is located immediately upstream of promoter 1 and is active only in mature NK cells, while promoter 3 is located 9 kb upstream of promoter 1 and is active only in fetal NK cells. Promoter 3 contains a site that exhibits enhancer-like activity. Neither of the Nkrp1c transcripts initiated by promoters 2 and 3 represents coding exons. The promoter/enhancer regions of Nkrp1c and Nkrp1b/d are conserved (33).
Background
NK cells are large granular lymphocytes capable of recognizing and killing certain tumor and virally infected cells, but not normal cells. NK cells belong to the innate arm of the immune system, functioning as one of the first lines of defense against microbial infection and tumor cells. Mature NK cells exist in a ready state, able to launch an immediate response. Once activated, NK cells perform several effector functions: (1) Spontaneous killing. NK cells kill via exocytosis of perforin- and granzyme-containing cytoplasmic granules to lyse target cells, similar to CD8+ cytotoxic T lymphocytes (CTL) (34). NK cells also kill through activation of FasL- and TRAIL-mediated pathways (35;36). (2) Cytokine production. Activated NK cells produce cytokines including TNF-α and IFN-γ, which mediate antiviral and cytotoxic effects, and prime the activity of other immune cells (37). The production of cytokines upon activation occurs rapidly, enabled in part by NK cell maintenance of stored cytokine transcripts that are immediately available for initiation of cytokine synthesis (38). (3) Chemokine production. NK cells help to attract other hematopoietic cells by secreting chemokines such as macrophage inflammatory protein (MIP)-1α, MIP-1β, and the factor regulated on activation-normal T cell expressed and secreted (RANTES) (37;39)
 
Figure 6. (A) NK cell inhibitory receptors include the Ly49 receptor family, the human killer cell immunoglobulin-like receptor (KIR) L family, and the CD94/NKG2A complex, and signaling is mediated through an immunoreceptor tyrosine-based inhibitory motif (ITIM) in their cytoplasmic domains. The ITIM is phosphorylated by Src family kinases and recruits phophatases such as SHP1, SHP2, and SHIP-1, leading to the inhibition of NK cell activation. Activating receptors include a subset of mouse Ly49 receptors, human KIR S family proteins, the CD94/NKG2C complex, and the NKG2D receptor, which interact with immunoreceptor tyrosine-based activation motif (ITAM)-containing molecules. Syk and ZAP70 are recruited to phosphorylated ITAMs via their SH2 domains, leading to NK cell activation and cytolytic function. (B) Potential NK cell-target cell interactions that induce NK activation or produce no response.
NK cells recognize abnormal cells as their targets, and distinguish them from normal healthy cells using several classes of constitutively expressed cell surface receptors, which regulate NK activation, proliferation and effector functions [reviewed in (40)] (Figure 6). NK cell receptor genes do not require somatic recombination, and NK cells develop and function normally in mice and humans that are unable to rearrange their antigen receptor genes (41;42). The observation that NK cells could lyse MHC-negative targets, and that NK cells appeared to be actively inhibited from responding to tumor cells expressing MHC class I, led to the formation of the “missing-self hypothesis” (43;44). It proposed that NK cells survey for targets with downregulated MHC class I, which frequently occurs on tumor or virally infected cells, and predicted the existence of negative regulatory receptors which interact with MHC molecules and prevent destruction of normal host cells. Both inhibitory and activating NK cell receptors have since been identified. The inhibitory MHC class I-specific receptors include members of the mouse Ly49 receptor family, the human killer cell immunoglobulin-like receptor (KIR) L family, and the CD94/NKG2A complex found in mice and humans. NK cell activating receptors include a subset of mouse Ly49 receptors, human KIR S family proteins, the CD94/NKG2C complex, and the NKG2D receptor, and recognize MHC class I-related and non-MHC ligands.
 
NK cell function is controlled by the integration of signals from numerous activating and inhibitory receptors (40). Stimulation of activating receptors will lead to killing only if that signaling outbalances signals from inhibitory receptors (which generally predominate). Thus, when multiple receptors or a sufficiently potent activating NK receptor is stimulated, NK cells are capable of eliminating targets even if their inhibitory receptors for MHC class I are ligated (45;46). Signaling from inhibitory receptors is mediated through an immunoreceptor tyrosine-based inhibitory motif (ITIM) in their cytoplasmic domains, which is phosphorylated by Src-family kinases and recruits phosphatases such as SHP1 (mutated in spin), SHP2, and SHIP-1 (mutated in styx), that presumably oppose phosphorylation pathways that activate NK cells. In contrast, NK cell activating receptors lack the intracellular ITIM, typically containing instead positively charged transmembrane residues that mediate interactions with immunoreceptor tyrosine-based activation motif (ITAM)-containing molecules such as DNAX-activating protein (DAP)10, DAP12, CD3ζ, and FcRγ adapter proteins. Upon tyrosine phosphorylation of the ITAM, the tyrosine kinases Syk and ZAP70 are recruited via their SH2 domains and initiate downstream signaling, ultimately leading to calcium influx, degranulation, and synthesis of cytokines and chemokines.
 
NKR-P1C lacks an ITIM and contains a charged transmembrane residue, consistent with its function as an activating receptor. In mice, antibodies to NKR-P1C (or against NKR-P1A in rats) mediate redirected lysis by NK cells against Fc receptor-bearing target cells (47;48).  Redirected lysis by NK cells is the phenomenon where an antibody against an activating NK receptor can induce lysis of an irrelevant target cell; redirected lysis requires the presence of FcR on the target cell which bind to the Fc region of the antibody, linking the NK and target cells. Thus, NKR-P1C can activate the cytolytic process. Antibody-dependent crosslinking of the rat NKR-P1A receptor has been shown to result in phosphatidylinositol (PI) turnover and calcium influx (49), while crosslinking NKR-P1C with the NK1.1 antibody in mice results in IFN-γ production by NK cells (50).
 
As mentioned above, the ability of NKR-P1C to activate NK cells upon receptor engagement depends on its association with FcRγ, a component of the NK cell-specific FcγRIII receptor (51). FcRγ protects the α chain of the receptor complex from degradation in the endoplasmic reticulum (26). NK cells from mice lacking FcRγ fail to produce IFN-γ or kill target cells in redirected lysis or antibody-dependent cellular cytotoxicity (ADCC, essentially the reverse of redirected lysis) assays (21;52). NKR-P1C ligation is thought to result in phosphorylation of the tyrosines in the ITAM of FcRγ, leading to recruitment tyrosine kinases that mediate downstream signal transduction (53;54). In particular, the Syk tyrosine kinase is activated downstream of the FcγRIII in NK cells activated by sensitive target cells, and dominant-negative kinase-inactive forms of Syk inhibit FcR-initiated killing (55-57). Lck tyrosine kinase has been shown to associate directly with NKR-P1C through a CxCP motif in the NKR-P1C cytoplasmic domain, and to be required for calcium mobilization and killing downstream of NKP-P1C engagement (27;28). Interestingly, NKR-P1B, an inhibitory NK receptor found in the SJL/J and SWR mouse strains which signals through a cytoplasmic ITIM to SHP1 (1;3;19;58), also binds in the same manner as NKR-P1C to Lck (28). The association of NKR-P1B with SHP1 requires an intact Lck binding site. It has been proposed that Lck phosphorylation of the ITIM in NKR-P1B creates a docking site for the SH2 domain of SHP1, leading to dephosphorylation and inhibition of proximal kinases. In contrast, Lck association with NKR-P1C may lead to phosphorylation of tyrosines in the ITAM of the FcRγ protein, resulting in recruitment of Syk kinase and further NK activating signaling. However, it has been shown in conflicting reports that NK cells from Lck-deficient mice possess either normal or impaired killing activity in redirected lysis or ADCC assays (28;59). The reason for the discrepancy is unknown.
 
Initial studies suggested that NKR-P1 receptors bind to lectin ligands, but further studies proved this not to be the case. So far, only ligands for the inhibitory NKR-P1B or D receptors, and the activating NKR-P1F receptor have been identified. These ligands, members of the osteoclast inhibitory lectin (Ocil) or C-type lectin-related (Clr) family, also have a C-type lectin-like structure, and their genes are interspersed among the genes encoding NKR-P1 receptors within the NKC (60;61). Ocil/Clr-b is considered a “self” ligand and is recognized by NKR-P1B/D, and Ocilrp2/Clr-g is recognized by NKR-P1F (6;7). Ocil/Clr-b is reportedly broadly expressed on hematopoietic cell types, but is lost from some tumor cell lines which are sensitive to NK cell killing (6;7). Thus, Ocil/Clr-b may function in an MHC-independent “missing self” NK recognition system. The interaction between Ocilrp2/Clr-g and NKR-P1F has not been fully characterized, but has been reported to costimulate T cell proliferation in response to antigenic stimulation (62;63).
 
Allelic polymorphisms, resulting from the presence or absence of a gene, point mutations that alter receptor specificity or signaling, deletions, or insertions, exist for many of the genes within the NKC. Several such polymorphisms result in dramatic differences in NK cell recognition and function. For example, resistance to mouse cytomegalovirus (MCMV) is conferred to C57BL/6 mice by the presence of the Ly49h gene, which is lacking in the susceptible BALB/c strain (64;65). NKR-P1C also shows polymorphism between the C57BL/6 and BALB/c strains, as revealed by positivity for the α-NK1.1 epitope in the C57BL/6, but not BALB/c strain. The presence of a serine instead of a threonine at residue 191 in C57BL/6 mice largely accounts for this difference (see Protein Prediction), but the functional consequences are unknown (20). It has been postulated that the Nkrp1b/c region may have undergone recent directed evolution under pressure from rat cytomegalovirus (RCMV) immune evasion of the NKR-P1B-Ocil/Clr-b interaction (66;67).
Putative Mechanism
Lack of expression of NK1.1 in mice, suggests that the Unnatural mutation results in lack of NKR-P1C protein. However, the loss of the NK1.1 epitope does not necessarily mean that NKR-P1C is not present. As described above, binding of the NK1.1 antibody depends on residue 191 and also on the conformation of the CTLD in NKR-P1 receptors. Due to their small size, glycines are often present at positions that require tight folding and packing of the protein, and the substitution of a glycine for an aspartic residue at amino acid 100 may allow the NKR-P1C CTLD domain to fold differently and disrupt NK1.1 binding to NKR-P1C without altering expression levels or localization of the protein. The S191 amino acid that appears to be so important in conferring NK1.1 recognition is located in β3 of the CTLD. Loop 3, which in other NK receptors confers ligand-binding specificity (12;17), connects β2’ to β3. Thus, altering the conformation of the CTLD so it does not recognize the NK1.1 epitope may also change the ligand-binding properties of the NKR-P1C receptor. 
 
Although altering residue 100 may disrupt binding to the NK1.1 epitope, this residue is located some distance away from loop 3 and β3 (Figure 4) and is one of the flanking, charged residues of the conserved β0 motif important in forming the receptor dimer interface (10). In CD69, these residues form salt bridges across the dimer interface that probably aid in dimer stabilization (15). It is likely that altering the D100 residue in NKR-P1C will disrupt the formation of the homodimeric receptor at the cell surface.
Primers Primers cannot be located by automatic search.
Genotyping
Unnatural 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.  This protocol has not been tested.
 
Primers
Unnatural (F): 5’- TACCAGGGATTACGGTCAGCCAAG -3’
Unnatural (R): 5’-GATAACAAGCTTCCATCGGGAGGTC -3’
 
PCR program
1) 95°C             2:00
2) 95°C             0:30
3) 56°C             0:30
4) 72°C             1:00
5) repeat steps (2-4) 29X
6) 72°C             7:00
7) 4°C              ∞
 
Primers for sequencing
Unnatural_seq(F): 5’- TGAGATCGCAATCACTTGGC -3’
 
The following sequence of 1465 nucleotides (from Genbank genomic region NC_000072 for linear genomic sequence of Klrb1c) is amplified:
 
3897                                                              tacc
3901 agggattacg gtcagccaag gcttacaaca accatctctg aagatgctat atcaccatat
3961 cttatgttag aggaaattgc tgcatagaaa cttgttgagg gtctcatggt tggtagttaa
4021 taaggctggt aacttctttg aaactatgtg acctggaagc ctgagatcgc aatcacttgg
4081 ccatactgcc tagggatgct agtacacttg atcattaact taagaaaacg ctgtgtcagt
4141 taactttgtc tacagtagaa aaaagtattc cttgttattt atatttcttt gatgtgtcat
4201 aacagtaatc ttgctcctgc acatgttaat aaaagtgtgg gttagtaaaa actactgtgt
4261 gtaaccagct gtggtgaagc aagaacaggg tagaatacac catgtcttcc tctcacttca
4321 gattgttcag ttaatttaga gtgcccacaa gactggcttt tacaccgaga taagtgcttt
4381 cacgtttctc aagtttccaa cacttgggag gaaggtcaag ctgactgtgg tagaaaagga
4441 gccactttgc tgctcattca agaccaagaa gaactggtaa gctgttcaat aaaacagaac
4501 acattagtat gttctttaag ctacaccaac atctctagac ttaacagatc ttccacacaa
4561 aggaaatgga gactttacat tatagaagct catagtgtga aagactaatt ggacatgtga
4621 gccaataact gcatgtaagg agagtgtcct ttagaacatt aatagaggtc acagatttca
4681 caaaggagga agaattaact gttggggatt caggatgggt acaggtttca gtaggctggt
4741 agaatgtgag tcactaatac acagactcag ggttaagtcc tctgctctgt ttggattgca
4801 gagattccta ctggactcaa taaaggaaaa atataattca ttttggattg gactaaggtt
4861 cacattgcca gacatgaact ggaagtggat aaatggcaca actttcaatt ctgatgtgtg
4921 agtactagaa gagctagttt gaatattcat ttctgttgca ggggcttttg gggattcccc
4981 ttggggctaa ttgtaaagac atggagaatc tgtgtagtat aagcctgtta gcctttctgc
5041 tggggacagt ccctgagcag ccagctattc cacctgcctc ccctgccccc tgggtcttca
5101 ttgcagggct ctctgctctg cttcctgctg ctcccctttc ttcctctggc tcaggtctcc
5161 agctctgttt catccaaagc cttctccttt ctgattaaaa aaaaaaaaaa aaaaaaaaaa
5221 gttaattagt atagtgggag gggcttctga gctcactcag gaaccaatca gcaagcagaa
5281 ggactacctc ccatttataa tctagtgttg acctttggcc aaattgaggg ctatgagacc
5341 tcccgatgga agcttgttat c
 
Primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated A is indicated in red.
References
Science Writers Eva Marie Y. Moresco, Nora G. Smart
Illustrators Diantha La Vine
AuthorsOwen M. Siggs, Bruce Beutler
Edit History
2011-09-01 4:49 PM (current)
2011-08-29 12:03 PM
2011-08-29 12:00 PM
2011-05-10 4:10 PM
2011-05-10 3:35 PM
2011-05-10 3:31 PM
2011-01-28 6:56 PM
2010-12-30 10:50 AM
2010-12-30 10:50 AM
2010-12-23 11:38 AM
2010-07-27 10:33 AM
2010-04-26 11:21 AM
2010-02-03 4:37 PM