Phenotypic Mutation 'ganymede' (pdf version)
Mutation Type small insertion
Base Change
Gene Tap2
Gene Name transporter 2, ATP-binding cassette, sub-family B (MDR/TAP)
Synonym(s) Abcb3, Ham-2, HAM2, Ham2, MTP2, PSF2, Tap-2
Chromosomal Location 34,203,527-34,216,321 bp (+)
MGI Phenotype FUNCTION: The membrane-associated protein encoded by this gene is a member of the superfamily of ATP-binding cassette (ABC) transporters. ABC proteins transport various molecules across extra- and intra-cellular membranes. ABC genes are divided into seven distinct subfamilies (ABC1, MDR/TAP, MRP, ALD, OABP, GCN20, White). This protein is a member of the MDR/TAP subfamily. Members of the MDR/TAP subfamily are involved in multidrug resistance. The protein encoded by this gene is involved in antigen presentation. This protein forms a heterodimer with Tap1 in order to transport peptides from the cytoplasm to the endoplasmic reticulum. Mutations in the human gene may be associated with ankylosing spondylitis, insulin-dependent diabetes mellitus, and celiac disease. [provided by RefSeq, Jul 2008]
PHENOTYPE: Homozygous mutant mice have no CD8+ T cells, although their numbers of CD4+ T cells and B cells are normal. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_011530; MGI: 98484

Mapped Yes 
Amino Acid Change
Institutional SourceBeutler Lab
Ref Sequences
Ensembl: ENSMUSP00000025197 (fasta)
Gene Model not available
SMART Domains

signal peptide 1 28 N/A INTRINSIC
transmembrane domain 55 77 N/A INTRINSIC
transmembrane domain 97 119 N/A INTRINSIC
Pfam:ABC_membrane 151 418 2.4e-52 PFAM
AAA 494 678 2.58e-19 SMART
Meta Mutation Damage Score Not available question?
Is this an essential gene? Probably nonessential (E-score: 0.204) question?
Phenotypic Category
Phenotypequestion? Literature verified References
CTL killing - decreased
FACS CD8+ T cells - decreased
immune system
NK cell response - decreased
Candidate Explorer Status CE: no linkage results
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI

All alleles(1) : Chemically induced(1)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00095:Tap2 APN 17 34215378 missense probably benign 0.09
IGL00802:Tap2 APN 17 34209130 missense probably damaging 0.96
IGL01291:Tap2 APN 17 34209210 missense probably benign 0.01
IGL01337:Tap2 APN 17 34205412 unclassified probably benign
IGL01549:Tap2 APN 17 34214329 missense probably benign 0.12
IGL02433:Tap2 APN 17 34205419 unclassified probably benign
IGL02488:Tap2 APN 17 34214642 unclassified probably benign
IGL02657:Tap2 APN 17 34205458 missense probably damaging 0.99
IGL02677:Tap2 APN 17 34212047 missense probably benign 0.20
IGL03183:Tap2 APN 17 34205425 unclassified probably benign
date UTSW 17 34212354 missense probably damaging 0.99
date2 UTSW 17 34214032 nonsense probably null
hebe UTSW 17 small insertion
juventas UTSW 17 small insertion
Palm UTSW 17 34215940 missense possibly damaging 0.64
3370:Tap2 UTSW 17 34209279 splice site probably null
ANU05:Tap2 UTSW 17 34209210 missense probably benign 0.01
FR4976:Tap2 UTSW 17 34205699 unclassified probably benign
R0595:Tap2 UTSW 17 34212354 missense probably damaging 0.99
R0841:Tap2 UTSW 17 34215940 missense possibly damaging 0.64
R1145:Tap2 UTSW 17 34215940 missense possibly damaging 0.64
R1145:Tap2 UTSW 17 34215940 missense possibly damaging 0.64
R1296:Tap2 UTSW 17 34211915 missense probably benign 0.12
R1567:Tap2 UTSW 17 34214091 missense probably benign 0.00
R1656:Tap2 UTSW 17 34205953 missense possibly damaging 0.64
R1693:Tap2 UTSW 17 34209212 missense probably benign 0.00
R2246:Tap2 UTSW 17 34208801 missense possibly damaging 0.82
R2251:Tap2 UTSW 17 34211954 missense probably damaging 0.98
R2937:Tap2 UTSW 17 34212354 missense possibly damaging 0.80
R4682:Tap2 UTSW 17 34214032 nonsense probably null
R5262:Tap2 UTSW 17 34214016 missense probably benign
R6052:Tap2 UTSW 17 34214709 missense probably damaging 1.00
R6151:Tap2 UTSW 17 34212047 missense probably benign 0.00
R6196:Tap2 UTSW 17 34214410 missense possibly damaging 0.50
R7020:Tap2 UTSW 17 34214414 missense possibly damaging 0.78
R7677:Tap2 UTSW 17 34205520 missense probably benign 0.01
R7694:Tap2 UTSW 17 34205697 missense probably benign
R8129:Tap2 UTSW 17 34205698 missense probably benign 0.01
R8256:Tap2 UTSW 17 34216032 missense probably benign 0.04
Z1177:Tap2 UTSW 17 34205668 missense probably benign 0.00
Mode of Inheritance Autosomal Recessive
Local Stock Sperm, gDNA
MMRRC Submission 032642-UCD
Last Updated 2019-02-12 11:39 AM by Diantha La Vine
Record Created 2009-08-17 12:00 AM
Record Posted 2009-12-03
Phenotypic Description
The ganymede phenotype was discovered among ENU-mutagenized G3 mice in an in vivo screen for mutants with defects in natural killer (NK) cell or CD8+ cytotoxic T lymphocyte (CTL) effector function (In vivo NK cell and CD8+ T cell cytotoxicity screen).  Mice were immunized with irradiated 5E1 cells (syngeneic class I MHC-deficient cells transformed by human adenovirus type 5 early region 1), and one week later, injected with three target cell populations: control C57BL/6J cells, antigen-specific CTL target cells (C57BL/6J splenocytes externally loaded with a peptide derived from the adenovirus E1B protein), and NK cell-specific target cells (syngeneic class I MHC-deficient cells). The reduction of NK target and CTL target populations relative to the control population after 48 hours reflects the in vivo cytotoxic ability of NK cells and CTLs. Homozygous ganymede mice failed to kill both antigen-specific targets and class I MHC-deficient cells, demonstrating impaired CTL and NK cell function (Figure 1). Homozygous ganymede mice also display a near complete absence of CD8+ T cells in the blood, and lack MHC class I expression on lymphocytes (Figure 2).
Nature of Mutation
The candidate genes Tap1, Tap2, Tapbp (tapasin), and B2m (β2-microglobulin), encoding proteins required for MHC class I expression, were sequenced. A single C nucleotide insertion was identified within a stretch of Cs located between nucleotides 1083 and 1087 of the Tap2 transcript, in exon 5 of 12 total exons. The sequence of Cs is increased from five to six nucleotides (highlighted in gray).  The resulting frameshifted transcript is predicted to encode aberrant amino acids from position 312 to 361, followed by a premature stop at position 362 in the mouse TAP2 protein (red text).

Wild type:
                          <--exon 5  exon 6-->
305  -A--E--K--V--Y--N--P--R--H--Q-  -A-...

305  -A--E--K--V--Y--N--P--P--P--S-- G--...-G--A--M--* 362
The ganymede mutation arose spontaneously, and not as a result of ENU-induced mutagenesis. Another mutant line (hebe) derived from an independent G0 founder exhibited the same phenotype as ganymede homozygotes. Sequence analysis has demonstrated that the hebe phenotype is caused by the same mutation found in ganymede mice. 
Illustration of Mutations in
Gene & Protein
Protein Prediction

Figure 3. Domain organization and predicted membrane topology of TAP. Top, domain structure of TAP2. Below, Topography of TAP1/TAP2. Both TAP1 and TAP2 contain a 6-helix core transmembrane domain (TMD) and a cytosolic nucleotide binding domain (NBD). Tapasin binding N-terminal accessory domains consist of 4 and 3 transmembrane helices for TAP1 and TAP2, respectively. The Walker A (A), Walker B (B), and signature/C-loop motif (C) sequences, involved in ATP binding and/or hydrolysis, are color coded to indicate to which ATPase site they belong (green=consensus site; orange=degenerate site). The ganymede mutation results in aberrant amino acids from position 312 to 361, followed by a premature stop at position 362.The image is interactive. Other mutations found in TAP2 are noted in red. Click on each allele for more information.

The transporter associated with antigen processing (TAP) pumps cytosolic peptides into the endoplasmic reticulum (ER) lumen for loading onto class I major histocompatibility (MHC) molecules and presentation to T lymphocytes. TAP is a member of the ATP-binding cassette (ABC) transporter family, ubiquitous proteins that shuttle a variety of substrates, including ions (see record for mayday), sugars, amino acids, peptides, vitamins, lipids, antibiotics, and drugs, across cellular membranes (1;2). ABC transporters function either as importers, present only in prokaryotes, or exporters, present in all kingdoms of life. Regardless of the direction of transport, all ABC transporters possess a modular architecture, the core of which consists of two hydrophobic transmembrane domains (TMDs) and two cytosolic nucleotide-binding domains (NBDs; also called ABCs) (3). These four domains may be contributed by one, two, three, or four distinct polypeptides; the most common arrangement for exporters (e.g. TAP) is a homo- or heterodimeric complex composed of two “half-transporters” each containing one TMD fused to one NBD. Outside of this “translocator unit,” accessory domains or proteins exist to provide regulatory or other functionalities to the transporters (3). Whereas the sequence and structural motifs of NBDs are highly conserved among members of the ABC transporter family, the sequence and structure of TMDs exhibit substantial variation that reflects the diversity of translocated substrates. 
TAP is a heterodimer of the homologous TAP1 (724 amino acids in mice) and TAP2 proteins (702 amino acids in mice), each of which contains a six-helix TMD and a C-terminal NBD (Figure 3) (4;5). TAP1 and TAP2 also contain N-terminal accessory domains, non-essential for peptide transport, that bind to tapasin, the specialized chaperone that bridges TAP and class I MHC molecules in the peptide loading complex (5;6) (see Background). The tapasin-binding accessory domains of TAP1 and TAP2 consist of four and three transmembrane helices, respectively (4;5), where the first N-terminal helix is essential for tapasin binding (7). Structural information about these domains is discussed below.
The structural arrangement of the TAP TMDs is predicted to follow that of the bacterial transporters Sav1866 (a putative drug transporter) and MsbA (a lipid flippase) (8;9), which share approximately 40% sequence similarity and 15% identity with TAP (10). The crystal structures of the bacterial transporters reveal that each TMD consists of a bundle of six long helices that extend into the cytoplasm (Figure 4, PDB ID 2HYD). The helices exhibit a domain-swapped organization in which two helices from one protein in the dimer form a bundle with four helices from the second protein subunit and vice versa. Thus, each TMD is formed by helices from both proteins of the dimer. In TAP, peptides would be translocated through the pore formed by the TMDs, which appears to have a V-shaped opening that normally faces the cytosol. This opening is predicted to alternate between facing the cytosolic side and facing the ER lumenal side of the membrane, powered by the ATP-dependent opening and closing of the NBDs (see below) (11).
In addition to residues that map to cytosolic loops and transmembrane domains of both TAP1 and TAP2 (12), at least three sites predicted to lie near the base of the opening have been shown to mediate binding to antigenic peptide. These consist of a TAP2 polymorphic site identified in the rat that alters TAP peptide specificity (13-16), and sites identified because they are either cleaved by reactive peptides or crosslink to cysteine-containing peptides (17). The rat TAP2 polymorphism is a complex nucleotide sequence polymorphism resulting in 25 amino acid changes across the N-terminal two-thirds of the protein that contains the transmembrane domains (16). Among these changes, substitution of the Ala-Glu dipeptide at position 217-218 (in the rTap2a-encoded allele) with Thr-Met (in the rTap2u allele) results in a preference for substrates with hydrophobic instead of basic C-termini (13;14).  In humans, coding sequence polymorphisms in TAP2 have not been shown to directly influence selectivity for C-terminal residues of peptide substrates (18;19). However, two human TAP2 isoforms that differ in their C-termini due to alternative splicing of distinct 11th exons and 3’ UTRs reportedly confer discrete peptide selectivities (20). Several mouse TAP2 isoforms generated by alternative splicing are listed in Ensembl, but the functional differences between these isoforms are unknown.
The three-dimensional structure of TAP2 has not been described, but the crystal structure of the isolated homologous TAP1 NBD bound to ADP reveals the typical NBD fold comprising two subdomains (Figure 5; PDB ID 1JJ7) (21). The ATPase subdomain, containing two β-sheets and six α-helices, is homologous to other RecA-like ATPases and binds nucleotides, and is joined by a hinge-like loop (Q-loop) to an α-helical bundle called the helical subdomain. In the TAP heterodimer, two ATP molecules are clasped within two ATPase sites formed at the interface of the NBDs. Both NBDs make contacts with both ATP molecules, and contribute to ATP hydrolysis (22-24). Several sequence motifs in the TAP NBDs, characteristic of all ABC transporters, form critical interactions that facilitate ATP binding and hydrolysis (Table 1) [reviewed in (11)]. These include the Walker A, Walker B, and switch motifs provided by one NBD (from its ATPase subdomain), and the signature motif provided by the second NBD (from its helical subdomain) (25).  Only when the NBDs close are the nucleotide and water molecules appropriately positioned for hydrolysis (24).
Table 1. Conserved sequence motifs in TAP NBDs
Walker A (P-loop)
contacts α- and β-phosphates of bound nucleotide; coordinates ATP-associated Mg2+
Walker B
critical E residue positions hydrolytic water molecule
switch (H-loop)
contacts γ-phosphate
signature (C-loop)
contacts γ-phosphate and mediates NBD dimerization
positions and polarizes the hydrolytic water molecule
links ATPase and helical subdomains
Nonconsensus substitutions are in red text.
In TAP1, the Walker B and switch motifs each contain one amino acid substitution in the consensus sequences; in TAP2 the signature motif contains two substitutions (Table 1). In the context of the NBD dimer, all three nonconsensus motifs map to one ATPase site, while the second ATPase site is formed exclusively from consensus motifs (26). Such asymmetry of ATPase sites, where one site contains sequence motifs that follow the consensus (called the consensus site), and the second site contains one or more sequence motifs that deviate from consensus (called the degenerate site), is a common feature of many eukaryotic ABC transporters (11). Biochemical and mutagenesis experiments demonstrate that the consensus ATPase site has higher activity than the degenerate site and is the principal driver of NBD closure and ATP hydrolysis (26). Consistent with this, TAP can still function when mutations prevent ATP hydrolysis in the degenerate site, but TAP loses all peptide transport function when the equivalent mutations are introduced to the consensus site (27-31)
ABC transporters including TAP are believed to share the same basic transport mechanism in which a substrate-binding cavity formed by the two TMDs alternately faces the inside or outside of the membrane (11;32).  When the NBDs are bound to ATP and dimerized, the TMDs form an outward facing (opposite the cytosolic side) cavity. Following ATP hydrolysis, the NBDs separate and the TMDs shift to face inward. Recent work supports the following model for the mechanism of TAP (Figure 6), proposed by Procko et al (11). In the resting state, the TAP NBDs are open (i.e. not dimerized) and the peptide-binding site faces the cytosol. Via their ATPase subdomains, TAP1 is bound to ATP, while TAP2 is bound to ADP [for which in isolation it has a strong preference (29)]. Peptide binding in the cytosol-facing cavity causes a conformational change transmitted to the NBDs via a coupling helix, which fits in a groove between the subdomains of each NBD, and connects each TMD to its respective NBD (9). This conformational change triggers the exchange of ADP for ATP on TAP2, dimerization of the NBDs, and thus the closing of the peptide binding site to the cytosol and opening to the ER lumen. Peptide affinity is lowest in the ATP-bound closed NBD conformation of TAP (17), and peptide is thereby released into the ER at this stage. ATP hydrolysis in the consensus ATPase site alone is sufficient to destabilize the closed NBDs, as described above, allowing the NBD dimer to separate and returning TAP to the resting conformation. Consistent with the requirement for a functional degenerate ATPase site for optimal transport activity, hydrolysis may also occur at the degenerate ATPase site and further favor NBD opening.
The Tap2ganymede allele encodes a mutant TAP2 protein with aberrant amino acids from position 312-361, followed by a premature stop at position 362. Based on the predicted membrane topology of TAP2 (4), the mutated region encompasses a portion of the 7th transmembrane helix and adjacent cytosolic loop (Figure 3).
TAP2 is expressed in all cell types and is found in complex with TAP1 in the ER membrane.
T cell receptors recognize antigens in the context of MHC class I and MHC class II molecules. MHC class I molecules are present on most nucleated cells, and mainly carry peptides derived from endogenous proteins for display to CD8+ cytotoxic T cells. In contrast, MHC class II molecules are generally restricted to antigen presenting cells, and present peptide fragments of exogenous proteins to CD4+ T helper cells. Some antigen presenting cells are also able to present antigens from exogenous pathogens on MHC class I via an alternative presentation pathway termed "cross-presentation" (33).
TAP is essential for the transport of peptides into the ER for loading onto MHC class I molecules and display at the cell surface. Cell lines with reduced cell surface expression of MHC class I molecules and defective antigen presentation provided the first clue to the existence of a transporter for cytoplasm to ER translocation of antigenic peptides (34-37). These cell lines possessed normal intracellular levels of MHC class I heavy chain and β2-microglobulin expression and function, as indicated by their ability to efficiently present exogenously added peptides or peptides targeted to the ER by an ER retention signal. The genetic defect mapped to the MHC locus (35), and subsequently four independent groups identified deletions in two genes encoding proteins homologous to ABC transporters (later named TAP1 and TAP2) (38-41). It was then shown that transfection of the deficient cell lines with genes for TAP1 and/or TAP2 restored their ability to present antigen (42-45).
Mature MHC class I heterotrimers consist of the MHC encoded polymorphic α-chain (heavy chain), the invariant β2-microglobulin subunit, and peptide. The construction of this complex requires an amazing series of coordinated enzymatic events (Figure 7). The pathway, constitutively active in all nucleated cells, begins in the cytoplasm with the degradation of intracellular proteins (both host and foreign) by the proteasome and peptidases [reviewed in (46)]. A small fraction of the generated peptides are translocated from the cytosol to the ER lumen by TAP for loading onto MHC class I by the peptide loading complex, consisting of TAP, MHC class I, and the chaperones calreticulin, ERp57, and tapasin. Some peptides must be further trimmed to the appropriate size in the ER lumen. Once bound to antigenic peptide, ER-resident chaperones are released, allowing peptide-bound MHC to migrate through the Golgi apparatus and on to the cell surface.
Peptide loading complex
Folding and loading of MHC class I is initiated when the MHC I heavy chain is cotranslationally inserted in the ER membrane and associates with calnexin and immunoglobulin-binding protein (BiP), chaperones that promote assembly with β2-microglobulin (47). After formation of a noncovalent dimer with β2-microglobulin, calnexin is replaced by calreticulin, which is thought to stabilize the complex and optimize peptide loading on MHC class I (48). The thiol reductase ERp57 is also added to the complex, and promotes intramolecular disulfide bond formation in the MHC class I heavy chain (49-51). ERp57 forms complexes with both calnexin and calreticulin, but only associates with the calreticulin-containing peptide loading complex; it is retained in the complex by a disulfide bond with tapasin (52).
Tapasin (also known as TAP-binding protein) is an essential component of the peptide loading complex (53). It is a type I membrane glycoprotein present in four copies in the complex, and performs multiple functions. Tapasin simultaneously binds to chaperone-associated MHC class I through its N-terminal ER lumenal domain, and to TAP through its C-terminal transmembrane and cytosolic stalk, acting to localize nascent MHC class I molecules close to the source of peptides (54-56). Cells and mice lacking tapasin display reduced levels of MHC class I cell surface expression due to a reduction in the efficiency of binding and in the optimization of peptide cargo, which selects peptides that confer high stability to MHC class I molecules (55;57-59). Tapasin also stabilizes TAP expression (54;60;61), and recruits and retains ERp57 in the peptide loading complex (49).
TAP substrates
The majority of substrates for TAP are derived from newly synthesized proteins, and TAP activity depends on continuing protein translation (62). Several cytosolic peptidases have been implicated in peptide generation, including tripeptidyl peptidase-II (TPP-II) (63), puromycin-sensitive aminopeptidase (PSA) (64), and bleomycin hydrolase (BH) (64), but the bulk are generated by the proteasome. The 20S/26S proteasome cotranslationally degrades an estimated one third to one half of newly synthesized proteins, typically damaged or unwanted proteins in the form of defective ribosomal products (DRiPs), polypeptides that never attain native structure as a result of errors in translation or post-translational processing that prevent proper folding (62;65;66). Perhaps surprisingly, the fraction of peptides presented on MHC class I out of the total number generated by the proteasome is exceedingly small (≤0.1%) (67). This is due in part to the fact that TAP cannot translocate many potential substrates because they are either too long or too short.
Peptides generated by the proteasome are 3-22 amino acids in length (68), whereas TAP preferentially binds peptides of 8-16 amino acids, and most efficiently transports peptides of 8-12 amino acids (69-71). Approximately 2% of proteasome-generated peptide fragments are of the appropriate size for TAP translocation and direct presentation by MHC class I molecules (72). For example, two percent of the products from proteasomal degradation of the model antigen ovalbumin are the immunodominant peptide SIINFEKL, but 6-8% of the products are SIINFEKL or an N-extended version of the peptide, suggesting that most MHC-presented peptides are derived from N-terminal trimming of extended peptides (72;73). The leucine aminopeptidase and other peptidases likely function redundantly in the trimming of N-extended peptides (74;75). However, cytosolic peptidases actually destroy most potential TAP substrates, and are the main reason so few peptides are ultimately presented by MHC class I (62;76). More than 99% of intracellular peptides are destroyed within one minute of their generation, before encountering TAP (77), due in particular to the action of the endopeptidase thimet oligopeptidase (76). To have a chance at MHC class I presentation, peptides of 8-11 amino acids, potential TAP substrates, must interact with TAP within a few seconds of generation.
Some peptides successfully bound and translocated by TAP are still too long for presentation by MHC class I. TAP transports peptides ranging in length from 7 to more than 20 amino acids, whereas class I MHC holds peptides of 8-10 residues (71). The ER aminopeptidase 1 (ERAP1) can trim peptides to the canonical 8-10 residues required for class I MHC binding (78-80). The functions of other ER aminopeptidases, including ERAP2 (80;81), in antigen processing remain to be determined.
The contribution of each peptide residue to affinity for TAP is well established. MHC class I substrates contain specific amino acids at only three anchor positions (positions 1, 2, and 9) that provide strong contacts to the MHC, while at other positions the sequence can freely vary. TAP substrate affinity must similarly be strong enough to capture substrates yet allow wide sequence diversity. The C-terminal amino acid and the first three N-terminal amino acids are the most important determinants of TAP substrate specificity (82-85). For the C-terminus, human TAP has highest affinity for hydrophobic and positively charged residues, moderate affinity for polar residues, and disfavors Asp and Gly. For the N-terminus, TAP prefers hydrophobic amino acids at position 3, and hydrophobic or charged residues at position 2. Aromatic or acidic residues at position 1, or proline at position 1 or 2 are highly disfavored. Amino acids at positions 4 to 8 have no effect on TAP affinity, but are the main determinants for TCR binding to MHC class I-associated peptide.
Regulation of antigen processing
Cytokines such as interferon (IFN)-γ activate the adaptive immune system during infection, causing a concerted upregulation of the machinery for antigen processing and presentation. MHC class I molecules are expressed at low levels in most cells, but are strongly induced by IFNγ. Both TAP1 and TAP2 transcripts are upregulated 10-20-fold within 12 hours of IFNγ stimulation. IFNγ also helps to overcome the low efficiency of presentation of proteasome-generated peptides (86). During interferon IFNγ stimulation, the three active proteasomal β-subunits are replacedby β-immunosubunits to generate an “immunoproteasome” (87). Immunoproteasomes show a different cleavage pattern compared with constitutive proteasomes, generating more peptides that have correct C-termini for MHC class I binding, as well as an overall increased efficiency in generating peptides (86). IFNγ stimulation also upregulates expression of peptidases involved in antigen processing, including leucine aminopeptidase, ERAP1, and ERAP2 (80;81;88).
TAP and disease
Peptide binding is required to stabilize MHC class I molecules, so mice with disrupted TAP1 or TAP2 genes assemble drastically reduced amounts of MHC class I molecules, and have nearly absent surface expression of MHC class I.  The cells of Tap1-/- mice (89) and mice with an ENU-induced point mutation in TAP2 (Tap2jasmine) (90) are deficient in cytosolic antigen presentation, and consequently CD8+ T cells fail to develop in these animals. Similarly, human mutations in TAP1 (91;92), TAP2 (93;94), or tapasin (95) cause the rarely occurring bare lymphocyte syndrome type I (type I BLS, OMIM #604571), characterized a reduction in MHC class I surface expression to 1-3% of normal levels. Interestingly, a human TAP2 mutation has been reported that permits cell surface expression of empty MHC class I molecules to levels 3-5 times higher than observed in other TAP-deficient individuals (96), suggesting that TAP also promotes MHC class I surface expression independently of its peptide transport function. Consistent with this, the causative mutation destroys the ATP-binding site and activity of the TAP2 subunit, but is predicted to preserve interactions with TAP1 and tapasin (96).
Patients with type I BLS from TAP1 or TAP2 mutations develop chronic bacterial infections of the sinus and bronchi in late childhood and/or granulomatous skin lesions, but no severe combined immunodeficiency, diarrhea, nor persistent systemic infections (97;98). Considering the role of MHC class I proteins in presentation of viral peptides to CTLs, individuals with type I BLS surprisingly do not suffer from severe viral infections. Although the mechanisms of viral resistance remain to be demonstrated, a variety of contributing factors have been proposed [reviewed in (97)]: a normal humoral response; the presence of reduced but significant numbers of TCR αβ+ CD8+ T cells (92;93); MHC class I presentation of TAP-independent viral antigens to CTLs (99); expansion of TCR γδ+ T cells (96); and the presence of NK cells (94;100).
Polymorphisms in human TAP2 (or TAP1) have been associated with autoimmune diseases including type I diabetes mellitus (101-103), Graves’ disease (104), and multiple sclerosis (105). The mechanisms by which these polymorphisms predispose individuals to autoimmunity have not been elucidated. It was initially hypothesized that downregulation of TAP2 mRNA leading to reduced MHC class I cell surface expression may flag cells as targets for natural killer (NK) cells. However, TAP2- or TAP1-deficient humans (94;100) and mice (106;107) do not suffer from autoimmune diseases early in life because their NK cells acquire tolerance toward MHC class I-negative cells. Inhibition of NK cell activity in individuals lacking surface MHC class I has been shown to be mediated in part by reduced expression of NK activating receptors and elevated expression of NK inhibitory receptors (106;108).
Putative Mechanism
The frameshift caused by the ganymede mutation results in mutation of amino acids 312-361 followed by premature truncation in the fourth cytosolic loop of TAP2. These drastic alterations are predicted to result in a completely nonfunctional protein that abrogates all TAP-dependent peptide transport and consequent MHC class I expression. The lack of CD8+ T cells is due to death after failure to interact with peptide-MHC class I in the thymus, which is known to promote the development and maintenance of these T cells. Ganymede NK cells may display tolerance to MHC class I-deficient cells as a result of altered activating and inhibitory receptor expression acquired during development.
Primers Primers cannot be located by automatic search.
Ganymede genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide insertion.
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
ganymede_seq(F): 5’- AGGTGAACCTGGCATCTGG -3’
ganymede_seq(R): 5’- GTCCATCAGTGAATGACTGAAC -3’
The following sequence of 765 nucleotides (from Genbank genomic region NC_­­­­­000083 for linear genomic sequence of Tap2, sense strand) is amplified:
4299                                          tc ttccaggaga ccaagacagg
4321 tgaacctggc atctgggtct tgggtgctcg gccctttctg gagtcctcag gtctcctgtc
4381 tgcctccctg ctggagcctg gcagttttct cttagagcag ggtagaggtc tagcccagtc
4441 tctttgtaaa ggctgagggg atgagtcagc agggagaccc agaggaaggt tttgggttcc
4501 catcatcctt tctgccctcc ccaggggagc tgaactcgag gctgagctct gacacctctc
4561 tgatgagccg ctggctccct ttcaatgcca atatcctgct gcggagcctg gtgaaggtgg
4621 tggggctcta cttcttcatg ctccaggtat cgccccgact caccttcctc tccctgctgg
4681 acctgcccct cacgatagca gctgagaagg tgtacaaccc ccgccatcag gtatgtgtgc
4741 atgtcacagt gccctgagag aaggcaaaca gacaagcaga cagacaaata agtagatagg
4801 taggtaggta gatagatggg tggatggata ggtaggtagt agacagacag acagacagac
4861 agacaaataa gtagataggt aggtagatag ataaataggt aggtaggtag atagatagat
4921 tgatagacag ataatcggta gacagacaga acacctctta tcagtcccta cagtccattc
4981 tatgcttgca caacattttg ctagttcagt cattcactga tggacatttg agttgaggag
5041 gttggaggcc cactggatct ata
Primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the stretch of 5 Cs into which an extra C is inserted is indicated in red.
Science Writers Eva Marie Y. Moresco
Illustrators Eva Marie Y. Moresco, Diantha La Vine
AuthorsPhilippe Krebs, Bruce Beutler
Edit History
2011-01-11 10:24 AM (current)
2011-01-10 1:21 PM
2011-01-07 3:22 PM
2011-01-07 3:21 PM
2011-01-07 3:06 PM
2011-01-07 3:04 PM
2011-01-07 2:28 PM
2011-01-07 2:26 PM
2011-01-07 1:43 PM
2011-01-07 1:39 PM
2010-12-23 2:03 PM
2010-07-27 10:32 AM
2010-04-26 11:19 AM
2010-02-03 10:09 AM