| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,1
* Rayne institute, Centre for Molecular medicine, Department of Medicine, University College of London, London WC1E 6JJ, UK; and
Immunology Group, Institute of Cell and Molecular Science, Barts and The London School of Medicine and Dentistry, London, EC1A 7ED, UK
1Correspondence: Rayne Institute, Centre for Molecular Medicine, Department of Medicine, University College of London, 5 University St., London WC1E 6JJ, UK. E-mail: k.paulsson{at}ucl.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT BACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
Key Words: loading complex • tapasin • TAP • peptide • COPI • transport • optimization
|
BACKGROUND |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
. MHC class I molecules are constitutively expressed at almost all nucleated cells and present peptides usually eight or nine amino acids in length (2
, 3)
. Antigenic peptides are derived from all kinds of proteins synthesized inside the cell, self-proteins, viral proteins, bacterial proteins, and altered self-proteins. The degradation of intracellular proteins is mainly performed by multicatalytic protease complexes termed proteasomes (4)
. The proteasomes generate peptides with carboxyl-terminal ends suitable for fitting the class I binding groove, but the amino-terminal ends need further trimming (5)
. The peptides are transported to the lumen of the endoplasmic reticulum (ER) by the transporter associated with antigen processing (TAP) (6)
. It has been suggested that the aminopeptidase ERAAP trims the amino-terminal ends of class I ligands in the ER, making them fit the class I binding groove. Recently it was shown that decreased expression of ERAAP correlates with reduced surface expression of MHC class I (7)
. IFN-
up-regulates ERAAP as well as many other components of the MHC class I antigen processing machinery. MHC class I heavy chain (HC), ß2-microglobulin (ß2-m) dimers are loaded with high-affinity peptides as the last step of their ER maturation process. Subsequently, the antigenic peptide–MHC class I complexes are transported to the cell surface and presented to cells of the immune defense. A schematic illustration of the MHC class I peptide loading and expression at the cell surface is presented in Fig. 1
.
|
During assembly and folding, newly synthesized MHC class I molecules undergo extensive quality control in the ER performed by an array of chaperones and assisting proteins, including the seemingly dedicated protein tapasin (8)
. The time spent by MHC class I in the ER depends on how early the HC-ß2-m dimer is loaded with high-affinity peptide. The formation and stability of MHC class I molecules are critically dependent on HC, ß2-m, and peptide (9
, 10)
. The nascent HC polypeptide is initially chaperoned by the general chaperone BiP and the ER lectin calnexin. The binding of BiP to HC can be biochemically detected only when maturation is arrested, indicating the transient nature of this interaction (11
, 12)
. Kinetic analyses have suggested a sequential interaction of chaperones with MHC class I, with BiP acting on the HC before and/or simultaneously with calnexin (12)
. In human cells, HLA binding to ß2-m results in an exchange of calnexin for the soluble homologue calreticulin (11
, 13)
. Late-stage maturation involves binding to a large complex of proteins, including at least calreticulin, tapasin, TAP, and ERp57. This complex has been termed the loading complex (LC) since it promotes peptide loading onto MHC class I (14)
. Tapasin is a 48 kDa transmembrane protein and is stably bound to TAP. It binds directly to class I HC-ß2-m dimers and thereby mediates class I interaction with the LC (15
, 16)
. Calreticulin has lectin-like activity and functions as a molecular chaperone, binding to monoglucosylated glycan groups of many N-linked glycoproteins in the ER. Moreover, a polypeptide-based, nonlectin binding-dependent function for calreticulin in MHC class I quality control has recently been suggested (17)
. The importance of calreticulin for the quality of class I molecules was highlighted when cells from calreticulin-deficient mice were shown to have impaired peptide loading onto class I molecules and low stability of cell surface-expressed class I molecules (18)
. The incorporation of class I molecules into the LC was not significantly impaired in these cells. Nevertheless, another recent study using a deglycosylation approach showed that fewer class I molecules were associated with tapasin and the LC in the absence of calreticulin association (19)
. The molecular mechanism for calreticulin in MHC class I maturation needs to be studied further before its exact role can be determined.
ERp57 is a thiol oxidoreductase and, similar to calnexin and calreticulin, acts on various substrates (20
21
22
23)
. ERp57 has been suggested to facilitate disulfide bond formation in class I heavy chains and has been reported to associate with MHC class I at a stage before incorporation into the LC as well as after (24
25
26)
. Disulfide bonded intermediates formed between tapasin and ERp57 in the LC have been reported and a shuttling mechanism of disulfide bonds in the LC, to allow for optimal MHC class I peptide loading, was suggested (27)
. In addition to its oxidoreductase activity, ERp57 is a cysteine protease (22)
. The cysteine protease activity of ERp57 may result in peptide trimming of class I ligands and perhaps trimming of peptides already loaded onto class I, but these options remain to be elucidated.
The specificity of tapasin for MHC class I is striking; unlike other chaperones such as BiP, calnexin, calreticulin, and ERp57, MHC class I is the only known target molecule. Different alleles of MHC class I depend in different degrees on tapasin for efficient peptide loading, but the critical function of tapasin in class I antigen presentation in general has been firmly established (16
, 28
29
30)
. It has been shown that certain natural single residue polymorphisms in HLA molecules dramatically affects tapasin dependence (31)
. The innate stability of the HC-ß2-m dimer without peptide as well as the peptide binding motif preferred by the class I allele might influence the allelic differences in tapasin dependence. The peptide binding preferences and promiscuity between different class I alleles in relation to tapasin association reliance are worth investigating. The mechanisms responsible for how tapasin influence the MHC class I maturation is one of the major questions in the field of antigen presentation today.
|
TAPASIN INFLUENCES THE BINDING OF PEPTIDE TO TAP |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
. Tapasin is binding to TAP via the transmembrane domains of both TAP1 and TAP2 (33
, 34)
. The estimated ratio of tapasin to TAP varies between studies. We have shown a 1:1 ratio, but other studies have found a ratio of as much as up to four tapasin molecules per TAP heterodimer (15
, 35)
. The expression of TAP has been suggested to be influenced by tapasin (36)
. However, the influence of tapasin on TAP expression is debated and may vary in different cell types. Experiments with the tapasin-deficient 721.220 cell line showed an increased amount of TAP expression after transfection with tapasin cDNA (36)
. However, results in our group have showed similar levels of expression of TAP in 721.220 and the related cell line 721.221, which is naturally expressing tapasin (37)
. Another experiment pointing in this direction was undertaken using ConA-stimulated blasts from tapasin knockout mice. These cells did not show reduced TAP expression compared with tapasin +/– cells (38)
.
A feature that indeed seems to vary in different cell types is the influence of tapasin on the peptide transport exerted by TAP. In 721.220 cells the absence of tapasin clearly decreases the peptide transport by TAP. Separate examination of the peptide binding, peptide translocation, and loading onto MHC class I showed that the peptide binding to TAP is significantly reduced in 721.220 cells compared with tapasin-expressing 721.221 cells (27)
. The peptide binding is restored by transfection of tapasin cDNA. Moreover, peptide transport is profoundly impaired in immortalized fibroblasts from tapasin knockout mice, though only moderately affected in spleen lymphoblasts from these mice (38)
. This clearly indicates cell type-specific variation in the TAP dependence on tapasin.
ATP binding to TAP is known to be essential for translocation of peptide across the ER membrane (39)
. Nucleotide binding to TAP was also shown to regulate dissociation of MHC class I from tapasin in the human cell line T2 transfected with cDNA for rat TAP (40)
. A conformational change first in TAP and then in tapasin as a result of nucleotide binding to TAP was suggested to account for class I release from tapasin. However, we have found the function of tapasin in the LC to be independent of nucleotides. In an assay with permeabilized T1 microsomes, peptides were added and allowed to enter the ER. Addition of apyrase (which degrades nucleotides) to the microsomes was not found to alter the association of MHC class I to tapasin nor did it alter the peptide induced dissociation of MHC class I from tapasin (41)
. Species specificity of TAP is a likely explanation for this discrepancy. Species specificity of TAP has been shown in terms of the peptide preferences of TAP, which is known to differ between species, as reviewed in ref 42
. Tapasin has also been demonstrated to exhibit species specificity in that mouse tapasin is not able to fully replace human tapasin for maturation of HLA B molecules in mouse cells (29)
.
|
TAPASIN INTERACTION PREVENTS IMMATURE MHC CLASS I MOLECULES FROM EXITING TO THE CELL SURFACE |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
, 43)
. In insect cells transfected with components of the MHC class I antigen processing machinery it was shown that, without coexpressed tapasin, empty HLA molecules were allowed to transport to the cell surface (43)
. Cotransfecting tapasin cDNA into these cells resulted in retrieval of empty class I in the ER. Another study showed that HLA-A 2.1 molecules were rendered incapable of interacting with the LC by a point mutation exchanging threonine to lysine at position 134 (T134K) (44
, 45)
. These T134K molecules were transported to the cell surface while lacking stably bound peptides and were deficient in presenting endogenous antigens to CD8+ T lymphocytes. A different approach to study tapasin and class I maturation was undertaken using H2Kb transfected 721.220 cells. Expression of H2Kb resulted in rapid egress to the cell surface of unstable H2Kb molecules, again pointing out the role for tapasin in successful stabilization of class I molecules (46)
.
ER retention at a stage after tapasin release was shown using GFP-tagged H2Ld molecules (47)
. A high ER diffusion coefficient in combination with ER localization of peptide-loaded class I molecules was suggested to result from limited access to ER exit sites or by association with a peptide optimizing machinery (47)
. Support for the ER exit site theory came when peptide-loaded class I molecules were found to accumulate at specific ER exit sites after dissociation from the TAP complex (48
, 49)
. Tapasin was excluded at these ER exit sites (50)
. It was also shown that antibody against H2Ld coprecipitated a putative COPII vesicle cargo receptor, BAP31. Intriguingly, clusters of wild-type HLA-A2 and the HLA-A2 mutant T134K, the latter known not to be able to bind TAP, were shown to accumulate at distinct regions of the ER (49
, 51)
. A supply of high-affinity peptides resulted in colocalization of the mutant T134K with the wild-type HLA-A2, indicating that only properly folded peptide optimized class I molecules gain access to these distinct ER exit sites. In conclusion, these findings suggest the existence of different mechanisms for the control of ER retention of MHC class I. The control mechanisms vary depending on the maturation stage of the class I molecule and, more precisely, on the quality of loaded peptide. Figure 2
shows a schematic model illustrating the different fates of MHC class I molecules during late-stage maturation.
|
|
QUALITY CONTROL DOWNSTREAM OF THE ER |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
, 44
, 46)
. The transport of premature class I molecules observed in the absence of tapasin may result from the lack of direct ER retention or failure of recycling of escaped premature class I molecules back to the ER, or from a combination of the two. Transport to the cell surface of only high stability class I molecules in wild-type cells has generally been attributed to ER retention of immature class I, but a recycling mechanism of MHC class I from late secretory compartments to the ER has been suggested from several studies (52
53
54
55)
. In a study proposing immature MHC class I recycling, the intermediate compartment (IC, morphologically defined as a tubulovesicular membrane region downward of the ER in the secretory pathway) was found to harbor a substantial amount of class I molecules lacking high-affinity peptides (53)
. When protein synthesis was inhibited by treatment with cycloheximide for 2 h, a significant proportion of MHC class I was still detected in the ER fractions. The ER fractions recovered from cycloheximide-treated cells grown under temperature block have, on the other hand, very little MHC class I. Temperature block inhibits protein transport in both the retrograde and anterograde direction, resulting in protein accumulation in the IC (56
, 57)
. From these studies it was suggested that in cycloheximide-treated cells grown at 37°C, the class I molecules detected in the rough ER resulted from recycling of nonoptimally loaded MHC class I molecules from the IC (53)
. Similarly, studies of other proteins such as misfolded VSV G-protein and T cell receptor 
(TCR) have showed recycling from post-ER compartments back to the ER (58
, 59)
.
Recently the interaction of tapasin with COPI-coated vesicles was revealed (54)
. Tapasin is a transmembrane protein with a double lysine motif in its carboxyl-terminal cytoplasmic tail (15
, 35)
. COPI-coated vesicles ferry cargo molecules in a retrograde manner from the Golgi to the ER and are recruited by carboxyl-terminally double lysine motif containing proteins (60
61
62
63)
. Both classical and nonclassical MHC class I molecules have been found associated with COPI-coated vesicles (54
, 63)
. Classical MHC class I molecules were found associated with COPI-coated vesicles only in the presence of tapasin (54)
. In tapasin mutant cells, neither TAP nor MHC class I were detected in association with COPI-coated vesicles, indicating a direct role for tapasin to mediate the MHC class I transport by COPI-coated vesicles. Studies with 721.220 cells transfected with a cDNA for mutant tapasin, in which the carboxyl-terminal double lysine motif was replaced by a double alanine motif, abrogated the interaction between tapasin and COPI. The association of TAP and MHC class I with the double alanine mutant tapasin remained intact. Subcellular fractionation as well as immunoelectron microscopy showed tapasin to be distributed in fractions of the ER and Golgi. The biochemical data also showed that tapasin-associated, peptide-receptive MHC class I molecules are found throughout the early secretory pathway (54)
. From this study, we proposed that immature MHC class I molecules are retrieved from the Golgi network to the ER by the tapasin interaction with COPI. In this way, tapasin-associated unstable MHC class I molecules that fail optimal peptide loading and escape the ER are allowed to recycle between the ER and the Golgi complex until optimal peptide is loaded. Studies of HLA-G, a nonclassical MHC class I molecule, also revealed a COPI interaction (63)
. The HLA-G–COPI interaction could be inhibited by addition of high-affinity peptide, indicating the need for COPI recycling only of HLA-G loaded with suboptimal peptide (63)
. Moreover, another nonclassical MHC class I molecule, H2-M3 has been shown to depend on tapasin for proper cell surface expression. H2-M3 presents N-formylated peptides transported by TAP, and tapasin has been shown to be significant for the maturation and expression of H2-M3–peptide complexes (64
, 65)
. H2-M3 molecules are not expressed at the cell surface under normal cellular conditions, but instead are retained in the ER. H2-M3 molecules are efficiently retained in the ER in tapasin-deficient cells, indicating a distinct retention mechanism apart from tapasin for this nonclassical MHC class I molecule (64
, 65)
. Together, these results from different systems support the proposal that tapasin is immensely important for recycling as a means for peptide optimization, and not only for direct ER retention (37)
. Figure 3
illustrates tapasin-mediated recycling of suboptimally loaded MHC class I molecules in COPI-coated vesicles.
|
|
PEPTIDE LOADING |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
. We have shown that significantly fewer class I molecules associate with tapasin in the TAP-deficient T2 cells than in the TAP-expressing T1 cell line (41)
. The total amount of class I HC is similar but the dimer level is reduced in T2 cells compared with T1. The ratio of tapasin-associated class I molecules to total amount of HC-ß2-m dimer is the same in T2 as in T1 cells. This suggests that class I dimers require stabilizing TAP transported peptides already at a stage prior to tapasin association and that tapasin preferably interacts with MHC class I preloaded with suboptimal peptides, as suggested by others (55)
.
The need for suboptimally loaded class I molecules to achieve high-affinity peptides before tapasin release and transport to the cell surface is demonstrated by the low off-rate of tapasin-associated class I molecules in T2 cells (41)
. Indeed, it was recently shown that tapasin-dependent peptide optimization occurs over time (67)
. Class I molecules were shown to improve their peptide cargo both quantitatively and qualitatively in the presence of tapasin over time. Moreover, tapasin but not calreticulin interacts with both unloaded and peptide loaded class I molecules, again suggesting that tapasin not only chaperones empty class I molecules but has a role in late-stage class I maturation as well (68)
. The peptide binding groove of MHC class I molecules without peptide has been characterized as being in the molten globule state (69)
, and, in the absence of high-affinity peptide, class I molecules are indeed highly unstable. The total amount of HC-ß2-m dimers in T2 cells is lower than in T1 cells (41)
, further pointing to class I molecules as being unstable in the absence of sufficient amounts of suitable peptides even in the presence of ER chaperones.
|
PROSPECTS AND PREDICTIONS |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
. If the peptide is of high affinity, the complex reaches a specialized ER exit site and is subsequently transported to the cell surface, possibly in COPII-coated vesicles (48)
. MHC class I molecules unable to interact with tapasin and that failed to achieve high-affinity peptide at this stage will not be given the opportunity of optimizing bound peptide in the LC. Instead, these suboptimally loaded class I’s will eventually exit the ER. Class I molecules leaving the ER without high-affinity peptide will, due to instability, be degraded. These class I molecules are excluded from the ER exit sites used by high-affinity, peptide-loaded class I and must use an alternative ER exit route. On the other hand, MHC class I molecules loaded with suboptimal peptide that are able to interact with tapasin will not exit the ER as immature complexes but instead bind to the LC (14)
. Loading of high-affinity peptide at this stage dissociates the class I molecule from the LC and allows exit through the ER exit sites for high-affinity, peptide-loaded class I (49)
. However, if the class I molecule spend too long in the LC without achieving a high-affinity peptide, the MHC class I may escape the ER and end up in the Golgi. The transport mechanism for the escape of these improperly loaded class I molecules from the ER to the Golgi is not clarified, but may result from cisternal maturation over time, active transport, or natural flow down the secretory pathway. The class I molecules incorporated into the LC in the ER are bound to tapasin. Tapasin and TAP remain bound to the MHC class I lacking high-affinity peptide after leaving the ER. At this stage, tapasin and TAP represent what we propose to call the core LC. Biochemical data from subcellular fractionation experiments show tapasin, TAP, and MHC class I to be present in Golgi fractions (54)
. The presence of TAP in both ER and cis-Golgi compartments has also been illustrated by immunoelectron microscopy studies (70)
. Moreover, the complex of TAP, MHC class I and tapasin coprecipitate together with COPI subunits, indicating a retrograde transport mechanism back to the ER of the core LC and the MHC class I bound to it. Finally, using chemically modified class I binding peptides for cross-linking revealed that a large proportion of MHC class I in the Golgi fractions was unloaded or loaded with low-affinity peptide (54)
. Together, these data support that tapasin mediates COPI recycling back to the ER of MHC class I molecules, allowing another round of peptide editing. Defining the organization on temporal as well as spatial level of the LC is now important for thoroughly understanding peptide optimization and class I maturation.
Peptide profiles from tapasin-deficient cells show reduced amounts of peptide. Overlays of the spectra from eluted peptides show significant overlap, although there are several specific peaks resulting from peptides eluted from the tapasin-deficient cells and as well as others resulting from the tapasin wild-type cells (71)
. The peptide profiles suggest the presence of tapasin-uncontrolled peptide loading as well as tapasin-controlled peptide loading of class I, with the variation in peptide spectra giving rise to a wider array of peptides eventually presented to CD8+ T lymphocytes. If indeed tapasin-associated class I molecules have a distinct peptide repertoire, the properties of these peptides are important to determine and will provide insight into peptide selection during class I loading. HLA-B27 is known to be less dependent on tapasin association for the cell surface expression of fairly stable peptide HLA–B27 complexes (29)
. This allele has also been demonstrated to be involved in autoimmune diseases such as ankylosing spondylitis (72)
. Further peptide profiling of peptides eluted from alleles such as HLA-B27 has the potential to reveal specific motifs of peptides easily loaded without binding to the LC. These peptides could then be considered for studies of antigen presentation and mechanisms in autoimmune disease and virally induced disease. Peptide profiling further has great commercial potential for identifying viral peptides for future vaccination studies against viruses as well as tumors.
In addition to peptide profiling experiments, there are several topics of great interest to investigate about the MHC class I antigen processing machinery. Today the field would greatly benefit from studies of the newly discovered aminopeptidase ERAAP and its role in generating suitable class I ligands, the thiol oxidoreductase ERp57, the importance of disulfide bond formation, and possibly disulfide bond breaking during class I maturation. Experiments on MHC class I transport out of the ER and the factors regulating its export have been initiated; future work will be very interesting to follow. The recycling of class I molecules in COPI vesicles is, as we suggest, a key event for quality control in terms of allowing peptide optimization of class I molecules. The generated tapasin knockout mice (38
, 73)
are interesting to study and will confirm whether the recycling theory holds in systems other than the gamma irradiation-generated 721.221 and 721.220 cell lines.
|
ACKNOWLEDGMENTS |
|---|
Received for publication August 22, 2003. Accepted for publication September 12, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTBACKGROUNDTAPASIN INFLUENCES THE BINDING...TAPASIN INTERACTION PREVENTS...QUALITY CONTROL DOWNSTREAM OF...PEPTIDE LOADINGPROSPECTS AND PREDICTIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Gruda, H. Achdout, N. Stern-Ginossar, R. Gazit, G. Betser-Cohen, I. Manaster, G. Katz, T. Gonen-Gross, B. Tirosh, and O. Mandelboim Intracellular Cysteine Residues in the Tail of MHC Class I Proteins Are Crucial for Extracellular Recognition by Leukocyte Ig-Like Receptor 1 J. Immunol., September 15, 2007; 179(6): 3655 - 3661. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. G. Santos, E. C. Campbell, S. Lynch, V. Wong, A. N. Antoniou, and S. J. Powis Major Histocompatibility Complex Class I-ERp57-Tapasin Interactions within the Peptide-loading Complex J. Biol. Chem., June 15, 2007; 282(24): 17587 - 17593. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Almeciga, Z. C. Wang, J. Zuniga, M. Fernandez-Vina, O. Clavijo, H. Araujo, V. Romero, J. Henry, S. Ferrone, and E. J. Yunis Allorecognition of an HLA-A*01 Aberrant Allele by an HLA Identical Family Member Carrying the HLA-A*0101 Allele J. Immunol., December 15, 2006; 177(12): 8643 - 8649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Montserrat, B. Galocha, M. Marcilla, M. Vazquez, and J. A. Lopez de Castro HLA-B*2704, an Allotype Associated with Ankylosing Spondylitis, Is Critically Dependent on Transporter Associated with Antigen Processing and Relatively Independent of Tapasin and Immunoproteasome for Maturation, Surface Expression, and T Cell Recognition: Relationship to B*2705 and B*2706 J. Immunol., November 15, 2006; 177(10): 7015 - 7023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Paulsson, M. Jevon, J. W. Wang, S. Li, and P. Wang The double lysine motif of tapasin is a retrieval signal for retention of unstable MHC class I molecules in the endoplasmic reticulum. J. Immunol., June 15, 2006; 176(12): 7482 - 7488. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Lu, D. G. Kavanagh, and A. B. Hill Cellular and Molecular Requirements for Association of the Murine Cytomegalovirus Protein m4/gp34 with Major Histocompatibility Complex Class I Molecules. J. Virol., June 1, 2006; 80(12): 6048 - 6055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. Boyle, A. K. Gillingham, S. Munro, and J. Trowsdale Selective Export of HLA-F by Its Cytoplasmic Tail. J. Immunol., June 1, 2006; 176(11): 6464 - 6472. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. F. de Almeida, I. F. Carvalho, C. S. Cardoso, J. V. Cordeiro, J. E. Azevedo, J. Neefjes, and M. de Sousa HFE cross-talks with the MHC class I antigen presentation pathway Blood, August 1, 2005; 106(3): 971 - 977. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
