Tài liệu: Ai cần tớ giúp....

Figure 1: Major histocompatibility complex (MHC) class I-restricted antigen processing and presentation in a virally infected target cell. Viral proteins within the cytosol are degraded via the proteasome, resulting peptides are transported into the ER lumen through TAP. Antigenic peptides (yellow circles) are loaded on to MHC class I molecules which then move to the cell surface, where they can be recognised by CD8+ cytotoxic T lymphocytes (CTL).
Figure 2: Generation of mature effector cytotoxic T lymphocytes (CTL) [adapted from (52)]. (a) Activation of CD8+ precursor CTL (p-CTL), into antigen-activated p-CTL occurs after interaction with antigen-major histocompatibility complex (MHC) class I complexes on target cells. Further proliferation and differentiation into mature CTL requires IL-2 secreted from TH1 cells that have been activated by an antigen-MHC class II interaction occurring on specialised APC. (b) Activation of p-CTL into mature CTL by dendritic cells (DC) can occur when exogenous antigen cross-primes the DC leading to both MHC class I and class II presentation.
Concay

Figure 1: Major histocompatibility complex (MHC) class I-restricted antigen processing and presentation in a virally infected target cell. Viral proteins within the cytosol are degraded via the proteasome, resulting peptides are transported into the ER lumen through TAP. Antigenic peptides (yellow circles) are loaded on to MHC class I molecules which then move to the cell surface, where they can be recognised by CD8+ cytotoxic T lymphocytes (CTL).
Figure 2: Generation of mature effector cytotoxic T lymphocytes (CTL) [adapted from (52)]. (a) Activation of CD8+ precursor CTL (p-CTL), into antigen-activated p-CTL occurs after interaction with antigen-major histocompatibility complex (MHC) class I complexes on target cells. Further proliferation and differentiation into mature CTL requires IL-2 secreted from TH1 cells that have been activated by an antigen-MHC class II interaction occurring on specialised APC. (b) Activation of p-CTL into mature CTL by dendritic cells (DC) can occur when exogenous antigen cross-primes the DC leading to both MHC class I and class II presentation.
Concay

Figure 3: Toxin pathways in eukaryotic target cells leading to antigen presentation. Different protein toxins (orange circles; see text) can carry antigenic epitopes (yellow squares) through defined pathways (as indicated by the solid lines labeled 1-3), from the plasma membrane to intracellular sites from where membrane translocation occurs. In the cytosol, the toxin-antigen fusions undergo proteasomal processing, and antigenic peptides are delivered to the TAP transporter for translocation into the lumen of the ER (dashed lines). The peptides are then loaded onto newly synthesized MHC class I molecules and the complexes are routed to the plasma membrane where they can interact with CD8+ T lymphocytes (black broken line).
Concay

Figure 3: Toxin pathways in eukaryotic target cells leading to antigen presentation. Different protein toxins (orange circles; see text) can carry antigenic epitopes (yellow squares) through defined pathways (as indicated by the solid lines labeled 1-3), from the plasma membrane to intracellular sites from where membrane translocation occurs. In the cytosol, the toxin-antigen fusions undergo proteasomal processing, and antigenic peptides are delivered to the TAP transporter for translocation into the lumen of the ER (dashed lines). The peptides are then loaded onto newly synthesized MHC class I molecules and the complexes are routed to the plasma membrane where they can interact with CD8+ T lymphocytes (black broken line).
Concay

Accessory proteins and the assembly of human class I MHC molecules: a molecular and structural perspective
Marlene Bouvier,
School of Pharmacy, University of Connecticut, 372 Fairfield Road U-92, Storrs, CT 06269, USA
Volume 39, Issue 12 , January 2003, Pages 697-706
Abstract
The cell-surface presentation of antigenic peptides by class I major histocompatibility complex (MHC) molecules to CD8+ T-cell receptors is part of an immune surveillance mechanism aimed at detecting foreign antigens. This process is initiated in the endoplasmic reticulum (ER) with the folding and assembly of class I MHC molecules which are then transported to the cell surface via the secretory pathway. In recent years, several accessory proteins have been identified as key components of the class I maturation process in the ER. These proteins include the lectin chaperones calnexin (CNX) and calreticulin (CRT), the thiol-dependent oxidoreductase ERp57, the transporter associated with antigen processing (TAP), and the protein tapasin. This review presents the most recent advances made in characterizing the biochemical and structural properties of these proteins, and discusses how this knowledge advances our current understanding of the molecular events underlying the folding and assembly of human class I MHC molecules in the ER.
Author Keywords: Class I MHC molecules; Chaperone protein; Endoplasmic reticulum; Class I loading complex; Antigen presentation
Abbreviations: 2m, 2-microglobulin; CNX, calnexin; CRT, calreticulin; ER, endoplasmic reticulum; MHC, major histocompatibility complex; TAP, transporter associated with antigen processing

Article Outline
1. Introduction
2. Quality control in the endoplasmic reticulum
3. The structure of calnexin and calreticulin
4. A molecular mechanism for glycoprotein folding by calnexin, calreticulin, and ERp57
5. The class I heavy chain/2m heterodimer
6. The class I loading complex
7. Molecular and structural insights in the class I loading complex
Acknowledgements
References

1. Introduction
One of the primary functions of the immune system is to protect against infections by controlling the spread and virulence of invading pathogens. Central to maintaining immunity is the cell-surface presentation of antigenic peptides to specific receptors on CD8+ cytotoxic T-lymphocytes by class I major histocompatibility complex (MHC) molecules on infected cells. This cell-mediated recognition process results in the elimination of infected cells and is critically important to maintain immunity against viruses and tumors.
Human class I MHC molecules are composed of a ~44 kDa transmembrane heavy chain glycoprotein, a 12 kDa soluble protein called 2-microglobulin (2m), and a short peptide of 8?"10 residues derived from endogenous proteins as well as virus- and tumor-specific proteins. The class I heavy chain is encoded by three polymorphic genes within the MHC. A large solvent-exposed groove in the class I heavy chain consisting of two long -helices (1 and 2 domains) forms a binding site for antigenic peptides (Bjorkman et al., 1987a). The 3 domain is packed against the underside of the 1 and 2 domains and anchors the class I heavy chain to the cell membrane. The protein 2m is noncovalently associated with the class I heavy chain and is proximal to the cell membrane. The antigenic peptide binds in the groove in an extended conformation and interacts with conserved and polymorphic MHC residues along the binding site through its main- and side-chain atoms. These interactions are important to maintain class I MHC molecules stably assembled ( Parker; Guo; Ruppert; Saito and Bouvier). Some of the peptide side-chains, particularly those from the middle positions of the sequence, are oriented upward towards the solvent. It is this region of the peptide that is presented by class I MHC molecules at the surface of infected cells for recognition by T-cell receptors ( Bjorkman et al., 1987b).
The folding and assembly of class I MHC molecules occurs in the endoplasmic reticulum (ER) and begins with translation of the class I heavy chain which then associates with 2m to form a class I heavy chain/2m heterodimer. This intermediate complex is conformationally unstable and requires the binding of an antigenic peptide to complete the maturation process. In a way that is similar to the folding of other glycoproteins in the ER, maturation of class I MHC molecules is highly dependent on interactions with the lectin chaperones calnexin (CNX) and calreticulin (CRT), and the thiol-dependent oxidoreductase ERp57, to proceed efficiently. However, in a way that is distinct, maturation of class I MHC molecules requires the ad***ional participation of two dedicated proteins called the transporter associated with antigen processing (TAP) and tapasin. Collectively, these ER-resident proteins associate transiently with immature forms of class I MHC molecules until they acquire their native structures and are sufficiently stable to be transported to the cell surface via the secretory pathway.
This review presents the most recent advances made in characterizing the structure of accessory proteins involved in the maturation of human class I MHC molecules in the ER. This knowledge is discussed in the context of our current understanding of the molecular events that define the biogenesis of these molecules.
2. Quality control in the endoplasmic reticulum
Upon co-translational insertion in the ER membrane, class I heavy chain acquires a single Asn-linked oligosaccharide that is processed to the form Glc1Man9GlcNAc2 and serves as a recognition element for the lectin chaperones CNX and CRT (Hammond; Nauseef; Wada; Peterson and Hebert). These homologous lectins are part of a well-characterized system of ER-resident proteins that ensures the structural integrity of glycoproteins prior to their release into the secretory pathway ( Ellgaard and Helenius, 2001). This highly coordinated protein machinery also includes ERp57 ( Oliver and Oliver), to catalyze the formation of native disulfide bonds in glycoprotein substrates, and the enzymes UDP-glucose:glycoprotein glucosyltransferase and glucosidase II ( Parodi and Ellgaard), to regulate interactions between glycoprotein substrates and CNX and CRT. More specifically, these two enzymes act successively to catalyze the ad***ion and removal of the outer glucose residue in the Glc1Man9GlcNAc2 glycan moiety which consequently regulate cycles of substrate binding and release from CNX and CRT. In a way that is similar to the more classical chaperone proteins, CNX and CRT have also been shown to associate with nonnative polypeptide segments of glycoproteins as evidenced by their ability *****ppress the aggregation of unfolded proteins in vivo and in vitro (Saito; Ihara and Danilczyk). Overall, the ER has evolved a quality control system that prevents the transport of misfolded and incompletely folded glycoproteins to the Golgi complex.
3. The structure of calnexin and calreticulin
CNX is a 65 kDa type I membrane protein and CRT is a 46 kDa protein present in a soluble form in the lumen of the ER. Interestingly, CRT shares a high degree of sequence homology with the lumenal domain of CNX. The structure of CRT has been described as an organization of three distinct domains with loosely defined boundaries (Smith and Fliegel): the N-domain (residues 1?"180), the P-domain (residues 181?"290), and the C-domain (residues 291?"400). The N-domain is the most conserved region among species of CRT and shares extensive sequence similarity with CNX. The N-domain also contains a high-capacity, low-affinity zinc binding site (14 mol of zinc/mol of CRT and Kd=0.3 mM) (Khanna et al., 1986) consistent with the presence of His and Cys residues in this region of the protein, although no consensus zinc-binding motif has been identified. The central P-domain contains two distinct proline-rich sequences, the type 1 (17 residues) and type 2 (14 residues) sequences, that repeat in a "111222" arrangement. These sequences are highly conserved in CNX repeating in a "11112222" tandem arrangement. The types 1 and 2 sequences have been shown by NMR and X-ray crystallographic studies to be structural motifs ( Ellgaard and Ellgaard; Schrag et al., 2001). Finally, the C-domain contains a large number of acidic residues clustered primarily within the last 56 residues consistent with the ability of CRT to bind a high molar amounts of calcium ions with low-affinity (~25?"50 mol of calcium/mol of CRT and Kd=1?"2 mM) (Ostwald; Treves and Baksh). This calcium-binding property enables CRT to be one of the major proteins regulating the process of calcium-homeostasis in the ER. The acidic stretch of amino acids present in the C-domain of CRT is not conserved in the lumenal portion of CNX.
The crystal structure of a proteinase K-resistant fragment of the lumenal domain of CNX (residues 47?"468) has now been reported (Schrag et al., 2001). The structure reveals a strikingly asymmetric protein comprising a compact globular core region formed by two series of antiparallel -strands organized into a convex and con**** face from which extends a highly elongated domain. The globular region is formed by residues from both the N- and C-domain and has been described as a jelly-roll fold, whereas the long arm is comprised of residues from the central proline-rich P-domain and adopts an extended -hairpin fold. A putative binding site for monoglucosylated glycans was modeled in the globular region of CNX based on the location of a single bound glucose molecule in the crystal structure. It has been suggested that it may represent the lectin binding site for glycoprotein substrates. A putative binding site for a calcium ion was also located in the globular region of CNX, distant from the carbohydrate binding site. Although the coordination of this calcium ion is incomplete in the crystal structure, some of the residues involved are highly conserved in CRT suggesting that it may correspond to the high-affinity calcium binding site that has been characterized in CRT. No distinguishing structural features that may enable CNX to bind nonnative polypeptide segments have been identified in the crystal structure. Although the spatial organization of the N-, P-, and C-domain of CRT has not yet been described, it has been shown from sedimentation analysis that the protein also adopts a distinctively asymmetric structure in solution consistent with its anomalous position on gel filtration columns ( Bouvier and Li). The hydrodynamic properties of CRT have been attributed to the extended -hairpin fold that the P-domain adopts in solution as determined by NMR studies of a recombinant form of this isolated domain ( Ellgaard and Ellgaard). That the P-domain of CRT adopts a structure in solution that is essentially identical, except for its overall length, to that observed in the crystal structure of CNX is consistent with both proteins having a high degree of sequence similarity in this middle domain. Given their identical lectin specificity and shared chaperone function, it is expected that CRT possesses the same overall topology as that of CNX, as also supported by rotary shadow electron microscopy images of CRT (Mabuchi and Bouvier, unpublished results).
4. A molecular mechanism for glycoprotein folding by calnexin, calreticulin, and ERp57
The elongated molecular shape of CNX and CRT strongly suggests that these proteins are likely to be characterized by high intrinsic structural plasticity, most particularly in the P-domain. Consistent with this view, an analysis of CRT by rotary shadowing electron microscopy revealed that a region of the protein appears highly flexible as evidenced by its pronounced susceptibility to molecular deformations (Mabuchi and Bouvier, unpublished results). Molecular flexibility is consistent with the function of CNX and CRT as chaperone proteins and the underlying requirement to interact dynamically and transiently with many different proteins in the ER. Recent biochemical and NMR studies investigating interactions between an isolated form of the P-domain of CRT and ERp57 point in support of this analysis (Frickel et al., 2002). These studies have shown that the P-domain interacts with ERp57 through several residues located at the tip of the -hairpin fold. The CRT/ERp57 complex was determined to have a 1:1 molar stoichiometry with a dissociation constant Kd=(9±3) M and a first-order exchange rate koff>1000 s-1 at 20 °C. A direct interaction between either CNX or CRT and ERp57 was also established by gel filtration chromatography (Zapun et al., 1998), cross-linking and native gel analysis ( Oliver et al., 1999), and by a solid-phase binding assay ( Leach et al., 2002). These results, combined with the knowledge of the CNX crystal structure, have allowed to propose a molecular mechanism describing how CNX and CRT in concert with ERp57 may facilitate the folding of glycoproteins in the ER ( Frickel et al., 2002). The model is based upon recognizing that the distinct topology of CNX and CRT, in which the elongated P-domain protrudes as a curved extension from the globular core, creates a partially closed environment for glycoprotein substrates to reside in as they undergo productive folding interactions. It has been suggested that glycoprotein substrates could be sequestered into this environment upon recognition of the Glc1Man9GlcNAc2 glycan moiety by the lectin binding site of CNX and CRT. In this structural organization, it has been proposed that the apparent flexibility of the P-domain may provide a mechanism whereby CRT can effectively optimize interactions between ERp57 and glycoprotein substrates to catalyze the formation and rearrangement of intramolecular disulfide bonds. Overall, the molecular model proposed to describe the folding of newly synthesized glycoproteins in the ER is based on the unique structural features of CNX and CRT and on cooperative interactions between these lectin chaperones and ERp57. As mentioned above, the CNX/ERp57 and CRT/ERp57 complexes are part of a system of chaperone proteins and specialized enzymes that act collectively to modulate the folding of glycoproteins in the ER until they fold into their native structures.
5. The class I heavy chain/beta-2m heterodimer
Based on our current understanding, the chaperones CNX and CRT function at different stages of the folding and assembly pathway of human class I MHC molecules in the ER. Upon co-translational insertion in the ER membrane, class I heavy chain interacts with CNX which serves to facilitate the folding of class I heavy chain and stabilize it from aggregation as well as promote its assembly with 2m (Degen; Noessner and Sadasivan). It has been shown that the single Asn-linked glycan at position 86 (1 domain) in class I heavy chain is essential for the initial recognition by CNX, but that interactions with nonnative polypeptide segments are also important as folding progresses ( Zhang et al., 1995). Class I heavy chain has two intrachain disulfide bonds (2 and 3 domains) which are formed at this initial step ( Tector; Tector and Antoniou) presuming that CNX recruits ERp57 to catalyze this process ( Fig. 1). In fact, recent results have suggested that the CNX/ERp57 complex may form prior to its active participation in the assembly of class I MHC molecules (Diedrich et al., 2001) consistent with the general role that this complex plays in the folding of glycoproteins in the ER. Interestingly, the requirement for CNX to associate with class I heavy chain is somewhat unclear since the assembly, transport, and cell-surface expression of class I MHC molecules is unaffected in a mutant cell line lacking CNX ( Scott and Sadasivan). It has been suggested that another ER-resident chaperone, possibly immunoglobulin-binding protein (BiP), may substitute for CNX ( Noessner and Degen). BiP belongs to the Hsp70 family of molecular chaperones and has also been found associated with newly synthesized class I heavy chains in the ER ( Degen and Noessner). Since very few studies aimed at characterizing interactions between BiP and class I heavy chain have been reported, the possibility that CNX and BiP function redundantly in relation to the folding of class I heavy chain remains somewhat elusive. In any case, the CNX/ERp57 complex is thought to facilitate the correct oxidation and folding of class I heavy chains into a conformation compatible with recognition by 2m ( Degen; Noessner; Tector and Tector). Whether CNX dissociates from class I heavy chain prior to or concomitantly with the binding of 2m also remains unclear but, except for one report ( Carreno et al., 1995), results have shown that CNX is absent in the class I heavy chain/2m heterodimer ( Sugita; Zhang; Noessner; Tector; Sadasivan; Tector and Diedrich). The association of class I heavy chain with 2m represents a critical step in the assembly of class I MHC molecules and is likely to involve conformational changes in class I heavy chain. It has been proposed that these conformational changes may be the molecular event that triggers the release of CNX from class I heavy chain ( Degen and Noessner).
The class I heavy chain/2m heterodimer is an intermediate complex that awaits the binding of an antigenic peptide as a final step in completing the assembly of class I MHC molecules. These so-called "empty" class I MHC molecules have been shown to be highly unstable molecules in solution from in vitro reconstitution studies using recombinant class I heavy chains and 2m (Bouvier and Wiley, 1998). Results from biochemical and biophysical studies indicated that the soluble class I heavy chain/2m heterodimer exhibits properties similar to those of molten globules. Remarkably, the peptide binding site (1 and 2 domains) was identified as the most destabilized region, while the 3 domain was shown to be independently stable. This study also revealed allele- and locus-specific differences in the ability of class I heavy chains to associate with 2m and to form soluble class I heavy chain/2m heterodimers. Overall, these results are consistent with studies carried out in cell systems, including mouse cells, indicating that class I heavy chains that are associated with 2m adopt an "open" conformation ( Smith; Carreno; Solheim and Harris). These results are also consistent with the knowledge that these intermediate molecules associate with chaperone proteins in the ER. Indeed, numerous studies have shown that the chaperone CRT interacts with the "open" form of class I heavy chain/2m heterodimer prior to the step of peptide binding ( Sadasivan; Solheim and Tector). Based on experiments carried out in both human and mouse cells, the association between CRT and the class I heavy chain/2m heterodimer is mediated, at least in part, through the Asn-linked glycan at position 86 (1 domain) in class I heavy chain ( Sadasivan; Zhang; Harris and Turnquist).
To this point, our understanding of the folding and assembly of class I MHC molecules indicates that although CNX and CRT are functionally homologous proteins and that both interact directly with the class I heavy chain, these proteins have distinct roles in the maturation process with, most significantly, the recognition of different class I folding intermediates. Furthermore, although it has been shown that the class I heavy chain/2m/CRT heterotrimer is present as an independent complex in the ER (Sadasivan; Solheim and Harris), the vast majority of these molecules are simultaneously assembled with three ad***ional accessory proteins; TAP, tapasin, and ERp57.
Được Milou sửa vào 06:08 ngày 18/06/2003

Accessory proteins and the assembly of human class I MHC molecules: a molecular and structural perspective
Marlene Bouvier,
School of Pharmacy, University of Connecticut, 372 Fairfield Road U-92, Storrs, CT 06269, USA
Volume 39, Issue 12 , January 2003, Pages 697-706
Abstract
The cell-surface presentation of antigenic peptides by class I major histocompatibility complex (MHC) molecules to CD8+ T-cell receptors is part of an immune surveillance mechanism aimed at detecting foreign antigens. This process is initiated in the endoplasmic reticulum (ER) with the folding and assembly of class I MHC molecules which are then transported to the cell surface via the secretory pathway. In recent years, several accessory proteins have been identified as key components of the class I maturation process in the ER. These proteins include the lectin chaperones calnexin (CNX) and calreticulin (CRT), the thiol-dependent oxidoreductase ERp57, the transporter associated with antigen processing (TAP), and the protein tapasin. This review presents the most recent advances made in characterizing the biochemical and structural properties of these proteins, and discusses how this knowledge advances our current understanding of the molecular events underlying the folding and assembly of human class I MHC molecules in the ER.
Author Keywords: Class I MHC molecules; Chaperone protein; Endoplasmic reticulum; Class I loading complex; Antigen presentation
Abbreviations: 2m, 2-microglobulin; CNX, calnexin; CRT, calreticulin; ER, endoplasmic reticulum; MHC, major histocompatibility complex; TAP, transporter associated with antigen processing

Article Outline
1. Introduction
2. Quality control in the endoplasmic reticulum
3. The structure of calnexin and calreticulin
4. A molecular mechanism for glycoprotein folding by calnexin, calreticulin, and ERp57
5. The class I heavy chain/2m heterodimer
6. The class I loading complex
7. Molecular and structural insights in the class I loading complex
Acknowledgements
References

1. Introduction
One of the primary functions of the immune system is to protect against infections by controlling the spread and virulence of invading pathogens. Central to maintaining immunity is the cell-surface presentation of antigenic peptides to specific receptors on CD8+ cytotoxic T-lymphocytes by class I major histocompatibility complex (MHC) molecules on infected cells. This cell-mediated recognition process results in the elimination of infected cells and is critically important to maintain immunity against viruses and tumors.
Human class I MHC molecules are composed of a ~44 kDa transmembrane heavy chain glycoprotein, a 12 kDa soluble protein called 2-microglobulin (2m), and a short peptide of 8?"10 residues derived from endogenous proteins as well as virus- and tumor-specific proteins. The class I heavy chain is encoded by three polymorphic genes within the MHC. A large solvent-exposed groove in the class I heavy chain consisting of two long -helices (1 and 2 domains) forms a binding site for antigenic peptides (Bjorkman et al., 1987a). The 3 domain is packed against the underside of the 1 and 2 domains and anchors the class I heavy chain to the cell membrane. The protein 2m is noncovalently associated with the class I heavy chain and is proximal to the cell membrane. The antigenic peptide binds in the groove in an extended conformation and interacts with conserved and polymorphic MHC residues along the binding site through its main- and side-chain atoms. These interactions are important to maintain class I MHC molecules stably assembled ( Parker; Guo; Ruppert; Saito and Bouvier). Some of the peptide side-chains, particularly those from the middle positions of the sequence, are oriented upward towards the solvent. It is this region of the peptide that is presented by class I MHC molecules at the surface of infected cells for recognition by T-cell receptors ( Bjorkman et al., 1987b).
The folding and assembly of class I MHC molecules occurs in the endoplasmic reticulum (ER) and begins with translation of the class I heavy chain which then associates with 2m to form a class I heavy chain/2m heterodimer. This intermediate complex is conformationally unstable and requires the binding of an antigenic peptide to complete the maturation process. In a way that is similar to the folding of other glycoproteins in the ER, maturation of class I MHC molecules is highly dependent on interactions with the lectin chaperones calnexin (CNX) and calreticulin (CRT), and the thiol-dependent oxidoreductase ERp57, to proceed efficiently. However, in a way that is distinct, maturation of class I MHC molecules requires the ad***ional participation of two dedicated proteins called the transporter associated with antigen processing (TAP) and tapasin. Collectively, these ER-resident proteins associate transiently with immature forms of class I MHC molecules until they acquire their native structures and are sufficiently stable to be transported to the cell surface via the secretory pathway.
This review presents the most recent advances made in characterizing the structure of accessory proteins involved in the maturation of human class I MHC molecules in the ER. This knowledge is discussed in the context of our current understanding of the molecular events that define the biogenesis of these molecules.
2. Quality control in the endoplasmic reticulum
Upon co-translational insertion in the ER membrane, class I heavy chain acquires a single Asn-linked oligosaccharide that is processed to the form Glc1Man9GlcNAc2 and serves as a recognition element for the lectin chaperones CNX and CRT (Hammond; Nauseef; Wada; Peterson and Hebert). These homologous lectins are part of a well-characterized system of ER-resident proteins that ensures the structural integrity of glycoproteins prior to their release into the secretory pathway ( Ellgaard and Helenius, 2001). This highly coordinated protein machinery also includes ERp57 ( Oliver and Oliver), to catalyze the formation of native disulfide bonds in glycoprotein substrates, and the enzymes UDP-glucose:glycoprotein glucosyltransferase and glucosidase II ( Parodi and Ellgaard), to regulate interactions between glycoprotein substrates and CNX and CRT. More specifically, these two enzymes act successively to catalyze the ad***ion and removal of the outer glucose residue in the Glc1Man9GlcNAc2 glycan moiety which consequently regulate cycles of substrate binding and release from CNX and CRT. In a way that is similar to the more classical chaperone proteins, CNX and CRT have also been shown to associate with nonnative polypeptide segments of glycoproteins as evidenced by their ability *****ppress the aggregation of unfolded proteins in vivo and in vitro (Saito; Ihara and Danilczyk). Overall, the ER has evolved a quality control system that prevents the transport of misfolded and incompletely folded glycoproteins to the Golgi complex.
3. The structure of calnexin and calreticulin
CNX is a 65 kDa type I membrane protein and CRT is a 46 kDa protein present in a soluble form in the lumen of the ER. Interestingly, CRT shares a high degree of sequence homology with the lumenal domain of CNX. The structure of CRT has been described as an organization of three distinct domains with loosely defined boundaries (Smith and Fliegel): the N-domain (residues 1?"180), the P-domain (residues 181?"290), and the C-domain (residues 291?"400). The N-domain is the most conserved region among species of CRT and shares extensive sequence similarity with CNX. The N-domain also contains a high-capacity, low-affinity zinc binding site (14 mol of zinc/mol of CRT and Kd=0.3 mM) (Khanna et al., 1986) consistent with the presence of His and Cys residues in this region of the protein, although no consensus zinc-binding motif has been identified. The central P-domain contains two distinct proline-rich sequences, the type 1 (17 residues) and type 2 (14 residues) sequences, that repeat in a "111222" arrangement. These sequences are highly conserved in CNX repeating in a "11112222" tandem arrangement. The types 1 and 2 sequences have been shown by NMR and X-ray crystallographic studies to be structural motifs ( Ellgaard and Ellgaard; Schrag et al., 2001). Finally, the C-domain contains a large number of acidic residues clustered primarily within the last 56 residues consistent with the ability of CRT to bind a high molar amounts of calcium ions with low-affinity (~25?"50 mol of calcium/mol of CRT and Kd=1?"2 mM) (Ostwald; Treves and Baksh). This calcium-binding property enables CRT to be one of the major proteins regulating the process of calcium-homeostasis in the ER. The acidic stretch of amino acids present in the C-domain of CRT is not conserved in the lumenal portion of CNX.
The crystal structure of a proteinase K-resistant fragment of the lumenal domain of CNX (residues 47?"468) has now been reported (Schrag et al., 2001). The structure reveals a strikingly asymmetric protein comprising a compact globular core region formed by two series of antiparallel -strands organized into a convex and con**** face from which extends a highly elongated domain. The globular region is formed by residues from both the N- and C-domain and has been described as a jelly-roll fold, whereas the long arm is comprised of residues from the central proline-rich P-domain and adopts an extended -hairpin fold. A putative binding site for monoglucosylated glycans was modeled in the globular region of CNX based on the location of a single bound glucose molecule in the crystal structure. It has been suggested that it may represent the lectin binding site for glycoprotein substrates. A putative binding site for a calcium ion was also located in the globular region of CNX, distant from the carbohydrate binding site. Although the coordination of this calcium ion is incomplete in the crystal structure, some of the residues involved are highly conserved in CRT suggesting that it may correspond to the high-affinity calcium binding site that has been characterized in CRT. No distinguishing structural features that may enable CNX to bind nonnative polypeptide segments have been identified in the crystal structure. Although the spatial organization of the N-, P-, and C-domain of CRT has not yet been described, it has been shown from sedimentation analysis that the protein also adopts a distinctively asymmetric structure in solution consistent with its anomalous position on gel filtration columns ( Bouvier and Li). The hydrodynamic properties of CRT have been attributed to the extended -hairpin fold that the P-domain adopts in solution as determined by NMR studies of a recombinant form of this isolated domain ( Ellgaard and Ellgaard). That the P-domain of CRT adopts a structure in solution that is essentially identical, except for its overall length, to that observed in the crystal structure of CNX is consistent with both proteins having a high degree of sequence similarity in this middle domain. Given their identical lectin specificity and shared chaperone function, it is expected that CRT possesses the same overall topology as that of CNX, as also supported by rotary shadow electron microscopy images of CRT (Mabuchi and Bouvier, unpublished results).
4. A molecular mechanism for glycoprotein folding by calnexin, calreticulin, and ERp57
The elongated molecular shape of CNX and CRT strongly suggests that these proteins are likely to be characterized by high intrinsic structural plasticity, most particularly in the P-domain. Consistent with this view, an analysis of CRT by rotary shadowing electron microscopy revealed that a region of the protein appears highly flexible as evidenced by its pronounced susceptibility to molecular deformations (Mabuchi and Bouvier, unpublished results). Molecular flexibility is consistent with the function of CNX and CRT as chaperone proteins and the underlying requirement to interact dynamically and transiently with many different proteins in the ER. Recent biochemical and NMR studies investigating interactions between an isolated form of the P-domain of CRT and ERp57 point in support of this analysis (Frickel et al., 2002). These studies have shown that the P-domain interacts with ERp57 through several residues located at the tip of the -hairpin fold. The CRT/ERp57 complex was determined to have a 1:1 molar stoichiometry with a dissociation constant Kd=(9±3) M and a first-order exchange rate koff>1000 s-1 at 20 °C. A direct interaction between either CNX or CRT and ERp57 was also established by gel filtration chromatography (Zapun et al., 1998), cross-linking and native gel analysis ( Oliver et al., 1999), and by a solid-phase binding assay ( Leach et al., 2002). These results, combined with the knowledge of the CNX crystal structure, have allowed to propose a molecular mechanism describing how CNX and CRT in concert with ERp57 may facilitate the folding of glycoproteins in the ER ( Frickel et al., 2002). The model is based upon recognizing that the distinct topology of CNX and CRT, in which the elongated P-domain protrudes as a curved extension from the globular core, creates a partially closed environment for glycoprotein substrates to reside in as they undergo productive folding interactions. It has been suggested that glycoprotein substrates could be sequestered into this environment upon recognition of the Glc1Man9GlcNAc2 glycan moiety by the lectin binding site of CNX and CRT. In this structural organization, it has been proposed that the apparent flexibility of the P-domain may provide a mechanism whereby CRT can effectively optimize interactions between ERp57 and glycoprotein substrates to catalyze the formation and rearrangement of intramolecular disulfide bonds. Overall, the molecular model proposed to describe the folding of newly synthesized glycoproteins in the ER is based on the unique structural features of CNX and CRT and on cooperative interactions between these lectin chaperones and ERp57. As mentioned above, the CNX/ERp57 and CRT/ERp57 complexes are part of a system of chaperone proteins and specialized enzymes that act collectively to modulate the folding of glycoproteins in the ER until they fold into their native structures.
5. The class I heavy chain/beta-2m heterodimer
Based on our current understanding, the chaperones CNX and CRT function at different stages of the folding and assembly pathway of human class I MHC molecules in the ER. Upon co-translational insertion in the ER membrane, class I heavy chain interacts with CNX which serves to facilitate the folding of class I heavy chain and stabilize it from aggregation as well as promote its assembly with 2m (Degen; Noessner and Sadasivan). It has been shown that the single Asn-linked glycan at position 86 (1 domain) in class I heavy chain is essential for the initial recognition by CNX, but that interactions with nonnative polypeptide segments are also important as folding progresses ( Zhang et al., 1995). Class I heavy chain has two intrachain disulfide bonds (2 and 3 domains) which are formed at this initial step ( Tector; Tector and Antoniou) presuming that CNX recruits ERp57 to catalyze this process ( Fig. 1). In fact, recent results have suggested that the CNX/ERp57 complex may form prior to its active participation in the assembly of class I MHC molecules (Diedrich et al., 2001) consistent with the general role that this complex plays in the folding of glycoproteins in the ER. Interestingly, the requirement for CNX to associate with class I heavy chain is somewhat unclear since the assembly, transport, and cell-surface expression of class I MHC molecules is unaffected in a mutant cell line lacking CNX ( Scott and Sadasivan). It has been suggested that another ER-resident chaperone, possibly immunoglobulin-binding protein (BiP), may substitute for CNX ( Noessner and Degen). BiP belongs to the Hsp70 family of molecular chaperones and has also been found associated with newly synthesized class I heavy chains in the ER ( Degen and Noessner). Since very few studies aimed at characterizing interactions between BiP and class I heavy chain have been reported, the possibility that CNX and BiP function redundantly in relation to the folding of class I heavy chain remains somewhat elusive. In any case, the CNX/ERp57 complex is thought to facilitate the correct oxidation and folding of class I heavy chains into a conformation compatible with recognition by 2m ( Degen; Noessner; Tector and Tector). Whether CNX dissociates from class I heavy chain prior to or concomitantly with the binding of 2m also remains unclear but, except for one report ( Carreno et al., 1995), results have shown that CNX is absent in the class I heavy chain/2m heterodimer ( Sugita; Zhang; Noessner; Tector; Sadasivan; Tector and Diedrich). The association of class I heavy chain with 2m represents a critical step in the assembly of class I MHC molecules and is likely to involve conformational changes in class I heavy chain. It has been proposed that these conformational changes may be the molecular event that triggers the release of CNX from class I heavy chain ( Degen and Noessner).
The class I heavy chain/2m heterodimer is an intermediate complex that awaits the binding of an antigenic peptide as a final step in completing the assembly of class I MHC molecules. These so-called "empty" class I MHC molecules have been shown to be highly unstable molecules in solution from in vitro reconstitution studies using recombinant class I heavy chains and 2m (Bouvier and Wiley, 1998). Results from biochemical and biophysical studies indicated that the soluble class I heavy chain/2m heterodimer exhibits properties similar to those of molten globules. Remarkably, the peptide binding site (1 and 2 domains) was identified as the most destabilized region, while the 3 domain was shown to be independently stable. This study also revealed allele- and locus-specific differences in the ability of class I heavy chains to associate with 2m and to form soluble class I heavy chain/2m heterodimers. Overall, these results are consistent with studies carried out in cell systems, including mouse cells, indicating that class I heavy chains that are associated with 2m adopt an "open" conformation ( Smith; Carreno; Solheim and Harris). These results are also consistent with the knowledge that these intermediate molecules associate with chaperone proteins in the ER. Indeed, numerous studies have shown that the chaperone CRT interacts with the "open" form of class I heavy chain/2m heterodimer prior to the step of peptide binding ( Sadasivan; Solheim and Tector). Based on experiments carried out in both human and mouse cells, the association between CRT and the class I heavy chain/2m heterodimer is mediated, at least in part, through the Asn-linked glycan at position 86 (1 domain) in class I heavy chain ( Sadasivan; Zhang; Harris and Turnquist).
To this point, our understanding of the folding and assembly of class I MHC molecules indicates that although CNX and CRT are functionally homologous proteins and that both interact directly with the class I heavy chain, these proteins have distinct roles in the maturation process with, most significantly, the recognition of different class I folding intermediates. Furthermore, although it has been shown that the class I heavy chain/2m/CRT heterotrimer is present as an independent complex in the ER (Sadasivan; Solheim and Harris), the vast majority of these molecules are simultaneously assembled with three ad***ional accessory proteins; TAP, tapasin, and ERp57.
Được Milou sửa vào 06:08 ngày 18/06/2003

6. The class I loading complex
The macromolecular complex consisting of the class I heavy chain/2m heterodimer in association with the proteins CRT, TAP, tapasin, and ERp57, is known as the "class I loading complex" to reflect the peptide-receptive nature of this protein-stabilized class I folding intermediate. These interactions stabilize class I heavy chain/2m heterodimers and retain them in the ER until the assembly process is completed by the binding of an antigenic peptide.
TAP is a member of the ATP-binding cassette (ABC) family of transporters. It is formed of the TAP1 and TAP2 subunits, both of which are encoded by genes located within the MHC. Each subunit consists of an N-terminal membrane-spanning domain and a C-terminal cytosolic ABC ATPase domain (Kelly et al., 1992). Given the structural complexity of this integral membrane protein, our current knowledge of its three-dimensional structure is rather limited, except for a recent report describing the crystal structure of the ATPase domain of the TAP1 subunit ( Gaudet and Wiley, 2001). TAP serves as a transporter of short peptides, generated from the degradation of cytosolic proteins by the proteasome, into the lumen of the ER. Our understanding of the mechanism by which TAP binds and transports peptides across the ER membrane is still mostly unclear, except that it has been described as a multistep process consisting of an ATP-independent peptide-binding step that appears to trigger the ATP-dependent transport of peptides into the lumen of the ER ( Androlewicz; van; Abele and Nijenhuis). In ad***ion to its critical role as a peptide transporter, experimental data using both human and mouse cells have indicated that TAP associates with the class I heavy chain/2m heterodimer and shows a marked preference for its "open" rather than "closed" (mature) form ( Ortmann; Suh and Carreno). Since TAP can function as a peptide transporter independently of this association, it was concluded that it may play an active role in the assembly of class I MHC molecules in the ER by, for example, maintaining a pool of free peptides at the site of class I heavy chain/2m heterodimers as a way to possibly enhance the loading process. More recently, the identification of a novel protein, refer to as tapasin, as a component of the class I loading complex ( Ortmann et al., 1997) has complicated the vali***y of this interpretation ( Lehner and Peh). A more accurate view of the functional significance of the association between TAP and the class I heavy chain/2m heterodimer may be to provide a mechanism of retention in the ER to ensure the assembly of high-affinity class I MHC molecules ( Lehner et al., 1998).
Tapasin is a 48 kDa type I membrane protein encoded by a gene located within the MHC and has been characterized as a member of the immunoglobulin superfamily of proteins (Ortmann; Frangoulis; Herberg and Mayer). Tapasin consists of a large N-terminal region localized in lumen of the ER (residues 1?"392), a single transmembrane domain (residues 393?"417), and a short C-terminal cytosolic tail with an ER-retention motif (residues 418?"428). To this date, no X-ray crystallographic or NMR analysis of tapasin have been reported, but the structure of a recombinant form of the lumenal domain of human tapasin has recently been probed using biochemical and biophysical approaches ( Chen et al., 2002). Results from sedimentation analysis revealed that the protein is monomeric in solution with hydrodynamic dimensions suggesting a moderately asymmetric structure. Proteolysis studies showed most notably that the lumenal domain of tapasin is organized into two stable core regions, an N-terminal domain of 9 kDa (residues ~1?"85) and a C-terminal domain of 34 kDa (residues ~94?"392) that are loosely linked together by a more flexible region (residues ~86?"93). This experimental analysis is consistent with an hypothetical three-dimensional model of tapasin in which the first N-terminal ~100 residues have been suggested to form a distinct domain, adopting as yet an unidentified fold, that is connected to a larger C-terminal domain of ~300 residues predicted to resemble a class I heavy chain (Mayer and Klein, 2001). Studies carried out in various cell systems have attributed numerous roles to tapasin in the assembly of class I MHC molecules, including stabilization and retention in the ER of class I heavy chain/2m heterodimers ( Ortmann; Grandea; Grandea; Schoenhals and Barnden), mediating interactions between the class I heavy chain/2m heterodimer and TAP ( Sadasivan et al., 1996), association with TAP in a way that enhances the ability of TAP to transport peptides ( Lehner and Bangia), and e***ing of antigenic peptides ( Li; Sijts; Lewis and Williams) in association with ERp57 ( Dick et al., 2002). Although many aspects on the function of tapasin remain to be clarified, there is sufficient evidence *****ggest that this protein is critically important for the cell-surface expression of class I MHC molecules.
Immunoprecipitation studies using cell lines with specific mutations have been extremely useful in identifying key accessory proteins of the class I loading complex. However, the study of how and in which sequential order these proteins interact with the "open" form of the class I heavy chain/2m heterodimer has been complicated by the cooperative nature of their interactions within the loading complex and consequently the interdependency of their functions (Yu and Bangia). Nevertheless, significant progress has been made in recent years and the overall picture that emerges from these efforts suggests that human and mouse class I loading complexes can be assembled by alternative pathways, each with some variations in the nature and precise sequence of molecular events ( Sadasivan; Solheim; Androlewicz; Diedrich and Harris). These variations appear to reflect, at least in part, differences in the requirement of some class I alleles to interact with CRT, TAP, tapasin, and ERp57 ( Neisig; Peh; Peh; Lehner; Lewis; Turnquist and Turnquist). The molecular basis of these differential dependencies may lie in relation to the class I polymorphism that is largely distributed within the most destabilized region of the class I heavy chain/2m heterodimer ( Bouvier and Wiley, 1998), namely the peptide binding site. Alternatively, it has been suggested ( Lehner and Androlewicz) that the abundance and nature, i.e. high- or low-affinity, of class I-restricted antigenic peptides available in the ER may have a significant influence on the folding and assembly pathway utilized by class I alleles. In summary, maturation of class I MHC molecules in the ER requires the concerted action of several accessory proteins, of which TAP and tapasin are uniquely dedicated to this process. It is clear that some class I alleles, or specific intracellular con***ions, can favor alternative assembly pathways that are more, or less, dependent on interactions with the class I accessory proteins.
===========
Fig. 1. A model showing interactions between calnexin (green) and ERp57 (orange) that illustrates how both proteins cooperate together to modulate the correct oxidation and folding of class I heavy chain (red) in the endoplasmic reticulum (ER). Class I heavy chain is bound to the lectin binding site (N-/C-domain) of calnexin via its single monoglucosylated Asn-linked glycan (blue). The P-domain of calnexin interacts with ERp57 and positions the thiol-dependent oxidoreductase optimally to catalyze disulfide bond formation and rearrangement in class I heavy chain through mixed disulfides. This figure has been adapted from Frickel et al. (2002).
===============
7. Molecular and structural insights in the class I loading complex
To this date, efforts in elucidating structural aspects of the class I loading complex have been primarily focused on characterizing a soluble form of the class I heavy chain/beta-2m heterodimer (Bouvier and Wiley, 1998) and on individual class I accessory proteins ( Bouvier and Stafford, 2000; Ellgaard and Ellgaard; Schrag; Gaudet; Li and Chen) such that the spatial organization of this multiprotein complex is as yet unknown. However, mutational studies in both human and mouse class I heavy chains have identified numerous residues and structural elements that are apparently critical to maintain interactions within the class I loading complex ( Lewis; Harris; Kulig; Yu; Suh and Beissbarth; Harris and Harris; Paquet and Williams, 2002; Turnquist et al., 2002c). Identifying which class I accessory proteins is directly affected by each of these mutations has been somewhat difficult due to the cooperative nature of interactions between the class I accessory proteins.
The putative binding sites can be mapped within an hypothetical structure of a soluble class I heavy chain/2m heterodimer (Fig. 2A), and several observations can be made from this analysis. First, the majority of the sites define a surface along one side of the class I heavy chain that is opposite to that occupied by 2m. This segregation of the sites underlines the important role played by 2m in maintaining class I heavy chain in a conformation that is compatible with recognition by the class I accessory proteins. Second, a significant number of the sites (residues 128?"136) are clustered within a solvent-exposed loop in the 2 domain. Interestingly, proteolysis studies of a recombinant, soluble class I heavy chain/beta-2m heterodimer have identified several sites of cleavage within this particular loop region (Bouvier and Wiley, 1998) thus emphasizing the more solvent-exposed and/or destabilized nature of this region. Consistent with these results, evidence have been provided that this loop forms a binding site for one or more of the class I accessory proteins ( Peace; Lewis; Elliott; Lewis; Yu; Harris and Paquet). Moreover, this particular region of class I heavy chain, including the short -helix in the 2 domain, have been suggested to undergo peptide-induced conformational changes that are responsible for the dissociation of all accessory proteins from the class I heavy chain/2m heterodimer as a final step in the maturation process ( Elliott and Yu). Third, a second cluster of putative interaction sites (residues 219?"233) forms a distinct loop in the membrane proximal 3 domain. It has been suggested that this loop may also represent a binding site for one or more of the class I accessory proteins, including most probably, the protein tapasin ( Harris; Kulig; Suh; Yu and Paquet). Finally, another putative interaction surface in the class I heavy chain/beta-2m heterodimer includes the single Asn-linked glycan at position 86 (1 domain) which is likely to be more specifically critical for interactions with the chaperone CRT owing to its lectin specificity.

Another important point that can be made from Fig. 2A is in relation to the conserved disulfide bond (Cys 101/Cys 164) located in the 2 domain. A recent study has shown that this disulfide bond isomerizes within the loading complex and that the proteins tapasin and ERp57 conjugate together to regulate this process ( Dick et al., 2002). It has been suggested that isomerization of this disulfide bond modulates the conformational state of the peptide binding site during folding and assembly and may serve as a mechanism to e*** the peptide repertoire in favor of high-affinity peptides ( Dick et al., 2002). In view of these results, the region of class I heavy chain that comprises this disulfide bond is likely to form a site of interaction for tapasin and ERp57. More specifically, it has been shown that the free residue Cys 95 in tapasin, located in the flexible linker region of the protein ( Chen et al., 2002), interacts via a labile disulfide bond involving residue Cys 57 located in the N-terminal thioredoxin motif of ERp57 ( Dick et al., 2002). It has been suggested that the C-terminal thioredoxin motif of ERp57 or, possibly, the disulfide-linked residues Cys 7 and Cys 71 located in the N-terminal domain of tapasin ( Chen et al., 2002), are actively engaged in catalyzing the isomerization reaction ( Dick et al., 2002). Collectively these results are consistent with evidence suggesting that the first N-terminal 50 residues of tapasin are essential to stabilize the class I loading complex ( Bangia et al., 1999).
Based on the data presented above, three distinct regions corresponding to putative interaction surfaces for CRT, tapasin, and ERp57, have been mapped in the peptide binding site (Fig. 2B). Although not shown in this figure, but compatible with the suggested spatial organization, the C-terminal domain of tapasin (residues ~94?"392) ( Chen et al., 2002) may interact with the solvent-exposed loop (residues 219?"233) in the 3 domain of class I heavy chain (see Fig. 2A). This would be consistent with the knowledge that specific residues within the C-domain of tapasin, residues 334?"342, are critical for association with class I heavy chain ( Turnquist et al., 2002a). The challenge remains to more precisely define the spatial organization of the class I loading complex possibly through in vitro reconstitution experiments using recombinant, soluble forms of its components in combination with functional, biochemical, and structural analysis. The significance of studying the molecular cell biology of antigen presentation is important not only in relation to activation of CD8+ T-cell and natural killer cell receptors, but also to better understand pathogenic processes that interfere with the biogenesis of class I MHC molecules in the ER.

Acknowledgements
I would like to thank members of my laboratory for contributing to several aspects of the work presented in this review, for preparing Fig. 2, and for their valuable comments on the manuscript. Dr. Lars Ellgaard is also acknowledged for his critical reading of the manuscript. This work is supported by NIH grant AI45070.
==============
Fig. 2. A. An hypothetical structure of the "open" form of soluble class I heavy chain (red)/2m (blue) heterodimer. Sites of interaction in the class I heavy chain that are apparently critical to maintain interactions within the class I loading complex are highlighted in green, including the disulfide bond Cys 101/Cys 164 in the 2 domain. The 1, 2, and 3 domains of class I heavy chain are indicated. The structure is based on the coordinates of HLA-A*0201 (Khan et al., 2000) in which the peptide was omitted; B. A top view of the peptide binding site, as labeled in Fig. 2A, showing putative interaction surfaces for calreticulin (CRT), tapasin, and ERp57. Tapasin, more specifically its N-terminal domain (residues ~1?"85) and linker region (residues ~86?"93), interact with ERp57 in a region of the 2 domain that includes the disulfide bond Cys 101/Cys 164 and the solvent-exposed loop formed by residues 128?"136. Residue Cys 95 of tapasin was found to form a covalent bond with residue Cys 57 in the N-terminal thioredoxin motif of ERp57 ( Dick et al., 2002). In this structural arrangement, the C-terminal domain of tapasin (residues ~94?"392) may interact with residues 219?"233 in the 3 domain of class I heavy chain (see Fig. 2A). CRT is likely to interact in the region of the 1 domain that comprises the single Asn-linked glycan.
================
Concay
Được Milou sửa vào 06:08 ngày 18/06/2003

6. The class I loading complex
The macromolecular complex consisting of the class I heavy chain/2m heterodimer in association with the proteins CRT, TAP, tapasin, and ERp57, is known as the "class I loading complex" to reflect the peptide-receptive nature of this protein-stabilized class I folding intermediate. These interactions stabilize class I heavy chain/2m heterodimers and retain them in the ER until the assembly process is completed by the binding of an antigenic peptide.
TAP is a member of the ATP-binding cassette (ABC) family of transporters. It is formed of the TAP1 and TAP2 subunits, both of which are encoded by genes located within the MHC. Each subunit consists of an N-terminal membrane-spanning domain and a C-terminal cytosolic ABC ATPase domain (Kelly et al., 1992). Given the structural complexity of this integral membrane protein, our current knowledge of its three-dimensional structure is rather limited, except for a recent report describing the crystal structure of the ATPase domain of the TAP1 subunit ( Gaudet and Wiley, 2001). TAP serves as a transporter of short peptides, generated from the degradation of cytosolic proteins by the proteasome, into the lumen of the ER. Our understanding of the mechanism by which TAP binds and transports peptides across the ER membrane is still mostly unclear, except that it has been described as a multistep process consisting of an ATP-independent peptide-binding step that appears to trigger the ATP-dependent transport of peptides into the lumen of the ER ( Androlewicz; van; Abele and Nijenhuis). In ad***ion to its critical role as a peptide transporter, experimental data using both human and mouse cells have indicated that TAP associates with the class I heavy chain/2m heterodimer and shows a marked preference for its "open" rather than "closed" (mature) form ( Ortmann; Suh and Carreno). Since TAP can function as a peptide transporter independently of this association, it was concluded that it may play an active role in the assembly of class I MHC molecules in the ER by, for example, maintaining a pool of free peptides at the site of class I heavy chain/2m heterodimers as a way to possibly enhance the loading process. More recently, the identification of a novel protein, refer to as tapasin, as a component of the class I loading complex ( Ortmann et al., 1997) has complicated the vali***y of this interpretation ( Lehner and Peh). A more accurate view of the functional significance of the association between TAP and the class I heavy chain/2m heterodimer may be to provide a mechanism of retention in the ER to ensure the assembly of high-affinity class I MHC molecules ( Lehner et al., 1998).
Tapasin is a 48 kDa type I membrane protein encoded by a gene located within the MHC and has been characterized as a member of the immunoglobulin superfamily of proteins (Ortmann; Frangoulis; Herberg and Mayer). Tapasin consists of a large N-terminal region localized in lumen of the ER (residues 1?"392), a single transmembrane domain (residues 393?"417), and a short C-terminal cytosolic tail with an ER-retention motif (residues 418?"428). To this date, no X-ray crystallographic or NMR analysis of tapasin have been reported, but the structure of a recombinant form of the lumenal domain of human tapasin has recently been probed using biochemical and biophysical approaches ( Chen et al., 2002). Results from sedimentation analysis revealed that the protein is monomeric in solution with hydrodynamic dimensions suggesting a moderately asymmetric structure. Proteolysis studies showed most notably that the lumenal domain of tapasin is organized into two stable core regions, an N-terminal domain of 9 kDa (residues ~1?"85) and a C-terminal domain of 34 kDa (residues ~94?"392) that are loosely linked together by a more flexible region (residues ~86?"93). This experimental analysis is consistent with an hypothetical three-dimensional model of tapasin in which the first N-terminal ~100 residues have been suggested to form a distinct domain, adopting as yet an unidentified fold, that is connected to a larger C-terminal domain of ~300 residues predicted to resemble a class I heavy chain (Mayer and Klein, 2001). Studies carried out in various cell systems have attributed numerous roles to tapasin in the assembly of class I MHC molecules, including stabilization and retention in the ER of class I heavy chain/2m heterodimers ( Ortmann; Grandea; Grandea; Schoenhals and Barnden), mediating interactions between the class I heavy chain/2m heterodimer and TAP ( Sadasivan et al., 1996), association with TAP in a way that enhances the ability of TAP to transport peptides ( Lehner and Bangia), and e***ing of antigenic peptides ( Li; Sijts; Lewis and Williams) in association with ERp57 ( Dick et al., 2002). Although many aspects on the function of tapasin remain to be clarified, there is sufficient evidence *****ggest that this protein is critically important for the cell-surface expression of class I MHC molecules.
Immunoprecipitation studies using cell lines with specific mutations have been extremely useful in identifying key accessory proteins of the class I loading complex. However, the study of how and in which sequential order these proteins interact with the "open" form of the class I heavy chain/2m heterodimer has been complicated by the cooperative nature of their interactions within the loading complex and consequently the interdependency of their functions (Yu and Bangia). Nevertheless, significant progress has been made in recent years and the overall picture that emerges from these efforts suggests that human and mouse class I loading complexes can be assembled by alternative pathways, each with some variations in the nature and precise sequence of molecular events ( Sadasivan; Solheim; Androlewicz; Diedrich and Harris). These variations appear to reflect, at least in part, differences in the requirement of some class I alleles to interact with CRT, TAP, tapasin, and ERp57 ( Neisig; Peh; Peh; Lehner; Lewis; Turnquist and Turnquist). The molecular basis of these differential dependencies may lie in relation to the class I polymorphism that is largely distributed within the most destabilized region of the class I heavy chain/2m heterodimer ( Bouvier and Wiley, 1998), namely the peptide binding site. Alternatively, it has been suggested ( Lehner and Androlewicz) that the abundance and nature, i.e. high- or low-affinity, of class I-restricted antigenic peptides available in the ER may have a significant influence on the folding and assembly pathway utilized by class I alleles. In summary, maturation of class I MHC molecules in the ER requires the concerted action of several accessory proteins, of which TAP and tapasin are uniquely dedicated to this process. It is clear that some class I alleles, or specific intracellular con***ions, can favor alternative assembly pathways that are more, or less, dependent on interactions with the class I accessory proteins.
===========
Fig. 1. A model showing interactions between calnexin (green) and ERp57 (orange) that illustrates how both proteins cooperate together to modulate the correct oxidation and folding of class I heavy chain (red) in the endoplasmic reticulum (ER). Class I heavy chain is bound to the lectin binding site (N-/C-domain) of calnexin via its single monoglucosylated Asn-linked glycan (blue). The P-domain of calnexin interacts with ERp57 and positions the thiol-dependent oxidoreductase optimally to catalyze disulfide bond formation and rearrangement in class I heavy chain through mixed disulfides. This figure has been adapted from Frickel et al. (2002).
===============
7. Molecular and structural insights in the class I loading complex
To this date, efforts in elucidating structural aspects of the class I loading complex have been primarily focused on characterizing a soluble form of the class I heavy chain/beta-2m heterodimer (Bouvier and Wiley, 1998) and on individual class I accessory proteins ( Bouvier and Stafford, 2000; Ellgaard and Ellgaard; Schrag; Gaudet; Li and Chen) such that the spatial organization of this multiprotein complex is as yet unknown. However, mutational studies in both human and mouse class I heavy chains have identified numerous residues and structural elements that are apparently critical to maintain interactions within the class I loading complex ( Lewis; Harris; Kulig; Yu; Suh and Beissbarth; Harris and Harris; Paquet and Williams, 2002; Turnquist et al., 2002c). Identifying which class I accessory proteins is directly affected by each of these mutations has been somewhat difficult due to the cooperative nature of interactions between the class I accessory proteins.
The putative binding sites can be mapped within an hypothetical structure of a soluble class I heavy chain/2m heterodimer (Fig. 2A), and several observations can be made from this analysis. First, the majority of the sites define a surface along one side of the class I heavy chain that is opposite to that occupied by 2m. This segregation of the sites underlines the important role played by 2m in maintaining class I heavy chain in a conformation that is compatible with recognition by the class I accessory proteins. Second, a significant number of the sites (residues 128?"136) are clustered within a solvent-exposed loop in the 2 domain. Interestingly, proteolysis studies of a recombinant, soluble class I heavy chain/beta-2m heterodimer have identified several sites of cleavage within this particular loop region (Bouvier and Wiley, 1998) thus emphasizing the more solvent-exposed and/or destabilized nature of this region. Consistent with these results, evidence have been provided that this loop forms a binding site for one or more of the class I accessory proteins ( Peace; Lewis; Elliott; Lewis; Yu; Harris and Paquet). Moreover, this particular region of class I heavy chain, including the short -helix in the 2 domain, have been suggested to undergo peptide-induced conformational changes that are responsible for the dissociation of all accessory proteins from the class I heavy chain/2m heterodimer as a final step in the maturation process ( Elliott and Yu). Third, a second cluster of putative interaction sites (residues 219?"233) forms a distinct loop in the membrane proximal 3 domain. It has been suggested that this loop may also represent a binding site for one or more of the class I accessory proteins, including most probably, the protein tapasin ( Harris; Kulig; Suh; Yu and Paquet). Finally, another putative interaction surface in the class I heavy chain/beta-2m heterodimer includes the single Asn-linked glycan at position 86 (1 domain) which is likely to be more specifically critical for interactions with the chaperone CRT owing to its lectin specificity.

Another important point that can be made from Fig. 2A is in relation to the conserved disulfide bond (Cys 101/Cys 164) located in the 2 domain. A recent study has shown that this disulfide bond isomerizes within the loading complex and that the proteins tapasin and ERp57 conjugate together to regulate this process ( Dick et al., 2002). It has been suggested that isomerization of this disulfide bond modulates the conformational state of the peptide binding site during folding and assembly and may serve as a mechanism to e*** the peptide repertoire in favor of high-affinity peptides ( Dick et al., 2002). In view of these results, the region of class I heavy chain that comprises this disulfide bond is likely to form a site of interaction for tapasin and ERp57. More specifically, it has been shown that the free residue Cys 95 in tapasin, located in the flexible linker region of the protein ( Chen et al., 2002), interacts via a labile disulfide bond involving residue Cys 57 located in the N-terminal thioredoxin motif of ERp57 ( Dick et al., 2002). It has been suggested that the C-terminal thioredoxin motif of ERp57 or, possibly, the disulfide-linked residues Cys 7 and Cys 71 located in the N-terminal domain of tapasin ( Chen et al., 2002), are actively engaged in catalyzing the isomerization reaction ( Dick et al., 2002). Collectively these results are consistent with evidence suggesting that the first N-terminal 50 residues of tapasin are essential to stabilize the class I loading complex ( Bangia et al., 1999).
Based on the data presented above, three distinct regions corresponding to putative interaction surfaces for CRT, tapasin, and ERp57, have been mapped in the peptide binding site (Fig. 2B). Although not shown in this figure, but compatible with the suggested spatial organization, the C-terminal domain of tapasin (residues ~94?"392) ( Chen et al., 2002) may interact with the solvent-exposed loop (residues 219?"233) in the 3 domain of class I heavy chain (see Fig. 2A). This would be consistent with the knowledge that specific residues within the C-domain of tapasin, residues 334?"342, are critical for association with class I heavy chain ( Turnquist et al., 2002a). The challenge remains to more precisely define the spatial organization of the class I loading complex possibly through in vitro reconstitution experiments using recombinant, soluble forms of its components in combination with functional, biochemical, and structural analysis. The significance of studying the molecular cell biology of antigen presentation is important not only in relation to activation of CD8+ T-cell and natural killer cell receptors, but also to better understand pathogenic processes that interfere with the biogenesis of class I MHC molecules in the ER.

Acknowledgements
I would like to thank members of my laboratory for contributing to several aspects of the work presented in this review, for preparing Fig. 2, and for their valuable comments on the manuscript. Dr. Lars Ellgaard is also acknowledged for his critical reading of the manuscript. This work is supported by NIH grant AI45070.
==============
Fig. 2. A. An hypothetical structure of the "open" form of soluble class I heavy chain (red)/2m (blue) heterodimer. Sites of interaction in the class I heavy chain that are apparently critical to maintain interactions within the class I loading complex are highlighted in green, including the disulfide bond Cys 101/Cys 164 in the 2 domain. The 1, 2, and 3 domains of class I heavy chain are indicated. The structure is based on the coordinates of HLA-A*0201 (Khan et al., 2000) in which the peptide was omitted; B. A top view of the peptide binding site, as labeled in Fig. 2A, showing putative interaction surfaces for calreticulin (CRT), tapasin, and ERp57. Tapasin, more specifically its N-terminal domain (residues ~1?"85) and linker region (residues ~86?"93), interact with ERp57 in a region of the 2 domain that includes the disulfide bond Cys 101/Cys 164 and the solvent-exposed loop formed by residues 128?"136. Residue Cys 95 of tapasin was found to form a covalent bond with residue Cys 57 in the N-terminal thioredoxin motif of ERp57 ( Dick et al., 2002). In this structural arrangement, the C-terminal domain of tapasin (residues ~94?"392) may interact with residues 219?"233 in the 3 domain of class I heavy chain (see Fig. 2A). CRT is likely to interact in the region of the 1 domain that comprises the single Asn-linked glycan.
================
Concay
Được Milou sửa vào 06:08 ngày 18/06/2003

References
Abele and Tampe, 1999. R. Abele and R. Tampe , Function of the transport complex TAP in cellular immune recognition. Biochim. Biophys. Acta 1461 (1999), pp. 405â?"419. Abstract | Full Text + Links | PDF (779 K)
Androlewicz, 1999. M.J. Androlewicz , The role of tapasin in MHC class I antigen assembly. Immunol. Res. 20 (1999), pp. 79â?"88. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Androlewicz and Cresswell, 1994. M.J. Androlewicz and P. Cresswell , Human-transporter associated with antigen processing possesses a promiscuous peptide-binding site. Immunity 1 (1994), pp. 7â?"14. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | $Order Document
Antoniou et al., 2002. A.N. Antoniou, S. Ford, M. Alphey, A. Osborne, T. Elliott and S.J. Powis , The oxidoreductase ERp57 efficiently reduces partially folded in preference to fully folded MHC class I molecules. EMBO J. 21 (2002), pp. 2655â?"2663. Abstract-EMBASE | Abstract-Elsevier BIOBASE | Abstract-BIOTECHNOBASE | Abstract-MEDLINE | $Order Document | Full Text via CrossRef
Baksh and Michalak, 1991. S. Baksh and M. Michalak , Expression of calreticulin in Escherichia coli and identification of its Ca2+ binding domains. J. Biol. Chem. 266 (1991), pp. 21458â?"21465. Abstract-MEDLINE | Abstract-BIOTECHNOBASE | Abstract-EMBASE | $Order Document
Bangia et al., 1999. N. Bangia, P.J. Lehner, E.A. Hughes, M. Surman and P. Cresswell , The N-terminal region of tapasin is required to stabilize the MHC class I loading complex. Eur. J. Immunol. 29 (1999), pp. 1858â?"1870. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | $Order Document | Full Text via CrossRef
Barnden et al., 2000. M.J. Barnden, A.W. Purcell, J.J. Gorman and J. McCluskey , Tapasin-mediated retention and optimization of peptide ligands during the assembly of class I molecules. J. Immunol. 165 (2000), pp. 322â?"330. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract-BIOTECHNOBASE | Abstract-MEDLINE | $Order Document
Beissbarth et al., 2000. T. Beissbarth, J. Sun, P.B. Kavathas and B. Ortmann , Increased efficiency of folding and peptide loading of mutant MHC class I molecules. Eur. J. Immunol. 30 (2000), pp. 1203â?"1213. Abstract-BIOTECHNOBASE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | Abstract-MEDLINE | $Order Document
Bjorkman et al., 1987a. P.J. Bjorkman, M.A. Saper, B. Samraoui, W.S. Bennett, J.L. Strominger and D.C. Wiley , Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329 (1987a), pp. 506â?"512. Abstract-INSPEC | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Bjorkman et al., 1987b. P.J. Bjorkman, M.A. Saper, B. Samraoui, W.S. Bennett, J.L. Strominger and D.C. Wiley , The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329 (1987b), pp. 512â?"518. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Bouvier and Wiley, 1994. M. Bouvier and D.C. Wiley , Importance of peptide amino acid and carboxyl termini to the stability of MHC molecules. Science 265 (1994), pp. 398â?"402. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Bouvier and Wiley, 1998. M. Bouvier and D.C. Wiley , Structural characterization of a soluble and partially folded class I major histocompatibility/2m heterodimer. Nat. Struct. Biol. 5 (1998), pp. 377â?"382.
Bouvier and Stafford, 2000. M. Bouvier and W.F. Stafford , Probing the three-dimensional structure of human calreticulin. Biochemistry 39 (2000), pp. 14950â?"14959. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | $Order Document | Full Text via CrossRef
Carreno et al., 1995. B.M. Carreno, J.C. Solheim, M. Harris, I. Stroynowski, J.M. Connolly and T.H. Hansen , TAP associates with a unique class I conformation, whereas calnexin associates with multiple class I forms in mouse and man. J. Immunol. 155 (1995), pp. 4726â?"4733. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | Abstract-BIOTECHNOBASE | $Order Document
Chen et al., 2002. Chen, M., Stafford, W.F., Diedrich, G., Khan, A., Bouvier, M., 2002. A characterization of the lumenal region of human tapasin reveals the presence of two structural domains. Biochemistry 41, 14539â?"14545.
Danilczyk and Williams, 2001. U.G. Danilczyk and D.B. Williams , The lectin-chaperone calnexin utilizes polypeptide-based interactions to associate with many of its substrates in vivo. J. Biol. Chem. 276 (2001), pp. 25532â?"25540. Abstract-MEDLINE | $Order Document | Full Text via CrossRef
Degen et al., 1992. E. Degen, M.F. Cohen-Doyle and D.B. Williams , Efficient dissociation of the p88 chaperone from major histocompatibility complex class I molecules requires both 2-microglobulin and peptide. J. Exp. Med. 175 (1992), pp. 1653â?"1661. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Dick et al., 2002. T.P. Dick, N. Bangia, D.R. Peaper and P. Cresswell , Disulfide bond isomerization and the assembly of MHC class I-peptide complexes. Immunity 16 (2002), pp. 87â?"98. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Diedrich et al., 2001. G. Diedrich, N. Bangia, M. Pan and P. Cresswell , A role for calnexin in the assembly of the MHC class I loading complex in the endoplasmic reticulum. J. Immunol. 166 (2001), pp. 1703â?"1709. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-BIOTECHNOBASE | Abstract-EMBASE | $Order Document
Ellgaard and Helenius, 2001. L. Ellgaard and A. Helenius , ER quality control: towards an understanding at the molecular level. Curr. Opin. Cell Biol. 13 (2001), pp. 431â?"437. SummaryPlus | Full Text + Links | PDF (85 K)
Ellgaard et al., 2001a. L. Ellgaard, R. Riek, D. Braun, T. Hermann, A. Helenius and K. Wuthrich , Three-dimensional structure topology of the calreticulin P-domain based on NMR assignment. FEBS Lett. 488 (2001a), pp. 69â?"73. SummaryPlus | Full Text + Links | PDF (306 K)
Ellgaard et al., 2001b. L. Ellgaard, R. Riek, T. Hermann, P. Guntert, D. Braun, A. Helenius and K. Wuthrich , NMR structure of the calreticulin P-domain. Proc. Natl. Acad. Sci. U.S.A. 98 (2001b), pp. 3133â?"3138. Abstract-EMBASE | Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-Elsevier BIOBASE | $Order Document | Full Text via CrossRef
Elliott, 1997. T. Elliott , How does TAP associate with MHC class I molecules?. Immunol. Today 18 (1997), pp. 375â?"379. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Fliegel et al., 1989. L. Fliegel, K. Burns, D.H. MacLennan, R.A.F. Reithmeier and M. Michalak , Peripheral membrane proteins of sarcoplasmic and endoplasmic reticulum. Comparison of carboxyl-terminal amino acid sequences. J. Biol. Chem. 264 (1989), pp. 21522â?"21528. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Frangoulis et al., 1999. B. Frangoulis, I. Park, F. Guillemot, V. Severac, C. Auffray and R. Zoorob , Identification of the Tapasin gene in the chicken major histocompatibility complex. Immunogenetics 49 (1999), pp. 328â?"337. Abstract-BIOTECHNOBASE | Abstract-EMBASE | Abstract-MEDLINE | $Order Document | Full Text via CrossRef
Frickel et al., 2002. E.-M. Frickel, R. Riek, L. Jelesarov, A. Helenius, K. Wutrich and L. Ellgaard , TROSY-NMR reveals interaction between ERp57 and the tip of the calreticulin P-domain. Proc. Natl. Acad. Sci. U.S.A. 99 (2002), pp. 1954â?"1959. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | $Order Document | Full Text via CrossRef
Gaudet and Wiley, 2001. R. Gaudet and D.C. Wiley , Structure of the ABC ATPase domain of human TAP1, the transporter associated with antigen processing. EMBO J. 20 (2001), pp. 4964â?"4972. Abstract-MEDLINE | Abstract-EMBASE | Abstract-BIOTECHNOBASE | Abstract-Elsevier BIOBASE | $Order Document | Full Text via CrossRef
Grandea et al., 1997. A.G. Grandea, III, P.J. Lehner, P. Cresswell and T. Spies , Regulation of MHC class I heterodimer stability and interaction with TAP by tapasin. Immunogenetics 46 (1997), pp. 477â?"483. Abstract-MEDLINE | Abstract-BIOTECHNOBASE | $Order Document
Grandea et al., 2000. A.G. Grandea, III, T.N. Golovina, S.E. Hamilton, V. Sriram, T. Spies, R.R. Brutkiewicz, J.T. Harty, L.C. Eisenlohr and L. Van Kaer , Impaired assembly yet normal trafficking of MHC class I molecules in tapasin mutant mice. Immunity 13 (2000), pp. 213â?"221.
Guo et al., 1993. H.C. Guo, D.R. Madden, M.L. Silver, T.S. Jardetzky, J.C. Gorga, J.L. Strominger and D.C. Wiley , Comparisons of the P2 specificity pocket in three human histocompatibility antigens: HLA-A*6801, HLA-A*0201, and HLA-B*2705. Proc. Natl. Acad. Sci. U.S.A. 90 (1993), pp. 8053â?"8057. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Hammond et al., 1994. C. Hammond, I. Braakman and A. Helenius , Role of N-linked oligosaccharide recognition, glucose trimming, and calnexin in glycoprotein folding and quality control. Proc. Natl. Acad. Sci. U.S.A. 91 (1994), pp. 913â?"917. Abstract-BIOTECHNOBASE | Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Harris et al., 1998. M.R. Harris, Y.Y.L. Yu, C.S. Kindle, T.H. Hansen and J.C. Solheim , Calreticulin and calnexin interact with different protein and glycan determinants during the assembly of MHC class I. J. Immunol. 160 (1998), pp. 5404â?"5409. Abstract-BIOTECHNOBASE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract-MEDLINE | $Order Document
Concay

References
Abele and Tampe, 1999. R. Abele and R. Tampe , Function of the transport complex TAP in cellular immune recognition. Biochim. Biophys. Acta 1461 (1999), pp. 405â?"419. Abstract | Full Text + Links | PDF (779 K)
Androlewicz, 1999. M.J. Androlewicz , The role of tapasin in MHC class I antigen assembly. Immunol. Res. 20 (1999), pp. 79â?"88. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Androlewicz and Cresswell, 1994. M.J. Androlewicz and P. Cresswell , Human-transporter associated with antigen processing possesses a promiscuous peptide-binding site. Immunity 1 (1994), pp. 7â?"14. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | $Order Document
Antoniou et al., 2002. A.N. Antoniou, S. Ford, M. Alphey, A. Osborne, T. Elliott and S.J. Powis , The oxidoreductase ERp57 efficiently reduces partially folded in preference to fully folded MHC class I molecules. EMBO J. 21 (2002), pp. 2655â?"2663. Abstract-EMBASE | Abstract-Elsevier BIOBASE | Abstract-BIOTECHNOBASE | Abstract-MEDLINE | $Order Document | Full Text via CrossRef
Baksh and Michalak, 1991. S. Baksh and M. Michalak , Expression of calreticulin in Escherichia coli and identification of its Ca2+ binding domains. J. Biol. Chem. 266 (1991), pp. 21458â?"21465. Abstract-MEDLINE | Abstract-BIOTECHNOBASE | Abstract-EMBASE | $Order Document
Bangia et al., 1999. N. Bangia, P.J. Lehner, E.A. Hughes, M. Surman and P. Cresswell , The N-terminal region of tapasin is required to stabilize the MHC class I loading complex. Eur. J. Immunol. 29 (1999), pp. 1858â?"1870. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | $Order Document | Full Text via CrossRef
Barnden et al., 2000. M.J. Barnden, A.W. Purcell, J.J. Gorman and J. McCluskey , Tapasin-mediated retention and optimization of peptide ligands during the assembly of class I molecules. J. Immunol. 165 (2000), pp. 322â?"330. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract-BIOTECHNOBASE | Abstract-MEDLINE | $Order Document
Beissbarth et al., 2000. T. Beissbarth, J. Sun, P.B. Kavathas and B. Ortmann , Increased efficiency of folding and peptide loading of mutant MHC class I molecules. Eur. J. Immunol. 30 (2000), pp. 1203â?"1213. Abstract-BIOTECHNOBASE | Abstract-EMBASE | Abstract-Elsevier BIOBASE | Abstract-MEDLINE | $Order Document
Bjorkman et al., 1987a. P.J. Bjorkman, M.A. Saper, B. Samraoui, W.S. Bennett, J.L. Strominger and D.C. Wiley , Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329 (1987a), pp. 506â?"512. Abstract-INSPEC | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Bjorkman et al., 1987b. P.J. Bjorkman, M.A. Saper, B. Samraoui, W.S. Bennett, J.L. Strominger and D.C. Wiley , The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329 (1987b), pp. 512â?"518. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Bouvier and Wiley, 1994. M. Bouvier and D.C. Wiley , Importance of peptide amino acid and carboxyl termini to the stability of MHC molecules. Science 265 (1994), pp. 398â?"402. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Bouvier and Wiley, 1998. M. Bouvier and D.C. Wiley , Structural characterization of a soluble and partially folded class I major histocompatibility/2m heterodimer. Nat. Struct. Biol. 5 (1998), pp. 377â?"382.
Bouvier and Stafford, 2000. M. Bouvier and W.F. Stafford , Probing the three-dimensional structure of human calreticulin. Biochemistry 39 (2000), pp. 14950â?"14959. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | $Order Document | Full Text via CrossRef
Carreno et al., 1995. B.M. Carreno, J.C. Solheim, M. Harris, I. Stroynowski, J.M. Connolly and T.H. Hansen , TAP associates with a unique class I conformation, whereas calnexin associates with multiple class I forms in mouse and man. J. Immunol. 155 (1995), pp. 4726â?"4733. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | Abstract-BIOTECHNOBASE | $Order Document
Chen et al., 2002. Chen, M., Stafford, W.F., Diedrich, G., Khan, A., Bouvier, M., 2002. A characterization of the lumenal region of human tapasin reveals the presence of two structural domains. Biochemistry 41, 14539â?"14545.
Danilczyk and Williams, 2001. U.G. Danilczyk and D.B. Williams , The lectin-chaperone calnexin utilizes polypeptide-based interactions to associate with many of its substrates in vivo. J. Biol. Chem. 276 (2001), pp. 25532â?"25540. Abstract-MEDLINE | $Order Document | Full Text via CrossRef
Degen et al., 1992. E. Degen, M.F. Cohen-Doyle and D.B. Williams , Efficient dissociation of the p88 chaperone from major histocompatibility complex class I molecules requires both 2-microglobulin and peptide. J. Exp. Med. 175 (1992), pp. 1653â?"1661. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Dick et al., 2002. T.P. Dick, N. Bangia, D.R. Peaper and P. Cresswell , Disulfide bond isomerization and the assembly of MHC class I-peptide complexes. Immunity 16 (2002), pp. 87â?"98. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Diedrich et al., 2001. G. Diedrich, N. Bangia, M. Pan and P. Cresswell , A role for calnexin in the assembly of the MHC class I loading complex in the endoplasmic reticulum. J. Immunol. 166 (2001), pp. 1703â?"1709. Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-BIOTECHNOBASE | Abstract-EMBASE | $Order Document
Ellgaard and Helenius, 2001. L. Ellgaard and A. Helenius , ER quality control: towards an understanding at the molecular level. Curr. Opin. Cell Biol. 13 (2001), pp. 431â?"437. SummaryPlus | Full Text + Links | PDF (85 K)
Ellgaard et al., 2001a. L. Ellgaard, R. Riek, D. Braun, T. Hermann, A. Helenius and K. Wuthrich , Three-dimensional structure topology of the calreticulin P-domain based on NMR assignment. FEBS Lett. 488 (2001a), pp. 69â?"73. SummaryPlus | Full Text + Links | PDF (306 K)
Ellgaard et al., 2001b. L. Ellgaard, R. Riek, T. Hermann, P. Guntert, D. Braun, A. Helenius and K. Wuthrich , NMR structure of the calreticulin P-domain. Proc. Natl. Acad. Sci. U.S.A. 98 (2001b), pp. 3133â?"3138. Abstract-EMBASE | Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-Elsevier BIOBASE | $Order Document | Full Text via CrossRef
Elliott, 1997. T. Elliott , How does TAP associate with MHC class I molecules?. Immunol. Today 18 (1997), pp. 375â?"379. Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Fliegel et al., 1989. L. Fliegel, K. Burns, D.H. MacLennan, R.A.F. Reithmeier and M. Michalak , Peripheral membrane proteins of sarcoplasmic and endoplasmic reticulum. Comparison of carboxyl-terminal amino acid sequences. J. Biol. Chem. 264 (1989), pp. 21522â?"21528. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Frangoulis et al., 1999. B. Frangoulis, I. Park, F. Guillemot, V. Severac, C. Auffray and R. Zoorob , Identification of the Tapasin gene in the chicken major histocompatibility complex. Immunogenetics 49 (1999), pp. 328â?"337. Abstract-BIOTECHNOBASE | Abstract-EMBASE | Abstract-MEDLINE | $Order Document | Full Text via CrossRef
Frickel et al., 2002. E.-M. Frickel, R. Riek, L. Jelesarov, A. Helenius, K. Wutrich and L. Ellgaard , TROSY-NMR reveals interaction between ERp57 and the tip of the calreticulin P-domain. Proc. Natl. Acad. Sci. U.S.A. 99 (2002), pp. 1954â?"1959. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | $Order Document | Full Text via CrossRef
Gaudet and Wiley, 2001. R. Gaudet and D.C. Wiley , Structure of the ABC ATPase domain of human TAP1, the transporter associated with antigen processing. EMBO J. 20 (2001), pp. 4964â?"4972. Abstract-MEDLINE | Abstract-EMBASE | Abstract-BIOTECHNOBASE | Abstract-Elsevier BIOBASE | $Order Document | Full Text via CrossRef
Grandea et al., 1997. A.G. Grandea, III, P.J. Lehner, P. Cresswell and T. Spies , Regulation of MHC class I heterodimer stability and interaction with TAP by tapasin. Immunogenetics 46 (1997), pp. 477â?"483. Abstract-MEDLINE | Abstract-BIOTECHNOBASE | $Order Document
Grandea et al., 2000. A.G. Grandea, III, T.N. Golovina, S.E. Hamilton, V. Sriram, T. Spies, R.R. Brutkiewicz, J.T. Harty, L.C. Eisenlohr and L. Van Kaer , Impaired assembly yet normal trafficking of MHC class I molecules in tapasin mutant mice. Immunity 13 (2000), pp. 213â?"221.
Guo et al., 1993. H.C. Guo, D.R. Madden, M.L. Silver, T.S. Jardetzky, J.C. Gorga, J.L. Strominger and D.C. Wiley , Comparisons of the P2 specificity pocket in three human histocompatibility antigens: HLA-A*6801, HLA-A*0201, and HLA-B*2705. Proc. Natl. Acad. Sci. U.S.A. 90 (1993), pp. 8053â?"8057. Abstract-BIOTECHNOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Hammond et al., 1994. C. Hammond, I. Braakman and A. Helenius , Role of N-linked oligosaccharide recognition, glucose trimming, and calnexin in glycoprotein folding and quality control. Proc. Natl. Acad. Sci. U.S.A. 91 (1994), pp. 913â?"917. Abstract-BIOTECHNOBASE | Abstract-Elsevier BIOBASE | Abstract-MEDLINE | Abstract-EMBASE | $Order Document
Harris et al., 1998. M.R. Harris, Y.Y.L. Yu, C.S. Kindle, T.H. Hansen and J.C. Solheim , Calreticulin and calnexin interact with different protein and glycan determinants during the assembly of MHC class I. J. Immunol. 160 (1998), pp. 5404â?"5409. Abstract-BIOTECHNOBASE | Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract-MEDLINE | $Order Document
Concay