HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 15/8/1448 00-0720fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by LINDQUIST, J. A. Articles by TROWSDALE, J. Search for Related Content PubMed PubMed Citation Articles by LINDQUIST, J. A. Articles by TROWSDALE, J.

ER60/ERp57 forms disulfide-bonded intermediates with MHC class I heavy chain¹

Department of Pathology, University of Cambridge, Cambridge CB2 1QP, U.K.; and
^* Department of Molecular Immunology, German Cancer Research Center, 69120 Heidelberg, Germany

²Correspondence: Department of Pathology, University of Cambridge, Cambridge CB2 1QP, UK. E-mail: JL255{at}mole.bio.cam.ac.uk

SPECIFIC AIMS

In this study, we investigated the role of ER60 (ERp57/GRP58) in the formation of disulfide bonds within the major histocompatibility (MHC) class I heavy chain. We also examined the role of calnexin in recruiting ER60 into early folding complexes with the nascent class I heavy chain (HC).

PRINCIPAL FINDINGS

1. ER60/ERp57 forms disulfide-bonded intermediates with class I heavy chain
Mouse EL4 cells were metabolically labeled for 15 min with [³⁵S] cysteine/methionine. After labeling, the cells were washed in PBS containing N-ethyl-maleimide (NEM), a membrane-permeable alkylating agent, and lysed in 1% digitonin lysis buffer containing NEM to trap unreacted cysteines. Protein complexes containing ER60 were isolated by immunoprecipitation using anti-ER60 sera. To identify a disulfide-bonded intermediate, the complexes were resolved on a 2-dimensional gel system using nonreducing gels in the first dimension. Proteins were resolved on 8% polyacrylamide gels in capillary tubes. The gels were then extruded and equilibrated in buffer containing 100 mM dithiothreitol (DTT). The tube gels were overlaid onto 12% slab gels and resolved as normal SDS-PAGE. Disulfide-bonded complexes should resolve as the sum of their molecular weights in the first dimension and, when reduced (in the second dimension), would be visible as ‘spots’ running off the diagonal. These gels show that the majority of ER60 and HC molecules resolve as monomers running on the diagonal; under longer exposure, however, one can observe the appearance of spots corresponding in molecular weight to both ER60 and HC that are not seen in the negative control. These results indicate that ER60 does indeed form a covalent (S-S) intermediate with HC during the folding reaction, since the addition of DTT causes the proteins to resolve independently of one another in the second dimension. Quantification of this spot revealed that 1% of the total ER60 is disulfide bonded to heavy chain.

To confirm that the spot migrating at 45 kDa was indeed HC, we performed the reciprocal experiment using human cell lines, as the necessary reagents were not available for mouse. The T cell lines T1 (TAP-positive) or T2 (TAP-negative) were incubated with anti-class I antibodies (W6/32 and HC10) at +4°C to identify surface class I molecules. After washing to remove unbound antibody, the cells were lysed and the surface class I molecules depleted. This was deemed necessary as surface class I account for >90% of the total class I. The remaining class I complexes, consisting predominantly of newly synthesized molecules, were subsequently isolated using HC10, which recognizes an epitope in the 3 domain of class I molecules lacking ß₂m. Lysis was performed using Nonidet P-40 in place of digitonin to reduce the amount of noncovalently associated protein in the precipitate. Proteins were then resolved on 8% polyacrylamide gels under nonreducing conditions and blotted onto nitrocellulose membranes. After blocking, ER60-HC complexes were detected using affinity-purified [¹²⁵I]-anti-ER60 sera (Fig. 1A ). Covalently linked ER60-HC complexes were detected at 105 kDa in both T1 and T2 cells, as indicated. No band was observed at the corresponding molecular weight in either the isotype control or in antibody alone, demonstrating that this indeed was a specific interaction. Noncovalently associated ER60 was also observed.

View larger version (41K): [in this window] [in a new window]   Figure 1. Identification of covalently linked ER60-HC complexes by Western blotting. A) Early folding intermediates were immunoprecipitates from Nonidet P-40 lysates of T1 and T2 cells using the monoclonal antibody HC10, which recognizes HC lacking ß2m, and were resolved on 8% polyacrylamide gels [ProSieve®, FMC Bioproducts] under nonreducing conditions. The positions of the intact immunoglobulin (Ig) and the ER60-HC dimer are indicated. Molecular masses are in kDa. B) Late assembly complexes immunoprecipitated from digitonin lysates of T1 and T2 cells using an antibody specific for TAP2. Digitonin was used to maintain the stability of proteins within the complex. Proteins were resolved as in panel A.

To see whether these intermediates could be found in late assembly complexes, we performed anti-TAP immunoprecipitates. When resolved on nonreducing gels and probed with [¹²⁵I]-anti-ER60 (Fig. 1B ), we observed a band corresponding to the ER60-HC dimer in the TAP-positive cells (T1), but not the TAP-negative cells (T2). Quantification of this band shows that the ER60-HC dimer accounts for 33% of the TAP-associated ER60. Thus, we are able to isolate covalent ER60-HC dimers that form early in the class I assembly pathway, before ß₂m binds, and that persist into late assembly complexes with TAP.

2. ER60/ERp57 requires calnexin for recruitment to heavy chain
It had previously been demonstrated that ERp57 (ER60) and calnexin could bind one another independent of substrate. Since the binding of ER60 to nascent HC was sensitive to the presence of properly trimmed oligosaccharides (i.e., binding was inhibited by treatment with castanospermine) and could not be temporally distinguished from the binding of calnexin, we hypothesized that ER60 might require calnexin for recruitment to the nascent HC. To address this question, we used the calnexin-negative cell line CEM.NKR. These cells possess normal levels of ER60 and calreticulin but are devoid of calnexin. The loss of calnexin does not affect the expression of class I in these cells.

To see whether the loss of calnexin affected the association of ER60 with HC, we performed anti-class I immunoprecipitates using the monoclonal antibody HC10, which recognizes HC lacking ß₂m. Figure 2A , B clearly shows that in the presence of calnexin (CEM) ER60 is associated with HC, but ER60 is not found in the absence of calnexin (NKR). However, when calnexin is restored to these cells by transfection (1B9), the association of ER60 is also restored. These results clearly indicate that ER60 depends on calnexin for its recruitment to newly synthesized HC. The absence of ER60 in the NKR lane rules out the possibility that ER60 is binding to the immunoprecipitating antibody, and the antibody alone control rules out cross-reactivity of the secondary reagent. Another protein that cross-reacts with the anti-ER60 sera was observed at 72 kDa (Fig. 2B ). This protein is almost certainly ERp72, another member of the thiol-reductase family, which shares 42% sequence identity with ER60 (ERp57). The regions of homology are particularly strong around the thioredoxin domains, which contain the CGHC motifs.

View larger version (43K): [in this window] [in a new window]   Figure 2. Western blots of anti-HC immunoprecipitates probed with anti-ER60 sera. Cells were lysed in 1% digitonin lysis buffer and early folding complexes were isolated using the antibody HC10. A) To achieve maximum resolution between ER60 and the immunoglobulin heavy chain (IgH), samples were resolved on an 8% gel using an extended vertical gel system (24 cmx 18 cm). The region of the gel containing proteins >66 kDa is not presented. B) A similar experiment resolved on a 10% gel using a standard vertical gel apparatus (16 cmx 18 cm). The positions of ER60, ERp72, and IgH are indicated.

CONCLUSIONS

The formation of disulfide bonds is important for the folding and stability of many proteins, particularly those exposed to the extracellular environment like MHC class I. The importance of disulfide bonds is evident by their conservation not only among MHC molecules, but throughout the immunoglobulin (Ig) superfamily. The isolation of disulfide-bonded HC-ER60 dimers is the first example of a disulfide-bonded intermediate to be isolated from native ER proteins and confirms ER60’s involvement in maintaining disulfide bonds during class I assembly. Others have claimed to show this by using an in vitro translation system, but no data have been presented to show an actual involvement of ERp57 (ER60) in disulfide bond formation. Rather, it is inferred that ERp57 may be involved based on the kinetics of its association with HC.

Recently, Molinari and Helenius [Nature (1999) vol. 402, pp. 90–93] identified disulfide-bonded intermediates of PDI and ERp57 (ER60) with the viral glycoproteins E1 and p62 during Semliki Forest virus infection in mammalian cells, using techniques similar to those applied in this study. The independent identification confirms that these are indeed reaction intermediates, as evident by their relatively low abundance, less than 1% of the total protein.

Although it is argued that disulfide bonds are not essential for stability in the folding of Ig domains, the importance of correctly formed disulfide bonds for the assembly of MHC class I heavy chain has clearly been demonstrated. Mutation of the disulfide bond within either the 2 or the 3 domain abrogates cell surface expression and leads to the accumulation of misfolded HC within the ER, demonstrating the importance of correct disulfide bond formation to the stability and assembly of MHC class I molecules.

The finding that disulfide-bonded ER60-HC dimers are TAP associated demonstrates that not all TAP-associated HCs have correctly formed disulfide bonds. That these molecules should account for one-third of the TAP-associated ER60 was quite unexpected, as they represent a relatively small percentage of the total ER60. The implications of this result are unclear. These molecules may represent either a population of HC associated to TAP in the absence of ß₂m, as previously observed in a ß₂m-negative mouse cell line, or a subset of class I molecules that require more assistance in folding. The latter possibility is interesting to consider. Do some class I molecules need to remain partially folded in order to bind peptide? This speculation would be in line with the observation that the 12 domain undergoes a conformation change on the binding of peptide and that some class I molecules escape the ER quality control machinery by binding low-affinity peptides.

Another possibility is that these molecules represent class I molecules with an additional unpaired cysteine, such as HLA-B27. Unpaired cysteines can lead to unwanted cross-linking between the HC and other proteins, as has been observed for several mouse class I alleles. In humans, the cross-linking of HLA-B27 to form HC homodimers has been documented. These aberrant molecules have been implicated in the disease association between HLA-B27 and ankylosing spondylitis.

A final possibility is that TAP-associated ER60-HC dimers represent MHC class I molecules that are in the process of being unfolded. It is known that class I molecules that fail to fold properly are transported from the ER into the cytosol for degradation by the proteasome. However, for this to occur the disulfide bonds within the HC must first be broken.

Having shown that ER60’s recruitment into the class I assembly pathway is dependent on calnexin, we now propose a revised model for MHC class I assembly (Fig. 3 ). This model incorporates several recent observations, among which are the existence of ER60:calnexin complexes in the absence of substrate, the presence of both ER60:HC:calnexin complexes and ER60:HC:ß₂m:calreticulin complexes in TAP-deficient cells, and the existence of ER60:calreticulin complexes. It would also explain the presence of ER60 in anti-TAP immunoprecipitates from the calnexin-negative cell line CEM.NKR. If ERp57 (ER60) is only recruited onto HC by calreticulin as others propose, then there would be no requirement for ER60 in disulfide bond formation since calreticulin is known to bind only to those heavy chain:ß₂m complexes that possess fully oxidized disulfide bonds. However, our ability to isolate disulfide-bonded ER60-HC complexes using an antibody that recognizes only heavy chains lacking ß₂m in order to show that ER60’s binding in early complexes is dependent on calnexin and to isolate complexes of ER60 and HC from cells lacking ß₂m argue against such a proposal and favor a model in which ER60 is involved in both the early and late stages of class I assembly.

View larger version (21K): [in this window] [in a new window]   Figure 3. Schematic diagram. Proposed model of the MHC class I assembly pathway. Abbreviations: calnexin (CNX), calreticulin (CRT), heavy chain (HC), beta-2-microglobulin (ß2m), transporter associated with antigen processing (TAP), endoplasmic reticulum (ER). The model is intended to show that in binding to the nascent HC, calnexin may first bind and then recruit ER60 or CNX:ER60 may bind as a dimeric complex. In either case, binding is dependent on the presence of properly trimmed oligosaccharides. Calnexin-associated HCs have been shown to contain 0, 1, or 2 disulfide bonds. After calnexin has dissociated, the ER60:HC:ß2m complex may recruit calreticulin or, as is the case in calnexin-negative cells, CRT:ER60 dimers may bind directly to HC:ß2m complexes. Calreticulin only binds to fully oxidized HC. After the binding of calreticulin, the CRT:ER60:HC:ß2m complex may bind to a complex of TAP:tapasin or tapasin may first be recruited into the CRT:ER60: HC:ß2m complex, which would then bind to TAP. For simplification, the hydrolysis of ATP in both protein degradation and peptide translocation has been omitted. We propose that ER60 associates with class I after the oligosaccharides have been trimmed and remains associated throughout its subsequent maturation to orchestrate appropriate disulfide bond formation.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0720fje ; to cite this article, use FASEB J. (April 18, 2001) 10.1096/fj.00-0720fje

This article has been cited by other articles:

				J. E. Chambers, C. E. Jessop, and N. J. Bulleid Formation of a Major Histocompatibility Complex Class I Tapasin Disulfide Indicates a Change in Spatial Organization of the Peptide-loading Complex during Assembly J. Biol. Chem., January 25, 2008; 283(4): 1862 - 1869. [Abstract] [Full Text] [PDF]

				S. G. Santos, E. C. Campbell, S. Lynch, V. Wong, A. N. Antoniou, and S. J. Powis Major Histocompatibility Complex Class I-ERp57-Tapasin Interactions within the Peptide-loading Complex J. Biol. Chem., June 15, 2007; 282(24): 17587 - 17593. [Abstract] [Full Text] [PDF]

				Y. Zhang, E. Baig, and D. B. Williams Functions of ERp57 in the Folding and Assembly of Major Histocompatibility Complex Class I Molecules J. Biol. Chem., May 26, 2006; 281(21): 14622 - 14631. [Abstract] [Full Text] [PDF]

				D. B. Williams Beyond lectins: the calnexin/calreticulin chaperone system of the endoplasmic reticulum J. Cell Sci., February 15, 2006; 119(4): 615 - 623. [Abstract] [Full Text] [PDF]

				L. Sesma, B. Galocha, M. Vazquez, A. W. Purcell, M. Marcilla, J. McCluskey, and J. A. L. de Castro Qualitative and Quantitative Differences in Peptides Bound to HLA-B27 in the Presence of Mouse versus Human Tapasin Define a Role for Tapasin as a Size-Dependent Peptide Editor J. Immunol., June 15, 2005; 174(12): 7833 - 7844. [Abstract] [Full Text] [PDF]

				R. Abele and R. Tampe The ABCs of Immunology: Structure and Function of TAP, the Transporter Associated with Antigen Processing Physiology, August 1, 2004; 19(4): 216 - 224. [Abstract] [Full Text] [PDF]

				M.-E. Paquet, M. Cohen-Doyle, G. C. Shore, and D. B. Williams Bap29/31 Influences the Intracellular Traffic of MHC Class I Molecules J. Immunol., June 15, 2004; 172(12): 7548 - 7555. [Abstract] [Full Text] [PDF]

				Z. Frenkel, M. Shenkman, M. Kondratyev, and G. Z. Lederkremer Separate Roles and Different Routing of Calnexin and ERp57 in Endoplasmic Reticulum Quality Control Revealed by Interactions with Asialoglycoprotein Receptor Chains Mol. Biol. Cell, May 1, 2004; 15(5): 2133 - 2142. [Abstract] [Full Text] [PDF]

				S. Ford, A. Antoniou, G. W. Butcher, and S. J. Powis Competition for Access to the Rat Major Histocompatibility Complex Class I Peptide-loading Complex Reveals Optimization of Peptide Cargo in the Absence of Transporter Associated with Antigen Processing (TAP) Association J. Biol. Chem., April 16, 2004; 279(16): 16077 - 16082. [Abstract] [Full Text] [PDF]

				E. S. Trombetta The contribution of N-glycans and their processing in the endoplasmic reticulum to glycoprotein biosynthesis Glycobiology, September 1, 2003; 13(9): 77R - 91R. [Abstract] [Full Text] [PDF]

				N. S. Dangoria, M. L. DeLay, D. J. Kingsbury, J. P. Mear, B. Uchanska-Ziegler, A. Ziegler, and R. A. Colbert HLA-B27 Misfolding Is Associated with Aberrant Intermolecular Disulfide Bond Formation (Dimerization) in the Endoplasmic Reticulum J. Biol. Chem., June 21, 2002; 277(26): 23459 - 23468. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 15/8/1448 00-0720fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by LINDQUIST, J. A. Articles by TROWSDALE, J. Search for Related Content PubMed PubMed Citation Articles by LINDQUIST, J. A. Articles by TROWSDALE, J.