Skip Navigation


Bioscience Horizons Advance Access originally published online on February 10, 2009
Bioscience Horizons 2009 2(1):13-21; doi:10.1093/biohorizons/hzp003
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
2/1/13    most recent
hzp003v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Berry, J.-L.
Right arrow Articles by Bulleid, N. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Berry, J.-L.
Right arrow Articles by Bulleid, N. J.
Related Collections
Right arrow Biochemistry
Right arrow Cell Biology
Right arrow Genetics
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2009 The Author(s)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

ERp57 is involved in the oxidative folding of the low-density lipoprotein receptor in the endoplasmic reticulum

Jamie-Lee Berry and Neil J. Bulleid*

Faculty of Life Sciences, The University of Manchester, Manchester M13 9PT, UK

* Corresponding author: The Michael Smith Building, The Faculty of Life Sciences, The University of Manchester, Oxford Road, Manchester M13 9PT, UK. Tel: +44 1612755103. Email: neil.bulleid{at}manchester.ac.uk

Supervisor: Professor Neil J. Bulleid, Faculty of Life Sciences, The University of Manchester, Manchester M13 9PT, UK.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results and Discussion
 Limitations and Further...
 Funding
 References
 
This work explores the role of the thiol-oxidoreductase ERp57 in the post-translational oxidative folding of the low-density lipoprotein receptor (LDL-R), a cell-surface glycoprotein responsible for the uptake of cholesterol from plasma. The LDL-R provides a general model to analyse oxidative folding of multi-domain proteins in the endoplasmic reticulum; yet its folding pathway is also of specific interest as a high proportion of mutations in disulphide-rich domains of the protein are evident in familial hypercholesterolemia. Previous studies have suggested that the LDL-R forms a set of distinct non-native disulphide intermediates during folding, which are extensively isomerized prior to secretion of the native conformer. In addition, ERp57 has been suggested to be predominantly reduced in vivo and to form a mixed disulphide with the LDL-R. In this study, the LDL-R was expressed in both wild-type cells and those lacking the thiol-oxidoreductase ERp57 under conditions that prevent disulphide formation. The protein was then allowed to fold under oxidizing conditions, and samples taken at various timepoints. The electrophoretic mobility of folding intermediates from knock-out cells was compared with that of wild-type cells. The results show that dissimilar disulphide intermediates form between the two cell types, particularly during early stages of folding. A mutant form of ERp57, able to form but unable to resolve mixed disulphides, was also found to form mixed disulphides with the LDL-R. The results signify the requirement for ERp57 in oxidative folding of the LDL-R and also suggest that non-native disulphide intermediates may be central to the process of multi-domain protein folding.

Key words: ERp57, disulphides, oxidoreductase, protein folding


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results and Discussion
 Limitations and Further...
 Funding
 References
 
The formation of disulphides in eukaryotic cells occurs in the endoplasmic reticulum (ER). Conditions in the ER are highly optimized for the formation of disulphides, owing first to its oxidizing environment relative to the cytosol, but also to the presence of a number of thiol-oxidoreductases—a broad group of folding enzymes possessing an active site analogous to that in the cytosolic reductase thioredoxin.1 The enzymes can catalyse disulphide bond formation; however, these enzymes can also reduce substrates allowing for the isomerization of incorrectly paired cysteines during protein folding.2 Figure 1 depicts some of the main players involved in maintenance of an oxidizing ER environment, based on their known functions in vivo. The ER environment is generally buffered by a balanced ratio of reduced to oxidized glutathione (GSH:GSSG),3 among other sources of oxidizing equivalents from parallel oxidation pathways that have been proposed to involve Ero1, Erv24 and flavin-containing mono-oxidase.5 There is also much conflicting evidence for the distinct roles of oxidoreductases as oxidases or isomerases in vivo. One protein, the low-density lipoprotein receptor (LDL-R), provides a good model on which to study the action of oxidoreductases in cells, due to the discovery of an unusual set of disulphide intermediates during its folding pathway.6


Figure 1
View larger version (71K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Possible mechanisms by which the endoplasmic reticulum (ER) may maintain an environment optimized for disulphide bond formation in proteins. Disulphides are introduced into nascent chains soon upon entry into the ER, most likely by a thiol-oxidoreductase, such as protein disulphide isomerase (PDI). PDI may be maintained in its oxidized form by the sulfhydryl oxidase Ero1, with oxygen as the final electron acceptor.13 If the substrate forms native disulphides, it may be released to further stages of secretion. If not, the non-native protein may be reduced again by a member of the PDI family, or alternatively passed on to the calnexin/calreticulin cycle for isomerization to native disulphides in the case of N-glycosylated substrates. As ERp57 is associated with calnexin/calreticulin, it functions to unscramble non-native disulphides.

 
The LDL-R is a multi-domain, cell-surface membrane glycoprotein. Various classes of mutations in the gene (LDL-R) are evident in familial hypercholesterolemia (FH), several of which instigate the misfolding of disulphide-rich domains in the receptor structure.7 For example, the structure includes three domains with epidermal growth factor (EGF) homology, each forming three intra-domain disulphides, and within which 54% of mutations causing FH are located.8, 9 Characterized mutations of LDL-R affecting the correct formation of disulphides have been implicated in FH, and these illustrate the importance of this process for structural and functional integrity.10

It has been suggested that gene-fusion events in evolution have led to a situation where modules in multi-domain proteins become immediately foldable, independent of their neighbours.11 Indeed, vectorial folding is a logical scenario which one would expect to make the process more efficient. For the LDL-R at least, findings to the contrary raise important questions about the other factors involved in the folding pathway, which may be responsible for the extensive isomerization of disulphides.6 One enzyme, ERp57, has been shown previously to form a mixed disulphide with the LDL-R. ERp57 is a thiol-oxidoreductase for N-glycosylated, heavily disulphide-bonded clients of the calnexin/calreticulin cycle. It appears also to exhibit preference for a subset of substrates that share common structural domains, such as the EGF-like domains in the LDL-R.12 ERp57 has also been shown to be reduced at steady state in vivo,13 further supporting its role as an isomerase.

The aim of this study was to define a role for ERp57 in the post-translational folding of the LDL-R, by comparing the set of folding intermediates observed in wild-type and ERp57–/– cells according to their electrophoretic mobility. It was predicted that in the absence of ERp57, the protein would form a non-native disulphide intermediate from which it would be unable to recover. The LDL-R coding sequence was ligated into the pSPUTK vector downstream of the β-globin enhancer to improve the expression levels. It was then allowed to translate fully in both cell types under reducing conditions, before folding was followed in an oxidative environment. Marked differences were observed during the early stages of folding, and the protein ultimately reached a conformation in each cell type with a distinct hydrodynamic volume. These results build on previous work to provide insight into multi-domain protein folding and evidence for the role of ERp57 as an isomerase.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results and Discussion
 Limitations and Further...
 Funding
 References
 
Culture and Preparation of Semi-intact Cells
Cells were cultured in Dulbecco's Modification of Eagle's Medium with a 10% (v/v) foetal calf serum and 1% (v/v) L-glutamine supplement. They were incubated at 37°C and 5% CO2 to 70–90% confluence. Semi-intact cells were prepared from cells grown in culture so that the folding of proteins translated in vitro could be followed in an environment equivalent to the intact cell. Cells were made permeable, and endogenous mRNA was degraded according to a previously described method.14 Cell lines used were HT1080 human fibroblasts (ATCC, Maryland, USA), wild-type murine fibroblasts (MF), ERp57–/– MF and V5-ERp57 (C2,7A) HT1080. In the latter cell type, ERp57 has a V5 tag and its active site cysteine is mutated to alanine, disabling its ability to resolve the mixed disulphides it forms with substrate.12

Analysis of Samples by Electrophoresis
Protein samples in this study were analysed by SDS–PAGE, at 22 mA per gel. Proteins were resolved in 7.5% polyacrylamide gel.15 Gels were fixed in 10% (v/v) acetic acid and 10% (v/v) methanol and dried onto filter paper in a vacuum for 1 h at 80°C. Gels were exposed to Kodak Biomax MR Film (GRI, UK) for a period of ~1 week for imaging.

Cloning and Modification of LDL-R
The LDL coding sequence was amplified from pcDNA 3.1 (Invitrogen, UK) by several cycles of the polymerase chain reaction (PCR) and modified by the following primers. Forward: 5'GAACTCTCTAGAGCCACCATGGGGCCCTGGGGC3'; reverse: 5'GCACGCGGTACCTCACGCCACGTCATC3'. The primers incorporated the appropriate XbaI and KpnI restriction enzyme recognition sites to enable ligation into the pSPUTK vector. Each PCR proceeded in a total aqueous volume of 100 µl containing 5 units of Taq polymerase with the provided buffer (1x) (Bioline, UK). Deoxy-nucleotides (dATP, dCTP, dTTP, dGTP; Bioline) were each added to a final concentration of 0.25 mM, and primers to a concentration of 1 µM. Reactions were performed containing both 5 and 10 µg of LDL-R in pcDNA 3.1. These were denatured for 1 min at 94°C, annealed at 55°C for 1 min and allowed to extend at 72°C for 30 cycles. Controls were performed both in the absence of polymerase and in the absence of DNA.

PCR products were washed using DNA Clean and Concentrator Kit (Qiagen, Germany) according to the manufacturer's protocol. The purified DNA and 5 µg of pSPUTK were then digested with XbaI and KpnI. Restriction digests were performed in a total aqueous volume of 50 µl, containing 20 units of each enzyme and provided ‘Buffer L’ to 2x (Promega, UK). Digestion proceeded for 4 h at 37°C. The cut PCR product and plasmid were isolated by electrophoresis on a 1% agarose gel, and DNA was extracted using the QIAGEN®II Gel Extraction Kit (Qiagen) protocol. Concentrations of eluted DNA were measured by nano-spectrophotometry. Ligation reactions were each performed in a total aqueous volume of 20 µl, with 18 units of T4 DNA ligase and the provided buffer (1:10 dilution) (Promega). The amount of DNA was varied in each reaction mixture, between 100 and 160 ng. Control experiments were conducted omitting either DNA insert, or T4 DNA ligase, from the reaction.

Transformation of Ligation Products into Escherichia coli cells
Escherichia coli XL1 Blue cells were inoculated into 200 ml Luria Bertani (LB) broth and grown up to OD600 0.5–0.6. Cells were centrifuged at 4000 g for 10 min at 4°C and re-suspended to 100 ml in 10% glycerol (ice-cold). Fifty microlitres of cell culture were combined with 2 µl of each ligation reaction in E. coli Pulser® cuvettes and chilled on ice for 5 min. Cells were electroporated, added to 250 µl of LB broth and shaken for 1 h at 37°C. Cells from each ligation reaction were plated onto separate LB and ampicillin (100 µg ml–1) agar plates and left overnight. Single colonies from the non-control ligations were picked and inoculated into 30 ml fresh LB broth with ampicillin (100 µg ml–1). Cell pellets from centrifugation were re-suspended in 100 µl GTE (50 mM glucose, 1 mM EDTA, 25 mM Tris–HCl, pH 8.0) containing RNase (100 mg ml–1). Cultures were treated with 200 µl NaOH–SDS [0.2 M NaOH, 1% (w/v) SDS] and 150 µl potassium acetate [29% (v/v) acetic acid and KOH to pH 4.8]. Supernatant was retained after centrifugation, and plasmids were purified by washing once with phenol and twice with chloroform.

TNT Lysate (Promega, UK) Translation
A coupled transcription and translation of LDL-R in each of the circular vectors (pcDNA 3.1 and pSPUTK) was set up using the TNT Lysate Translation System (Promega) according to the manufacturer's protocol. This minimizes any experimental error that may be encountered when performing separate reactions. Products were analysed by SDS–PAGE electrophoresis, and pixel density of the image was measured using AIDA Image Analyzer software.

Transcriptions in vitro
The pSPUTK vector was cut with HpaI, and linear DNA was transcribed according to a previous method,16 using 30 units of SP6 polymerase per 50 µl of reaction mixture. Reagents were from Promega. RNA transcripts were purified by washing once with phenol and twice with chloroform and RNA was precipitated in ethanol for 60 min with 0.3 M sodium acetate (pH 5.2) at –20°C. The mRNA was re-suspended in DEPC-treated water.

Translations in vitro
The mRNA was translated with 16.5 µl of Rabbit Reticulocyte Lysate (Flexilysate, Promega) per 25 µl translation, containing 1 µl of mRNA, 20 µM amino acids minus methionine (Bioline), 20 mM potassium chloride and 10 µCi EasyTag EXPRESS-35 Protein labelling mix [35S] (NEN/PerkinElmer). Translations were performed either in the presence or absence of ~105 selectively permeabilized (SP) cells and incubated for 1 h at 30°C. Samples not containing SP cells were placed on ice, reduced with 5 mM dithiothreitol (DTT) and boiled with equal volume of 5x SDS sample buffer for 3 min at 100°C before analysis by SDS–PAGE. Otherwise, translation mixtures were centrifuged to terminate translation and isolate the SP cells, which were then washed in KHM (110 mM potassium acetate, 2 mM magnesium acetate, 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) to remove the lysate. Cells were then re-suspended in ~10 µl, 5x SDS buffer (156 mM Tris pH 6.8, 7.5% SDS, 10% glycerol, 0.02% bromophenol blue), boiled and analysed by SDS–PAGE. The LDL-R products from the post-translational folding assay were immunoisolated before SDS–PAGE.

Post-translational Folding Assay
Translation mixtures were set up to a 25 µl volume per timepoint, including SP cells. Seven timepoints were taken at 0, 1, 5, 15, 30, 60 and 90 min from one total translation mixture. Prior to the 0 timepoint, the total mixture was allowed to translate for 30 min under reducing conditions in the presence of 5 mM DTT. This should enable full translation of adequate levels of the polypeptide in a completely reduced form. After 30 min, oxidized glutathione was added to a final concentration of 15 mM, in order to recreate the oxidizing environment of the ER and to allow folding of the polypeptide under conditions reflecting those in vivo. To cease the reaction at each timepoint, a 25 µl aliquot was taken and placed on ice, then N-ethylmaleimide (NEM) was added immediately to a concentration of 25 mM to prevent further disulphide exchange by the transfer of alkyl groups to cysteine sulphydryls. This experiment was conducted twice: once with wild-type MF cells and once with ERp57–/– MF cells.

Immunoisolation of the LDL-R
Cells from each translation were isolated by centrifugation, washed in an excess of KHM buffer and then isolated for a second time. Supernatant containing any remaining translation mixture was extracted and discarded. Cell pellets were then re-suspended in 0.5 ml lysis buffer containing 25 mM NEM. Samples were pre-cleared to prevent non-specific binding by adding Protein A Sepharose (PAS) beads to 1%, and incubating at 4°C for 30 min with constant agitation to keep the beads suspended. After 30 min, samples were centrifuged for 5 min, the supernatant was removed and PAS beads were added again to 1%, this time coupled with polyclonal anti-LDL-R antibody 121 (1:300 dilution). Samples were incubated at 4°C overnight to allow specific binding of the LDL-R to the antibody. The beads were isolated by centrifugation and washed three times in lysis buffer, before re-suspension in 15 µl 5x SDS loading buffer. Samples were boiled for 2–3 min at 100°C and agitated to release the LDL-R into solution. Beads were settled by centrifugation, and the supernatants were analysed by SDS–PAGE under non-reducing conditions. Antibody was provided by Ineke Braakman, University of Utrecht, Netherlands.

Analysis of Mixed Disulphides
Four translations were conducted under non-reducing conditions, two of which were in the presence of SP wild-type HT1080 cells and two in the presence of SP V5-ERp57 (C2,7A) cells. After 1 h, NEM was added to each sample to 25 mM to prevent further thiol exchange. One of the samples from each of the cell types was immunoisolated, using anti-V5 antibody (1:300 dilution) (Invitrogen) conjugated to Protein G Sepharose beads. The other two samples were washed with KHM and boiled in 5x SDS sample buffer without immunoisolation. Samples were analysed under non-reducing conditions by SDS–PAGE.


   Results and Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
Limitations and Further...
 Funding
 References
 
LDL-R Expression is Optimized in pSPUTK
The LDL-R coding sequence was modified by oligonucleotide primers to enable its ligation into the mammalian pSPUTK vector, downstream of the β-globin enhancer. In addition, downstream of the XbaI consensus site, a Kozak ribosomal binding region was included in the primer before the translation initiation codon.

Figure 2A shows analysis of translation products by SDS–PAGE. Coupling transcription and translation reactions minimize experimental error and loss of product during purification. The LDL-R formed a single major translation product when expressed from pcDNA 3.1 and pSPUTK. The resulting image was analysed by 2D Densitometry using AIDA Image Analyzer software (Fig. 2B) and translation was shown to be improved by ~23% after cloning and ligation of LDL-R downstream of the pSPUTK β-globin enhancer. Translation of LDL-R from pcDNA 3.1 minus the enhancer appeared to yield less protein product. This increase can also be attributed to the 5' sequence included in the modified gene.


Figure 2
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Translation of LDL-R in vitro and quantification of protein levels. (A) Coupled transcription and translations were conducted from LDL-R in pcDNA 3.1 (lane 1) and pSPUTK (lane 2). (B) Levels of protein from coupled reactions were compared with 2D Densitometry and the results are shown in the bar chart. Expression of the coding sequence from pSPUTK increased protein levels by ~23%.

 
The LDL-R Folds via a Non-native Disulphide Intermediate
Figure 3A shows the post-translational folding of the LDL-R in wild-type and ERp57–/– MF cells. In the wild-type, at 0 min the protein appears completely reduced, forming a concise band on the gel. Upon the addition of oxidized glutathione, the protein migrates a greater distance through the gel and the product becomes much less distinct, appearing as a smear. Between 15 and 30 min, the samples decrease in electrophoretic mobility. This decrease in mobility is due to an increased hydrodynamic volume, i.e. the molecules are less compact and take up more space in solution, so migrate more slowly through the gel network. Here there is a small, yet observable shift in mobility towards that of the reduced form. After 90 min in an oxidizing environment, the protein molecules occupy less hydrodynamic volume than the reduced form, yet failed to migrate as far through the gel, therefore occupy a greater hydrodynamic volume, than the conformers observed between 1 and 10 min. The product at 90 min does not appear completely reduced, yet migrates at a rate closer to the reduced form than the earlier folding intermediates. These results indicate that the protein folds to a compact form early in the time course, but then becomes less compact, probably due to a rearrangement in the disulphide bonds.


Figure 3
View larger version (112K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Post-translational folding of the LDL-R in semi-intact MF cells. (A) LDL-R was first translated with a rabbit reticulocyte lysate under reducing conditions (5 mM DTT) for 30 min in the presence of either semi-intact wild-type (left panel), or ERp57–/– (right panel), MF cells. At 0 min, the LDL-R was allowed to fold by the addition of oxidized glutathione (15 mM) to permit disulphide exchange. Samples were taken at the timepoints shown, lysed in the presence of 25 mM NEM and the LDL-R was immunoisolated with a polyclonal anti-LDL-R antibody conjugated to Protein A Sepharose beads. Samples were analysed by SDS–PAGE. (B) The experiment was repeated, and samples taken from each cell type at each timepoint were analysed beside each other by SDS–PAGE under non-reducing conditions.

 
From previous published results,6 and those in this study, it is clear that the LDL-R folds via a non-native disulphide intermediate. It should be noted, however, that the shift in electrophoretic mobility towards that of the ‘native’ conformer in the MF cells is less pronounced than previously observed in HeLa cells.6 Although this may be purely due to differences in activity between the cell types, it should be kept in mind when drawing conclusions from these results.

In the ERp57–/– cells, LDL-R appears to fold similarly to the wild-type. It was predicted that the receptor would be unable to recover from a non-native disulphide intermediate without the oxidoreductase present. This seems not strictly to be the case, as the results suggest that isomerization of disulphides is taking place in the knock-outs (Fig. 3A, right-hand panel). In fact, the results shown in Figure 3A do not suggest any obvious defect in recovery from the non-native intermediate with ERp57 absent. However, the products do appear to form a slightly different pattern on each gel, suggesting differences in disulphide intermediates between the two cell types. This was confirmed by the results shown in Figure 3B. Here the folding intermediates from the wild-type and ERp57–/– cells were compared directly as the samples from each timepoint were run adjacent to one another on the same gel. There is no difference in mobility between the reduced forms of the LDL-R in each cell type at 0 min; however, intermediates formed between 1 and 5 min in wild-type cells decrease more dramatically in hydrodynamic volume, compared with samples from ERp57–/– cells. Interestingly, the intermediate reached at 5 min in the wild-type cells has formed a different, more compact, disulphide intermediate to the species formed with the ERp57–/– cells. Therefore, it may be proposed that alternative non-native disulphides are formed during early stages of folding without ERp57. This suggests a role for ERp57 in the formation of non-native disulphides as well as their isomerization.

In the ERp57–/– cells, the LDL-R reduces slightly in mobility after 15 min like the wild-type; however, the ultimate conformers at 60 and 90 min have a greater hydrodynamic volume than those from the wild-type, insinuating that a different conformation has been reached. The disulphides that form, therefore, may be central to the folding process as opposed to forming haphazardly. Indeed, it is possible that non-native disulphide intermediates are an important step enabling the protein to eventually reach the native conformation. Their formation may give rise to large conformational loops, keeping parts of the protein apart in space that may otherwise form favourable interactions during early stages of folding, thereby allowing the individual domains to fold correctly. A disulphide intermediate was previously observed in the tailspike endorhamnosidase from bacteriophage p22,17 whereas the native protein contains no disulphides. Such evidence supports an evolutionary role for non-native disulphide intermediates during protein folding.

The LDL-R Forms a Mixed Disulphide with ERp57 in HT1080 Cells
To confirm an interaction between ERp57 (57 kDa) and the LDL-R, the mRNA was translated in either untransfected or a stable transfected cell line expressing a mutant form of V5-tagged ERp57, in which the second active site cysteine was replaced with alanine to trap mixed disulphides (C2,7A).12 Without immunoisolation, a band corresponding to LDL-R was observed in both cell types with an approximate molecular weight of 97 kDa (Fig. 4, lanes 1 and 3). Upon immunoisolation of the V5-tagged ERp57 from cell lysates with an anti-V5 antibody, two proteins with a molecular weight greater than LDL-R were observed after analysis by SDS–PAGE run under non-reducing conditions (Fig. 4, lanes 4 and 5). When the untransfected cells were lysed and treated with the anti-V5 antibody, no protein was precipitated (lane 2). This was expected because HT1080 cells did not express the V5-tagged ERp57.


Figure 4
View larger version (65K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Mixed disulphides form between ERp57 and LDL-R. LDL-R was translated in both wild-type (lane 1) and V5-ERp57 (C2,7A) cells (lane 3) and cell lysates were analysed by SDS–PAGE. Next the LDL-R was translated in the wild-type (lane 2) and V5-ERp57 (C2,7A) (lane 4) cells and lysates were immunoisolated with an anti-V5 antibody to isolate the tagged ERp57. Products were analysed under non-reducing conditions. Two higher molecular weight bands appear in lane 4 (clearer after 7 days image exposure in lane 5) corresponding to ERp57 and the LDL-R as a co-precipitant, trapped in a mixed disulphide state.

 
The presence of immunoisolated products with a higher molecular weight than the LDL-R confirms that ERp57 forms mixed disulphides with the protein. Degradation of endogenous mRNA during semi-intact cell preparation, before the addition of LDL-R mRNA to the system, ensures that the most likely interacting partner causing an increase in molecular weight of the precipitated V5-tagged ERp57 is the LDL-R. However, two distinct higher molecular weight species were immunoisolated from the V5-ERp57 (C2,7A) cells. The 160 kDa product corresponds well to conjoined LDL-R and ERp57, yet the band over 220 kDa is of unknown origin. We do not, however, know the stoichiometry of the interaction in these mutant cells. Mixed disulphides are too transient to detect normally, so the ratio may be 1:1 in the wild-type but it is plausible that more than one ERp57 molecule becomes trapped with each LDL-R molecule in the mutant. In the system used here, although it can be assumed with a degree of conviction that an interaction takes place, it cannot be inferred whether single or multiple interactions will occur in vivo.

A Possible Redundancy Mechanism in ERp57–/– Cells
When the translation products from the different cell types were analysed adjacent to one another (Fig. 3B), it became evident that the wild-type samples form more compact structures throughout the folding pathway. This may be a knock-on effect of the alternative non-native disulphides that form at earlier folding stages in the presence of ERp57. Alternatively, this may be because ERp57 is required to isomerize the folding intermediates, a process defective in the ERp57–/– cells. The samples from ERp57–/– cells, however, are ultimately forming structures with greater hydrodynamic volume than those from earlier stages of the folding pathway. This indicates that the protein in the knock-out cells does undergo some form of isomerization to try to recover from the non-native disulphide intermediate. However, the native structure is not attained, as samples at 90 min maintain distinct conformers to those from wild-type cells. Incorrectly folded LDL-R polypeptides, as a result of incorrect cysteine pairing, would be retained for continuous futile cycles of re-association with calnexin/calreticulin.18, 19 Of course this would be in vain in the absence of ERp57, but it is conceivable that another, less well-characterized oxidoreductase acts in its place upon deletion. ERp72, for example, has been shown to share 37% sequence homology (Fig. 5) with ERp57.20 ERp72 has been shown to form mixed disulphides with substrates of ERp57,20, 21 yet its depletion has no effect on certain substrates where ERp57 has a specialized role.22 Therefore, there is indication that they may share substrates, along with evidence to the contrary; however, it is feasible that upon ERp57 deletion ERp72 is a good candidate replacement.


Figure 5
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. A schematic showing the similar domain organization observed for PDI-family proteins. Using the example of ERp72, it is unlikely that eukaryotes have evolved without a possible compensatory mechanism in the event of ERp57 dysfunction. ERp72 shares 37% homology with ERp57, which shares 29% identity and 56% similarity with PDI.24 All share striking similarity with respect to domain organization,20 except ERp72 has an extra catalytic a domain and ERp57 has a basic (+) as opposed to acidic (–) tail. The b' domains affiliate with substrate specificity and/or targeting to substrates, for example, ERp57 has been shown to interact with calnexin via its b' domain.25

 
Although thought to be tightly regulated in an oxidizing state by Ero1 (Fig. 1), protein disulphide isomerase (PDI) has shown to exhibit distinct catalytic activity between active sites of its a and a' domains. Mutational studies have shown that although the a' domain is an efficient oxidase, the a domain acts as an isomerase.23 PDI itself may be implicated in the situation without ERp57, perhaps in complex with BiP. It may be the case that a compensatory mechanism is less efficient than the calnexin/calreticulin and ERp57 system, hence the difference in folding intermediates; yet it may provide a means to an end, perhaps resulting in a receptor with reduced function.


   Limitations and Further Direction
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
 Limitations and Further...
Funding
 References
 
It has been reported previously that ERp57 is primarily reduced at steady state, so its apparent involvement in the formation of non-native disulphides may not ring true in vivo. The large injection of oxidized glutathione into the system here may have caused an initial surge in levels of oxidized ERp57, permitting thiol exchange with the LDL-R in a way that would not occur normally. The subsequent rebalance of the system by an increase in reduced glutathione would then increase levels of reduced ERp57, akin to the situation in cells, allowing it to isomerize the disulphides (see Fig. 1). The system used in this study is perhaps not an accurate reflection of conditions in the ER. Indeed, it is more likely that a primarily oxidized enzyme such as archetypical PDI will have a key role in the formation of disulphides in LDL-R; yet it is not unreasonable to propose the involvement of a proportion of oxidized ERp57 in vivo. It should also be noted that the difference in electrophoretic mobility of the LDL-R from two different cell types, namely ERp57–/– MF and wild-type MF, may not be definitively attributed to the presence or absence of ERp57. It may be useful to reproduce the results in a further study, using various cell types.

A substitute oxidoreductase may be acting in the absence of ERp57, yet may not be able to access the LDL-R in the calnexin cycle while ERp57 is present. Therefore, this may not be stearically possible in vivo in cases where ERp57 is dysfunctional but not absent. A more suitable approach to study ERp57 function may be to introduce a loss of function mutation into the active site so that the protein is still physically present and interacting with calnexin and calreticulin. With this approach, we may see the LDL-R reaching a non-native intermediate that cannot be isomerized, confirming the requirement for ERp57 in vivo.

To study a compensatory mechanism further, one method would be to block entry into the calnexin cycle with the glucosidase inhibitor castanospermine, then look at LDL-R folding. Substrate binding to calnexin or calreticulin is dependent upon the presence of a monoglucosylated oligosaccharide side chain; castanospermine inhibits the ER glucosidase activity and therefore prevents trimming of the glycan chain.19 If the set of folding intermediates matches the wild-type, it is likely another oxidoreductase and ER retention system can act in place of ERp57 and the calnexin cycle. In this study, the LDL-R was able to enter the calnexin cycle, making a compensatory mechanism unlikely. Also, the only oxidoreductase known to conserve the residues interacting with calnexin is ERp72, which has been shown not to interact with calnexin in ERp57–/– cells.21 Before we can assume that the native conformer has been reached in the wild-type cells, it would be necessary to carry out immunoisolation using a conformational specific antibody which only binds to the native LDL-R conformer. The results shown here, however, strongly suggest that a distinct set of disulphide intermediates arise during folding of the LDL-R in the absence of ERp57.


   Funding
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
 Limitations and Further...
 Funding
References
 
This work was funded by The Wellcome Trust (grant #074081) and the University of Manchester.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
 Limitations and Further...
 Funding
 References
 

  1. Sevier C, Kaiser C. Formation and transfer of disulphide bonds in living cells. Nat Rev Mol Cell Biol (2002) 3:836–847.[CrossRef][Web of Science][Medline]
  2. Jessop C, Chakravarthi S, Watkins R, et al. Oxidative protein folding in the mammalian endoplasmic reticulum. Biochem Soc Trans (2004) 32:655–658.[CrossRef][Web of Science][Medline]
  3. Chakravarthi S, Jessop C, Bulleid NJ. The role of glutathione in disulphide bond formation and endoplasmic reticulum-generated oxidative stress. EMBO Rep (2006) 7:271–275.[CrossRef][Web of Science][Medline]
  4. Sevier C, Cuozzo J, Vala A, et al. A flavoprotein oxidase defines a new endoplasmic reticulum pathway for biosynthetic disulphide formation. Nat Cell Biol (2001) 3:874–882.[CrossRef][Web of Science][Medline]
  5. Suh J, Poulsen L, Ziegler D, et al. Yeast flavin-containing monooxygenase generates oxidizing equivalents that control protein folding in the endoplasmic reticulum. Proc Natl Acad Sci USA (1999) 96:2687–2691.[Abstract/Free Full Text]
  6. Jansens A, Duijin E, Braakman I. Coordinated non-vectorial folding in a newly synthesized multi-domain protein. Science (2002) 298:2401–2403.[Abstract/Free Full Text]
  7. Hobbs H, Brown M, Goldstein J. Molecular genetics of the LDL receptor gene in familial hypercholesterolemia. Hum Mutat (1992) 1:445–466.[CrossRef][Medline]
  8. Jeon H, Meng W, Takagi J, et al. Implications for familial hypercholesterolemia from the structure of the LDL receptor YWTD-EGF domain pair. Nat Struct Biol (2001) 8:499–504.[CrossRef][Web of Science][Medline]
  9. Leren T, Solberg K, Rødningen OK, et al. Two novel point mutations in the EGF precursor homology domain of the LDL receptor gene causing familial hypercholesterolemia. Hum Genet (2005) 96:241–242.[CrossRef]
  10. Saha S, Boyd J, Werner J, et al. Solution structure of the LDL receptor EGF-AB pair: a paradigm for the assembly of tandem calcium binding EGF domains. Structure (2001) 9:451–456.[Medline]
  11. Netzer W, Hartl F. Recombination of protein domains facilitated by co-translational folding in eukaryotes. Nature (1997) 388:343–349.[CrossRef][Medline]
  12. Jessop C, Chakravarthi S, Garbi N, et al. ERp57 is essential for efficient folding of glycoproteins sharing common structural domains. EMBO J (2007) 26:28–40.[CrossRef][Web of Science][Medline]
  13. Jessop C, Bulleid N. Glutathione directly reduces an oxidoreductase in the endoplasmic reticulum of mammalian cells. J Biol Chem (2004) 279:55341–55347.[Abstract/Free Full Text]
  14. Wilson R, Allen A, Oliver J, et al. The translocation, folding, assembly and redox-dependent degradation of secretory and membrane proteins in semi-permeabilised mammalian cells. Biochemical J (1995) 307:679–687.[Web of Science][Medline]
  15. Duffy M, Noormohammadi A, Bassegio N, et al. Polyacrylamide gel-electrophoresis separation of whole-cell proteins. In: Methods in Molecular Biology—RJ Miles, Nicholas RAJ, eds. (1998) 104. Totowa, NJ: Humana Press.
  16. Gurevich V, Pokrovskaya I, Obukhova T, et al. Preparative in vitro mRNA synthesis using SP6 and T7 RNA polymerases. Anal Biochem (1991) 195:207–213.[CrossRef][Web of Science][Medline]
  17. Robinson A, King J. Disulphide-bonded intermediate on the folding and assembly pathway of a non-disulphide bonded protein. Nat Struct Biol (1997) 4:450–455.[CrossRef][Web of Science][Medline]
  18. Brook D. Introduction: molecular chaperones of the ER: their role in protein folding and genetic disease. Semin Cell Dev Biol (1999) 10:441–442.[CrossRef][Web of Science][Medline]
  19. Kleizen B, Braakman I. Protein folding and quality control in the endoplasmic reticulum. Curr Opin Cell Biol (2004) 16:343–349.[CrossRef][Web of Science][Medline]
  20. Maattanen P, Kozlov G, Gehring K, et al. ERp57 and PDI: multifunctional protein disulfide isomerases with similar domain architectures but differing substrate-partner associations. Biochem Cell Biol (2006) 84:881–889.[Web of Science][Medline]
  21. Soldà T, Garbi N, Hammerling G, et al. Consequences of ERp57 deletion on oxidative folding of obligate and facultative clients of the calnexin cycle. J Biol Chem (2006) 281:6219–6226.[Abstract/Free Full Text]
  22. Zhang Y, Balg E, Williams D. Functions of ERp57 in the folding and assembly of major histocompatibility complex class I molecules. J Biol Chem (2006) 281:14622–14631.[Abstract/Free Full Text]
  23. Kulp M, Frickel E, Ellgaard L, et al. Domain architecture of protein-disulphide isomerase facilitates its dual role as an oxidase and an isomerase in Ero1p-mediated disulphide formation. J Biol Chem (2006) 281:876–884.[Abstract/Free Full Text]
  24. High S, Fabienne L, Russell S, et al. Glycoprotein folding in the endoplasmic reticulum: a tale of three chaperones? FEBS Lett (2000) 476:38–41.[CrossRef][Web of Science][Medline]
  25. Russell S, Ruddock L, Salo K, et al. The primary substrate binding site in the b' domain of ERp57 is adapted for endoplasmic reticulum lectin association. J Biol Chem (2004) 279:18861–18869.[Abstract/Free Full Text]
Submitted on 30 September 2008; accepted on 20 January 2009


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
2/1/13    most recent
hzp003v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Berry, J.-L.
Right arrow Articles by Bulleid, N. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Berry, J.-L.
Right arrow Articles by Bulleid, N. J.
Related Collections
Right arrow Biochemistry
Right arrow Cell Biology
Right arrow Genetics
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?