Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Andersson, H. Articles by Garoff, H. Search for Related Content PubMed PubMed Citation Articles by Andersson, H. Articles by Garoff, H.

 Previous Article  |  Next Article 

Journal of Virology, June 2003, p. 6676-6682, Vol. 77, No. 12
0022-538X/03/$08.00+0     DOI: 10.1128/JVI.77.12.6676-6682.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Lectin-Mediated Retention of p62 Facilitates p62-E1 Heterodimerization in Endoplasmic Reticulum of Semliki Forest Virus-Infected Cells

Helena Andersson^* and Henrik Garoff

Department of Biosciences at Novum, Karolinska Institute, S-141 57 Huddinge, Sweden

Received 22 November 2002/ Accepted 25 March 2003

Top Abstract Introduction Materials and Methods Results Discussion References

 
The Semliki Forest virus (SFV) spike subunits p62 and E1 are made from a common coding unit in the order p62-E1. The proteins are separated by the host signal peptidase upon translocation into the endoplasmic reticulum (ER). Shortly thereafter, p62 and E1 form heterodimers. Heterodimerization preferentially occurs between subunits derived from the same translation product, so-called cis heterodimerization. As the p62 protein has the capacity to leave the ER in the absence of E1, it has been postulated that there exists a retention mechanism for the p62 protein, putatively at or near the translocon, in the ER in order to promote cis heterodimerization (B. U. Barth and H. Garoff, J. Virol. 71:7857-7865, 1997). Here we show that there exists such a mechanism, that it is at least in part mediated by the ER chaperones calnexin and calreticulin, and that the retention is important for efficient cis heterodimerization.

Top Abstract Introduction Materials and Methods Results Discussion References

 
The spike subunits p62 and E1 of Semliki Forest virus (SFV) are made sequentially, together with the viral capsid (C), from a common coding unit, the 26 S subgenome in the endoplasmic reticulum (ER). Translation starts with C and then follows the p62 protein and finally the E1 protein (15). The C protein is cleaved by C autoprotease activity, whereas the membrane polyprotein is separated into individual proteins upon translocation into the ER lumen by the host signal peptidase (1, 14, 17, 23, 26). Although not initially associated, p62 and E1 heterodimerize fast and very efficiently and are thereafter transported through the secretory pathway, where p62 is cleaved into the E2 and E3 proteins, to the plasma membrane (PM), where virus budding takes place (11, 16, 31, 32). The E1 subunit carries the membrane fusion function of the virus, whereas p62/E2 serves as an E1 chaperone controlling its activity and preventing premature fusion activation (2, 7, 22, 38, 39). It has been previously shown that p62/E1 heterodimerization in ER preferentially occurs between subunits derived from the same translation product, translocated at one site, so-called cis heterodimerization (4). trans heterodimerization is, however, possible though inefficient when p62 and E1 are made from separate coding units (4). When E1 is expressed in the absence of p62, it is retained in the ER and the bulk of the protein aggregates rapidly, showing that E1 is dependent on the p62 protein for folding and transport to the PM (2). The p62 protein in the absence of E1, on the other hand, is transported in a modified form called E2* to the PM (3). As this does not occur in wild-type (wt) virus-infected cells, we asked how it is possible for p62 to wait for the subsequent synthesis of E1 and heterodimerize rather than to escape, forming E2*. We hypothesized that there would exist a retention mechanism for p62, putatively at the translocation site, to await the emergence of the E1 polypeptide and to ensure efficient heterodimerization. This model was tested biochemically, and our results indicated that there indeed exists a retention mechanism for p62 and that the retention is mediated at least in part by the ER chaperones calnexin and/or calreticulin. The lectin-mediated retention is not absolutely required for successful p62/E1 heterodimerization but facilitates cis heterodimerization and thus facilitates efficient particle production.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cells, virus, DNA constructs and antibodies. BHK-21 cells were grown in G-MEM (with L-glutamine without tryptose phosphate broth) (Invitrogen Ltd., Paisley, United Kingdom) supplemented with 10% tryptose phosphate broth, 5% heat-inactivated fetal bovine serum, 20 mM HEPES (pH 7.3), and 2 mM glutamine. Penicillin (100 U/ml) and streptomycin (100 µg/ml) were also added to media for passage of cells. Cells were incubated at 37°C and 5% CO₂. wt SFV was the laboratory strain, SFV4 (24). Plasmids used are described in Table 1 (4, 5, 12, 24, 35). Plasmids pSFV p62N13Q, pSFV p62N200Q, and pSFV p62N262Q were made by PCR (19, 40). In all plasmids the SFV sequences can be transcribed by SP6 polymerase into RNA molecules (the viral genomes) that are competent for replication in cells. Antibodies used were anti-E2 monoclonal antibodies UM 5.1 (obtained from H. Snippe, Eijkman-Winkler Institute for Microbiology, Utrecht University Hospital, Utrecht, The Netherlands) (6, 37) and K-26/98 (obtained from A. Salmi, Virology Department, University of Turku, Turku, Finland) (3) and the anticalnexin polyclonal antibody SPA-860 (Stressgen Biotechnologies Corp., Victoria, Canada).

View this table: [in this window] [in a new window]

TABLE 1. Plasmids used in this study

Infection and transfection of cells. wt virus was used to infect BHK-21 cells at a multiplicity of infection of 10. Infection was done by incubating cells at 37°C in a 35-mm petri dish with virus suspension in 500 µl of MEM (MEM with Earle's salts, without L-glutamine) (Invitrogen Ltd.) supplemented with 20 mM HEPES, 2 mM glutamine, and 0.2% bovine serum albumin (Invitrogen Ltd.) for 1 h. After this, the medium was replaced with G-MEM. Transcription of RNA from plasmid was done in a 50-µl mixture as described earlier (24), only that G(5')ppp(5')G was used instead of m⁷G(5')ppp(5')G. Two-fifths of the mixture (20 µl) was used to transfect 10⁷ BHK cells by electroporation in a 0.4-cm cuvette (850 V/25 µF) (25). Efficient transfection was achieved by thoroughly suspending cells beforehand by trypsinization and gentle mixing. After transfection the cells were diluted 20-fold with G-MEM, distributed into 35-mm plates, and incubated at 37°C in a 5% CO₂ atmosphere.

Metabolic labeling of viral proteins. For the labeling of one dish of cells, 100 µCi of [³⁵S]methionine (Amersham Biosciences, Amersham, Little Chalfont, United Kingdom; specific activity, >37 TBq/mM, >1,000 Ci/mM) was used. Labeling was carried out for different lengths of time in 500 µl of MEM lacking methionine. The incubation of cells in this medium was started 30 min before labeling to prestarve the cells of methionine. Labeling was terminated by changing the medium to MEM containing 1 mM unlabeled methionine. Chase in this medium was continued for different lengths of time. Castanospermine (CST) was used at a final concentration of 1 mM (stock solution, 100 mM) and was added 2 h prior to labeling, during labeling and chase. Cells infected with wt virus were labeled 6 h after starting the infection. Transfected cells were labeled 6.5 h postelectroporation.

Preparation of cell lysates, immunoprecipitation, and Endo H digestion. After pulse-chasing of cells, these were put onto ice, the medium was removed, and cells were washed once with cold phosphate-buffered saline containing 20 mM N-ethylmaleimide to block free sulfhydryl groups and were solubilized with 300 µl of NP-40 lysate buffer (1% NP-40, 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2 mM EDTA, 10 µg of phenylmethylsulfonyl fluoride/ml, and 20 mM N-ethylmaleimide). The lysate was centrifuged at 5,000 x g for 5 min at 4°C. The supernatant was transferred to a fresh tube. One-third (100 µl) of an NP-40 lysate sample was used for protein A-mediated immunoprecipitation. Briefly, lysate was mixed with 50 µl of protein A Sepharose slurry (Amersham Biosciences) and antibody, incubated approximately 3 h at 4°C, and then washed as described earlier (37). For endoglycosidase H (Endo H_f, 10⁶ U/ml; New England Biolabs, Beverly, Mass.) treatment of a cell lysate, 15 µl of the latter was mixed with 2 µl of 0.5 M sodium citrate, pH 5.5, and 2 µl of 10% sodium dodecyl sulfate (SDS). The mix was incubated for 2 min at 70°C. Two hundred units of Endo H_f was then added, and the sample was incubated at 37°C for 16 h. For Endo H treatment of an immunoprecipitated lysate sample, the precipitate was mixed with 50 µl of Endo H buffer (50 mM sodium citrate, pH 5.5-1% SDS), incubated for 2 min at 70°C, and centrifuged for 2 min at 16,100 x g. The supernatant with the viral proteins was divided in two parts that were either Endo H (200 U of Endo H_f) treated or mock treated for 16 h at 37°C.

SDS-PAGE. Samples of immunoprecipitates and lysates were diluted in equal volumes of SDS-gel loading buffer (125 mM NaHPO₄, pH 7.0, 40% glycerol, 6% SDS, 8 mM EDTA, 1.6 mM methionine, and 0.03% bromophenol blue) and run on 10% gels under nonreducing conditions. The electrophoresis system has been described earlier (34). After SDS-polyacrylamide gel electrophoresis (PAGE), the gel was treated for fluorography with 1 M sodium salicylate for 15 min and then dried. Quantitation of the radioactivity in the viral protein bands was done with a Fuji phosphorimage, type FUJIX BAS 2000 TR. Molar ratios were calculated after normalizing the photostimulated luminescence values of the proteins to their methionine contents. In the case of CST-treated cells, the trimming of sugar units is through endomannosidase-mediated deglucosylation (29). Trimming by this pathway in BHK cells is only about three-fourths as efficient as that by the normal pathway (H. Andersson and H. Garoff, unpublished data) (20). Therefore, when Endo H resistance in CST-treated cells was estimated, the measured fraction was corrected with a factor of 4/3.

Top Abstract Introduction Materials and Methods Results Discussion References

 
p62 retention in the ER is saturable. A first indication that there indeed would exist a p62 retention mechanism came from earlier expression studies that used an SFV p62 construct lacking other viral structural protein genes (3). These showed that, when synthesized alone, the p62 protein remained Endo H sensitive, indicating retention in the early secretory pathway. Interestingly, in the same study it was shown that, when synthesized at a higher level from an SFV C-p62 construct, p62 was trimmed into an Endo H-resistant form, reflecting transport through the Golgi, and proteolytically cleaved into a faster-migrating E2 protein, named E2*, in the late secretory pathway (3). This form differed from normal E2 made in context of the E2/E1 heterodimer, which was migrating slightly faster in PAGE and which in addition is sensitive to Endo H (3). These earlier findings were confirmed, and the results are shown in Fig. 1A and B. The difference in p62 retention in the two cases could be due either to the presence or absence of the C protein or to the different expression levels. When p62 is made without C, it is produced at a 10- to 20-fold-lower level than under wt conditions. This is due to the lack of a translation-enhancing RNA structure, which is located in the 5' coding region of the C gene, the first gene in the subgenome (13, 33). The possible role of C in promoting p62 export from the ER was investigated by coexpressing the C and p62 genes from separate coding units, thereby keeping the p62 gene expression level low while still expressing the C gene. By using electroporation, an approximately 100% transfection frequency is obtained, leading to coexpression of C and p62 genes in most cells (24). Expression of the C gene was confirmed by analyzing total cell lysates by SDS-PAGE (data not shown). Immunoprecipitation of p62 from the same lysate and subsequent analyses by SDS-PAGE showed that p62 remained Endo H sensitive and uncleaved and thus remained in a pre-Golgi compartment even in the presence of C (Fig. 1C). To determine whether the C protein only would affect p62 retention when translated in cis, we used a C-deletion mutant in which the nucleotide sequence coding for amino acids 11 to 29 within the translation-enhancer region had been deleted (SFV C11-29-p62) (12). This will ensure low expression levels while still translating a C molecule from the same coding unit as p62. A C protein with this deletion in a full-length construct is still able to produce virus particles, and thus all C interactions necessary for virus assembly are probably preserved (12). The p62 protein was, however, to a large extent retained also in this case (Fig. 1D). A very small number of p62 molecules seemed to be released after a prolonged chase (Fig. 1D). This could be due to a somewhat higher expression level than in the case of p62 expressed from SFV p62. These data suggest that there exists a concentration-dependent retention mechanism for p62 in the early secretory pathway independent of C. Thus, p62 is exported when this retention mechanism is saturated.

View larger version (51K): [in this window] [in a new window]

FIG. 1. The C protein does not promote export of p62. BHK-21 cells were electroporated with in vitro transcribed RNA coding for p62 (A), C-p62 (B), C and p62 (coelectroporation) (C), or C11-29-p62 (D). Cells were metabolically labeled for 15 min followed by chase for 5, 30, 60, and 120 min. Cells were lysed, immunoprecipitated with the monoclonal antibody K26/98 against p62, and either mock or Endo H treated. The Endo H-digested forms of p62 are indicated by H. The Endo H-resistant E2 form is designated E2*. Note that there is significant degradation of E2* from the SFV C-p62 construct during the chase. Note also that the gels were differently exposed so that the signals for all constructs are equally intense for comparison.

Lectin-mediated retention of p62 in the ER. As p62 is dependent upon the ER resident chaperones calnexin and its soluble homologue calreticulin for proper folding and as calnexin has been indicated to be involved in ER retention of subunits prior to oligomeric assembly, the role of these proteins in p62 retention was investigated (27, 30). One way to inhibit calnexin and calreticulin from interacting with substrate protein is to treat cells with CST (21). CST is an -glucosidase inhibitor and will thus inhibit glucose trimming of N-linked oligosaccharides on glycoprotein substrates. Trimming is a necessity for substrate attachment to the two lectins, as binding is specific for the monoglucosylated N-linked oligosaccharide (18). BHK-21 cells were transfected with in vitro-transcribed RNA coding for the p62 protein. At a time 2 h prior to metabolic labeling, cells were subjected to 1 mM CST treatment. This was continued during the pulse and chase. Cells were then lysed, and the p62 protein was immunoprecipitated and analyzed by SDS-PAGE (Fig. 2). In the presence of CST a fraction corresponding to, on average, 35% of the p62 molecules (38, 34, and 32% in three separate experiments) was trimmed to the Endo H-resistant E2* form, indicating transport out of the early secretory pathway (Fig. 2, lanes 7 and 8). This form is not observed in the absence of CST (Fig. 2, lanes 5 and 6). The amount of E2* did not increase significantly when the CST concentration was increased twofold, showing that 1 mM CST was enough to obtain maximum release of p62 (data not shown). Therefore, calnexin and calreticulin are, to some extent, able to prevent export of p62 from the ER. The reason why not all p62 is exported might be retention by alternative chaperones (see Discussion).

View larger version (28K): [in this window] [in a new window]

FIG. 2. CST treatment induces p62 transport and processing into E2*. Cells were electroporated with the SFV p62 construct and pulse-chased in the presence or absence of CST. Note that there is a slight decrease in mobility of p62 from CST-treated cells (lanes 3 and 7) from that of p62 from untreated cells (lanes 1 and 5), probably due to the presence of glucose residues in three N-linked oligosaccharides of p62. This form of p62 is termed p62CST. All other bands are marked as described in Fig. 1 legend.

Mapping of lectin-binding domain on p62. The prevailing model for substrate interaction with calnexin and calreticulin is through monoglucosylated oligosaccharides only, although there is some support for protein-protein-mediated interactions as well (10, 18). As the most N-terminal glycan of p62 is thought to be important for directing p62 into the lectin-mediated folding pathway, we were interested to determine if there also existed a particular glycan in p62 that would mediate ER retention (27). Therefore, each of the three acceptor sites for N-linked glycosylation in p62 was abolished by changing the asparagine to a glutamine. The three mutants, named SFV p62N13Q, SFV p62N200Q, and SFV p62N262Q, were then analyzed in pulse-chase experiments in the presence or absence of CST. None of the mutants produced detectable amounts of Endo H-resistant E2 molecules (Fig. 3a, lanes 7, 11 and 15). This, however, was also to a great extent the case when the mutants were analyzed in the presence of CST (Fig. 3a, lanes 8, 12, and 16). This indicates that the deletion of any glycan in the p62 protein leads to changes in the protein such that the protein does not become properly folded. The mutants might have increased affinity for calreticulin and calnexin or other chaperones and are therefore retained. It is thus concluded that the part of p62 interacting with calnexin and/or calreticulin mediating the retention could not be linked to a specific glycan. To check whether the p62 glycosylation mutants as well as p62 could indeed interact with calnexin, cells expressing the p62 gene or the different p62 gene constructs were metabolically labeled and analyzed by immunoprecipitation with anticalnexin antibody. For all p62 constructs, a coprecipitation of p62 and a protein with an apparent molecular mass of 90 kDa, probably representing the abnormally running 65-kDa calnexin, were observed when analyzed by SDS-PAGE (Fig. 3b, lanes 1 to 4) (36). This suggested that calnexin interacted with p62 in all cases. When the same analyses were performed on cells that had been treated with CST, only the putative calnexin band was seen (Fig. 3b, lanes 5 to 8). This shows that p62 as well as all three p62 glycosylation mutants can interact with calnexin and that this interaction is prevented by CST treatment.

View larger version (30K): [in this window] [in a new window]

FIG. 3. Processing and calnexin interaction of p62 and its glycosylation mutants in the presence or absence of CST. p62 gene mutants where the codon for asparagine was changed to glutamine in either position 13, 200, or 262 were made in pSFV p62, and the corresponding RNAs were synthesized and used for expression studies in the presence or absence of CST. Cells expressing the different constructs were metabolically labeled for 15 min and chased for 30 min (a) or labeled for 10 min and chased for 0 min (b). Cell lysates were immunoprecipitated with E2 monoclonal antibody K26/98 (a) or calnexin polyclonal antibody SPA-860 (b). Samples shown in panel a were also either mock or Endo H treated. The constructs SFV p62, SFV p62N13Q, SFV p62N200Q, and SFVp62N262Q are termed p62, N13Q, N200Q, and N262Q, respectively. Calnexin is marked cnx. All other bands are marked as in previous figures. Molecular mass markers (in kilodaltons) are indicated to the right of the gel in panel b.

Decreased p62/E1 heterodimerization in the presence of CST. Having established that there existed a lectin-mediated retention mechanism for p62 in the ER, we wanted to find out whether this was important for p62/E1 heterodimerization. Therefore, we analyzed whether CST treatment of wt virus-infected cells affected the cis heterodimerization process of E1 and p62 in ER. Decreased cis heterodimerization will result in an increased fraction of separated E1 and p62 molecules after synthesis. The separated subunits should behave like E1 and p62, which have been produced in cells expressing separate E1 and p62 coding units and which have been studied before (2-4). Accordingly, E1 will be retained in ER forming an ER pool, whereas p62 can be transported to the PM alone. However, p62 can also partially complex in trans with E1 in the ER pool and then be transported. These alterations in handling the subunits in the cell will affect their intracellular processing and thereby provide several independent biochemical assays for measuring heterodimerization apart from direct coimmunoprecipitation. Thus, the retained E1 will stay Endo H sensitive, whereas the complexed and transported E1 will become Endo H resistant. The noncomplexed p62 will during transport be processed by cleavage and sugar trimming into Endo H-resistant E2* and the complexed p62 into normal E2, which will remain Endo H sensitive. Therefore, decreased heterodimerization should be accompanied by E2* formation, an increased fraction of Endo H-sensitive E1, and a decreased ratio between the Endo H-resistant E1 and E2 when measured in pulse-chase experiments. The latter effect is expected when pulse-labeled p62 complexes in trans with E1 from the ER pool, which will contain both labeled and unlabeled E1 subunits.

We first studied whether E2* was produced in CST-treated cells that had been infected with wt virus and pulse-labeled, by immunoprecipitation with a monoclonal anti-E2 antibody. By using the E2* of an SFV p62 vector-transfected cell sample as control in SDS-PAGE analysis (Fig. 4, lanes 4 and 8), we were able to detect E2* also in the sample of CST-treated wt SFV-infected cells (Fig. 4, lane 6). Because it migrated only slightly more slowly than E2, it tends to fuse with the upper part of the latter band. However, after Endo H treatment, which affects E2 but not E2*, the latter was clearly visible as a distinct band (Fig. 4, lane 7). The E2* band was not observed in wt virus-infected control cells. This was evident by the gap between the p62 and E2 band (Fig. 4, lane 2). The absence of E2* in control cells was less evident after Endo H treatment due to the smearing of p62 (Fig. 4, lane 3). Thus, a fraction of p62 is able to escape the ER in unheterodimerized form when calnexin- and calreticulin-p62 interactions are inhibited. Quantitation showed that one-fifth of p62 had escaped heterodimerization and formed E2*. The effect of CST on p62/E1 heterodimerization was further studied by analyzing the amount of Endo H-resistant E1 (see E1 bands in Fig. 5, lane 2) as well as analyzing the relationship between the Endo H-resistant E1 and the E2 (see E1 and E2 bands in Fig. 5, lane 2) directly in total cell lysates. Note that all E2 is Endo H sensitive and thus migrates as a faster band in the analysis of the Endo H-treated sample. A modest 11% decrease in Endo H-resistant E1 formation in the presence of CST from that found in untreated cells as well as an 11% decrease of the ratio of Endo H-resistant E1 to E2 in the presence of CST was observed. Thus, these analyses also showed that a fraction of E2/E1 has not succeeded in heterodimerizing in cis in the presence of CST; i.e., the p62-lectin interaction promotes efficient cis heterodimerization. Finally, the CST effect on p62/E1 heterodimerization was studied by a p62 and E1 coimmunoprecipitation assay with a p62 monoclonal antibody termed UM 5.1 (37). Consequently wt virus-infected cells were pulsed for 5 min, chased for 5 min, and then used for heterodimerization analyses. As can be seen in Fig. 6, the amount of E1 molecules immunoprecipitated decreased in the presence of CST from that in the absence of the drug, again indicating that heterodimerization is impaired in the presence of CST. By quantitation the decrease was determined to be on average 20% (18 and 22% in two separate experiments). Taken together, these data point to a role for calreticulin and calnexin in promoting efficient SFV p62/E1 heterodimerization in the ER.

View larger version (26K): [in this window] [in a new window]

FIG. 4. CST treatment of wt SFV-infected cells affects heterodimerization and results in E2* production. BHK-21 cells were infected with wt SFV. Cells were metabolically labeled for 15 min and chased for 30 min, and viral proteins were recovered with the E2 monoclonal antibody K26/98, which does not preserve the E2/E1 interaction (3). Therefore, the E1 protein is not precipitated. Mock and Endo H treatments were performed on the precipitates followed by SDS-PAGE. Cells transfected with SFV p62 (lanes 1, 4, 5 and 8) were processed in a manner similar to that of control.

View larger version (51K): [in this window] [in a new window]

FIG. 5. CST treatment of wt SFV-infected cells affects heterodimerization and increases the fraction of Endo H-sensitive E1. BHK-21 cells were infected with wt SFV and were metabolically labeled for 15 min and chased for 60 min. Cells were lysed; lysates were either mock or Endo H treated. The Endo H-sensitive forms of p62, E2, and E1 are indicated by H. Note that the E2* form seen in the previous figure is too weak to be detected when analyzing total lysate.

View larger version (69K): [in this window] [in a new window]

FIG. 6. Decreased coimmunoprecipitation of E1 and p62 in CST-treated cells. Cells infected with wt SFV were incubated with or without CST, metabolically labeled for 5 min, and chased for 5 min. Cells were lysed and immunoprecipitated with the E2 monoclonal antibody UM 5.1, which preserves heterodimers.

Chaperone saturation leads to enhanced E2* formation. To study whether the relatively minor effect seen on p62/E1 heterodimerization in CST-treated cells was due to compensatory p62 retention by other chaperones, we made cotransfections with the SFV C-p62-E1 nsp and the SFV C-p62^SQL constructs. The former construct contains the complete structural gene region but lacks the nonstructural polymerase region. Consequently, it can only replicate in SFV C-p62^SQL-cotransfected cells where polymerase is available. The rationale of this approach was that p62^SQL would saturate ER chaperones, just as when expressed from the SFV C-p62 construct. This would lead to an increased formation of E2* from the SFV C-p62-E1 nsp construct, should alternative chaperones be involved in p62 retention. The advantage of using p62^SQL for saturation is that it cannot form E2* due to the SQL mutation, and hence all E2* detected will be derived from the p62 expressed by the SFV C-p62-E1 nsp construct. p62^SQL that escapes retention forms p62^SQL*, migrating more slowly than p62. The analyses showed that as much as four-fifths of p62 expressed by the SFV C-p62-E1 nsp construct escaped retention and formed E2* in the presence of p62^SQL (Fig. 7). The result indicated that the reason why p62/E1 heterodimerization is not completely inhibited in CST-treated cells could indeed be due to p62 retention by other ER chaperones. Taken together, these results suggest that the lectin chaperone-mediated p62 retention is necessary and perhaps even essential for efficient p62/E1 heterodimerization.

View larger version (36K): [in this window] [in a new window]

FIG. 7. Enhanced E2* production in cells cotransfected with SFV C-p62SQLand SFV C-p62-E1 nsp RNAs. Cotransfected cells were metabolically labeled for 15 min and were chased for 30 min. Mock- and Endo H-treated immunoprecipitates of cell lysate were analyzed by SDS-PAGE.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Based on previous findings that p62/E1 heterodimerization preferentially occurs between subunits generated from the common C-p62-E1 translation product (cis heterodimerization) and that p62 molecules at low expression levels are retained intracellularly, it was postulated that there would exist a p62 retention mechanism in the ER (3). p62 retention would be necessary to retain p62 in the ER, most likely at the translocon, until the E1 protein had been synthesized and folded into a conformation competent for p62/E1 heterodimerization. The p62 retention would be relieved upon p62/E1 heterodimerization, leading to concomitant export of the heterodimer to the PM. Thus, p62 retention would ensure efficient cis heterodimerization and thereby transport of the fusion-mediating E1 protein, which is totally dependent on heterodimerization for cell surface transport (2).

As the ER chaperone calnexin, which is a type I transmembrane protein, has been shown to be associated with nascent chains, to interact with ribosomes, and to be involved in the folding of p62, calnexin was considered a likely molecule to mediate p62 retention (8, 9, 27). When treating cells expressing the p62 gene with CST, p62 retention was partially relieved. This suggests that retention of p62 could at least to some extent be mediated by calnexin or its soluble homologue calreticulin. It has been shown that, when cells are treated with CST, calnexin-dependent proteins are handled by other chaperones instead (41). Thus, there is a redundancy in the ER-folding machinery where different chaperones and folding factors complement each other, serving as backup systems should one system fail or be overloaded. This could explain why only 35% of the p62 molecules were exported in cells treated with CST.

Although the p62 retention that we observed when producing this subunit at a low level and separately from E1 was consistent with a retention model, the efficient PM transport with an increased rate of synthesis, as when p62 is produced from an SFV C-p62 construct, was not. A reasonable explanation for this apparent discrepancy is that, when p62 is produced at a high level from an SFV C-p62 construct, the backup system also gets saturated, leading to export of most p62 molecules. The reason why p62 release and concomitant E2* formation were not taking place during a wt virus infection is probably that p62 and E1 rapidly heterodimerize. This will effectively minimize the time that p62 occupies the chaperones in the ER.

To directly see whether the observed p62 lectin-mediated retention was necessary for cis heterodimerization as proposed by the model, cells infected with wt SFV were treated with CST. This treatment resulted in a decrease of p62/E1 heterodimerization as measured by a decrease of Endo H-resistant E1, an increase in E2 in relation to Endo H-resistant E1, and formation of E2*. These results were confirmed by coimmunoprecipitation of newly synthesized E1 with an E2 monoclonal antibody. In the presence of CST, E1 precipitation was decreased, showing a decreased heterodimerization efficiency. It could be argued that treatment of cells with CST would affect the folding of E1 and thereby decrease the amount of E1 precipitating with the E2 antibody. It has, however, been shown that E1 primarily makes use of the molecular chaperone BiP for folding (27).

Another finding that seemed inconsistent with the p62/lectin-retention model was that heterodimerization could to a great extent still take place when p62-lectin interactions were inhibited. How can a minor effect on heterodimerization by lectin chaperone inhibition be reconciled with a model where lectin-binding mediates cis heterodimerization? The largest effect on the cellular processes when inhibiting calnexin and calreticulin with CST is on protein folding. It has been shown that 80% of p62 and E1 synthesized in cells treated with CST will aggregate (28). Likewise, we find considerable aggregation of p62 and E1 and a 60% reduction in SFV spike subunit incorporation into viral particles from cells treated with CST compared to the result for particles from untreated cells (H. Andersson and H. Garoff, unpublished data). Thus, a significant part of subunits aggregates, probably to a large extent prior to oligomerization and would, if aggregation could be prevented, fail to oligomerize. This together with the redundancy of folding factors in the ER compensating for lectin chaperones during CST treatment makes the effect of lectin chaperone inhibition on cis heterodimerization minor. As shown in the coexpression experiment where p62^SQL that was synthesized from the SFV C-p62^SQL construct supposedly saturated ER chaperones, there was indeed formation of a large amount of E2* derived from the SFV C-p62-E1 nsp construct, thus demonstrating the importance of chaperones in cis heterodimerization. Furthermore, alternative chaperones could retain unheterodimerized spike subunits in the ER, thereby promoting or at least giving the subunits an opportunity to heterodimerize in trans.

Taken together, we have shown that there exists a retention mechanism in the ER for the SFV p62 protein. This retention is important and possibly essential for efficient cis heterodimerization of the two spike subunits p62 and E1. The retention is mediated either by the ER resident lectin chaperone calnexin or calreticulin or both, although involvement of additional factors cannot be excluded.

 
We thank Ylva Rabo for cell cultures, Mathilda Sjöberg for critical reading of the manuscript, and Bernd Uwe Barth, Kerstin Forsell, and Maarit Suomalainen for SFV plasmids. We are also grateful to A. Salmi and H. Snippe for the generous gift of antibodies.

This work was supported by Swedish Research Council grant B 5107-20006266/2000 to H.G.

 
* Corresponding author. Mailing address: Department of Biosciences at Novum, Karolinska Institute, S-141 57 Huddinge, Sweden. Phone: 46-8-608 91 20/7. Fax: 46-8-774 55 38. E-mail: Helena.Andersson{at}biosci.ki.se.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Aliperti, G., and M. J. Schlesinger. 1978. Evidence for an autoprotease activity of sindbis virus capsid protein. Virology 90:366-369.[CrossRef][Medline]
2
2. Andersson, H., B. U. Barth, M. Ekstrom, and H. Garoff. 1997. Oligomerization-dependent folding of the membrane fusion protein of Semliki Forest virus. J. Virol. 71:9654-9663.[Abstract]
3
3. Barth, B. U., and H. Garoff. 1997. The nucleocapsid-binding spike subunit E2 of Semliki Forest virus requires complex formation with the E1 subunit for activity. J. Virol. 71:7857-7865.[Abstract]
4
4. Barth, B. U., J. M. Wahlberg, and H. Garoff. 1995. The oligomerization reaction of the Semliki Forest virus membrane protein subunits. J. Cell Biol. 128:283-291.[Abstract/Free Full Text]
5
5. Berglund, P., M. Sjoberg, H. Garoff, G. J. Atkins, B. J. Sheahan, and P. Liljestrom. 1993. Semliki Forest virus expression system: production of conditionally infectious recombinant particles. Bio/Technology 11:916-920.[CrossRef][Medline]
6
6. Boere, W. A., T. Harmsen, J. Vinje, B. J. Benaissa-Trouw, C. A. Kraaijeveld, and H. Snippe. 1984. Identification of distinct antigenic determinants on Semliki Forest virus by using monoclonal antibodies with different antiviral activities. J. Virol. 52:575-582.[Abstract/Free Full Text]
7
7. Bron, R., J. M. Wahlberg, H. Garoff, and J. Wilschut. 1993. Membrane fusion of Semliki Forest virus in a model system: correlation between fusion kinetics and structural changes in the envelope glycoprotein. EMBO J. 12:693-701.[Medline]
8
8. Chen, W., J. Helenius, I. Braakman, and A. Helenius. 1995. Cotranslational folding and calnexin binding during glycoprotein synthesis. Proc. Natl. Acad. Sci. USA 92:6229-6233.[Abstract/Free Full Text]
9
9. Chevet, E., H. N. Wong, D. Gerber, C. Cochet, A. Fazel, P. H. Cameron, J. N. Gushue, D. Y. Thomas, and J. J. Bergeron. 1999. Phosphorylation by CK2 and MAPK enhances calnexin association with ribosomes. EMBO J. 18:3655-3666.[CrossRef][Medline]
10
10. Danilczyk, U. G., and D. B. Williams. 2001. The lectin chaperone calnexin utilizes polypeptide-based interactions to associate with many of its substrates in vivo. J. Biol. Chem. 276:25532-25540.[Abstract/Free Full Text]
11
11. de Curtis, I., and K. Simons. 1988. Dissection of Semliki Forest virus glycoprotein delivery from the trans-Golgi network to the cell surface in permeabilized BHK cells. Proc. Natl. Acad. Sci. USA 85:8052-8056.[Abstract/Free Full Text]
12
12. Forsell, K., M. Suomalainen, and H. Garoff. 1995. Structure-function relation of the NH2-terminal domain of the Semliki Forest virus capsid protein. J. Virol. 69:1556-1563.[Abstract]
13
13. Frolov, I., and S. Schlesinger. 1994. Translation of Sindbis virus mRNA: effects of sequences downstream of the initiating codon. J. Virol. 68:8111-8117.[Abstract/Free Full Text]
14
14. Garoff, H., A. M. Frischauf, K. Simons, H. Lehrach, and H. Delius. 1980. Nucleotide sequence of cdna coding for Semliki Forest virus membrane glycoproteins. Nature 288:236-241.[CrossRef][Medline]
15
15. Garoff, H., K. Simons, and B. Dobberstein. 1978. Assembly of the Semliki Forest virus membrane glycoproteins in the membrane of the endoplasmic reticulum in vitro. J. Mol. Biol. 124:587-600.[CrossRef][Medline]
16
16. Green, J., G. Griffiths, D. Louvard, P. Quinn, and G. Warren. 1981. Passage of viral membrane proteins through the Golgi complex. J. Mol. Biol. 152:663-698.[CrossRef][Medline]
17
17. Hahn, C. S., E. G. Strauss, and J. H. Strauss. 1985. Sequence analysis of three Sindbis virus mutants temperature-sensitive in the capsid protein autoprotease. Proc. Natl. Acad. Sci. USA 82:4648-4652.[Abstract/Free Full Text]
18
18. Hebert, D. N., B. Foellmer, and A. Helenius. 1995. Glucose trimming and reglucosylation determine glycoprotein association with calnexin in the endoplasmic reticulum. Cell 81:425-433.[CrossRef][Medline]
19
19. Horton, R. M., H. D. Hunt, S. N. Ho, J. K. Pullen, and L. R. Pease. 1989. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77:61-68.[CrossRef][Medline]
20
20. Karaivanova, V. K., P. Luan, and R. G. Spiro. 1998. Processing of viral envelope glycoprotein by the endomannosidase pathway: evaluation of host cell specificity. Glycobiology 8:725-730.[Abstract/Free Full Text]
21
21. Kaushal, G. P., and A. D. Elbein. 1994. Glycosidase inhibitors in study of glycoconjugates. Methods Enzymol. 230:316-329.[Medline]
22
22. Kielian, M., S. Jungerwirth, K. U. Sayad, and S. DeCandido. 1990. Biosynthesis, maturation, and acid activation of the Semliki Forest virus fusion protein. J. Virol. 64:4614-4624.[Abstract/Free Full Text]
23
23. Liljestrom, P., and H. Garoff. 1991. Internally located cleavable signal sequences direct the formation of Semliki Forest virus membrane proteins from a polyprotein precursor. J. Virol. 65:147-154.[Abstract/Free Full Text]
24
24. Liljestrom, P., and H. Garoff. 1991. A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. Bio/Technology 9:1356-1361.[CrossRef][Medline]
25
25. Liljestrom, P., S. Lusa, D. Huylebroeck, and H. Garoff. 1991. In vitro mutagenesis of a full-length cDNA clone of Semliki Forest virus: the small 6,000-molecular-weight membrane protein modulates virus release. J. Virol. 65:4107-4113.[Abstract/Free Full Text]
26
26. Melancon, P., and H. Garoff. 1986. Reinitiation of translocation in the Semliki Forest virus structural polyprotein: identification of the signal for the E1 glycoprotein. EMBO J. 5:1551-1560.[Medline]
27
27. Molinari, M., and A. Helenius. 2000. Chaperone selection during glycoprotein translocation into the endoplasmic reticulum. Science 288:331-333.[Abstract/Free Full Text]
28
28. Molinari, M., and A. Helenius. 1999. Glycoproteins form mixed disulphides with oxidoreductases during folding in living cells. Nature 402:90-93.[CrossRef][Medline]
29
29. Moore, S. E., and R. G. Spiro. 1990. Demonstration that Golgi endo-alpha-D-mannosidase provides a glucosidase-independent pathway for the formation of complex N-linked oligosaccharides of glycoproteins. J. Biol. Chem. 265:13104-13112.[Abstract/Free Full Text]
30
30. Rajagopalan, S., Y. Xu, and M. B. Brenner. 1994. Retention of unassembled components of integral membrane proteins by calnexin. Science 263:387-390.[Abstract/Free Full Text]
31
31. Schlesinger, M. J., and S. Schlesinger. 1986. Formation and assembly of alphavirus glycoproteins, p. 121-148. In M. J. Schlesinger and S. Schlesinger (ed.), The Togaviridae and Flaviviridae. Plenum Press, New York, N.Y.
32
32. Schlesinger, S., and M. J. Schlesinger. 1972. Formation of Sindbis virus proteins: identification of a precursor for one of the envelope proteins. J. Virol. 10:925-932.[Abstract/Free Full Text]
33
33. Sjoberg, E. M., M. Suomalainen, and H. Garoff. 1994. A significantly improved Semliki Forest virus expression system based on translation enhancer segments from the viral capsid gene. Bio/Technology 12:1127-1131.[CrossRef][Medline]
34
34. Smyth, J., M. Suomalainen, and H. Garoff. 1997. Efficient multiplication of a Semliki Forest virus chimera containing Sindbis virus spikes. J. Virol. 71:818-823.[Abstract]
35
35. Suomalainen, M., P. Liljestrom, and H. Garoff. 1992. Spike protein-nucleocapsid interactions drive the budding of alphaviruses. J. Virol. 66:4737-4747.[Abstract/Free Full Text]
36
36. Wada, I., D. Rindress, P. H. Cameron, W. J. Ou, J. J. Doherty II, D. Louvard, A. W. Bell, D. Dignard, D. Y. Thomas, and J. J. Bergeron. 1991. SSR alpha and associated calnexin are major calcium binding proteins of the endoplasmic reticulum membrane. J. Biol. Chem. 266:19599-19610.[Abstract/Free Full Text]
37
37. Wahlberg, J. M., W. A. Boere, and H. Garoff. 1989. The heterodimeric association between the membrane proteins of Semliki Forest virus changes its sensitivity to low pH during virus maturation. J. Virol. 63:4991-4997.[Abstract/Free Full Text]
38
38. Wahlberg, J. M., R. Bron, J. Wilschut, and H. Garoff. 1992. Membrane fusion of Semliki Forest virus involves homotrimers of the fusion protein. J. Virol. 66:7309-7318.[Abstract/Free Full Text]
39
39. Wahlberg, J. M., and H. Garoff. 1992. Membrane fusion process of Semliki Forest virus. I: Low pH-induced rearrangement in spike protein quaternary structure precedes virus penetration into cells. J. Cell Biol. 116:339-348.[Abstract/Free Full Text]
40
40. Yon, J., and M. Fried. 1989. Precise gene fusion by PCR. Nucleic Acids Res. 17:4895.[Free Full Text]
41
41. Zhang, J. X., I. Braakman, K. E. Matlack, and A. Helenius. 1997. Quality control in the secretory pathway: the role of calreticulin, calnexin and BiP in the retention of glycoproteins with C-terminal truncations. Mol. Biol. Cell 8:1943-1954.[Abstract/Free Full Text]

---

Journal of Virology, June 2003, p. 6676-6682, Vol. 77, No. 12
0022-538X/03/$08.00+0     DOI: 10.1128/JVI.77.12.6676-6682.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Andersson, H. Articles by Garoff, H. Search for Related Content PubMed PubMed Citation Articles by Andersson, H. Articles by Garoff, H.