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

J. Biol. Chem., Vol. 276, Issue 47, 44239-44246, November 23, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/47/44239    most recent M105967200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Plemper, R. K. Articles by Cattaneo, R. Search for Related Content PubMed PubMed Citation Articles by Plemper, R. K. Articles by Cattaneo, R. Social Bookmarking                      What's this?

Measles Virus Envelope Glycoproteins Hetero-oligomerize in the Endoplasmic Reticulum^*

Richard K. Plemper, Anthea L. Hammond, and Roberto Cattaneo

From the Molecular Medicine Program, Mayo Foundation, Rochester, Minnesota 55905

Received for publication, June 27, 2001, and in revised form, September 4, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The endoplasmic reticulum (ER) was investigated as the initial oligomerization site for the envelope glycoproteins H and F of measles virus (MV), a clinically relevant member of the Paramyxoviridae family, and consequences of this interaction for viral replication were studied. Both proteins were tagged at their cytosolic tails with RRR and KKXX motifs, respectively, resulting in their efficient retention in the ER. Co-transfection of the retained constructs with transport competent MV glycoproteins revealed a dominant negative effect on their biological activity indicating intracellular complex formation and thus retention. Pulse-chase analysis and co-immunoprecipitation experiments demonstrated that this effect is based on both homo- and hetero-oligomerization in the ER. Recombinant viruses additionally expressing ER-retained F showed an altered cytopathic phenotype accompanied by greatly reduced particle release. Similar mutant viruses additionally expressing ER-retained H could not be rescued indicating an even greater negative effect of this protein on virus viability. Our study suggests that both homo- and hetero-oligomerization of MV glycoproteins occur in the ER and that these events are of significance for early steps of particle assembly.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Measles virus (MV)¹ is a negative-stranded RNA virus of the Paramyxoviridae family. Despite the existence of an effective live attenuated vaccine based on the MV-Edmonston strain (MV-Edm), MV remains among the 10 most potent global pathogens, killing over 1 million children annually in countries where vaccination is not routine (1).

The MV envelope consists of hemagglutinin (H) and fusion (F) glycoproteins. Whereas H binds to the MV receptors CD46 (2, 3) and signaling lymphocytic activation molecule (4, 5), F carries a hydrophobic fusion peptide that mediates membrane fusion upon receptor binding of H (6). MV H is thought to exist at the viral surface as a tetramer consisting of a dimer of two covalently linked dimers (7), whereas F is considered to trimerize. Upon synthesis as an inactive precursor F₀, the protein becomes proteolytically activated in the trans-Golgi network by furin yielding a large transmembrane F₁ and a small F₂ fragment.

The mechanism of MV-induced membrane fusion may involve receptor-induced conformational changes in H and then F, pointing to a dynamic interaction between these proteins. This was first supported through studies showing that expression of H or hemagglutinin neuraminidase (HN) and F from the same paramyxovirus is necessary for efficient fusion, inferring their type-specific interaction (8-10). Furthermore, specific mutations in HN lead to loss of fusion (10-12), and direct cell surface interactions between MV H and F (13) and human parainfluenza virus 2 (HPIV-2) HN and F (14) have been demonstrated by co-immunoprecipitation experiments.

Although compelling evidence for an interaction between H/HN and F exists, it remains unclear at which stage of cellular processing this occurs. Upon synthesis, both proteins are integrated into the endoplasmic reticulum (ER) membrane. Because the ER possesses a high density of molecular chaperones (15), the virus might exploit these for the efficient formation of envelope glycoprotein complexes. Indeed, evidence for an ER association between HPIV-2 and -3 HN and F has been presented (16, 17). However, in these studies soluble derivatives of the glycoproteins carrying KDEL sequences (18) were used, which may show folding patterns distinct from the native transmembrane proteins. In contrast, another study demonstrated that full-length simian virus 5 (SV5) and HPIV-3 HN and F proteins retained in the ER did not affect transport of homotypic HN and F species (19). For these experiments, the authors added RRRRR (20) and KKXX motifs (21) to the cytosolic tails of HN and F, respectively. In several studies theses retention motifs have proven to be valuable tools to study virus assembly (22, 23) and the cell biology of secretory proteins (24, 25).

All studies following the interaction of paramyxovirus glycoproteins retained in the ER so far were based on the fate of plasmid-encoded, transiently expressed glycoproteins. Therefore, the effect of intracellularly retained viral glycoproteins on productive complex assembly in the context of a viral infection is unclear. We show that the H and F glycoproteins of MV interact in the ER and that this interaction has consequences for viral replication. By examining the influence of transiently expressed ER-retained F and H on transport of their unmodified homologues, strong homotypic and heterotypic interactions of MV glycoproteins in the ER were identified, which significantly reduced their biological activity. During virus infection, these interactions were found to be important for MV release and cytopathicity, implicating the ER as a site of MV glycoprotein complex formation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Culture, Generation of MV Stocks, and Virus Growth Kinetics-- Vero (African green monkey kidney), HT1080 (human bone fibrosarcoma), and PA317 (mouse hybridoma) cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, penicillin, and streptomycin at 37 °C and 5% CO₂. Stably transfected HT1080 cells were grown in the presence of 2.5 µg/ml puromycin and 293-3-46 helper cells (26) in 1.2 µg/ml geneticin. For transient transfection LipofectAMINE (Life Technologies, Inc.) was used, and cells were analyzed 18-24 h post-transfection.

To prepare virus stocks, Vero cells were infected at a multiplicity of infection (m.o.i.) of 0.1 plaque-forming units (pfu)/cell and incubated at 37 °C until particles were released by three freeze-thaw cycles. Titers were determined by 50% tissue culture infective dose (TCID₅₀) titration on Vero cells according to the Spearman-Karber method (27).

For virus growth kinetics, Vero cells (4 × 10⁵ per time point) were infected with an m.o.i. of 0.03 pfu/cell and incubated at 37 °C. At the indicated time points, virus titers in cleared supernatants and freeze-thawed cell lysates were determined by TCID₅₀ titration.

Plasmid Construction-- Parental plasmids for mutagenesis and all experiments were pCG-H_Edm and pCG-F_Edm encoding MV-Edm H and F under the control of the cytomegalovirus promoter (28). Site-directed mutagenesis was performed using the quick change system (Stratagene) and mutant constructs confirmed by DNA sequencing and Western analysis. In all primer sequences altered or added nucleotides are underlined.

Primers used to generate pCG-H_Edm-ER were 5-CTTAGGGTGCAAGATCATCGATAATGCGACGTCGGCGCCGTTCACCACAACGAGACCGGATAAATGC and to generate pCG-F_Edm-ER were 5-CCTATGTAAGGTCGCTCAAGAGTAAGACTCATTAATGATCCTCTACAACTCTTG. If indicated, the sequence encoding the FLAG epitope (DYKDDDDK) was inserted upstream of the MV-H_Edm, MV-H_Edm-ER, and MV-F_Edm stop codon for detection purposes. Primers used were 5-CGGGAAGATGGAACCAATCGCAGAGACTACAAGGATGACGATGACAAGTAGGGCTGCTAGTGAACCAATCTC to generate MV-H_Edm-C_FLAG and MV-H_Edm-ER C_FLAG and 5-CATCAAAATCCTATGTAAGGTCGCTCGACTACAAGGATGACGATGACAAGTGATCCTCTACAACTCTTGAAACAC to generate MV-F_Edm-C_FLAG. Biological activity of all tagged proteins was unchanged when compared with the parental versions in transient transfection.

To integrate H_Edm-ER and F_Edm-ER in a DNA copy of the measles genome, p(+)MV (26), MluI, and AatII restriction sites were introduced upstream and downstream of the open reading frames adhering to the reported "rule of six" requirement (29). Primers used for pCG-H_Edm-ER C_FLAG were 5-CTTAGGGTGCAAGATCATCGATAACGCGTATGCGGCGTC GGCGCCGTTCACCACAAC to generate the MluI site and 5-GATGACAAGTAGGGCTGCTAGTGAACGACGTCACCAATCTCATGATGTCACCCAGAC to generate the AatII site. Primers used for pCG-F_Edm-ER were 5-CAGAACCCAGACCCCGGCCCAC ACGCGTGCGCCCCCAACCCCCGACAACC to generate the MluI site and 5-CCCTCTGGCCGAACAATATCGGTAGACGTCAAAAG GATCCACTAGTTCTAGAGTCG to generate the AatII site. The MluI AatII fragments were ligated into MluI AatII digested p(+)MPeGFPV and p(+)MHeGFPV plasmids carrying GFP in post-P or post-H position (30), thereby replacing GFP. This resulted in plasmids p(+)MV-(P)H_Edm-ER and p(+)MV-(P)F_Edm-ER and plasmids p(+)MV-(H)H_Edm-ER and p(+)MV-(H)F_Edm-ER. Measles genomes containing the GFP gene in pre-N position and the H_Edm-ER-C_FLAG or F_Edm-ER open reading frame in post-H position were generated by cloning the GFP-containing SacII NotI fragment of p(+)MV-GFP (31) into SacII NotI-digested plasmids p(+)MV-(H)H_Edm-ER and p(+)MV-(H)F_Edm-ER, resulting in plasmids p(+)MV-GFP (H)H_Edm-ER and p(+)MV-GFP (H)F_Edm-ER.

Western Analysis and Deglycosylation-- For Western analysis, 4 × 10⁵ cells were transfected with 2.5 µg of plasmid DNA or infected with an m.o.i. of 0.03 pfu/cell and incubated at 37 °C. At the indicated times, cells were washed in phosphate-buffered saline (PBS) and lysed for 10 min at 4 °C in lysis buffer (50 mM Tris, pH 8.0; 62.5 mM EDTA; 0.4% deoxycholate; 1% Igepal (Sigma)) containing protease inhibitors (Complete mix (Roche Molecular Biochemicals)) and 1 mM phenylmethylsulfonyl fluoride (PMSF). Unless otherwise stated, 2.5 µg of total protein (determined using the DC Protein-Assay Kit (Bio-Rad)) was mixed with urea buffer (200 mM Tris, pH 6.8; 8 M urea; 5% SDS; 0.1 mM EDTA; 0.03% bromphenol blue; 1.5% dithiothreitol) for 25 min at 50 °C. For nonreducing conditions, dithiothreitol was omitted. Samples were fractionated on SDS-polyacrylamide gels, blotted to polyvinylidene difluoride membranes (Millipore), probed with the indicated antibody, and analyzed by enhanced chemiluminescence (Amersham Pharmacia Biotech). For endoglycosidase H (Endo H) treatment, lysates were mixed with denaturing buffer (final concentration 0.5% SDS; 1% -mercaptoethanol) for 25 min at 50 °C and then with deglycosylation buffer (final concentration 50 mM sodium citrate, pH 5.5) and 0.5 units of Endo H for 12 h at 37 °C. Urea buffer was added and samples subjected to Western analysis.

Metabolic Labeling and Immunoprecipitation-- Transfected Vero cells were incubated for 30 min in labeling medium lacking cysteine, methionine, and ammonium sulfate and then labeled with 100 µCi/ml [³⁵S]methionine (Amersham Pharmacia Biotech) for 45 min at 37 °C. Subsequently, cells were incubated in chase medium containing 10% fetal calf serum at 37 °C for the indicated chase periods. For direct immunoprecipitation, cells were lysed in radioimmunoprecipitation assay buffer (10 mM Tris, pH 7.4; 1% deoxycholate; 1% Triton X-100; 0.1% SDS; 150 mM sodium chloride) containing protease inhibitors and 1 mM PMSF for 15 min at 4 °C and centrifuged for 25 min at 20,000 × g and 4 °C. Lysates were incubated with antibodies directed against MV H (Chemicon) or the FLAG epitope (M2, Sigma) for 90 min at 4 °C and then immune complexes adsorbed to protein G-agarose (Life Technologies, Inc.) for 90 min at 4 °C. Precipitates were washed in lysis buffer, incubated in urea buffer for 25 min at 50 °C, and fractionated on SDS-polyacrylamide gels. Dried gels were exposed to Biomax films (Eastman Kodak Co.).

Confocal Microscopy-- Transfected Vero cells (5 × 10³ cells in chamber slides (Chamber Slides; Lab Tec)) were fixed with methanol at 20 °C, washed with PBS, and permeabilized with PBS containing 1% Triton X-100. After blocking with nonspecific goat antiserum (Sigma), cells were incubated with antibodies specific for MV H or F tails (32) and subsequently with Texas Red-coupled anti-rabbit serum (Sigma). As ER marker, fluorescein isothiocyanate-coupled concanavalin A (Molecular Probes) was added to the secondary antibody. Cells were washed, fixed with ProLong Antifade Kit (Molecular Probes), and analyzed with a Zeiss LSM 510 confocal microscope.

Sucrose Gradient Fractionation-- Transfected Vero cells (2 × 10⁶) were scraped in resuspension buffer (10 mM Tris, pH 7.4; 10 mM sodium chloride; 1% Triton X-100; 0.5% sodium deoxycholate) containing protease inhibitors and 1 mM PMSF, and 600 µg of total protein was layered onto a 10-26% sucrose gradient prepared in resuspension buffer containing 0.1% Triton X-100. For control, proteins were treated with 0.5% SDS on ice prior to loading. Gradients were centrifuged in an SW41 rotor for 18 h at 38,000 rpm at 10 °C, and 10 equal fractions were collected. Proteins were precipitated with trichloroacetic acid, resuspended in urea buffer, and subjected to Western analysis. The location of marker proteins catalase (240,000 kDa), aldolase (158,000 kDa), and bovine serum albumin (68,000 kDa) (all Sigma) was determined by silver staining (Silver staining kit; Bio-Rad) of SDS-polyacrylamide gels.

Syncytium Formation-- Unless otherwise stated, Vero cells were co-transfected in duplicate with 1.0 µg each of plasmid DNA encoding F_Edm and H_Edm or GFP for control, and 3.0 µg of F_Edm-ER or H_Edm-ER and incubated at 32 °C to prevent them from reaching over-confluency. The number of syncytia in 20 representative fields of view was determined at the indicated times.

Co-immunoprecipitation Experiments-- Cells were co-transfected with 2.5 µg each of plasmid DNA encoding F_Edm and H_Edm or the ER-retained versions. Cells were washed in PBS, scraped in co-immunoprecipitation buffer (10 mM Hepes, pH 7.4; 50 mM sodium pyrophosphate; 50 mM sodium fluoride; 50 mM sodium chloride; 5 mM EDTA; 5 mM EGTA; 100 µM sodium vanadate; 1% Triton X-100) containing protease inhibitors and 1 mM PMSF, and centrifuged for 25 min at 20,000 × g and 4 °C. Protein complexes were immunoprecipitated from 300 µg of total protein using MV H-specific antibodies (Chemicon). As internal standard, 5 µg of total protein was directly mixed with urea buffer. After absorption of immune complexes to protein G-agarose (Life Technologies, Inc.), samples were washed in buffer 1 (100 mM Tris, pH 7.6; 500 mM lithium chloride; 0.1% Triton X-100; 1 mM dithiothreitol) and then in buffer 2 (20 mM Hepes, pH 7.2; 2 mM EGTA; 10 mM magnesium chloride; 0.% Triton X-100; 1 mM dithiothreitol), resuspended in urea buffer for 25 min at 50 °C, and subjected to Western analysis using antibodies specific for F tail.

Rescue of Mutant Viral Particles-- Recombinant MV particles were generated essentially as described (26). The helper cell line 293-3-46 stably expressing MV-N, MV-P, and T7-polymerase was co-transfected by calcium phosphate precipitation (ProFection; Promega) with plasmids encoding the MV genome and L polymerase. Helper cells were overlaid on Vero cells 76 h post-transfection and incubated until syncytia appeared or, for GFP-expressing viruses, fluorescence was detectable. Infectious centers were passaged on fresh Vero cells. The integrity of recombinant viruses was verified by reverse transcriptase-polymerase chain reaction and DNA sequencing.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

MV Glycoproteins Carrying ER Retention Signals at Their Cytosolic Tails Are Stably Expressed and Accumulate in the ER-- To generate ER-retained versions of full-length MV-Edm glycoproteins, a 5-amino acid RRRRR sequence was added to the cytosolic amino terminus of H_Edm and a KSKTH sequence to the carboxyl terminus of F_Edm (Fig. 1A).

View larger version (48K): [in this window] [in a new window]   Fig. 1.   MV glycoproteins carrying retention signals are efficiently retained in the ER. A, scheme of MV H and F glycoproteins showing the five amino acid extensions added to the cytosolic amino and carboxyl terminus. B, pulse chase analysis of HEdm-ER and unmodified HEdm transiently expressed in Vero cells. Numbers indicate chase times in minutes. C, endoglycosidase H (EndoH) treatment and Western analysis of antigenic material shown in B, 180-min chase time. Co contains mock-transfected cell lysates for control. res, resistant; sens, sensitive. D, Western analysis of FEdm and FEdm-ER transfected Vero cells. E and F, confocal analysis of HEdm and HEdm-ER 30 h post-transfection and of FEdm and FEdm-ER 40 h post-transfection. For general ER staining, fluorescein isothiocyanate-labeled concanavalin A (ConA) was used.

Expression and stability first of H_Edm-ER was assessed by transient expression in Vero cells and pulse-chase analysis (Fig. 1B). Whereas the stability of H_Edm-ER was similar to that observed for H_Edm, H_Edm-ER lacked the Golgi-specific conversion from core to complex N-linked oligosaccharide chains characterized by an ~4 kDa increase in molecular mass over time (33), suggesting its inhibited export from the ER. This was further assessed by treating the samples corresponding to 180-min chase time with endoglycosidase H (Endo H) prior to immunoblotting (Fig. 1C). All of the H_Edm-ER material was sensitive to Endo H treatment, supporting its efficient retention in the ER.

Because most oligosaccharides on F₀ are highly sensitive to Endo H (34), Endo H resistance cannot be used to monitor its ER export. Instead, processing of MV F from F₀ to F₁, mediated by furin in late Golgi compartments (35), was assessed after transient transfection of Vero cells; most unmodified F_Edm was proteolytically activated, but no antigenic material corresponding to an F₁ fraction could be detected for F_Edm-ER supporting its ER localization (Fig. 1D).

To investigate further the ER retention of H_Edm-ER and F_Edm-ER, their co-localization with an ER-specific lectin, concanavalin (Con) A, was compared with that of the unmodified parental proteins by confocal microscopy (Fig. 1, E and F). Localization of H_Edm-ER overlapped almost exclusively with ConA staining, confirming the ER retention of the H_Edm-ER protein. In cells expressing H_Edm, however, only some overlap with Con A staining was observed corresponding to the normal steady state level of H_Edm in the ER.

Slightly greater co-staining of F_Edm with ConA was observed (Fig. 1F), presumably reflecting a slightly higher steady state level of transport-competent F_Edm than H_Edm in the ER. In cells expressing F_Edm-ER, however, almost complete co-localization with ConA was observed, confirming its ER retention.

Homo-oligomerization Capacity of ER-localized MV Proteins Is Unchanged-- To analyze whether the presence of the ER retention signals influences native folding of H_Edm-ER and F_Edm-ER, their ability to homo-oligomerize was monitored. Covalently linked H-H homodimers were detected by separation of transiently expressed material under reducing and nonreducing conditions (Fig. 2A). Both H_Edm and H_Edm-ER migrated exclusively as dimers under nonreducing conditions, whereas treatment of cell lysates with 1.5% dithiothreitol resulted in migration of both variants as monomers.

View larger version (49K): [in this window] [in a new window]   Fig. 2.   ER-retained MV glycoproteins homo-oligomerize as efficiently as the unmodified proteins. A, Western analysis of Vero cells expressing HEdm and HEdm-ER under reducing (1.5% DTT) and nonreducing (- -) conditions. Numbers correspond to the molecular weight of protein standard (in thousands). B, sucrose gradient centrifugation and Western analysis of FEdm and FEdm-ER in the presence and absence of 0.5% SDS. Lys contains 1 µg of total protein lysates for comparison. Ten gradient fractions from top (1) to bottom (10) are loaded. The positions of marker proteins bovine serum albumin (BSA), aldolase (Ald), and catalase (Cat) separated on independent gradients are indicated.

Because MV F protein is thought to exist as a noncovalently linked trimer, sucrose gradient fractionation of F_Edm-ER or F_Edm proteins was employed to detect oligomeric versions (Fig. 2B). The majority of unmodified F₀ was found in fraction 5 corresponding to the molecular weight of an F trimer as determined by comparison with molecular weight marker proteins. Cleavage of F₀ into F₁ seemed to reduce the stability of the trimers because an increasing amount of F₁ material was found in earlier fractions of the gradient corresponding to monomeric F protein. F_Edm-ER was not proteolytically processed, and its F₀ form was predominantly found in fraction 5, although some antigenic material corresponding to higher order complexes could be detected. Treatment of both F_Edm and F_Edm-ER with 0.5% SDS resulted in a shift of protein to the monomeric form sedimenting in fraction 3 of the gradient because of destruction of noncovalent complexes. Thus, F and H homo-oligomerization capacities are unchanged by their ER retention.

Intracellularly Retained MV Glycoproteins Reduce Transport Kinetics of Their Unmodified Homologues-- We assessed the biological consequence of the ER localization of F_Edm-ER and H_Edm-ER by studying their effect on syncytium induction mediated by unmodified MV H and F when transiently co-expressed in Vero cells (Fig. 3A).

View larger version (46K): [in this window] [in a new window]   Fig. 3.   HEdm-ER and FEdm-ER have a dominant negative effect on the biological activity of unmodified MV glycoproteins. A, kinetics of syncytium formation following transfection of Vero cells with 1 µg of plasmid DNA encoding HEdm, FEdm, and a 3-fold excess of either HEdm-ER or FEdm-ER. For comparison, cells transfected only with HEdm and FEdm, HEdm-ER and FEdm, or FEdm-ER and HEdm are shown. Columns reflect the average number of syncytia scored in three independent experiments. 4000 syncytia per well represent complete lysis which precluded further counting. B, representative fields of view of cells 20 h post-transfection with 0.5 µg of plasmid DNA encoding HEdm and FEdm and with a 5-fold excess of either HEdm-ER or FEdm-ER. For control, cells transfected with HEdm and FEdm only and mock-transfected cells (Co) are shown. C and D, Western analysis of total cell lysates derived from B using MV F- (C) or MV H-specific (D) antibodies. Antigenic material shown in D was subjected to Endo H treatment as indicated.

Expression of either H_Edm-ER with unmodified F_Edm or, reciprocally, F_Edm-ER with unmodified H_Edm resulted in minimal syncytium formation, consistent with efficient intracellular retention of H_Edm-ER and F_Edm-ER. Expression of both unmodified H_Edm and F_Edm with a 3-fold excess of H_Edm-ER resulted in a dominant negative effect on syncytium formation, with a delay in development of cytopathic effects. Expression of F_Edm-ER with unmodified H_Edm and F_Edm also delayed the onset of syncytia but to a lesser degree than that observed with H_Edm-ER. This difference is probably because of the covalent and therefore more stable dimerization of MV H in comparison to the noncovalent homo-oligomerization observed for MV F.

By using a 5-fold excess of the ER-retained constructs to establish more stringent retention of the unmodified proteins, only very few isolated syncytia could be detected in contrast to the massive formation of syncytia in cells expressing the unmodified proteins alone (Fig. 3B). Subsequent Western analysis of cell lysates revealed that in the presence of F_Edm-ER, proteolytic processing of F_Edm to F₁ was strongly inhibited, whereas in its absence, F_Edm was efficiently processed to F₁ indicating transport to post-ER compartments (Fig. 3C).

Similarly, in the cells co-expressing H_Edm-ER (Fig. 3D), H_Edm showed a strongly reduced rate of Golgi-type carbohydrate chain trimming, indicated by its nearly complete sensitivity to Endo H treatment. In the absence of H_Edm-ER however, ~50% of H_Edm was found to be resistant to Endo H treatment. Thus, both F_Edm and H_Edm are strongly retained in the ER in the presence of F_Edm-ER and H_Edm-ER, respectively.

Efficient Hetero-oligomerization of MV Glycoproteins Occurs in the ER-- To address whether hetero-oligomerization of the ER-retained glycoproteins with their unmodified counterparts also contributes to the dominant negative phenotype, their influence on transport kinetics of H_Edm and F_Edm was investigated by pulse-chase analysis (Fig. 4, A and B). FLAG epitope-tagged versions of H_Edm and F_Edm proteins were used for these experiments to achieve identical immunoprecipitation conditions. Transport kinetics of the tagged glycoproteins were shown not to be influenced by the FLAG epitope itself (Ref. 7 and data not shown).

View larger version (52K): [in this window] [in a new window]   Fig. 4.   MV H and F proteins form complexes in the ER. A and B, pulse chase analysis of Vero cells transfected with equal amounts of HEdm and FEdm or FEdm-ER (A) or FEdm and HEdm or HEdm-ER (B). Numbers correspond to chase times in minutes. FLAG epitope-tagged versions of HEdm (A) and FEdm (B) were transfected, and immunoprecipitation was carried out using antibodies directed against the FLAG epitope. Quantifications are based on either the increase in Golgi-type carbohydrate chains (A) or the decrease in intensity of the F0 fractions (B). C, co-immunoprecipitation of FEdm or FEdm-ER with either HEdm or HEdm-ER using antibodies directed against MV H followed by Western analysis of MV F. Lanes marked IP contain material co-precipitated from 100 µg of total cell lysate, whereas total cell lysates (TL) indicate direct loading of 1 µg of total lysate without prior immunoprecipitation. Cells transfected with FEdm or HEdm alone were analyzed for control. Numbers indicate the relative percentage of precipitated material in relation to the total lysate. In case of unmodified proteins H and F, the sum derived from both F fractions, F0 and F1, was used for quantification.

In the presence of equivalent amounts of H_Edm and F_Edm-ER, kinetics of carbohydrate chain conversion of H_Edm to the Golgi-specific pattern was significantly reduced, with far less than 50% conversion after 180 min (Fig. 4A). In contrast, when co-expressed with F_Edm, 50% carbohydrate chain conversion of H_Edm was observed after less than 120 min. When the reciprocal analysis of F_Edm transport was carried out, F_Edm was processed efficiently, with furin cleavage ~75% complete by 120 min in the presence of unmodified H_Edm (Fig. 4B). After co-expression with H_Edm-ER, however, processing to F₁ was less than 30% complete after 120-min chase time. Thus, both H_Edm and F_Edm are significantly retained in the ER because of a heterotypic interaction with H_Edm-ER and F_Edm-ER, respectively. Interestingly, the amount of metabolically labeled F_Edm immunoprecipitated in the presence of H_Edm-ER was consistently reduced compared with that in cells co-expressing H_Edm. Conceivably, this reflects a slightly reduced F_Edm protein expression level in the presence of H_Edm-ER.

To assess hetero-oligomerization of MV glycoproteins in the ER by a more direct biochemical approach, we followed the interaction of F_Edm or F_Edm-ER with H_Edm or H_Edm-ER by co-immunoprecipitation of MV F with H, using anti-F antibodies for detection by Western analysis (Fig. 4C). The relative efficiency of co-precipitation was determined for each sample by comparison with directly loaded total cell lysates. F_Edm-ER not only co-precipitated with H_Edm, but this interaction was significantly more efficient than that of the unmodified parental proteins F_Edm with H_Edm. Similarly, the reciprocal co-immunoprecipitation of F_Edm with H_Edm-ER was substantially more efficient than that of F_Edm with H_Edm (Fig. 4C). For control, cells were co-transfected with both ER-retained glycoproteins; as expected very efficient co-precipitation was observed under these conditions. Thus, the dominant negative effect of the ER-retained species on syncytium formation is attributable to both homotypic interaction of F_Edm-ER with F_Edm or H_Edm-ER with H_Edm and heterotypic oligomerization of F_Edm-ER trimers with H_Edm or H_Edm-ER complexes with F_Edm.

Mutant MV Additionally Expressing F_Edm-ER Shows Reduced Cytopathicity and Release-- To address the interference function of H_Edm-ER, which displayed the stronger negative effect on syncytium formation, in a viral infection, an MV-susceptible HT1080 cell line stably expressing H_Edm-ER was generated. When infected with MV-Edm, however, generation of cell-associated infectious MV particles was unaffected, although release of virus was reduced at early time points (data not shown). Assessing the relative levels of nuclear and virally encoded H proteins demonstrated that the virus overcomes any inhibitory effects of H_Edm-ER by producing a vast excess of virally encoded H_Edm.

To express ER-retained glycoproteins at higher levels in the context of a viral infection, the genes encoding H_Edm-ER and F_Edm-ER were thus inserted as additional transcription units downstream of either P or H in a positive strand copy of the MV genome (Fig. 5A). Because of a transcription gradient described for MV, levels of gene expression from the P position are about three times higher than from the H position (36). In repeated attempts none of these recombinant MVs could be recovered, whereas parallel rescues of unmodified MV-Edm were successful (data not shown).

View larger version (42K): [in this window] [in a new window]   Fig. 5.   Virally encoded ER-retained MV glycoproteins alter the cytopathic effect and viral spread. A, scheme of recombinant MV plasmids containing HEdm-ER or FEdm-ER additional transcription units. The position of the GFP gene in the MV genome in constructs p(+)MV-GFP(H)HEdm-ER and p(+)MV-GFP(H)HEdm-ER is indicated. B, spread of mutant viral particles in tissue culture followed by the appearance of green fluorescence. Vero cells were infected with newly rescued viral particles and incubated at 37 °C. C, growth kinetics of MV-GFP and MV-GFP(H)FEdm-ER in Vero cells. Cells were infected with an m.o.i. of 0.03 pfu/cells. Plotted are titers determined in three independent experiments for both cell associated and released viral particles. D-F, Western analysis of cell culture supernatant (D) and total cell lysates (E and F) derived from Vero cells infected with MV-GFP and MV-GFP(H)FEdm-ER at the indicated time points postinfection. Equal volumes of supernatant or equal amounts of total protein were loaded on the gels. D and E, MV F specific antibodies were used for detection. Antigenic material in F analyzed with MV H-specific antiserum was subjected to Endo H treatment as indicated.

To facilitate detection of infection, full-length genomes carrying GFP in addition to H_Edm-ER and F_Edm-ER in post-H position were generated (Fig. 5A). Whereas MV-GFP (H)H_Edm-ER could not be rescued, MV-GFP (H)F_Edm-ER was recovered, and the integrity of the insert in the viral genome was confirmed for several independent clones by reverse transcriptase-polymerase chain reaction and DNA sequencing (data not shown). When a single infectious center of either MV-GFP (H)F_Edm-ER or MV-GFP was transferred to Vero cells, MV-GFP spread completely through the cell monolayer by 5 days post-infection (Fig. 5B). In contrast, MV-GFP (H)F_Edm-ER demonstrated a greatly reduced lateral spread characterized by the formation of very few defined syncytia even 7 days post-infection.

Release of MV-GFP (H)F_Edm-ER particles into the medium was greatly delayed in comparison with MV-GFP, with ~100-fold less particles released at 48-56 h post-infection (Fig. 5C). This finding mirrored our previous observation of delayed particle release from HT1080 H_Edm-ER cells. Consistent with the viral growth curve, barely detectable levels of F₁ in released MV-GFP (H)F_Edm-ER particles were found 64 h post-infection, whereas F₁ incorporated into MV-GFP was detected 48 h post-infection (Fig. 5D). In MV-GFP-infected cells, F₁ was detected 40 h post-infection, and the F₁ signal was stronger than the F₀ signal (Fig. 5E), a pattern of F expression consistent in MV infections. In MV-GFP (H)F_Edm-ER-infected cells, however, appearance of F₁ was greatly delayed as compared with F₀, with a faint F₁ band visible only after 48 h (Fig. 5E). In contrast, ER export of MV H, as measured by conversion from core to Golgi-specific complex N-linked oligosaccharide chains, was only slightly reduced (Fig. 5F), suggesting a stronger homotypic interaction in the context of this virus. Thus, in the presence of virally encoded F_Edm-ER protein, transport of MV F and release of viral particles are significantly impaired resulting in a diminished cytopathic effect.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Our study demonstrates that ER-localized MV H and F proteins interfere with the ER export of their unmodified homologues via both homo- and hetero-oligomerization. Importantly, ER retention of H and F has biological consequences, not only diminishing their ability to induce cell-cell fusion upon transient expression but also reducing particle release in the context of a viral infection.

Although a concern of our approach may be that ER retention of MV H and F results in formation of non-native multimers,the demonstration that both proteins are able to homo-oligomerize efficiently suggests that this is not the case. Furthermore, that ER-retained SV5 and HPIV-3 HN and F carrying similar retention signals do not interact with their unmodified counterparts (19) suggests that our contrasting findings are not a general consequence of the approach but rather reflect a specific property of MV glycoprotein complex formation.

Given the native conformation of our ER-retained MV glycoproteins and their inhibitory effect on biological activity of unmodified H and F through both homo- and hetero-oligomerization when transiently expressed, it was feasible to assess their impact on virus replication. When ER-retained and nonretained glycoproteins were both virally encoded, the retained proteins apparently had a strong negative effect on virus viability. Several recombinant MV genomes were constructed, but only that expressing the protein least efficient in ER retention, F_Edm-ER, at the lowest level could be recovered and propagated. In MV-GFP (H)F_Edm-ER infected cells release of viral particles was severely impaired, and secondary infections were greatly reduced, resulting in a delay of viral spread through a cell monolayer. Consistent with this, processing of F₀ to F₁ was markedly delayed, suggesting ER retention of the virus-encoded unmodified F protein. This delay was not mirrored when following cell-associated infectious titer, possibly reflecting activation by furin cleavage of the ER-retained particles during the harvesting process. Although our transient expression data demonstrate both homo- and hetero-oligomerization of MV H and F in the ER, ER export of H in MV-GFP (H)F_Edm-ER-infected cells was not significantly delayed. Thus under these conditions hetero-oligomerization of H_Edm with F_Edm-ER cannot be observed, possibly because of a combination of stronger homotypic than heterotypic interaction in the ER and a lower expression level of F_Edm-ER compared with unmodified F. The recombinant particles in which this ratio is reversed, however, could not be rescued.

In contrast to our findings, ER-retained HN and F proteins of the related SV5 and human parainfluenza virus type 3 (HPIV-3) do not interact with their unmodified homologues when transiently expressed (19). For these paramyxoviruses, it is therefore likely that HN and F interact functionally only at the cell surface, a strategy that may prevent fusion in inappropriate cellular compartments after furin cleavage. That MV H and F proteins do interact in the ER suggests MV must either adopt a different mechanism to prevent premature fusion or must not require such a strategy at all. It is intriguing to speculate that distinct entry mechanisms for different paramyxoviruses may confer virus-specific requirements for fusion regulation. The receptor for both SV5 and HPIV-3, and all paramyxoviridae with neuraminidase activity, is the abundant ganglioside sialic acid (37, 38). In contrast, MV uses a specific protein, either CD46 or signaling lymphocytic activation molecule as a receptor (2-4). These proteins may be less available for early fusion than sialic acid, rendering the problem of inappropriate intracellular less pronounced for MV.

In conclusion, the data obtained in our study show that MV H and F glycoprotein oligomers associate in the ER and that this interaction has relevance for efficient virus replication. Thus, the ER may be considered an assembly site of functional MV glycoprotein complexes. Although we have not observed any escape mutants of the virus expressing F_Edm-ER so far, it is intriguing whether extended passaging results in the appearance of such mutants with altered glycoprotein complex stability. Analysis of these mutations may allow further insight into requirements for the interaction of H and F and thus into the earliest steps of MV particle assembly.

	ACKNOWLEDGEMENTS

We thank L. Hangartner and M. Billeter for providing plasmids encoding full-length MV genomes, B. Fuchs and S. Vongpunsawad for assistance, and A. Fielding for critical reading of the manuscript.

	FOOTNOTES

^* This work was supported by grants from the Siebens and Mayo Foundations (to R. C.), the Fraternity of the Eagles (to R. K. P.), and an Emmy-Noether postdoctoral fellowship from the Deutsche Forschungsgemeinschaft (to R. K. P.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Molecular Medicine Program, Mayo Foundation, 200 1st St. SW, Rochester, MN 55905. Tel.: 507-538-1105; Fax: 507-284-8388; E-mail: plemper.richard@mayo.edu.

Published, JBC Papers in Press, September 4, 2001, DOI 10.1074/jbc.M105967200

	ABBREVIATIONS

The abbreviations used are: MV, measles virus; MV-Edm, MV-Edmonston strain; H, hemagglutinin; F, fusion; HN, hemagglutinin neuraminidase; HPIV, human parainfluenza virus; ER, endoplasmic reticulum; m.o.i., multiplicity of infection; pfu, plaque-forming units; GFP, green fluorescent protein; PBS, phosphate-buffered saline; PMSF, phenylmethylsulfonyl fluoride; Endo H, endoglycosidase H; SV5, simian virus 5.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. D. Muhlebach, V. H. J. Leonard, and R. Cattaneo The Measles Virus Fusion Protein Transmembrane Region Modulates Availability of an Active Glycoprotein Complex and Fusion Efficiency J. Virol., November 15, 2008; 82(22): 11437 - 11445. [Abstract] [Full Text] [PDF]

				J. K. Lee, A. Prussia, T. Paal, L. K. White, J. P. Snyder, and R. K. Plemper Functional Interaction between Paramyxovirus Fusion and Attachment Proteins J. Biol. Chem., June 13, 2008; 283(24): 16561 - 16572. [Abstract] [Full Text] [PDF]

				P. Plattet, P. Cherpillod, D. Wiener, L. Zipperle, M. Vandevelde, R. Wittek, and A. Zurbriggen Signal Peptide and Helical Bundle Domains of Virulent Canine Distemper Virus Fusion Protein Restrict Fusogenicity J. Virol., October 15, 2007; 81(20): 11413 - 11425. [Abstract] [Full Text] [PDF]

				E. A. Corey and R. M. Iorio Mutations in the Stalk of the Measles Virus Hemagglutinin Protein Decrease Fusion but Do Not Interfere with Virus-Specific Interaction with the Homologous Fusion Protein J. Virol., September 15, 2007; 81(18): 9900 - 9910. [Abstract] [Full Text] [PDF]

				J. K. Lee, A. Prussia, J. P. Snyder, and R. K. Plemper Reversible Inhibition of the Fusion Activity of Measles Virus F Protein by an Engineered Intersubunit Disulfide Bridge J. Virol., August 15, 2007; 81(16): 8821 - 8826. [Abstract] [Full Text] [PDF]

				L. W. McGinnes and T. G. Morrison Inhibition of Receptor Binding Stabilizes Newcastle Disease Virus HN and F Protein-Containing Complexes J. Virol., March 15, 2006; 80(6): 2894 - 2903. [Abstract] [Full Text] [PDF]

				J. Doyle, A. Prussia, L. K. White, A. Sun, D. C. Liotta, J. P. Snyder, R. W. Compans, and R. K. Plemper Two Domains That Control Prefusion Stability and Transport Competence of the Measles Virus Fusion Protein J. Virol., February 1, 2006; 80(3): 1524 - 1536. [Abstract] [Full Text] [PDF]

				R. K. Plemper, J. Doyle, A. Sun, A. Prussia, L.-T. Cheng, P. A. Rota, D. C. Liotta, J. P. Snyder, and R. W. Compans Design of a Small-Molecule Entry Inhibitor with Activity against Primary Measles Virus Strains Antimicrob. Agents Chemother., September 1, 2005; 49(9): 3755 - 3761. [Abstract] [Full Text] [PDF]

				G. G. Welstead, E. C. Hsu, C. Iorio, S. Bolotin, and C. D. Richardson Mechanism of CD150 (SLAM) Down Regulation from the Host Cell Surface by Measles Virus Hemagglutinin Protein J. Virol., September 15, 2004; 78(18): 9666 - 9674. [Abstract] [Full Text] [PDF]

				J. A. Hengst and J. S. Bond Transport of Meprin Subunits through the Secretory Pathway: ROLE OF THE TRANSMEMBRANE AND CYTOPLASMIC DOMAINS AND OLIGOMERIZATION J. Biol. Chem., August 13, 2004; 279(33): 34856 - 34864. [Abstract] [Full Text] [PDF]

				R. K. Plemper, K. J. Erlandson, A. S. Lakdawala, A. Sun, A. Prussia, J. Boonsombat, E. Aki-Sener, I. Yalcin, I. Yildiz, O. Temiz-Arpaci, et al. A target site for template-based design of measles virus entry inhibitors PNAS, April 13, 2004; 101(15): 5628 - 5633. [Abstract] [Full Text] [PDF]

				W. Ding, S.-J. Kim, A. M. Nair, B. Michael, K. Boris-Lawrie, A. Tripp, G. Feuer, and M. D. Lairmore Human T-Cell Lymphotropic Virus Type 1 p12I Enhances Interleukin-2 Production during T-Cell Activation J. Virol., October 15, 2003; 77(20): 11027 - 11039. [Abstract] [Full Text] [PDF]

				V. von Messling and R. Cattaneo N-Linked Glycans with Similar Location in the Fusion Protein Head Modulate Paramyxovirus Fusion J. Virol., October 1, 2003; 77(19): 10202 - 10212. [Abstract] [Full Text] [PDF]

				N. Chazal and D. Gerlier Virus Entry, Assembly, Budding, and Membrane Rafts Microbiol. Mol. Biol. Rev., June 1, 2003; 67(2): 226 - 237. [Abstract] [Full Text] [PDF]

				R. K. Plemper and R. W. Compans Mutations in the Putative HR-C Region of the Measles Virus F2 Glycoprotein Modulate Syncytium Formation J. Virol., April 1, 2003; 77(7): 4181 - 4190. [Abstract] [Full Text] [PDF]

				K. N. Bossart, L.-F. Wang, M. N. Flora, K. B. Chua, S. K. Lam, B. T. Eaton, and C. C. Broder Membrane Fusion Tropism and Heterotypic Functional Activities of the Nipah Virus and Hendra Virus Envelope Glycoproteins J. Virol., October 11, 2002; 76(22): 11186 - 11198. [Abstract] [Full Text] [PDF]

				V. von Messling and R. Cattaneo Amino-Terminal Precursor Sequence Modulates Canine Distemper Virus Fusion Protein Function J. Virol., March 27, 2002; 76(9): 4172 - 4180. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/47/44239    most recent M105967200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Plemper, R. K. Articles by Cattaneo, R. Search for Related Content PubMed PubMed Citation Articles by Plemper, R. K. Articles by Cattaneo, R. Social Bookmarking                      What's this?