key: cord-0749954-w8aew9vs
authors: BOS, EVELYNE C.W.; LUYTJES, WILLEM; MEULEN, HANS VAN DER; KOERTEN, HENK K.; SPAAN, WILLY J.M.
title: The Production of Recombinant Infectious DI-Particles of a Murine Coronavirus in the Absence of Helper Virus
date: 1996-04-01
journal: Virology
DOI: 10.1006/viro.1996.0165
sha: 5b44af2cdde5e0824570443c614a8d52ebeed552
doc_id: 749954
cord_uid: w8aew9vs

Abstract We have studied the production and release of infectious DI-particles in vaccinia-T7-polymerase recombinant virus-infected L cells that were transfected with five different plasmids expressing the synthetic DI RNA MIDI-HD and the four structural proteins (M, N, S, and E) of the murine coronavirus MHV-A59. The DI cDNA contains the hepatitis delta ribozyme sequences to generate in the transfected cells a defined 3′ end. In EM studies of transfected cells virus-like particles (VLP) were observed in vesicles. Release of the particles into the medium was studied by immunoprecipitations of proteins released into the culture supernatant. Particle release was independent of S or N, but required M and E. Coexpression of E and M was sufficient for particle release. Coexpression of the structural proteins and the MIDI-HD RNA resulted in the production and release of infectious DI-particles. Infectivity of the DI-particles was determined by adding helper virus MHV-A59 to the medium containing the VLPs and using this mixture to infect new L cells. Intracellular RNA of several subsequent undiluted passages was isolated to detect the MIDI-HD RNA. Passage of the MIDI-HD RNA was dependent on the expression of the structural proteins of MHV-A59 in the transfected cells. In the absence of either E or M, MIDI-HD RNA could not be passaged to fresh L cells. We have thus developed a system in which we can produce coronavirus-like particles and an assay to test their infectivity.

5 leader sequence (Baric et al., 1988) . Furthermore, a domain located on the genome, at the 3 end of the Coronaviruses are enveloped viruses that have a posipolymerase 1B open reading frame most likely interacts tive-stranded RNA genome of 27-32 kb in a helical nuwith the N protein as it has been demonstrated that this cleocapsid form. During replication, a 3-coterminal domain is involved in encapsidation of the genome of nested set of mRNAs is produced, from which the differdefective interfering particles (Van der Most et al., 1991 ; ent proteins are translated (reviewed by Spaan et al., Fosmire et al., 1992) . N protein and the NC interact with 1988; Luytjes, 1995) . Coronaviruses bud in the intermedimembranes (Anderson and Wong, 1993) and with M ate compartment of the host cell (Krijnse Locker et al., (Sturman et al., 1980 (Sturman et al., ). 1994 Tooze et al., 1987) , inserting either two or three viral

The MHV S protein is cotranslationally glycosylated protein species into the membrane: the spike protein (S) , resulting in a S precursor protein of 150 kDa that forms the membrane protein (M), and in some coronaviruses homo-oligomers in the ER (Vennema et al., 1990) . The the hemagglutinin protein (HE) (reviewed by Spaan et al., homo-oligomers are either inserted into the virions in the 1988). Recently, an additional small membrane protein intermediate compartment or are transported to the cell (sM or E) has been identified in the virions of the pig surface through the constitutive pathway. In the Golgi (TGEV), avian (IBV), and murine coronavirus (MHV) (Tung stacks the high mannose sugar side-chains are trimmed et Liu and Inglis, 1991; Yu et al., 1994) . and modified, giving rise to an almost endo-H-resistant During virus assembly the helical nucleocapsid (NC) 180-kDa protein (Niemann and Klenk, 1981) . A portion of consisting of the genomic RNA and many N molecules the MHV-A59 S molecules is cleaved in the post-Golgi is enveloped, thereby forming an infectious coronavirus into two 90-kDa subunits (Sturman et al., 1985) . At the particle. We are particularly interested in understanding cell surface the spike protein can bind to the receptor murine coronavirus nucleocapsid formation and the interon neighboring host cells (Dveksler et al., 1991) and inaction between the viral membrane proteins and the NC. duce cell to cell fusion (Vennema et al., 1990) , but recep-The protein component of the NC, the N protein, is the tor-independent fusion has also been described (Galonly viral structural protein that is not synthesized on lagher et al., 1992) . Cleavage of MHV S is not absolutely membrane bound ribosomes. It binds specifically to the required for the induction of cell to cell fusion. However, expression of uncleaved S on the cell surface resulted MHV-A59 INFECTIOUS VIRUS LIKE PARTICLES The M protein of MHV-A59 (22-26.5 kDa) is an O-rum. MHV-A59 stocks were grown as described (Spaan et al., 1981) . Vaccinia virus vTF7.3 stocks (kindly provided glycosylated (Holmes et al., 1981) triple-spanning membrane protein (Armstrong et al., 1984) that forms large by Dr. B. Moss) were grown on RK13 cells. aggregates in the Golgi (Krijnse Locker et al., 1995) . Data reported by several groups indicate an important role for Construction of plasmids M in virus assembly (Holmes et al., 1981; Rottier et al., Standard DNA recombination procedures were used 1981; Holmes et al., 1987) . When expressed indepen- (Sambrook et al., 1989) . pMIDI-HD: The hepatitis delta dently, the protein accumulates beyond the budding comribozyme and the T7 terminator were introduced at the partment in the trans-Golgi network and is not trans-3-end of pMIDI (Van der Most et al., 1991) just downported to the plasma membrane (Rottier and Rose, 1987;  stream of the poly(A)-tail. The unique NheI site of pMIDI Krijnse-Locker et al., 1992) . When S and M are expressed was filled in with the Klenow fragment of DNA polymertogether, both proteins are retained in the trans-Golgi ase I. Vector (2.0) (Pattnaik et al., 1992; kindly provided (Opstelten et al., 1995) . Interactions, presumably lateral, by Dr. L. A. Ball) was digested with SmaI and XbaI to between S and M in the ER have been established both obtain the 250-bp fragment that contains the hepatitis in infected cells and in cells coexpressing both proteins delta ribozyme and the T7 terminator sequence. The fragand it has been suggested that this interaction plays an ment was cloned into the Klenow-treated NheI site of important role in the inclusion of the S oligomers into pMIDI. pTUM-M: The construction of pTUM-M was debudding virions (Opstelten et al., 1994 (Opstelten et al., , 1995 . scribed by Opstelten et al. (1993) . pTUM-N: was de-Not much is known yet about the function of the E scribed by Vennema et al. (1991) . pTUM-S: A BamHI protein (9.6 kDa). The E protein of TGEV is expressed at MHV-S containing fragment was cloned into the BamHI the cell surface (Tung et al., 1992) , the IBV E protein is site of pTUG3 (Vennema et al., 1991) . pIRES-E: The separt of the viral envelope (Liu et al., 1991) , and the E quence encoding the E gene (155 nt) was amplified by protein of MHV-A59 is acylated and was detected in viri-PCR from a cDNA clone of MHV-A59 mRNA5, pRG68 ons albeit in very low amounts (Yu et al., 1994) . (Bredenbeek, 1990 ) using oligo's c093 (containing an Although several interactions between the structural NcoI site at the AUG codon of E: 5CATGCCATGGCCTTTproteins of MHV have been investigated, it is not known AATTTATTCCTTAC3) and c094 (containing the stopcowhich of these are required for assembly and budding of don and an XbaI site downstream of it: 5 CTAGTCTAGinfectious virions. There is no reverse genetics approach ATTAGATATCATCCAC 3). The amplified fragment was available to study virus assembly. An infectious cDNA isolated from gel, digested with NcoI and XbaI, and inclone of MHV-A59 has yet to be constructed and targeted serted into the NcoI-XbaI-digested vector pIRES (Den recombination has so far only been successful at the 5 Boon et al., 1995) , containing the encephalomyocarditis and 3 ends of the genomic RNA (Van der Most et al.,

virus internal ribosomal entry site. 1992; Masters et al., 1994; Makino and Lai, 1989; Chang et al., 1994; Peng et al., 1995) . A full-length cDNA clone MHV-A59 infection of a naturally occurring defective interfering (DI) RNA of MHV-A59 has been extensively characterized (Van der Confluent monolayers of L cells were infected with Most et al., 1991) . MIDI contains the signals for replica-MHV-A59 in PBS-DEAE, supplemented with 3% FCS at a tion and packaging, but is dependent on helper virus multiplicity of infection (m.o.i.) of 10. After absorption for MHV-A59 for its propagation.

1 hr at 37Њ, virus was removed and cells were cultured Assembly and budding requirements for other envein DMEM supplemented with 3% FCS. Undiluted passage loped viruses have been studied with the use of viruswas performed as described before (Van der Most et al., like particles (VLP; Hobman et al., 1994; Qiu et al., 1994; . Mebatsion et al., 1995; Suomalainen et al., 1992) . In these systems domains in the structural proteins that are im-DNA transfection in the vaccinia T7 system portant for assembly can be located by insertion of mutated proteins into VLPs.

L cells (1 1 10 6 ) were seeded in 35-mm dishes. Sixteen We describe in this paper the assembly of virus-like hours later the cells were infected with the T7 RNA polyparticles of MHV-A59 by coexpressing the structural promerase expressing vaccinia virus recombinant (vTF7.3) teins using the vaccinia virus T7 system. Further, we at a m.o.i. of 5. At 1 hr postinfection the cells were show that a DI-genome can be packaged into these partitransfected with lipofectin containing the appropriate cles. Finally, we show that the DI particles are infectious. plasmids as recommended by GibcoBRL.

Mouse L cells were grown in Dulbecco's modified Ea-Three hours after DNA transfection, the cells were infected with MHV-A59 at an m.o.i. of 10. Actinomycine D gle medium (DMEM; Gibco) containing 10% fetal calf se-(20 mg/ml) was added to the medium at 4 hr post MHV RESULTS infection.

Coronavirus-like particles are detected in L-cells Isolation and analysis of viral RNA First, we studied whether virus-like particles were produced in the vTF7.3-infected L-cells that were cotrans-Intracellular RNA was isolated from infected and fected with four DNA constructs encoding the known transfected L cells 8 hr postinfection or transfection as structural proteins S, M, N, and E of MHV-A59. vTF7.3described previously (Spaan et al., 1981) . RNAs were infected cells that were mock transfected or transfected separated on 1% agarose/2.2 M formaldehyde gels with all structural proteins were fixed at 10 hr after trans- (Meinkoth and Wahl, 1984) , and hybridization was done fection. MHV-infected L-cells were fixed at 6 hr postinfecin dried gels using 5 end-labeled probes (Meinkoth and tion and prepared for electron microscopy analysis. Wahl, 1984) . Oligo 48 (5GTGATTCTTCCAATTGGCCATG In L-cells endogenous retrovirus type A particles con-3), which binds to the 3 end of the genome, and oligo taining a clear double membrane (reviewed by Kuff and c122 (5ATGCCATGCCGACCCCT 3), which binds to the Lueders, 1988) were observed in vesicles (Fig. 1) . MHV region between the hepatitis delta ribozyme and the T7 virions were detected in collecting or budding vesicles terminator, were used for hybridization. Oligonucleotides in the MHV-A59-infected cells. Virions were heterogewere labeled using [g-32 P]ATP (NEN-Dupont) and T4 poneous in size, but could easily be distinguished from the lynucleotide kinase.

retroviruses as the latter have a distinct morphology.

In vaccinia virus-infected cells that were not transfected, the retroviruses were also detected, together with several Cells were fixed in 1.5% glutaraldehyde in 0. were less electron dense than the retro-and coronavirus and subsequently flat embedded in epoxy resin LX-112 particles and did not have the typical retrovirus type A and polymerized at 60Њ. Ultrathin sections (60 nm) were particle double membrane. The coronavirus VLPs were stained with uracyl acetate followed by lead hydroxide absent in cells that did not express the recombinant and examined with a Philips EM-410LS electron microstructural proteins of MHV (Fig. 1 ). scope at 80 kV.

Virus-like particles are released from cells expressing Metabolic labeling of proteins and lysis of cells the structural proteins of MHV-A59 Cells were metabolically labeled with 100 mCi 35 S-labeled methionine and cysteine (Tran 35 S label, ICN Bio-Next, we determined whether virus-like particles were released by studying which proteins and which forms medicals) in medium lacking methionine from 4 to 8 hr posttransfection. The labeling medium was subsequently of the proteins were detected into the medium of transfected cells. MHV-A59 virions contain the M, N, E replaced with chase medium, containing four times the normal concentration of methionine and cysteine. At 12 proteins and the 180-and 90-kDa cleaved forms (S1 and S2) of the spike protein (Spaan et al., 1988 ; Yu et al., hr posttransfection the medium was collected. Cells were lysed in RIPA buffer (150 mM NaCl, 1.0% NP-40, 0.5% 1994). However, not all of these proteins can be used as markers for virion release into the medium of infected DOC, 0.1% SDS, 50 mM Tris, pH 8.0) and 2 mM PMSF. The lysate was centrifuged at 4Њ for 10 min at 13,000 cells. The N protein is detected in the medium independent of virion formation. The S1 subunit of the spike rpm to remove nuclei and cell debris. The medium was cleared by a 4-min centrifugation (4,000 rpm) and one-protein is found in the medium even when S alone is expressed in the cells, since it is not stably associated fifth volume of a 51 concentrated RIPA buffer containing 10 mM PMSF was added. Immunoprecipitations were to the membrane bound S2 subunit (Sturman et al., 1990) . Other viral membrane proteins like the 150-kDa S precur-performed on the supernatant using rabbit polyclonal MHV-A59 antiserum k134. After an overnight incubation sor protein can reach the medium only on membrane fragments when cells start to lyse. These fragments, at 4Њ, 50 ml Pansorbin cells (Calbiochem, La Jolla) and KCl to a final concentration of 0.5 M were added, followed however, are cleared by centrifugation. M is the only membrane protein that is known to be retained in the by an incubation for 1 hr at 4Њ. After washing the samples three times in RIPA, they were boiled in Laemmli sample trans-Golgi network when expressed alone ( Krijnse  Locker et al., 1992) , release of M into the medium is buffer for 2 min (Laemmli, 1970) . Samples were analyzed by SDS-PAGE on 12.5% gels.

taken as proof for release of membraneous particles. Likewise, the 180-kDa forms of the spike protein depends detected in the medium, clearance of the medium had been successful. on membraneous particles for release out of the cell.

To determine which proteins are required for particle Thus, only detection of M or the uncleaved mature spike release, we transfected L-cells with different combinaprotein (180 kDa) in the medium can be used as marker tions of three plasmids, each encoding a structural profor release of particles. tein as indicated above the lanes in Fig. 2 . When S or N MHV-A59 structural proteins were expressed in vTF7were not expressed in the cells, the M protein, which is infected cells and labeled with [ 35 S]methionine from 4 to a marker for particle formation, could still be detected in 8 hr posttransfection and subsequently chased for 4 hr. the medium. However, when E was omitted, neither S180 The medium was cleared by centrifugation to remove nor M were released into the supernatant. Omission of cells and cellular debris containing viral membrane pro-M also resulted in the absence of S180 in the medium. teins. Cell lysates and the medium from the same cells These data indicated that particle release was depenwere subjected to immunoprecipitation with the polydent on the expression of E and M. clonal rabbit antiserum k134. With this antibody the im-We next tested whether coexpression of E and M alone mature 150-kDa, the mature 180-kDa, and very little of was sufficient for particle release (Fig. 2C) . As indicated the 90-kDa spike proteins could be detected in the cell by the detection of M in the medium, particles were inlysates, in addition to N and the five forms of M ( Fig. 2A, deed formed and released, although less efficiently. first lane). M and 180-kDa S could also be detected in the medium of the cells, indicating that membraneous particles had been released from the transfected cells.

Assay to determine infectivity of the virus like The E protein could not be detected with this antibody.

particles We have not yet succeeded in producing an E-specific antibody. Since the immature ER-restricted 150-kDa form After having established that virus-like particles were of the spike protein and the M0 and M1 forms of M produced in transfected cells and subsequently released into the medium, we next analyzed whether the VLPs (Fig. 2B, first lane; Krijnse Locker et al., 1992) were not

A were subjected to immunoprecipitation with the polyclonal antibody k134. Which proteins were expressed in the vTF7.3-infected cells is indicated above the lanes. Release of M and 180-kDa S into the medium were used as markers for the presence of VLPs. C was exposed twice as long as B.

were able to package an MHV DI genome (Van der Most the sequence between the hepatitis delta ribozyme and the T7 terminator (data not shown). This indicated that et al., 1991) .

Following transfection of vTF7.3-infected cells with both the ribozyme and the termination signal were active in vivo. pMIDI, encoding MIDI RNA under the control of the T7 promoter (Van der Most et al., 1991) , a distinct RNA band

The pMIDI-HD construct was used in a simple but very sensitive protocol to study the production of infectious of 5.4 kb was detected only in cells that had been superinfected with MHV-A59 (Fig. 3A) . However, in the ab-VLPs. vTF7-infected L cells were transfected with pMIDI-HD and four different plasmids encoding the structural sence of MHV the DI RNA could not be detected. This is most likely due to the lack of a T7 terminator sequence proteins of MHV (M, N, S, and E), all under the control of the T7 promoter. The production of RNA-containing on pMIDI: the T7 transcripts that are produced are heterogeneous in length and cannot be detected by hybrid-particles was tested by mixing helper virus MHV-A59 with the medium of the transfected cells and adding this ization. Since nothing is known about possible 3 end constraints for RNA packaging we have introduced the mixture to a new monolayer of L cells. When the VLPs are infectious, they will be able to deliver the packaged cis-acting hepatitis delta ribozyme followed by the T7 terminator sequence into the cDNA clone behind the MIDI-HD RNA to the cytoplasm of the cells and the DI RNA will subsequently be replicated and packaged by poly(A)-tail in order to generate an RNA that has a 3 end that resembles the 3 end of MIDI RNA as much as the coinfecting helper virus MHV-A59. The presence of MIDI-HD in the intracellular RNA is thus used as a marker possible. Only four nonviral nucleotides are present downstream of the poly(A)-tail. The structure of the re-for the infectivity of the particles. The experimental setup is schematically presented in Fig. 4 . sulting construct, named pMIDI-HD is shown in Fig. 3B .

When cellular RNA of vTF7-infected, pMIDI-HD DNA-Two plates of L-cells were infected with vTF7.3. In one plate, pMIDI-HD was cotransfected with the plasmids transfected L-cells was isolated 8 hr after transfection, two equally abundant RNA species hybridizing to the encoding the structural proteins M, S, E, and N. In the other, pMIDI-HD was cotransfected with pUC20 DNA. MHV-specific 3 end probe were detected; RNA A and RNA B (Fig. 3A) . RNA A comigrates with MIDI RNA and

The amount of DNA for both plates was similar. The medium of the transfected cells was harvested 12 hr RNA B hybridizes to a probe that is complementary to both E and M were required for VLP release (Fig. 2) , we reasoned that the DI RNA would not be transferred from plasmids encoding either E or M were not included, MIDI-HD RNA could not be detected in the intracellular RNA after passaging with helper virus (Fig. 6 ). An endogposttransfection and the latter was mixed with helper virus MHV-A59 (m.o.i. of 10). A fresh monolayer of Lcells was infected with this mixture and serial undiluted passages were performed. Intracellular RNA of P0 cells and of the passages was analyzed in a hybridization assay. Figure 5 shows that MIDI-HD RNA was produced abundantly in transfected cells. After passaging of the material derived from the cells that were transfected with pMIDI-HD and the structural genes, MIDI-HD RNA could be detected in the intracellular RNA of P2, P3, and P4 cells (Fig. 5A) . However, no MIDI-HD RNA was observed when pMIDI-HD was cotransfected with pUC20 and subsequently passaged in the presence of helper virus (Fig.  5B) . This experiment showed that the virus-like particles can package the MIDI-HD RNA and that they are infec- Although we have not shown the expression of E by immunoprecipitation using an E-specific antibody, our data clearly suggest that not only M but also the expression of E is required for the production of VLPs. More importantly, coexpression of E and M appeared to be sufficient for the release of virus-like particles. Therefore, E and M must be important factors in virus budding.

Not much is known yet about the small membrane protein E, except that the acylated protein is found in virions in very low amounts (Yu et al., 1994) and that it is expressed at the cell surface (Tung et al., 1992) . Other enveloped viruses, like Influenzavirus, alphaviruses, and pestiviruses, also have small membrane proteins that play an important role in the biogenesis of infectious progeny (Pinto et al., 1992; Allison et al., 1995; Loewy et al., 1995) . The E protein might have similar functions. Possibly, an interaction between E and M induces the budding process. Both proteins are modified in the intermediate compartment; the M protein acquires GalNac (Tooze et al., 1988; Krijnse Locker et al., 1992) absence of M can induce virus budding (Fig. 2B) . The Above the lanes is indicated whether E or M expressing plasmids were present in P0 cells together with N, S, and pMIDI. The arrow points to only markers for VLP release are S180 and M and if the MIDI-HD RNA.

insertion of S into particles is dependent on an interaction with M (Opstelten et al., 1995) , the absence of the spike protein in the medium would not exclude the formaenous DI comigrating with RNA3 that is never observed tion and release of particles from cells expressing solely when MIDI is present in the cells was observed in these the E protein. RNA samples (Van der Most et al., 1992) . Expression of A function of M in coronavirus budding has been proendogenous DI is variable, but it is never seen before posed before. When hybridomas producing monoclonal passage 4. Presumably, expression of MIDI interferes antibodies to the M protein were infected with MHV-A59, with appearance of the endogenous DI, as seen in the no virions were produced (Holmes et al., 1987) . In MHV-P4 lane of the E//M/ experiment in Fig. 6 . It is therefore A59-infected cells treated with tunicamycin and in hybridunlikely that the endogenous DI would prevent expresomas expressing anti-S antibodies, S-deficient virions sion of MIDI in the E0/M/ and E//M0 experiments.

are produced (Holmes et al., 1981; Rottier et al., 1981 ; Instead, the presence of the endogenous DI would be Holmes et al., 1987) , indicating that S is not required for an extra indication of the absence of MIDI in the controls.

virion release from infected cells, but M is. These findings are consistent with the data presented in this paper, in DISCUSSION which omission of S does not prevent release of particles into the medium, whereas M is absolutely required. The In this paper we describe the production and infectivity of coronavirus-like particles, by coexpressing the struc-interaction between M and N (Sturman et al., 1980, Anderson and Wong, 1993) was thought to be important tural proteins of MHV-A59 and a DI genome. Infectivity of the VLPs was demonstrated by transfer of a DI genome during the budding process in the intermediate compartment. We show here that the nucleocapsid-M interac-to fresh cells. We present an assay that might be a powerful tool to study packaging, assembly, and budding by tion is not a prerequisite for budding, since even in the absence of RNA and N, virus-like particles are released expressing mutated structural proteins and RNAs. Questions concerning the protein-protein and RNA-protein into the medium (Fig. 2B) . In this aspect, the budding mechanism of MHV-A59 is distinct from that of the alpha-interactions that are required for these processes can now be addressed.

viruses where nucleocapsid-envelope protein interactions are the driving force for budding (Suomalainen et When we expressed the structural proteins of MHV-A59 in L cells with the use of the vaccinia vTF7.3 expres-al., 1992; Lopez et al., 1994; . The viral RNA itself, or its replication, is not directly sion system, we obtained morphological proof for the production of VLPs by electron microscopy (Fig. 1) .

involved in the budding process, since particles were

ization of sialyltransferases in N-and O-linked glycosylation

Oligomeric rearrangement of tick-borne encephalitis Characterization of the budding compartment of mouse hepatitis virus envelope proteins induced by an acidic pH

Membrane and phospholipid binding Krijnse Locker

part of its retention

The intracisternal A-particle gene protein, E1 glycoprotein, from a coronavirus. Nature (London) 308, family: Structure and functional aspects

Interactions between coroof the head of bacteriophage T4

Association of the infectious bronchitis transcription

virus 3c protein with the virion envelope

Isolation and identification of virus-specific mRNAs in cells infected with mouse hepatitis virus (MHV-A59)

Nucleocapsid-glycoprotein interactions required for assembly of Structure and genome expression

Proteolytic cleavage alphaviruses

The Coronaviridae'' of the murine coronavirus surface glycoprotein is not required for fusion

491-562. switching during coronavirus defective interfering RNA replication

Isolation of coronavirus envelope proteins and interaction with the viral nucleocapsid

Optimization of targeted RNA recombination and mapping of a novel nucleo

age of the E2 glycoprotein of murine coronavirus: activation of cellfusing activity of virions by trypsin and separation of two different Mebatsion

change of the coronavirus peplomer glycoprotein at pH8.0 and 37C correlates with virus aggregation and virus-induced cell fusion

Hybridization of nucleic acids immobilized on solid supports

993-1010. nucleocapsid interactions drive the budding of alphaviruses

Fusion formation by the uncleaved spike protein of murine coronavirus JHM variant cl-2

Sorting of progeny coronavirus from condensed secretory proteins at the exit from the formation between the spike protein and the membrane protein during mouse hepatitis virus assembly

Site of addition of N-195

and acetyl-galactosamine to the E1 glycoprotein of mouse hepatitis virus-A59

Coexpression and association of the spike and the membrane protein of mouse hepatitis virus

at the 3 end of the porcine transmissible gastroenteritis coronavirus genome is membrane-associated

Infectious defective interfering particles of VSV from transcripts of a encapsidation of coronavirus Defective Interfering RNAs

Homologous RNA recombination allows efficient introduction protein has ion channel activity

Expression of site-specific mutations into the genome of coronavirus MHV-A59 via synthetic co-replicating RNAs

Intracellular transport of recombinant coronavirus spike proteins; implications for virus assembly

Enhancement of the vaccinia virus/phage T7 RNA expressed from cloned cDNA localizes to the Golgi region

Mouse hepatitis virus gene 5b protein is a new virions envelope protein

 (1992). O-glycosilation of the coronavirus M protein; differential local-

Mutational analysis of the murine coronavirus spike protein: Effect released in the absence of RNA. However, in our system, [453] [454] [455] [456] [457] [458] [459] [460] [461] [462] [463] packaging of the DI RNA is not very efficient, since it can Bredenbeek, P. (1990) . Nucleic acid domains and proteins involved in only be detected after two passages (Fig. 5) . One obvious the replication of coronaviruses. Thesis, University of Utrecht. reason for the inefficiency might be that very few cells Chang, R. Y., Hofmann, M. A., Sethna, P. B., and Brian, D. A. (1994) . A cis-acting function for the coronavirus leader in defective interfering were transfected with all five plasmids and which is a RNA replication. J. Virol. 68, [8223] [8224] [8225] [8226] [8227] [8228] [8229] [8230] [8231] prerequisite for the production of infectious VLPs. Since Collins, A. R., Knobler, R. L., Powell, H., and Buchmeier, M. J. (1982) .

Monoclonal antibodies to murine hepatitis virus-4 (strain JHM) define Most et al., 1991) , inefficient packaging into the VLPs is the viral glycoprotein responsible for attachment and cell-cell fusion.unlikely to be due to defective signals on the DI genome.Virology 119, 358-371. Den Boon, J., Faaberg, K. S., Meulenberg, J. J. M., Wassenaar, A. L. M., A more likely alternative explanation for the inefficient Plageman, P. G. W., Gorbalenya, A. E., and Snijder, E. J. (1995). packaging is that replication, which does not occur in