key: cord-0762134-h02d5quu authors: Ricard, Cynthia S.; Koetzner, Cheri A.; Sturman, Lawrence S.; Masters, Paul S. title: A conditional-lethal murine coronavirus mutant that fails to incorporate the spike glycoprotein into assembled virions date: 1995-12-31 journal: Virus Research DOI: 10.1016/0168-1702(95)00100-x sha: e36bf83d8e36778452ded5ce3b89532d512696c9 doc_id: 762134 cord_uid: h02d5quu Abstract The coronavirus spike glycoprotein (S) mediates both the attachment of virus to the host cell receptor and membrane fusion. We describe here the characterization of a temperature-sensitive mutant of the coronavirus mouse hepatitis virus A59 (MHV-A59) having multiple S protein-related defects. The most remarkable of these was that the mutant, designated Albany 18 (Alb 18), assembled virions devoid of the S glycoprotein at the nonpermissive temperature. Alb18 also failed to bring about syncytia formation in cells infected at the nonpermissive temperature. Virions of the mutant assembled at the permissive temperature were much more thermolabile than wild type. Moreover, mutant S protein that was incorporated into virions at the permissive temperature showed enhanced pH-dependent thermolability in its ability to bind to the MHV receptor. Alb18 was found to have a single point mutation in S resulting in a change of serine 287 to isoleucine, and it was shown by revertant analysis that this was the lesion responsible for the phenotype of the mutant. particular member of the coronavirus family, the S protein can exhibit large intrastrain variation. Such differences in primary sequence, which occur mainly in the S1 portion of the molecule (Kusters et al., 1989; Parker et al., 1989; Gallagher et al., 1990; La Monica et al., 1991) are important for evasion of immune surveillance and are thought to play the major role in differences in tissue tropism and pathogenicity. An understanding of the S protein must balance this sequence diversity against the necessary conservation of essential structure that allows the functions of receptor binding and membrane fusion. As part of a genetic approach to the study of coronavirus structural proteins we have isolated a temperature-sensitive mutant of MHV-A59 that assembles virions devoid of spikes at the nonpermissive temperature. In this report we characterize the phenotype of this mutant and examine its molecular basis. Murine fibroblast cell lines 17 clone 1 (17C11), L2, and Sac-have been described previously (Frana et al., 1985) . Stocks of wild-type MHV-A59, the mutant Albl8 and its revertants, and vesicular stomatitis virus (VSV) (New Jersey serotype) were grown in 17C11 cells; infectious titer was measured by plaque assay in L2 ceils at both 33 ° and 39°C. Due to the high selective advantage of revertants of Albl8, only low passage stocks of Albl8 with efficiencies of plaquing (39°C/33°C) of less than 10 -4 were used in experiments. Eight independent spontaneous revertants (designated Albl8Rev) were isolated from Albl8. Albl8Revl arose in a stock that had been grown for 54 h at 39°C and was plaque purified four times. Albl8Rev2-Albl8Rev8 were each isolated as a plaque formed at 39°C from a virus stock begun from an individual plaque of Albl8 obtained at 33°C. For labeling of virion proteins, confluent monolayers of 17Cll or Sac-cells were infected at a multiplicity of 5 PFU/cell. Following absorption for 1 h at 33°C, ceils were re-fed with MEM containing 1/10 of the normal level of cystine (2.4 mg/l), 10% fetal bovine serum, and 2.0 /~Ci/ml [35S]cysteine (Amersham). Cultures were incubated at 33°C or 39°C until 95% of the wild type-infected cells at 39°C had formed syncytia (24 h for 17CI1 cells or 16 h for Sac-cells). Protein sample analysis was performed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (Laemmli, 1970) . Tube gels were fractionated, solubilized and quantitated by liquid scintillation counting as described previously (Sturman, 1977) . For Western blotting, proteins separated on slab gels were electroblotted to a PVDF membrane (Millipore) using a tank type transblotter (BioRad). Nonspecific binding to membranes was blocked by incubation in 0.1 M Tris-HCl (pH 7.5) containing 3% nonfat milk; primary antiserum was used at a dilution of 1 : 250 in 0.1 M Tris-HC1 (pH 7.5). Goat anti-S polyclonal antibody AO4 (Boyle et al., 1987) was generously provided by Dr. Kathryn Holmes (Uniformed Services University of the Health Sciences). Staining of bound antibody was performed with a biotin avidin horseradish peroxidase kit (Vector Labs). The binding of virion S protein to the MHV receptor was assayed essentially as described previously (Boyle et al., 1987) . Intestinal brush border membranes from BALB/c mice (kindly provided by Dr. Mark Frana and Dr. Kathryn Holmes, Uniformed Services University of the Health Sciences) were separated by SDS-PAGE, transblotted to PVDF, and blocked as for Western blots. Prior to the binding assay, purified virus samples were heat-inactivated at the indicated pH for 24 h at 40°C (Koetzner et al., 1992) ; control samples were held on ice at pH 6.5. Strips cut from the blot were incubated with virus samples for 1 h at room temperature and were then stained with goat anti-S protein antiserum AO4 and developed as for Western blots. Albl8 genomic RNA was extracted from purified virions as described previously (Koetzner et al., 1992) . Libraries of cDNA clones of the M and S genes of Albl8 were generated by a modification of the procedure of Gubler and Hoffman (Gubler and Hoffman, 1983) using primers complementary to nt 580-597 of the N gene (for M first strand cDNA) or to nt 1625-1641 of the S gene and to nt 5-24 beyond the S gene stop codon (for S first strand cDNA). An additional set of cDNA clones corresponding to nt 1136-2770 of the S gene was obtained by reverse transcription followed by PCR using a primer pair bordering this region; PCR was performed as described previously (Koetzner et al., 1992) . DNA sequencing was carried out by the dideoxy chain termination method (Sanger et al., 1977) using modified T7 DNA polymerase (Sequenase, U.S. Biochemical). A set of 31 oligonucleotide primers spanning the S gene and adjacent regions and a set of 7 oligonucleotide primers spanning the M gene and adjacent regions were used. In addition to sequencing of cDNA clones, direct sequencing of reverse transcription-PCR products was performed to verify portions of the Albl8 S and M genes and to sequence the relevant portions of the S genes of Albl8Revl-Albl8Rev8. The mutant Albl8 was identified among a collection of mutants obtained by nitrous acid mutagenesis of wild-type (heat-resistant) MHV-A59 . In this search mutants were selected for alterations in or lack of cytopathic effect on 17C11 cells at the nonpermissive temperature (39°C) as compared with that occurring at the permissive temperature (33°C). At temperatures of 33°C through 36°C, it was found that Albl8 infection resulted in polykaryon formation by 24 h post-infection. At 370C or above, however, no polykaryons were observed. This suggested that there was a sharp transition temperature above which Albl8 became unable to replicate. Roughly half of the mutants obtained in the search produced small plaques at the nonpermissive temperature; we have previously characterized two members of this set as N gene mutants (Koetzner et al., 1992; Masters et al., 1994) . The remaining half of the collection comprised tight temperature-sensitive mutants, having efficiencies of plaquing (39°C/33°C) on the order of 10 -4. For these, in all cases examined, plaques formed at the nonpermissive temperature were found to be revertants. Albl8 fell in this latter set of mutants but was unique in being RNA ÷, i.e., able to synthesize viral RNA at the nonpermissive temperature . Moreover, preliminary results indicated that Albl8 synthesized viral structural proteins at the nonpermissive temperature. These considerations led us to examine the possibility that Albl8-infected cells produced non-infectious virions at the nonpermissive temperature. Albl8-infected 17Cll ceils incubated at either the permissive or nonpermissive temperature were metabolically labeled with [35S]cysteine, and virus released into the culture supernatant was purified by sucrose gradient centrifugation (Sturman et al., 1980) . Material obtained from the 39°C culture behaved identically to material from the 33°C culture (and to wild-type controls) throughout this standard virus purification procedure. Remarkably, when analyzed by SDS-PAGE it was found that Albl8 virions formed at the nonpermissive temperature were devoid of both the 180 kDa and the 90 kDa forms of the S protein (Fig. 1) . These virions contained the other two viral structural proteins, N and M, in the same relative amounts and with the same electrophoretic mobilities as those of virions obtained at the permissive temperature. In Fig. 1 , since equal amounts of radioactivity of each sample were analyzed, the 39°C sample contained proportionately more of the N and M proteins. A similar experiment was performed using Sac-cells to test whether this result was host cell line-specific and to simplify the analysis of S, since all S protein in virions released from Sac-cells is cleaved to the 90 kDa form (Frana et al., 1985) . In addition, to address the possibility that loosely associated S protein on Albl8 virions produced at 39°C might have been stripped from virus particles during the multiple steps of the purification procedure, culture medium was instead collected and pelleted directly for analysis by both Western blotting and autoradiography. As shown in Fig. 2A , Western blotting detected equivalent amounts of 90 kDa S protein in virus released from Albl8-infected Sac-cells at 33°C and from wild type-infected cells at either 33°C or 39°C. In contrast, virus released from Albl8infected cells at 39°C contained no 90 kDa S protein, nor was there S-specific immunoreactive material of any other molecular weight. Examination of the same blot by autoradiography confirmed this result (Fig. 2B ). Because the samples in (Sturman et al., 1980) . Equal amounts of radioactivity for each sample were analyzed by electrophoresis on cylindrical 10% polyacrylamide gels, followed by fractionation and quantitation by liquid scintillation counting. this experiment had not been purified, there were a number of additional nonviral proteins present in the pelleted material, but no 90 kDa S protein could be detected in the Albl8 at 39°C sample, nor were any of the additional bands unique to this sample. Analysis of supernatant fractions by immunoprecipitation failed to detect the presence of soluble S protein (data not shown). These results confirmed the conclusion that Albl8 failed to incorporate S protein into assembled virions at the nonpermissive temperature. The phenotype of Albl8 strongly suggested that the lesion in this mutant resided in either the S protein or the M protein. In ceils infected with MHV the M protein localizes in the endoplasmic reticulum (ER) and Golgi region (Tooze et al., 1984; Tooze and Tooze, 1985) , and M is believed to be responsible for determining the assembly and budding of progeny virions . To examine the possibility that Albl8 contained a defect in the M protein leading to an altered ability to associate with S protein, the M gene of Albl8 was cloned and sequenced. No nucleotide difference was found between the M gene of Albl8 and that of wild-type virus in either the coding region or in the upstream intergenic region. This finding precluded a causative role for M protein in the phenotype of Albl8. The entire S gene of Alb18 as well as 150 nt upstream was next analyzed by sequencing of overlapping eDNA clones prepared from genomic RNA. In the coding region of the Albl8 S gene, a single nucleotide difference was found in comparison to the S gene of wild-type virus: nt 860 was changed from G to T. In the encoded amino acid sequence, this resulted in a change of residue 287 from serine to isoleucine (Fig. 3) . Two additional nucleotide changes were found in the Albl8 S coding region in comparison to the previously published sequence for MHV-A59 (Luytjes et al., 1987 ; GenEMBL accession Nr. Pl1224). The first, a (Parker et al., 1989) . The proteolytic cleavage site separating the S1 and $2 portions of the molecule is marked with an arrow. change of A to G at nucleotide 293, resulted in a change from asparagine to serine at residue 98; the second, a change of G to A at nt 3045, was silent. Both of these changes were also found in the wild-type virus sequence, however, and thus were of no significance to the phenotype of Albl8. The same two nucleotide differences in our laboratory wild-type MHV-A59 strain have been reported by Gombold et al. (1993) . The 150 nt upstream of the S coding region were found to be identical to the previously published MHV-A59 sequence (Luytjes et al., 1988) . To determine the significance of the single coding change that was found in the Albl8 S gene, an analysis of revertants was carried out. Eight independent, spontaneously arising revertants of Albl8 were isolated on the basis of their ability to form plaques at the nonpermissive temperature. Revertants arose at a frequency on the order of 10 -5, and all formed wild-type-sized plaques at 39°C. In addition, all of the revertants resembled wild-type virus in thermal stability. Albl8 virus produced at the permissive temperature was 10-fold more thermolabile than wild-type virus when incubated at 40°C, pH 6.5, for 24 h (Table 1) . By contrast, the thermolability of each of the revertants under the same conditions was indistinguishable from that of wild-type virus (Table 1 ). This showed that temperaturesensitivity and thermolability of virions were two consequences of the same mutation in Albl8. The region encompassing nt 860 of the S gene was sequenced in all of the revertants. For one of these, Albl8Rev6, nt 860 was found to have exactly reverted from T to G, returning codon 287 to that of the wild-type serine. For each of the seven other revertants, nt 860 was changed from T to C, resulting in a conservatively substituted threonine residue at position 287 (Table 1) . These results clearly a Thermal inactivation was carried out for 24 h at 40°C and pH 6.5 as described previously (Koetzner et al., 1992) . Titers of surviving virus were determined on L2 cells at 33°C. established that the nucleotide change found at codon 287 was the lesion responsible for the phenotype of Albl8. To learn whether the temperature-sensitivity of Alb18 could be reversed, temperature shift experiments were performed. 17C11 cells were infected with Albl8 or wild-type virus at either the permissive temperature ( Fig. 4A and B ) or the nonpermissive temperature ( Fig. 4C and D) . At 8, 16, or 20 h post-infection, samples of virus released into the culture medium were taken for determination of infectious titer, and then the cultures were shifted to the nonpermissive or permissive temperature, respectively. Samples were again taken at 24 h, and their infectious titers were compared with those of unshifted control cultures. In cells infected with Albl8, shifting the temperature to 39°C after infection at 33°C resulted in the immediate and complete inhibition of further release of infectious virus following the shift (Fig. 4B ). This suggested that the mutant S protein underwent a change to an assembly-incompetent form as soon as the temperature was raised. Conversely, for cells infected with Alb18 at the nonpermissive temperature, the yield of infectious virions after a shift to the permissive temperature was equivalent to the yield from an infection at the permissive temperature that had proceeded for the same amount of time (Fig. 4D ). This showed that prior incubation at 39°C was not lethal to the intracellular virus and that replication'could resume once infected ceils were returned to 33°C. It also suggested that, following the downwards shift, newly synthesized S protein was required for the assembly of virions. If the population of Albl8 S protein previously synthesized at 39°C had acquired an assembly-competent form upon being shifted to 33°C, then we would have ex- pected to see a sharper burst of release of infectious virus assembled from previously synthesized components, and the final titers at 24 h for all shifted samples in Fig. 4D should have approached the same value. Further studies will be required to determine the molecular basis for the irreversibility of the defect in Alb18 S protein. Because the MHV S protein is an oligomer, it was of interest to examine the outcome of a mixed infection of Albl8 and wild-type virus. Coinfection of either 17Cll or Sac-cells with both Albl8 and wild-type virus, each at a multiplicity of 5 PFU/ceI1, resulted in a 20-fold reduction of infectious yield at 39°C compared with that obtained with wild-type virus alone (Table 2 , Experiment 1). The progeny of these mixed infections resembled wild-type virus in thermolability (data not shown). This suggested that at the nonpermissive temperature, Albl8 and wild-type S protein monomers were able to form mixed oligomers that were at least partially defective, and only those progeny having a sufficient concentration of wild-type homo-oligomeric S protein molecules were infectious. An alternative explanation of the data was that inhibition of wild-type infection by Albl8 was the result of a general phenomenon such as nonspecific aggregation between Albl8 S protein and other proteins in the membrane of the ER, including wild-type S protein. To examine this possibility, mixed infections were carried out at 39°C with Albl8 (or wild-type MHV) and the rhabdovirus VSV. Although VSV assembles at the plasma membrane rather than at intracellular membranes, its envelope glycoprotein follows the same secretory pathway as MHV S protein. The infectious yield of VSV in either 17Cll or Sac-cells was not reduced by coinfection with Albl8 irrespective of whether both viruses were infected simultaneously (data not shown) or even if Albl8 infection was allowed to proceed for 3 h prior to VSV infection (Table 2, Experiment 2). Therefore, the inhibition of wild-type MHV by Albl8 was a specific effect. The ability of mutant virion S protein to bind to the MHV receptor was evaluated using an in vitro solid phase assay developed previously by Holmes and wild type I II heat-treated heat-treated I I I I pH -~ 6.5 7.0 7.5 8.0 6.5 6.5 7.0 7.5 8.0 6.5 border membrane proteins were separated by SDS-PAGE on 7.5% polyacrylamide gels and were transferred to PVDF membranes. Blots were cut into strips and incubated with purified Albl8 or wild-type virus that had been inactivated for 24 h at 40°C at pH 6.5, 7.0, 7.5 or 8.0. Control (non-heat-inactivated) virus samples were held at 4°C at pH 6.5. Bound virus was detected by incubation with polyclonal goat antiserum AO4 followed by staining with horseradish peroxidase-coupied second antibody. coworkers (Boyle et al., 1987) . Receptor-containing intestinal membrane vesicles from BALB/c mice were solubilized and separated by SDS-PAGE followed by Western blotting. Virus samples were incubated with strips of the blot, and virions bound to the MHV receptor were visualized by staining with anti-S protein antisera. Owing to the absence of spikes on Albl8 virions released from cells at the nonpermissive temperature, it was only possible to analyze Albl8 virions produced at the permissive temperature. As shown in Fig. 5 , Albl8 virions exhibited the same ability as wild-type virions to bind to receptor (compare non-heat-treated lanes for each). To determine the stability of the Albl8 spikes under conditions whose effects on wild type were already known (Sturman et al., 1990) , virus samples were treated at pH 6.5, 7.0, 7.5 or 8.0 for 24 h at 40°C. Heat-treatment at pH 6.5 slightly decreased the ability of Alb18 virions to bind to receptor. This effect increased with increasing pH, and by pH 8.0 the receptor-binding ability of Albl8 was almost completely abrogated. By contrast, wild-type virions retained the ability to bind in this assay after all pH treatments. These data show that Albl8 virion spikes could be impaired by heat and mildly alkaline conditions much more readily than those of wild-type virions, indicating that the pH-dependent confor-mational change in S protein (Sturman et al., 1990) is more pronounced or more easily triggered in the mutant S protein of Albl8. The single amino acid change of serine 287 to isoleucine in the spike glycoprotein of the MHV mutant Albl8 confers a phenotype in which multiple functions of S protein are affected. The most salient feature of this mutant is that at the nonpermissive temperature virions are assembled devoid of S protein. In addition, the lesion results in two other major impediments: the inability of infected cells to form syncytia at the nonpermissive temperature and the thermolability of S protein even after it has been assembled into virions at the permissive temperature. The formation of spikeless virions by Albl8 indicates that S protein does not play an active role in MHV assembly, which is probably mediated by interactions between the M protein and the viral nucleocapsid (Sturman et al., 1980) . This supports the same conclusion reached previously from studies demonstrating that spikeless virions are released by MHV-infected ceils treated with tunicamycin, an inhibitor of N-linked glycosylation (Holmes et al., 1981; Rottier et al., 1981) . Similar observations have also been made with human coronavirus OC43 (Mounir and Talbot, 1992) and turkey enteric coronavirus (Dea et al., 1989) . The Albl8 phenotype recalls that of the well-studied VSV mutant tsO45, which also produces spikeless virions at its nonpermissive temperature (Gallione and Rose, 1985; Metsikkti and Simons, 1986 , and references therein). It has been shown that, in virions of tsO45 assembled at the nonpermissive temperature, the ectodomain of the VSV spike glycoprotein G has been proteolytically removed, but an anchor consisting of the transmembrane and cytoplasmic portions of the molecule remains embedded in the virion membrane (Metsikk6 and Simons, 1986) . We were not able to detect an analogous carboxy-terminal fragment of the Albl8 S protein in purified spikeless virion preparations tither directly or by immunoprecipitation with an anti-peptide antibody, which was raised against the carboxy terminus of S (data not shown). At present, however, we cannot completely rule out the possibility that the very cysteine-rich carboxy terminus of S is refractory to immunoprecipitation or standard conditions of SDS-PAGE, as is the VSV G protein fragment (Metsikk6 and Simons, 1986) . If S protein anchors were found in Albl8 virions formed at 39°C, this would require re-evaluation of the notion that S is only a passive participant in MHV assembly. The locus of the mutation that brings about the multiple changes in the S protein of Albl8 is distinct from that of any previously mapped MHV S mutation (Gallagher et al., 1991; Wang et al., 1992; Gombold et al., 1993; Fu and Baric, 1994) . Serine 287 fails a little more than a third of the distance from the mature amino terminus of the S1 half of the molecule (Fig. 3) . It is highly conserved, and it lies between two cysteine residues that are absolutely conserved in all reported coronavirus S sequences. The multiple effects of the Albl8 mutation likely cannot be accounted for at the level of primary sequence. The loss of serine 287 does not eliminate or create a consensus motif for N-linked glycosylation. Moreover, there is no evidence that two other potential post-translational modifications of serine, O-glycosylation or phosphorylation, occur in the MHV S protein; and a third, acylation, has been shown to occur in the $2, but not the S1, half of the molecule . It is noteworthy that the Albl8 mutation falls within the region to which the receptor-binding ability of MHV S has been recently mapped (amino acids 1-330; Kubo et al., 1994) . Conceivably, the lesion in Albl8 S results in a temperature-and pH-dependent conformational change that affects assembly, receptor binding and fusion. Further study of this mutant should provide useful insights into the structure and function of the MHV spike glycoprotein. In particular, it would be informative to use conformation-specific monoclonal antibodies to probe the structure of Albl8 S protein that is synthesized at the nonpermissive temperature. It would also be highly interesting if second-site revertants could be obtained to the Albl8 mutation, since these might yield evidence about intramolecular interactions in the S molecule. Genetic resistance to mouse hepatitis virus correlates with absence of virus-binding activity on target tissues Coronavirus IBV: structural characterization of the spike protein The cellular and molecular pathogenesis of coronaviruses Identification and location of the structural glycoproteins of a tissue culture-adapted turkey enteric coronavirus Assembly of coronavirus spike protein into trimers and its role in epitope expression Proteolytic cleavage of the E2 glycoprotein of murine coronavirus: host-dependent differences in proteolytic cleavage and cell fusion Map locations of mouse hepatitis virus temperature-sensitive mutants: confirmation of variable rates of recombination Neutralization-resistant variants of a neurotropic coronavirus are generated by deletions within the amino-terminal half of the spike glycoprotein Alteration of the pH dependence of coronavirus-induced cell-fusion: effect of mutations in the spike glycoprotein A single amino acid substitution in a hydrophobic domain causes temperature-sensitive cell-surface transport of a mutant viral glycoprotein Fusion-defective mutants of mouse hepatitis virus A59 contain a mutation in the spike protein cleavage signal A simple and very efficient method for generating cDNA libraries Tunicamycin resistant glycosylation of a coronavirus glycoprotein: demonstration of a novel type of viral glycoprotein Repair and mutagenesis of the genome of a deletion mutant of the coronavirus mouse hepatitis virus by targeted RNA recombination Localization of neutralizing epitopes and the receptor-binding site within the amino-terminal 330 amino acids of the murine coronavirus spike protein Phylogeny of antigenic variants of avian coronavirus IBV Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Localization of extensive deletions in the structural genes of two neurotropic variants of murine coronavirus JHM Primary structure of the glycoprotein E2 of coronavirus MHV-A59 and identification of the trypsin cleavage site Sequence of mouse hepatitis virus A59 mRNA2: Indications for RNA recombination between coronaviruses and influenza C virus Optimization of targeted RNA recombination and mapping of a novel nucleocapsid gene mutation in the coronavirus mouse hepatitis virus The budding mechanism of spikeless vesicular stomatitis virus particles Sequence analysis of the membrane protein gene of human coronavirus OC43 and evidence for O-glycosylation Disulfide bonds in folding and transport of mouse hepatitis coronavirus glycoproteins Sequence analysis reveals extensive polymorphism and evidence of deletions within the E2 glycoprotein gene of several strains of murine hepatitis virus Isolation of the subunits of the coronavirus envelope glycoprotein E2 by hydroxyapatite high performance liquid chromatography Viral protein synthesis in mouse hepatitis virus strain A59-infected cells: effects of tunicamycin DNA sequencing with chain terminating inhibitors Characterization of a coronavirus: I. Structural proteins: effect of preparative conditions on the migration of protein in polyacrylamide gels The molecular biology of coronaviruses The novel glycoproteins of coronaviruses Isolation of coronavirus envelope glycoproteins and interaction with the viral nucleocapsid Proteolytic cleavage of the E2 glycoprotein of murine coronavirus: Activation of cell-fusing activity of virions by trypsin and separation of two different 90 K cleavage fragments Temperature-sensitive mutants of MHV-A59 Conformational change of the coronavirus peplomer glycoprotein E2 at pH 8.0 correlates with virus aggregation and virus-induced cell fusion Infection of AtT20 murine pituitary tumor cells by mouse hepatitis virus strain A59: virus budding is restricted to the Golgi region Replication of coronavirus MHV-A59 in Sac-cells: determination of the first site of budding of progeny virions Biosynthesis and function of the coronavirus spike protein Sequence analysis of the spike protein gene of murine coronavirus variants: study of genetic sites affecting neuropathogenicity Purification of the ll0-kilodalton glycoprotein receptor for mouse hepatitis virus (MHV)-A59 from mouse liver and identification of a nonfunctional, homologous protein in MHV-resistant SJL/J mice Receptor for mouse hepatitis virus is a member of the carcinoembryonic antigen family of glycoproteins Mouse hepatitis virus gene 5b protein is a new virion envelope protein We wish to thank Dr. Kathryn V. Holmes (Uniformed Services University of the Health Sciences) for numerous valuable discussions and for generously providing mouse intestinal membrane vesicles, antiserum, and some of the oligonucleotides used in this study. We thank Tim Moran of the Molecular Genetics Core Facility of the Wadsworth Center for the synthesis of other oligonucleotides. This work was supported in part by Public Health Service grants GM 31698 (L.S.S.) and AI 31622 (P.S.M.) from the National Institutes of Health.