key: cord-274673-tjzlssal
authors: De Groot, Raoul J.; Van Leen, Robert W.; Dalderup, Mieke J.M.; Vennema, Harry; Horzinek, Marian C.; Spaan, Willy J.M.
title: Stably expressed FIPV peplomer protein induces cell fusion and elicits neutralizing antibodies in mice
date: 1989-08-31
journal: Virology
DOI: 10.1016/0042-6822(89)90619-3
sha: 
doc_id: 274673
cord_uid: tjzlssal

Abstract We have established bovine papilloma virus (BPV)-transformed mouse C127 cell lines that synthesize the peplomer protein of the feline infectious peritonitis virus (FIPV) strain 79-1146. For this purpose, a new cassette expression vector pHSL, which carries the Drosophila HSp70 promotor and the polyadenylation signal of the Moloney murine leukemia virus long terminal repeat, was constructed. Cocultivation of the BPV-transformed cell lines with FIPV-permissive feline fcwf-D cells resulted in polykaryocyte formation. Since it depended on the presence of fcwf-D cells, binding of E2 to the cell receptor may be required for membrane fusion. E2 was synthesized as a core-glycosylated protein of 180K which was only slowly transported from the endoplasmic reticulum to the medial Golgi: of the E2-molecules labeled during a 1-hr pulse about half was still completely sensitive to endoglycosidase H after a 2-hr chase, while the remaining E2 had been chased into multiple, partially endoglycosidase H-resistant forms. Immunofluorescence studies also indicated that most E2 was retained intracellularly. Mice immunized with whole lysates of the transformed cells produced FIPV-neutralizing antibodies as shown by plaque reduction.

Coronaviruses, a group of positive-stranded, enveloped RNA viruses, cause considerable economical losses by infecting livestock and other domestic animals. The molecular biology of these viruses is studied not only because of their unusual replication strategy (for review see Siddel et a/., 1983; Spaan et a/., 1988) but also for the development of effective vaccines by recombinant DNA technology.

Coronaviruses are characterized by large petalshaped peplomers protruding from the viral membrane. These surface projections, which consist of the peplomer protein E2, play an important role during the infection process, since they mediate receptor binding and membrane fusion (Sturman and Holmes, 1983) .

E2 is a large (180K to 200K) glycoprotein (Sturman and Holmes, 1983) . Sequence analyses of the E2 genes of several coronaviruses revealed that its protein moiety is 1 150 to 1450 amino acid residues in length and contains an N-terminal signal sequence, a C-terminal transmembrane anchor, and 21 to 35 potential Nglycosylation sites (Binns et al., 1985; De Groot et a/., 1987b,c; Jacobs eta/,, 1987; Luytjes eta/., 1987; Niesters eta/., 1986; Rasschaer-t and Laude, 1987; Schmidt eta/., 1987) . Besides glycosylation, synthesis of E2 entails acylation and for some coronaviruses, like mouse ' Present address: Gist Brocades N.V., Postbus 1,260O MA Delft. *To whom requests for reprints should be addressed. hepatitis virus (MHV) and avian infectious bronchitis virus (IBV), cleavage into two subunits (Frana et a/., 1985; Stern and Sefton, 1982; Sturman eta/., 1985) . In the infected cell some E2 is not incorporated into virions but transported to the plasma membrane where it can induce polykaryocyte formation Holmes, 1983, 1985) . Cavanagh (1983) provided evidence that each peplomer is a dimer or trimer of E2. The presence of two regions with heptad periodicity in the C-terminal half of the protein suggests that the monomers are held together by a complex interchain coiled coil (De Groot et a/., 1987c) .

Several observations suggest that E2 is the principal antigen eliciting protective immunity. The N-terminal half of E2 was required for a protective immune response against IBV in chickens (Cavanagh et a/., 1986) while mice could be protected against a lethal challenge with MHV by vaccination with purified E2 (Hasony and MacNaughton, 1981) or E2-derived synthetic peptides (Talbot et a/., 1988; M. Koolen, personal communication) .

Passive immunization with E2specific monoclonal antibodies also provided protection against MHV (Buchmeier et a/., 1984; Wege et al., 1984) . However, the situation is much more complex in the case of feline infectious peritonitis virus (FIPV), which causes a fatal disease in cats (FIP), involving antibody-mediated early death (Pedersen and Boyle, 1980; Weiss and Scott, 198 1) .

Mammalian cell lines expressing the FIPV peplomer gene would provide a convenient source of protein to dissect the role of E2 in FIP. Moreover, such cell lines could be used to study virus assembly and membrane fusion. Here we report synthesis of FIPV E2 in bovine papilloma virus (BPV)-transformed Cl 27-cells. It is shown that the expression product induces fusion of FIPV-permissive feline cells and is immunogenic in mice. In addition, we describe two new BPV "cassette" expression vectors allowing (i) easy cloning of genes between promotor and termination sequences, (ii) easy replacement of promotor and termination sequences, and (iii) both transient and stable expression.

Plasmid pdBPV-MMTneo(342-12) (Law et al,, 1983 ) was obtained from the American Type Culture Collection. Plasmids pA4 LTR and pBN247 were kindly provided by Dr. A. J. M. Berns (Dutch Cancer Institute, Amsterdam). The Cl 27 cells used for transfection experiments were obtained from Dr. P. Howley (NIH, Bethesda). Cells were propagated in DMEM with 10% fetal calf serum. Transfection was carried out using the calcium phosphate precipitation technique (Graham and Van Der Eb, 1973) . To enhance the transfection efficiency, the cells were treated with 25% DMSO (Stow and Wilkie, 1976) or 20% glycerol (Frost and Williams, 1978) 4 hr after transfection. Cell lines were established by isolating foci as described by Law et al. (1983) . Fe/is catus whole fetus (fcwf)-D cells were obtained from Dr. N. C. Pedersen (School of Veterinary Medicine, University of California, Davis).

For Northern blot analysis RNA was isolated as described by Spaan et al. (1981) . Glyoxal-denatured RNA was fractionated on 0.8% agarose gels in 10 mM sodium phosphate, pH 8.0 (McMaster and Carmichael, 1977) transferred to nylon membranes (GeneScreen Plus; NEN, Boston), and hybridized according to the manufacturer's recommendations.

Standard recombinant DNA techniques were performed as described . Cytoplasmic dot hybridization was performed according to White and Bancroft (1982) .

The expression vector pHSL was derived from the vector pdBPV-MMtneo(342-12).

In order to facilitate the construction of this new vector (see Fig. l) , the large BamHl fragment (7945 bp) representing the entire BPV-1 genome was deleted from pdBPV-MMtneo. The remaining fragment, designated pMTN, was recircular-ized. As terminator sequences we chose the long terminal repeat (LTR) of Moloney murine leukemia virus (Van Beveren et a/., 1980) . The 0.6-kb PvulllHindlll fragment containing the LTR was isolated from plasmid pA4 LTR. By linker addition the Pvull site was converted into a Bg/ll site. Similarly, the HindIll site was changed into a BamHl site. The resulting fragment was ligated to the large fragment of BgllIIBamHI-digested pMTN, yielding plasmid pMTL. The 0.7-kb EcoRIl BamHl fragment, carrying the Drosophila heat-shock (HSp70) promotor (Torok and Karch, 1980) was excised from pBN247 and, after converting the BarnHI site into a Bg/ll site by linker addition, ligated to the large EcoRIIBgllI fragment from pMTL. The resulting vector, pHSL, was used in our experiments.

A full-length cDNA copy of the E2 gene of FIPV strain 79-1 146 (see below) was inserted into the 8g/ll site of pHSL, while the BPV-1 genome was inserted into the BamHl site. The construct containing the BPV genome in the same transcriptional orientation as the FIPV E2 gene was designated pHSFILB(+); the construct containing the BPV genome in the opposite orientation was named pHSFlLB(-). Cell lines made with these plasmids were indicated by RM(+) and RM(-), respectively.

Tailoring of the FIPV peplomer gene Plasmid pB1 contains a full-length cDNA copy of the E2 gene of FIPV 79-l 146 (De Groot et al., 1987b) . Before cloning this gene into the expression vectors redundant noncoding sequences located at the 5' and 3' ends were removed. For trimming of the 5'end (extending 350 nucleotides from the E2 initiation codon), 16 ng of the synthetic oligonucleotide YTGTGCCATGAT-TGTGCT 3'(corresponding to position -6 to +l 1 of the E2 gene) was annealed in 10 m/l/l Tris-HCI, pH 7.5, 50 mlVl NaCl (total volume 30 ~1) to 1 pg of single-stranded DNA from a recombinant M 13 mp8 phage containing the (-)sense strand of the 5' terminal 1.4-kb Pstl fragment of pB1. Double-stranded DNA was synthesized by incubation with 0.05 mM dNTPs and 3 units Klenow DNA polymerase in a final volume of 40 ~1, for 15 min at room temperature. The DNA was precipitated with isopropanol, washed with 70% ethanol, and resuspended in 10 /*I 10 m/l/l Tris-HCI, pH 8, 1 mM EDTA. Subsequently, 15 ~1 of Sl buffer (30 mM potassium acetate, pH 4.6,0.25 n/l NaCI, 1 mMZnSO,, 5% glycerol) containing 7 units Sl nuclease (Pharmacia) was added; digestion was for 20 min at room temperature (Eghterdarzadeh and Henikoff, 1986) . After phenol extraction and ethanol precipitation, the DNA was digested with Pstl. Thus a 1058-bp fragment was generated corresponding to the 5' end of the E2 gene. This DNA was cloned into HindIll (blunt-ended with Klenow)/Pstl-digested pUC8 in the presence of BarnHI linkers. The ligation mixture was used for transformation of JM 109 according to Hanahan (1983) . Recombinant clones were selected by colony hybridization. The FIPV-specific inserts were sequenced. Plasmid pl A contained the 5' end sequence of the FIPV peplomer gene, starting 5 nucleotides upstream of the AUG co-don (1 nucleotide was lost probably by nibbling of the Sl nuclease), preceded by a BarnHI linker. Plasmid pl A was digested with Pstl and the 3000-bp Pstl fragment of B 1 was inserted in the correct orientation. The Accl site located 98 nucleotides downstream of the stop codon was used to remove the redundant sequences at the 3' end. As a result, the complete FIPV E2 gene was contained in a 4.3-kb BarnHI fragment.

To monitor E2-induced cell fusion, 1 O6 fcwf-D cells and 1 O5 BPV-transformed Cl 27 cells were seeded into a 35-mm-diameter petri dish and allowed to adhere and grow for 16 hr at 37". For induction of the heat-shock promotor the cells were incubated for 2 hr at 42" and then reincubated at 37". Cell fusion was assessed by light microscopy.

Protein analysis FIPV 79-l 146-infected cells were labeled with [35S]methionine (Amersham) from 6 to 6.5 hr after infection as described previously (De Groot et a/., 1987a). After 2 hr incubation at 42', RM(+)19 cells were labeled at 37" by growing in cysteine-free DMEM, containing 100 $Ci/ml [35S]cysteine (Amersham). Cells were lysed in PBS, 0.5YoTriton X-l 00. E2 was immunoprecipitated using ascitic fluid from an FIPV-infected cat (De Groot et a/., 1987a). Digestion with endoglycosidase H was performed as described by Rose and Bergmann (1983) . Proteins were denatured in Laemmli sample buffer, containing 5% 2-mercaptoethanol, and subsequently analyzed on 5% SDS-polyacrylamide gels.

lmmunofluorescence Indirect surface immunofluorescence was carried out with RM(+)l9 cells, fixed in 3% paraformaldehyde as described by Rose and Bergmann (1983) . For intracellular immunofluorescence, the cells were fixed for 20 min at -20" in 95% methanol, 5% acetic acid. Sera (1: 100) obtained from a cat before and after experimental infection with FIPV 79-l 146 were used as the primary antibodies, followed by FITC-conjugated rabbit anti-cat immunoglobulin G (1:80, Nordic).

Immunization RM(+)19 and AM2 cells were grown to confluence, subjected to heat shock, followed by a 5-hr incubation at 37". The cells were washed three times with PBS and harvested with a cell scraper. They were then suspended in PBS (10' cells/ml) and disrupted by freezethawing followed by 1 0-min sonification (bath sonicator, Type RK102H, Bandelin, West Germany). The cell suspensions were stored at -70".

The concentration of the cell lysates was adjusted to an equivalent of IO' cells/mouse/dose.

The first dose, consisting of lysate mixed with an equal volume of complete Freund's adjuvant, was applied subcutaneously. The mice were boosted intraperitoneally on Days 10 and 21 with lysates suspended in PBS (Marchioli et al., 1987) .

Twofold serial dilutions of heat-inactivated sera (150 ~1) were mixed with about 120 PFU of FIPV 79-l 146 in an equal volume of DMEM, 10% fetal calf serum. Incubation was for 1 hr at 37". These samples were used to infect monolayers of fcwf-D cells. After 1 hr absorption at 37", the cells were washed twice with PBS and finally maintained for 20 hr under an agar overlay. Plaques were counted after staining with 0.02% neutral red in PBS. Antibody titers, expressed as the reciprocal serum dilution causing 50% plaque reduction, were estimated according to the method of Reed and Munch (1938) .

At the time we started this research, no convenient BPV expression vectors were available. Therefore, we constructed the expression vector pMTL. This vector carries the mouse metallothioneine (MMT) I promotor (Hamer and Walling, 1982) while the Moloney murine leukemia LTR (Van Beveren et al., 1980) provides a polyadenylation signal. pMTL has been designed with the aim to allow for easy exchange of promotor and termination sequences. To construct vector pHSL the MMT promotor was replaced by the Drosophila HSp70 heat-shock promotor (Torok and Karch, 1980) (Fig. 1) . Any protein coding sequence can be cloned into the unique Bg/ll site downstream of the promotor. Vectors for stable expression were constructed by inserting the BPV-1 genome Sarver et al., 1981) into the unique BarnHI site downstream of the LTR. The versatility of these vectors was demonstrated by stable expression of the vesicular stomatitis virus (VSV) membrane glycoprotein and the nucleocapsid protein of MHV (not shown). A BPV-transformed cell line AM2, which expresses the MHV nucleocapsid protein under the control of the MMT promotor (Bredenbeek et a/., unpublished), served as a negative control in the experiments described below.

A full-length cDNA copy of the FIPV E2 gene was obtained from plasmid pB1 (De Groot et al., 1987b) .

To avoid possible interference with transcription and translation, noncoding sequences at the 5' and 3' end of the gene were removed.

To anticipate possible adverse effects related to the transcriptional orientation of the BPV sequences in the vector, we used both pHSFILB(+) and pHSFILB(-) (Fig.  1) 2. (a) Screening for E2 transcripts. Cytoplasmic RNA was extracted from 2 x 1 O5 BPV-transformed cells and spotted on nitrocellulose as described by White and Bancroft (1982) (rows b-f). About 1 pg of RNA isolated from FIPV-infected cells and 50 ng of an 1800bp HindIll fragment derived from the FIPV E2 gene were spotted as positive controls (a2 and a4, respectively). A duplicate filter was incubated for 30 min at 37" in 2x SSC, containing 10 @/ml RNase A. Filters were hybridized to a nick-translated FIPV E2-specific probe. (b) Glyoxal-denatured RNA extracted from noninduced (lane 2) and heat-shock-induced (lane 3) RM(-)I 7 cells was separated on 0.8% agarose gels, transferred to a nylon membrane, and hybridized to a nick-translated 4.5-kbBamHl fragment containing the complete FIPV E2 gene. To estimate molecular weights a glyoxal-denatured EcoRIl HindIll digest of phage X DNA was induced (lane 1). RM(-)] were isolated, established as continuous cell lines, and tested for accumulation of EZ-specific RNA by cytoplasmic dot hybridization (White and Bancroft, 1982) after a heat shock (Fig. 2a) . About 20% of the cell lines gave a positive reaction. There was no difference in the percentage of positive cell lines generated by pHSFILB(+) or (--).

To characterize the E2 transcripts, cytoplasmic RNA was extracted from some of these cell lines and analyzed by Northern blotting. The nick-translated 4.3-kb BamHl restriction fragment containing the complete E2 gene (Fig. 1 ) was used as a probe. A prominent RNA species of 4.9 kb in length was detected. This size is consistent with a full-length transcript initiated at the heat-shock promotor and terminated at the polyadenylation site in the LTR. The 4.9-kb RNA species was also observed in noninduced cells, indicating that the heatshock promotor is leaky (Fig. 2b) . Transcription was enhanced about two-to fivefold after induction. In one cell line, RM(+)15, the 4.9-kb RNAwas not present; instead an E2-specific RNA of about 2 kb was found (not shown).

To test for E2 synthesis and its transport to the cell surface, we used an assay based on E2-induced fusion activity. The eight "best hybridizing" cell lines were cocultivated with fcwf-D cells (FIPV permissive feline cells) and found to induce cell fusion (Fig. 3) . Syncytia were only formed after addition of fcwf-D cells. Fusion did not occur in fcwf-D cells cocultivated with AM2 cells (which express the MHV nucleocapsid protein) or RM(+)15 cells (which produce an aberrant E2 transcript; not shown).

For most cell lines heat shock was required to induce cell fusion; after 16 hr of cocultivation at 37" only few syncytia were observed. Syncytia appeared and increased in number 4 to 5 hr after heat shock. In contrast, one cell line RM(+)19 caused widespread cell fusion even without induction. Differences in fusion potential were taken to reflect differences in the level of E2 expression. Using this criterium the RM(+)19 cell line was considered the best producer and therefore selected for further study. The low-producing cell line RM(-)17 was included for comparison.

To analyze E2 synthesized in these cell lines, [35S]cysteine-labeled cell lysates were subjected to immunoprecipitation with an FIPV-specific antiserum. A protein of 180K was detected in RM(+)19 cells. This product was also present in RM(-)I7 cells, although in very low amounts, but it was not detected in lysates of AM2 cells (Fig. 4a) . The 180K protein comigrated with pulse-labeled E2 produced in FIPV-infected fcwf-D cells (Fig. 4b) . These results indicated that the 180K product was indeed the E2 protein. The bands >200K were observed in both the lysates of E2-producing cell lines and the AM2 cell line indicating that these products are nonspecifically precipitated by the FIPV polyclonal antiserum.

E2 could not be detected on the plasma membrane of RM(+)l9 cells by immunofluorescence microscopy using the serum of an FIPV-infected cat (not shown), but in permeabilized RM(+)l9 cells perinuclear and granular cytoplasmic staining was observed (Fig. 5) ; staining was not observed in AM2 cells (not shown). These results suggested that most E2 was retained intracellularly.

To study the transport of E2 in the transformed cells, we analyzed the sensitivity of its oligosaccharide (3). Cells were labeled with [%]cysteine from 0 to 5 hr after heat shock. Lysates were subjected to immunoprecipitation using an E2-specific serum. Samples were analyzed on 5% SDS-polyacrylamide gels. The 180K product is indicated by an arrow; m: molecular weight marker. (b) Pulse-chase labeling of E2 in RM(+)l9 cells. Cells were labeled from 1 to 2 hr after heat shock with [%]cysteine.

Subsequently, cells were either lysed immediately ("pulse") or after a 2-hr chase with DMEM containing 4 mNIunlabeled cysteine ("chase"). E2 was immunoprecipitated.

Half of each sample was digested with endo H (+); the other half was not digested (-). Samples were analyzed on 59/o polyacrylamide gels. E2 immunoprecipitatedfrom pulse-labeled FIPWnfected cells was used as marker. Relative amounts of E2 were estimated by densitometry of the autoradiogram using a scanning densitometer (Type DD2, Kipp and Sons, The Netherlands). chains to endoglycosidase H (endo H). Endo H-sensitive high-mannose oligosaccharides are added to the protein backbone during translocation across the membrane of the endoplasmic reticulum (ER). These precursor sugar chains are processed to endo H-resistant "complex" oligosaccharides in the medial Golgi (Dunphy and Rothman, 1985; Hubbard and Ivatt, 1981) . Figure 4b shows that E2 produced during a lhr pulse-labeling of RM(+)l9 cells did not contain complex oligosaccharide side chains. Digestion with endo H yielded a 160K product comigrating with pulse-labeled, endo H-treated E2 from FIPV-infected cells. About 50% of E2 in RM(+)19 cells was still fully endo H-sensitive after a 2-hr chase. The remaining E2 molecules appeared to be chased into products ranging in length from 160K to 180K after endo H-treatment. Prolonged chasing did not result in the appearance of a discrete endo H-resistant product. Rather, the material accumulated as multiple diffuse bands ranging in length from 180K to 200K (not shown).

To determine whether the BPV-transformed cell lines synthesize enough E2 for use as immunogen, mice were immunized on Day 1 and boosted on Days 10 and 21 with lysates of RM(+)19 and AM2 ceils. Sera collected at Days 1 (preserum), 21, 28, and 40 were tested for the presence of E2-specific antibody in a plaque reduction assay. Table 1 shows that mice immunized with RM(+)l9 cell lysates developed rising titers of FIPV neutralizing antibodies. No plaque reduc- FIG. 5 . Detection of the E2 protein by indirect immunofluorescence microscopy. RM(+)19 cells grown on coverslips were fixed with 95% methanol, 5% acetic acid and incubated with serum obtained from a cat before (a) and after(b) experimental infection with FIPV 79-1 146. tion was observed with sera from mice immunized with AM2 cell lysates.

Expression vectors containing either the complete genome or subgenomic fragments of BPV-1 have been used to establish continuous cell lines producing large quantities of exogenous proteins (Hsiung et al., 1984; Pavlakis and Hamer, 1983; Reddy et a/., 1987; Sambrook er al., 1985; Sarver et a/., 1985 Sarver et a/., , 1987 Zinn et al., 1982) . In these vectors the BPV sequences carry the functions for plasmid replication and stable mainte-nance in the transformed cells (Law eta/., 1981; Sarver eta/., 1981) .

We have set out to establish a number of BPV-transformed mammalian cell lines which express coronavirus structural proteins. Such cell lines may provide a continuous source of protein for use as vaccine, but may also aid in the study of protein function and virion assembly. To facilitate plasmid construction we have developed the cassette vectors pMTL and pHSL containing the mouse MMT-1 and Drosophila HSp70 promotor, respectively. Both vectors possess the Moloney murine leukemia virus LTR to provide a polyadenylation signal. The unique Bg/ll site downstream of the promo- a Mice were immunized with disrupted RM(+)19 or AM2 cells (five mice per group). Doses of either preparation were given on Days 0 (subcutaneously), 10, and 21 (intraperitoneally). Heat-inactivated sera were pooled and tested in a plaque reduction assay.

'The neutralizing titer is indicated as the reciprocal of the serum dilution causing 50% plaque reduction when incubated with 120 PFU of FIPV 79-l 146.

tor elements can be used to insert any gene of interest. For stable expression, the BPV genome can be introduced into the unique BarnHI site. However, the vectors are equally suited for transient expression; e.g., by inserting the SV40 origin into the BarnHI site they can be used for expression in COS cells (Subramani and Southern, 1983 ).

Here we have used pHSL-derived plasmids to establish cell lines synthesizing the peplomer protein of FIPV strain 79-l 146. Southern blot analysis of total cellular DNA extracted from our best producing cell line RM(+)l9 showed the presence of 100-l 50 integrated copies of pHSFILB(-t) DNA per equivalent of diploid mouse genome (not shown).

The cell lines induced syncytium formation when cocultivated with FIPV-permissive feline cells. This effect was concluded to be E2 specific, since it was not induced by the BPV-transformed cell line AM2 synthesizing the MHV nucleocapsid protein nor by the RM(+)15 cell line producing truncated E2 transcripts. In addition, syncytia formation was inhibited in the presence of FIPV antiserum; preimmune serum from the same cat did not inhibited cell fusion (data not shown). Fusion did not occur between murine cells but depended upon the presence of feline cells, suggesting that for membrane fusion to occur binding of E2 to the FIPV cell receptor is required. A similar observation was made for cell lines expressing the envelope glycoprotein of human immune deficiency virus: polykaryocyte formation was restricted to cells bearing the CD4 receptor (Sodroski et al., 1986) .

In RM(+)19 ceils E2 was synthesized as a core-glycosylated protein of 180K. Removal of the oligosaccharide side chains by endo H reduced the size of the expression product to 160K (about the size of the FIPV E2 apoprotein) (De Groot et a/., 1987b). As shown by the fusion assay some of the E2 was transported to the plasma membrane. However, immunofluorescence studies suggested that most E2 was retained intracel-lularly. This was confirmed in pulse-chase experiments: of the E2 molecules labeled during a 1 -hr pulse about half was still completely sensitive to endo H treatment after a 2-hr chase. The remaining material was chased into immature endo H-resistant products 160K to 180K in length, probably representing partially processed medial Golgi forms (Dunphy and Rothman, 1985) . These results indicate that in RM(+)19 cells E2 is only slowly transported from ER to Golgi. For comparison, the VSV glycoprotein expressed in BPV-transformed Cl 27 cells is processed to the mature endo Hresistant form in less than 30 min (Florkiewicz et a/., 1983) . The slow maturation of E2 in RM(+)19 cells could be host cell determined: the influenza virus hemagglutinin (HA) acquired terminal glycosylation with a half-time of more than 2 hr in BPV-transformed Cl27 cells (Sambrook eta/., 1985) , while in simian CV-1 cells terminal glycosylation was completed within 40 min (Copeland et a/., 1986) . However, the slow rate of transport was also observed in various cell lines infected with a vaccinia virus recombinant expressing E2 of FIPV (Vennema et al., submitted). Since coronaviruses bud through pre-Golgi membranes (Tooze et a/., 1988) a delayed intracellular transport of free E2 is likely to be a prerequisite for efficient virus assembly.

To explore the potential of our BPV-transformed cell lines as immunogen, we immunized mice with lysates of RM(+)19 cells. We did not determine the amount of E2 in these lysates, since quantitative assays are not yet available. However, there was enough E2 to elicit FIPV neutralizing antibodies. Although the titers were modest, we trust that higher titers can readily be obtained by using larger amounts of cell lysate and optimized immunization protocols. We would like to use RM(+)19 cells for immunization experiments in cats to further study the role of E2 in immunopathology and protection.

Cloning and sequencing of the gene encoding the spike protein of the coronavirus IBV

Murine hepatitis virus-4 (strain JHM)-induced neurologic disease is modulated in vivo by monoclonal antibody

Coronavirus IBV: Structural characterization of the spike protein

Coronavirus IBV: Virus retaining spike glycopolypeptide S2 but not Sl is unable to induce virus-neutralizing or hemagglutinininhibiting antibody, or induce chicken tracheal protection

Assembly of influenza hemagglutinin trimers and its role in intracellular transport

The intracellular RNAs of the feline coronavirus FIPV strain 79-l 146

cDNA cloning and sequence analysis of the gene encoding the peplomer protein of feline infectious peritonitis virus

Evidence for a coiled-coil structure in the spike protein of coronaviruses

Compartmental organization of the Golgi stack

Use of oligonucleotides to generate large deletions

Isolation of stable mouse cell lines that express cell surface and secreted forms of the vesicular stomatitis virus glycoprotein

Proteolytic cleavage of the E2 glycoprotein of murine coronavirus: Host-dependent differences in proteolytic cleavage and cell fusion

Mapping temperature-sensitive and host-range mutations of adenovirus type 5 by marker rescue

A new technique for the assay of infectivity of human adenovirus 5 DNA

Regulation in vivo of a cloned mammalian gene: Cadmium induces the transcription of a mouse metallothionein gene in SV40 vectors

Studies on transformation of Escherichia co/i with plasmids

Antigenicity of mouse hepatitis virus strain 3 subcomponents in C57 strain mice

Efficient production of hepatitis B surface antigen using a bovine papilloma virus-metallothionein vector

Synthesis and processing of asparagine-linked oligosaccharides

The nucleotide sequence of the peplomer gene of porcine transmissible gastroenteritis virus (TGEV): Comparison with the sequence of the peplomer protein of feline infectious peritonitis virus (FIPV)

A stable bovine papillomavirus hybrid plasmid that expresses a dominant selective trait

Mouse cells transformed by bovine papillomavirus contain onlyextrachromosomal viral DNA sequences

Primary structure of the glycoprotein E2 of coronavirus MHV-A59 and identification of the trypsine cleavage site

Molecular Cloning: A Laboratory Manual

Evaluation of pseudorabies virus glycoprotein gp50 as a vaccine for Aujeszky's disease in mice and swine: Expression by vaccinia virus and Chinese hamster ovary cells

Analysis of singleand double-stranded nucleic acids on polyacrylamide and agarose gels by using glyoxal and acridine orange

The peplomer protein sequence of the M41 strain of coronavirus IBV and its comparison with Beaudette strains

Regulation of a metallothionein-growth hormone hybrid gene in bovine papilloma virus

Immunologic phenomena in the effusive form of feline infectious peritonitis

The predicted primary structure of the peplomer protein E2 of the porcine coronavirus transmissible gastroenteritis virus

Expression of human uterine tissue-type plasminogen activator in mouse cells using BPV vectors

A simple method for estimating fifty percent endpoints

Altered cytoplasmic domains affect intracellular transport of the vesicular stomatitis virus glycoprotein

Lines of BPV-transformed murine cells that constitutively express influenza virus hemagglutinin

Bovine papilloma virus deoxyribonucleic acid: a novel eukaryotic cloning vector

Sustained high-level expression of recombinant human gamma interferon using a bovine papillomavirus vector

Stable expression of recombinantfactorVII1 molecules using a bovine papillomavirus vector

Nucleotide sequence of the gene encoding the surface projection glycoprotein of coronavirus MHV-JHM

The biology of coronaviruses

Role of the HTLV-III/LAV envelope in syncytium formation and cytopathicity

Coronaviruses: Structure and genome expression

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

Coronavirus proteins: Biogenesis of avian infectious bronchitis virus virion proteins

An improved technique for obtaining enhanced infectivity with herpes simplex virus type 1 DNA

The molecular biology of coronaviruses

The novel glycoproteins of coronaviruses

Proteolytic cleavage of the E2 glycoprotein of murine coronavirus: Activation of cell fusion activity of virions by ttypsin and separation of two different 90K cleavage fragments

Analysis of gene expression using simian virus 40 vectors

Vaccination against lethal coronavirus-induced encephalitis with a synthetic decapeptide homologous to a domain in the predicted peplomer stalk

Site of addition of N-acetyl-galactosamine to the El glycoprotein of mouse hepatitis virusA59

Nucleotide sequences of heat shock activated genes in Drosophila melanogaster. I. Sequences in the regions of the 5' and 3' ends of the HSp70 gene in the hybrid plasmid 56H8

Structure of Moloney murine leukemia viral DNA: Nucleotide sequence of the 5' long terminal repeat and adjacent cellular sequences

Hybridoma antibodies to the murine coronavirus JHM: Characterization of epitopes on the peplomer protein (E2)

Antibody-mediated enhancement of disease in feline infectious peritonitis: Comparisons with Dengue hemorrhagic fever

Cytoplasmic dot hybridization

Regulated expression of an extrachromosomal human b-interferon gene in mouse ceils