key: cord-0980731-mmgb7cxo
authors: To, L. T.; Bernard, S.; Bottreau, E.
title: Transmissible gastroenteritis coronavirus: surface antigens induced by virulent and attenuated strains
date: 1992-12-31
journal: Research in Virology
DOI: 10.1016/s0923-2516(06)80112-3
sha: 0e6279418b0659d03887b50be72d26bd5fadda2a
doc_id: 980731
cord_uid: mmgb7cxo

Summary Three strains of the transmissible gastroenteritis virus (TGEV) possessing different degrees of pathogenicity for piglets were examined for their capacity to express M and S glycoproteins on the infected cell surface using a microwell immunoperoxidase test. These two viral glycoproteins were easily detected on the plasma membrane of 0.1 % paraformaldehyde-fixed swine testis (ST) or pig kidney (RP.D) cells which were infected with high-passaged Purdue-115 and low-passaged D-52 strains and a high-passaged attenuated (188-SG) mutant of TGEV. No significant differences were found between attenuated and virulent strains with regard to the viral antigen expression on the membrane of infected cells over a 14-h period.

Transmissible gastroenteritis (TGE) is a highly contagious enteric infection of swine caused by a transmissible gastroenteritis coronavirus (TGEV) (Woode, 1969) . The causative agent of TGE belongs to the Coronaviridae, a family of enveloped viruses possessing a single-stranded co-linear RNA genome of positive polarity (for review, see Sturman and Holmes, 1983) . Three major structural proteins have been described for all coronaviruses: a high tool. wt. (220 kDa) glycoprotein (S) which forms the characteristic peplomers of the "corona", a small (29 kDa) transmembrane glycoprotein (M) and a phosphorylated protein (N, 47-50 kDa) associated with RNA (Garwes and Pocock, 1975; Garwes et al., 1976; Horzinek et al., 1982; Laude et al., 1986) . The peplomer glycoprotein is assumed to be involved in both virus adsorption to the cell and induction of virus-neutralizing antibody (Garwes et al., 1978) . The transmembrane glycoprotein is postulated to play a key role in alpha-interferon induction (Charley and Laude, 1988) . TGEV infection is followed by a very high mortality rate of up to 100 % in piglets which are less than 2 weeks old (Haelterman, 1972) . Sows that are naturally exposed to the virulent TGEV produce immune milk, which passively protects newborn pigs (Saif and Bohl, 1981; Bachman, 1982) . In contrast, attenuated TGEV does not induce complete protection by lactogenic immunity (Salf and Bohl, 1981) .

Since the virulence of TGEV has been shown to decrease by serial passages in tissue culture, many authors have tried to differentiate the highpassaged (HP) attenuated strains from the lowpassaged (LP) virulent strain by in vitro markers, such as the level of the thermosensitivity of replication (Furuuchi et al., 1975; Hess and Bachman, 1976) , the resistance to digestive enzymes, low pH and temperature (Laude et al., 1981) , and by comparing viral replication and synthesis of structural antigens (Nguyen et al., 1987) .

Using an HP attenuated mutant of TGEV (188-SG strain), which survives in the physicochemical environment of the digestive tract of adult pigs (Aynaud et al., 1985) , to study passive protection against TGEV infection in piglets, we found that this new TGEV mutant was capable of inducing protective lactogenic immunity and that it could be considered as candidate for an oral TGEV vaccine (Bernard et al., 1990; Aynaud et al., 1991) . However, the exact mechanism leading to the induction of protective immunity following oral immunization of sows with this mutant is still unknown.

In mouse hepatitis virus (MHV), a well studied coronavirus, the M protein migrates to the Golgi apparatus, but is not transported to the plasma membrane as readily as the S protein (Sturman and Holmes, 1983) . For porcine TGEV, the presence of the virus envelope S antigen on the surface of infected cells was demonstrated by immunofluorescence , while the presence of the M antigen on the plasma membrane has only been suspected by unspecified monoclonal antibodies (mAb) (Welch and Saif, 1988 ). There has not been any published report concerning the presence of N antigen on the plasma membrane of infected cells. However, our group and others (Laviada et al., 1990) have recently demonstrated the presence not only of S but also of M viral antigens on the membrane of ST cells infected with HP Purdue-115 strain of TGEV.

The purpose of the present study was to determine whether the LP virulent D-52 strain and HP attenuated 188-SG mutant were capable of expressing their M and S glycoproteins on the infected cell membrane in a similar way to the HP Purdue-115 strain. For the sake of comparison, the kinetics of expression of viral antigens on the plasma membrane and in the cytoplasm of ST and RP.D cells infected by these strains of TGEV was also studied with a view to discovering markers for differentiating HP attenuated strains from LP virulent strains.

RP.D is a previously described pig kidney cell line (Laude et al., 1981) . The McClurkin swine testis (ST) cell line was supplied by E.H. Bohl (Wooster, OH, USA). Minimal essential medium (MEM) supplemented with 10 °70 foetal calf serum, penicillin (100 IU/ml) and streptomycin (100 ~tg/ml) was used for cell growth.

Purdue-115 is an HP TGEV strain (Bohl et al., 1972) , D-52 is a virulent strain which was isolated from an acute case of TGE (P. Vannier, CNEVA, Laboratory of Porcine Pathology, Ploufragan, France) and passaged 5 times in RP.TG cells (Aynaud et al., 1985) and 188-SG is an HP attenuated mutant which was previously obtained in our laboratory through serial cycles of survivor selection in gastric juice (Aynaud et aL, 1985) .

For the experiments with inactivated virus, a viral suspension of each of these 3 strains was exposed to ultraviolet light (120 s, 2 mW/cm 2) (Charley et al., 1983) . Subsequent titration by plaque assay showed that the TGEV strains were fully inactivated following this treatment. Three mAb, anti-M (25/22), anti-S (51/13) (Delmas et al., 1986) and anti-N (22-6), were prepared and used as ascitic fluids following injection of BALB/c mice with the antibody-producing hybridomas .

Confluent monolayers of 2.5 x 105 cells/cm 2 in 96-well, flat-bottomed plastic plates (Falcon 3072, Becton Dickinson) were incubated with a volume of 0.1 ml of virus suspension at a multiplicity of infection (m.o.i.) of 10. After a 30-min incubation at 37°C under 5.5 070 CO 2, the inoculum in each well was removed and the cells were washed twice with phosphate-buffered saline (PBS). The monolayers were then overlaid with 0.1 ml of MEM containing 5 070 heat-inactivated (56°C, 30 min) normal calf serum and the plate was incubated at 37°C under 5.5 070 CO 2. The cell culture supernatant was harvested at the indicated time intervals and kept at -200C until titration.

An IPT which had been previously developed for the detection of surface viral antigens induced by Purdue-115 strain in infected ST cells was used. Briefly, the infected monolayers harvested at indicated times were washed twice with PBS and the cells fixed with 0.1 070 paraformaldehyde (Prolabo-France) at 4°C for 30 rain. After cell saturation with 5 % skimmed milk in PBS without calcium and magnesium for 15 min at room temperature, the monolayers were overlaid with 0.1 ml of each of 3 abovementioned mAb at working dilutions for 90 min at 4oc. The reagents were removed from the plates by rinsing twice with tap-water and twice with PBS containing 0.05 070 Tween-20 (scrva) and were then replaced with 0.1 ml/well of an optimal dilution of peroxidase-labelled goat anti-mouse Fc serum (ICN Immunobiologicals, Israel). After a further 90 min of incubation at 40C, the plates were washed as before and the enzymatic reaction was developed by incubation at 370C for 1 h with 2,2'azino-bis(3-ethyl-benzthiazoline-6-sulphonic acid) (ABTS; Boehringer Mannheim)/H20 2 substrate solution. The supernatant was transferred to another plate containing 0.02 ml of sodium dodecyl sulphate (SDS) to stop the enzymatic reaction and to permit the reading of the plate. The peroxidase was quantified by measuring the optical density at 415 nm with "Titertek Multiscan" (Flow Laboratories, Irvine, Scotland, UIO. Each antigen quantity, tested in quadruplicate, was expressed as the difference between the

For the detection of virus-induced antigens in cytoplasm, the infected cells were fixed with 80 % acetone at -20°C for 30 min and the IPT was applied as for surface antigens.

A plaque assay (Aynaud et al., 1985) was used to titrate the infectious virus in the cell culture supernatants sampled. Briefly, 2 to 3-day-old monolayer cultures of ST cells were produced by seeding 5 x 105 cells per 30-mm container (6-well trays). The cultures were inoculated with an appropriate TGEV dilution, and 2 ml MEM supplemented with 2 070 calf serum and 1 070 agarose (Indubiose) were added. Plaques were counted by neutral red staining following incubation at 37 to 38°C in 5.5 % CO2 for 48 h.

For the detection of M and S viral antigens in the culture supernatants, an ELISA immunocapture technique (Bernard et al., 1986) was used. Briefly, 96-well microtitre plates (Nunc-immunoplates, 4-42404) precoated with anti-M, anti-S and anti-N mAb, were incubated for 2 h at 37°C in carbonate buffer (pH 9.6). After washing, the plates were blocked overnight at 4°C with 1 ~/0 skimmed milk in PBS. Viral antigens were bound onto the precoated plates by incubating wells for 2 h at 37°C with supernatants from ST and RP.D cell cultures infected with either Purdue-115, D-52 or 188-SG strain. The peroxidase-labelled pig IgG polyclonal antibodies (Bernard and Lantier, 1985) were added for the next 2 h at 37°C. The enzymatic reactions were developed as mentioned above.

TGEV which was inactivated by ultraviolet irradiation failed to induce production of viral antigens while the infectious viruses did, as shown by IPT in infected ST cells ( fig. la and  b) . Also, neither infectious virus particles nor structural viral antigens could be detected by plaque assay and ELISA immunocapture in the cell culture supernatants sampled at the indicated time intervals. This experiment showed clearly that LP virulent D-52 strain and HP attenuated 188-SG mutant were also capable of expressing their glycoproteins on the plasmic membrane of infected ST cells, as previously described for HP Purdue-ll5 strain . strain. In contrast, RP.D cells infected with each of these 3 TGEV strains showed the same OD values for M and S antigens at 14 h p.i. 

We have recently developed a microwell IPT for detecting and quantifying the expression of viral S and M glycoproteins on the plasma membrane of ST cells infected with Purdue-115 strain of TGEV . In the present study, this technique was used to demonstrate and compare the expression of surface viral antigens in ST and RP.D cells infected with LP virulent D-52, HP Purdue-115 strains and HP attenuated 188-SG mutant. With this approach, we tried to find markers which would enable the differentiation of HP attenuated strains and LP virulent strains with regard to antigen expression on infected cell surface.

Of the 3 mutant viruses tested, the Purdue-115 is an HP attenuated strain (115 passages in ST-cell culture). However, undeg our experimental conditions, this strain was weakly virulent for newborn piglets (Shirai et al., 1988) . The 188-SG is an attenuated mutant previously obtained in our laboratory from the virulent Gep-II strain by 188 serial cycles of survivor selection in gastric juice of adult pigs (Aynaud et al., I985) . This mutant survives in the physico-chemical environment of the digestive tract of adult pigs, is nonpathogenic for piglets (Aynaud et al., 1985) and is capable of inducing lactogenic immunity in sows following oral immunization (Bernard et aL, 1990; Aynaud et al., 1991) . The original virulent D-52 strain is a mutant obtained from the virulent Gep-II strain by 5 passages in RP.TG cells (Aynaud et al., 1985) and is pathogenic for newborn piglets (Bernard, unpublished data) . Unlike the Gep-II strain, the virulent D-52 strain could be grown in in vitro cell culture.

No differences in the capacity to express surface viral glycoproteins ( fig. 1) were found between the 3 TGEV strains, as the presence of M and S glycoproteins was determined easily in infected ST and RP.D cells, while the presence of N antigen was not (data not shown). In contrast, the N (data not shown), M and S ( fig. 2) antigens were easily detected by IPT in the cytoplasm of TGEV-infected cells which were fixed with 80 °70 acetone. For the purpose of comparing the expression of viral antigens on the surface of infected cells, the anti-N mAb was used as a marker to ensure that after PFA fixation, the cell membrane would remain intact and only the viral antigens expressed on the plasmic membrane of infected cells would be detected. Experiments using inactivated virus have demonstrated that protein synthesis is a prerequisite for antigen expression on the cell membrane.

Our previous results indicated that the expression of M, S and N antigens appeared in multimodal patterns which peaks at 14, 16 and 18 h p.i. when ST cells were infected with 2 m.o.i. of Purdue-115 virus . Using the same m.o.i, of D-52 strain and 188-SG mutant, the patterns of expression of viral antigens in infected cells were also multimodal (data not shown). This phenomenon was due to incomplete infection of cell monolayers, which led to multi-cycle multiplication of virus. Laude et al. (1986) found that about 20 070 of ST cells expressed S antigen at 20 h p.i. when cells were infected at a m.o.i, of 2.5 x 10-2pFU/cell of Purdue-115 virus. In order to have glycoproteins appearing at the cell surface under single-cycle conditions of viral multiplication, a high m.o.i. (10 PFU/celI) was chosen to ensure that all cells were infected. It is interesting to note that the levels of expression of surface M and S antigens of the 3 virus strains were not significantly different when cells were infected with high m.o.i. (10 PFU/cell) ( fig. 2 ). This observation implies that the capacity to express glycoproteins on the cell membrane was not a marker for differentiating HP and LP TGEV strains. For all 3 mutant viruses used, surface virus antigen quantity was significantly lower with the RP.D cells than with the ST cells. This could be explained by the influence of cell culture systems on virus replication and synthesis of viral antigens (Nguyen et al., I987) . Furthermore, the appearance of the M and S antigens on the outer membrane of the cells could depend on an antigen-processing system, as previously described for other viruses (Long and Jacobson, 1989; Yewdell et al., 1981) .

With all our different combinations of viruses and cells, a lag was seen between the cytoplasmic antigens which had decreased in quantity 12 h p.i., while the surface antigens were still increasing. The decrease in cytoplasmic antigen expression can be explained since 12 h p.i. is the moment at which the virus progeny begin to be released from the cytoplasm of infected cells.

Concerning the production of infectious viruses and synthesis of structural antigens ( fig. 3 ) the infectious titres of HP Purdue-115 and LP virulent D-52 (about 109 PFU/ml) were higher than that of HP attenuated 188-SG mutant (about 107 PFU/ml) whereas the quantities of viral M and S antigens in the culture supernatants and cytoplasm of cells infected with these 3 viruses were similar. This experiment indicated clearly that the 188-SG mutant was characterized by a high structural antigen content and low infectivity in comparison with the 2 other viruses, since a high quantity of S antigen was detected at time 0. These observations are consistent with our previous results (Nguyen et al., 1987) of the comparison of viral replication and synthesis of structural antigens of these 3 strains of TGEV. The 188-SG mutant was characterized by low infectivity, delayed and restricted growth associated with low and delayed RNA synthesis and a high structural antigen content. In contrast, Purdue-115 and D-52 strains were characterized by high infectivity and a normal pattern of virus replication RNA and structural antigen synthesis.

In conclusion, no significant differences in in vitro expression of TGE viral antigens on plasma membranes were observed between the 3 virus strains and the 2 cell lines used which could explain the major differences existing between the virulence and the immunogenicity conferred by the different virus strains (Saif and Bohl 1981 ; Bernard et al., 1990; Aynaud et al., 1991) . Research on in vivo expression of TGEV antigens on the surface of intestinal cells of infected sows, especially in those undergoing oral immunization by HP attenuated 188-SG mutant, should be carried out in order to answer this question.

Induction of lactogenic immunity to transmissible gastroenteritis coronavirus using an attenuated mutant able to survive in the physicochemical environment of digestive tract

Transmissible gastroenteritis (TGE) of swine: survivor selection of TGE virus mutants in stomach juice of adult pigs

Comparative aspects of pathogenesis and immunity in animals

Lactogenic immunity to TGE of swine induced by the attenuated Nouzilly strain of TGE virus : passive protection of piglets and detection of serum and milk classes by ELISA

Detection of transmissible gastroenteritis coronavirus by a sandwich ELISA technique

A new conjugate for the ELISA quantitation of porcine IgA

Antibody response in serum, coiostrnm and milk of swine after infection or vaccination with transmissible gastroenteritis virus

Myxovirus and coronavirus induces in vitro stimulation of spontaneous cell-mediated cytotoxicity by porcine blood leukocytes

Induction of alphainterferon by transmissible gastroenteritis virus : role of transmembrane glycoprotein El

Antigenic structure of transmissible gastroenteritis virus: --II. Domains in the pepiomer glycoprotein

0975), Comparison of properties between virulent and attenuated strains of transmissible gastroenteritis

The polypeptide structure of transmissible gastroemeritis virus

0976), Isolation of subviral components from transmissible gastroenteritis virus

Antigenicity of structural components from transmissible gastroenteritis virus

On the pathogenesis of transmissible gastroenteritis of swine

Antigenic relationship among homologous structural components from porcine, feline and canine coronaviruses

In vitro differentiation and pH sensitivity of field and cell-cultureattenuated strains of transmissible gastroenteritis virus

Antigenic structure of transmissible gastroenteritis virus. --I. Properties of monoclonal antibodies directed against virion proteins

In vitro properties of low-and high-passaged strains of transmissible gastroenteritis coronavirus of swine

Expression of swine transmissible gastroenteritis virus envelope antigens on the surface of infected cells: epitopes externally exposed

0989), Pathways of viral antigen processing to CTL

Etude compar~e de 3 souches de coronavirus de la gastroent~rite transmissible: conditions de la r6plication viraie de la synth6se des anti-g6nes structuraux

Passive immunity against enteritic viral infection

Lactogenic immunity to transmissible gastroenteritis (TGE) of swine induced by the attenuated Nouzilly strain of TGE virus: neutralizing antibody classes and protection

The molecular biology of coronaviruses

Fixed-cell immunoperoxidase technique for the study of surface antigens induced by the coronavirus of transmissible gastroentefitis (TGEV)

Monoclonal antibodies to a virulent strain of transmissible gastroenteritis virus : comparison of reactivity with virulent and attenuated virus

Transmissible gastroenteritis

Expression of influenza A virus internal antigen on the surface of infected P815 cells

We are grateful to Dr Jean-Marie Aynaud for helpful advice.We also wish to acknowledge the Fondation Marcel M~rieux for awarding a scholarship to the leading author.

Trois souches du virus de la gastroent6rite transmissible (GET) poss6dant une pathog6nicit6 diff6rente pour les porcelets ont ~t6 examines, /t l'aide d'une technique d'immunop6roxidase en microplaque, pour leur capacit6 d'expression des glycopro-t6ines M et S/l la surface des cellules infect6es. Ces deux glycoprot6ines sont facilement d6tect6es sur la membrane plasmique des cellules ST (testicule du pore) et des cellules RP.D (rein de pore) infect6es par trois souches diff&entes de virus de la GET, et fix6es A la paraformald~hyde. Aucune difference d'expression des antig~nes viraux sur la membrane des cellules infect6es sont observables en fonction des souches virales et des lign6es cellulaires utilis~es.Mots-cl~s: Coronavirus, gastroent&ite transmissible, Virulence, Antig~nicit~; Expression, Souches HP et LP, immunoperoxidase, glycoprot6ines Met S.