key: cord-310467-v2wlu5eq
authors: Tô, Long-Thành; Bernard, Serge
title: Effect of fixation on the detection of transmissible gastroenteritis coronavirus antigens by the fixed-cell immunoperoxidase technique
date: 1992-10-02
journal: Journal of Immunological Methods
DOI: 10.1016/0022-1759(92)90192-v
sha: 
doc_id: 310467
cord_uid: v2wlu5eq

Abstract The effect of various fixatives and detergents on the in vitro detection of the viral determinants which are expressed in swine testis cells infected with the transmissible gastroenteritis coronavirus (TGEV) was studied using a microwell immunoperoxidase technique. When compared with glutaraldehyde and formaldehyde, 0.1% paraformaldehyde was found to be the fixative of choice for the detection of these determinants on the membranes of infected cells. Among dehydrating fixatives, 80% acetone or a mixture of acetone and ethanol or of acetone, methanol and ethanol were found to be the best fixatives for the detection of these viral determinants which are expressed in infected cells. In the case of acetone, the temperature of fixation and its concentration in the fixative preparation were found to be important. The treatment of 0.05% glutaraldehyde-fixed, infected cells with 0.1% saponin or 0.1% paraformaldehyde-fixed, infected cells with 1%NP-40 led to satisfactory detection of viral determinants. Using Triton X-100 to render cells permeable, the quantities of N and M antigen detected in TGEV-infected cells prefixed with either 0.05% glutaraldehyde or 0.1% paraformaldehyde were equal to those of 80% acetone-fixed, TGEV-infected cells while the quantity of S antigen detected was diminished. The effect of other detergents such as zwittergent, embigen BB, Chaps and N-lauroylsarcosine on the detection of viral determinants was also studied.

Transmissible gastroenteritis (TGE) is a highly contagious enteric infection of swine caused by transmissible gastroenteritis coronavirus (TGEV) (Woode, 1969 major structural proteins: the nucleocapsid (N), which is associated with the RNA genome; the peplomer glycoprotein (S) and the transmembrane glycoprotein (M) (Laude et al., 1990) .

In recent years there has been increasing interest in the development and application of the fixed-cell ELISA system for virological and immunological investigations. For TGE infections, this technique using infected cells as a source of antigen has been used to detect total anti-TGEV antibodies (Kodama et al., 1980a) or IgA antibodies (Kodama et al., 1980b) in the serum of infected swine. For other viruses, it has been used for screening hybridomas (Russell et al., 1983) , for diagnosis (Pauli et al., 1984; Groin and Bernard, 1985; Van Tiel et al., 1988) and for identification of epitopes on enveloped or naked viruses (Anderson et al., 1986; Boere et al., 1986; Oosterlaken et ai., 1989; Simkins et al., 1989 ). Recently, we described a quantitative fixed-cell immunoperoxidase technique for the study of surface viral antigens induced by TGEV (I"6 et al., 1991) .

As the name of the technique implies, many fixatives have been used by different authors to fix infected cells to the plastic plates. However, there have not been any published reports concerning the effect of fixatives on the detection of viral antigens in infected cells by this technique.

In the present study, we have established the effect of various fixatives and detergents on the detection of TGEV determinants in infected cells, by means of MAbs or PAbs in a quantitative fixed-cell immunoperoxidase technique (IPT),

The high-passaged Purdue-ll5 strain of TGEV (Bohl et al., 1972) and the swine testis (ST) cell line (McClurkin and Norman, 1966) , supplied by Dr. E.H. Bohl (Wooster, OH, USA) were used in this study. Eagle's minimal essential medium (EMEM) supplemented with 10% heat-inactivated fetal calf serum (FCS), penicillin (100 IU/ml) and streptomycin (100/tg/ml) was used for cell growth.

Confluent monolayers of 2.5 × 105 ceUs/cm 2 in 96-well, flat-bottomed plastic plates (Falcon 3072, Becton Dickinson) were inoculated with a volume of 0.1 ml/well of virus suspension at a multiplicity of infection (m.o.i) of 1. After 30 rain of incubation at 37°C under 5.5% CO 2, the inoculum in each well was removed, washed twice with phosphate buffered saline (PBS) and replaced with 0.1 ml/well of EMEM containing 5% heatinactivated FCS. The infected cells were incubated at 37°C under 5.5% CO 2 for an additional 15 h, The unfixed, TGEV-infecmd cells served as controls.

The MAbs were directed against M (25-22), S (51-13) and N (22-6) determinants of TGEV and had been prepared and used as ascitic fluids following injection of BALB/c mice with the antibody-producing hybridomas (Laude et al., 1986) . The anti-TGEV PAbs (53-12) obtained from the serum of a sow orally immunized with vaccine containing the Nouzilly strain of TGEV (Aynaud et al., 1985) , and having a neutralizing titer of 1:65,000, were used for comparative purposes. (Pauli et al., 1984) , 60% acetone, 25% ethanol and 15% methanol (Pauli et al., 1984) , 70% acetone and 30% ethanol (Laude et al., 1986) .

Fixation procedure The TGEV-infected cells were gently washed twice with PBS and the cells were fixed at -20°C for 30 min with the appropriate fixative (0.2 ml per well) which was previously frozen at -20°C unless otherwise stated. The fixed cells were washed twice with PBS before addition of 0.1 ml/well of blocking solution and incubation for 15 min at RT. 

An IiT previously developed for the detection of surface viral antigens induced by the Purdue-115 strain in infected ST cells (T6 et al., 1991) was used. Briefly, the fixed cells were washed gently with PBS and were incubated for 90 min at 4°C with 0.1 ml/well of a predetermined dilution of anti-M, anti-S, anti-N MAb or anti-TGEV PAbs. The reagents were removed from the plates by two rinses with tap-water and two washes with PBS containing 0.05% Tween 20 (Serva) and then replaced by 0.1 ml/well of a working dilution of peroxidase-labelled goat anti-mouse Fc serum (ICN ImmunoBiologicals, Israel) or peroxidaseu 197 labelled rabbit anti-swine (heavy and light chain specific) serum (Cappei, Organon Teknica, France) . After incubation at 4°C for 90 min, the plates were washed as before and the enzymatic reaction was developed by incubation at 37'C for 1 h with 2,2'-azino-bis (3-ethyl benzthiazoline-6 sulphonic acid) (ABTS (Boehringer, Mannheim, Germany))/H20 2 chromogen/substrate solution. The supernatants were transferred to another plate containing 0.02 ml of 10% sodium dodecyl sulphate (SDS) (Serva) to stop the enzymic reaction and to permit the reading of the plate. The optical density (OD) was measured at 415 nm by the ELISA reader (Titertek Multiskan) coupled to a microcomputer for data storage and statistical processing.

The OD signals obtained from the liT are composed of three components: (a) a specific OD signal generated by the second-step conjugate specifically bound to the antigen-bound mona clonal antibody; (b) non-specific OD generated by non-specifically bound second-step conjugate; and (c) nonspecific spontaneous OD due to other reasons, e.g., spontaneous decay of the substrate or the intrinsic OD of the cells. The first component is the one of interest. The other two contribute non-specifically to the OD signal and therefore should be corrected in order to determine the specific signal. The quantity of each antigen, tested in quadruplicate, was expressed as the difference between the OD at 415 nm of virus-infected and uninfected cells using the formula:

For comparative purposes the degree of relative antibody binding to a specified determinant was expressed as a percentage of antibody binding to unfixed infected cells.

The effects of cell fixation with FA, GA and PFA at various concentrations on antibody bind-ing to viral determinants in the TGEV-infected cells are shown in Fig. 1 . Antibody binding to the N determinant was close to zero when cells were fixed with all these three aldehyde fixatives. Fixation with FA at different concentrations led to irregular patterns for antibody binding to M and S determinants (Fig. la) . FA at high concentrations caused a decrease of antibody binding to S determinant but an increase of the M determinant up to 175% of controls at a FA concentration of 0.5%. The levels of antibody binding to M and S determinants were very low using fixation with GA at different concentrations (Fig. lb) . and $ (51-13) determinants were expressed as percentages compared with the value obtained with 0.1% paraformaldehyde-fixed, TGEV-infected cells. The degree of antibody binding to the N (22-6) determinant was expressed as a percentage compared with that obtained with 80% acetone-fixed, TGEV-infected cells.

infected cells which were fixed with 0.1% PFA and antibody binding profiles for these two determinants agreed well (Fig. lc) . It is clear that, among the aldehydes tested, the M and S determinants were readily detected in TGEV-infected cells which were fixed with 0.1% PFA.

The effects of cell fixation on antibody binding to viral determinants in the TGEV-infected cells and at different temperatures are shown in Fig. 2 . Antibody binding to M determinant was decreased whereas binding to S and N determinants increased in cells fixed with increasing acetone concentrations at -20°C (Fig. 2a) . When cell fixation occurred at 4°C, an increase in acetone concentration led to poor detection of M and S determinants although the latter could be detected in cells which were fixed with 70% or 80% acetone (Fig. 2b) . In contrast, the antibody binding to the N (22-6) determinant reached a plateau at 40%. The antibody binding profiles to the M, S and N determinants in cells fixed with acetone at RT showed similarities with the profiles obtained in cell fixation at 4°C although their OD values were lower (Fig. 2c ).

In conclusion, M and S determinants were not well detected when cell fixation occurred at 4°C and at RT while the N determinant was well detected in cells fixed with high acetone concentrations regardless of the fixation temperature. Fig. 4 . The effect of treating cells with 0.05% glutaraldehyde or 0.1% paraformaldehyde and then with various concentrations of saponin on antibody binding to viral determinants. The degree of antibody binding to the N (22-6) determinant was expressed as the optical density value obtained with TGEV-infected ceils prefixed with the above-mentioned fixative preparation, treated with saponin to permeate the cell membrane and then incubated with anti-N (22-6) MAb (block 1), relative to the optical density obtained with 80% acetone-fixed, TGEV-infected cells. The degree of antibody binding to the M (25-22) and S (51-13) determinants was expressed as the optical density value obtained with TGEV-infected cells prefixed with above-mentioned fixative preparation, treated with saponin and then incubated with MAbs having specificity for the M (25-22) (block 2) and S (51-13) determinants (block 3) relative to the optical densities of 0.1% paraformaldehyde-fixed, TGEV-infected cells. determinant was expressed as the optical density value obtained with TGEV-infectad cells prefLxed with the above-mentioned fixative preparation, treated with Triton X-100 to permeate the cell membrane and then incubated with a MAb having specificity for the N (22-6) determinant (block 1), relative to the optical density obtained with 80% acetone-fixed, TGEV-infected cells. The degree of antibody binding to the M (25-22) and S (51-13) determinants was expressed as the optical density value obtained with TGEV-infected cells prefixed with above-mentioned fixative preparation, treated with Triton X-100 and then incubated with MAbs having specificity for the M (25-22) determinant (block 2) and S (51-13) determinants (block 3) relative to the optical densities of 0.1% paraformaldehyde-fixed, TGEV-infected cells.

As shown in Fig. 3 , the degree of antibody binding to M, S and N determinants in methanol/acetone/ethanol-and methanol/acetone-fLxed cells showed similarities with the results obtained in 80% acetone-fixed cells. The extent of antibody binding to these determinants in methanol/acetone/FA-, absolute methanol-, ethanol/acetic acid-, ethanol/ether-, absolute ethanol-fixed cells was lower than that detected in 80% acetone-fixed cells. It can be noted that antibody binding to the M and S determinants in 0.1% PFA-fixed, TGEV-infected cells was greater than those of cells fixed with any of these dehydrating fixatives. Fig. 6 . The comparative effect of treating cells with 0.1% paraformaldehyde and then with different detergents on antibody binding to the viral determinants. The degree of antibody binding to the N (22-6) determinant was expressed as the optical density value obtained with TGEV-infected cells prefixed with above-mentioned fixative preparation, treated with indicated detergent to perorate the cell membrane and then incubated with a MAb having specificity for the N (22-6) determinant (block 1), relative to the optical density obtained with 80% acetone-fixed, TGEV-infected cells. The degree of antibody binding to the M (25-22) and S (51-13) determinants was expressed as the optical density value obtained with TGEV-infected cells prefixed with above-mentioned fixative preparation, treated with detergetit and then incubated with MAbs having specificity for the M (25-22) (block 2) and S (51-13) determinants (block 3) relative to the optical densities of 0.1% paraformaldehyde-fixed, TGEV-infected cells.

The degrees of antibody binding to viral determinants in cells which were fixed with the aidehyde fixative, e.g., 0.05% GA or 0.1% PFA and then permeated with different concentrations of saponin are shown in Fig. 4 . The degree of antibody binding to the M, S and N determinants in 0.05% GA-fixed cells which were permeated with 0.1% saponin was similar to that ob/ained in 80% acetone-fixed cells. With the exception of the 0.1% PFA control the degree of antibody binding to all other samples was lower than that detected in 80% acetone-fixed cells.

When using Triton X-100 (Fig. 5) , antibody binding to the M and N determinants in 0.05% GA-or 0.1% PFA-fixed cells which were then permeated with Triton X-100 at various concentrations showed similarities with the results obtained in 80% acetone-fixed cells. The degree of antibody binding to the S determinant in 0.05% GA-fixed cells which were then permeated with Triton X-100 at 1.0, 0.75 and 0.5% was lower than that obtained with 80% acetone-fixed, infected cells. The degree of antibody binding to the S determinant in 0.1% PFA-fixed cells which were permeated with different concentrations of Triton X-100 was similar to that detected in 80% acetone-fixed cells. The OD of cells incubated with anti-M MAb was equal to that obtained from 0.1% PFA-fixed, infected cells while the OD of cells incubated with anti-S was significantly lower than that of 0.1% PFA-f'Lxed, infected cells.

For some other permeabilizing agents (Fig. 6) , antibody binding to the N determinant in 0.1% PFA-fixed cells which were permeated with either 1% zwittergent, 1% empigen BB, 1% NP-40, 10 mM and 50 mM Chaps showed similarities with the results obtained in 80% acetone-fixed cells while antibody binding to this determinant in cells permeated with 10 mM N-lauroyl and 1 mM Chaps was lower than that of 80% acetonefixed, infected cells. The degree of antibody binding to the S and M determinants in cells which were permeated with 1% zwittergent 3-12 or 10 mM N-lauroyl was lower than that of 80% acetone-luted, infected cells. The degree of antibody binding to these determinants in cells permeated with 1% zwittergent 3-16, or 1% empigen BB, 1% NP-40 or Chaps was similar to that obtained in 80% acetone-fixed, infected cells.

The following experiments were carried out to identify differences in localisation of antigens detected in infected cells which were fixed with either 0.1% PFA at 4°C (Fig. 7a ) or with 80% acetone at -20°C (Fig. 7b) using two different fixation procedures. The degree of antibody binding to viral determinants in unfixed, infected control cells was compared with that measured in infected cells which were fixed (i) before their incubation with antibody or (ii) after their incubation with antibody. The antibodies used in this experiment were MAbs and PAbs as mentioned in the materials and methods section. Whichever PFA fixation procedure was used, the degree of antibody binding to the M, S, N determinants and to virions in TOEV-infected cells was not significantly different from that of controls. However, antibody binding to viral determinants varied between cells fixed with 80% acetone at -20°C and the control infected cells. The degree of antibody binding to the M determinant in cells fred according to the two different fixation procedures was similar to that obtained in the controls. The degree of antibody binding to the S determinant was lower than that with the controls suggesting that after acetone fixation the accessibility of this determinant to the MAb was affected. The degree of antibody binding to the N determinant in cells fixed with 80% acetone after their incubation with MAb was equal to that of controls but considerably lower than that of infected cells which were fixed with 80% acetone prior to their incubation with MAb. The anti-TGEV polyclonal binding profiles to viral antigens were similar to those obtained with the M and S determinants.

The present study has shown that GA fixation alters the accessibility of TGEV determinants to anti-M, anti-S and anti-N MAbs and that FA fixation has the opposite effect on the detection of M and S viral determinants compared with PFA fixation. In another study Smith et al. (1988) has shown that formalin fixation blocks, alters or destroys a specific antigen expressed in bovine viral disease virus (BVDV)-infected cells. Similar results were also obtained by Saunders (1977) for the detection of hog cholera virus antigen after GA f~ation at different concentrations.

To investigate the influence of fixation temperature on viral antigen detection, the fixation step was carried out at -20°C, 4°C or at RT. The detectable quantity of M antigen decreased with an increase in acetone concentration at all three test temperatures. The increase in acetone concentration parallelled the increase of detectable N antigen. Our results suggest that the acetone concentration in fixative preparation and the fixa-tion temperature are two important criteria for a fixation procedure. Similar results were also obtained (Groen et al., 1989) by comparing the use of ethanol (-70°C) or acetone (+ 20°C) fixation for the detection of Hantaan virus. In his study on the development and evaluation of an enzyme labelled antibody test for the rapid detection of hog cholera antibodies, Saunders (1977) showed that this test gave a good result only when the fixative solution contained less than 30% acetone.

It should be noted that 100% acetone is a routinely used fixative which gives excellent results in immunohistochemistry. In our experiments, cell fixation with a high concentration of acetone (80%) resulted in adequate detection of TGEV antigens. However, we could not use higher acetone concentrations to fix the cells because the microtiter plates would be distorted with acetone concentration higher than 80% or would melt with 90-100% acetone (personal observation). Therefore, we do not know if acetone concentrations higher than 80% would give the same results in fixed-cell immunoperoxidase.

A comparison of cell fixation with dehydrating fixatives and mixtures of these fixatives (Fig. 3) suggested that the highest quantity of antigen detected was in TGEV-infected cells which were fixed with either 80% acetone or mixtures of act/meth or act/meth/eth. It is possible that adequate detection of TGEV antigens in act/ meth-and act/meth/eth-fixed cells is due to the acetone in the fixative preparations since viral antigens are not so readily detected in infected cells which are fixed with methanol or ethanol alone. Similarly, comparisons of cell treatment with permeabilizing agents (Figs. 4, 5 and 6) lead to the conclusion that the TGEV antigens were readily detected in PFA-fixed cells which were then permeated with 1% NP-40 and in GA-fixed cells which were then permeated with 0.1% saponin.

Several authors have compared the effects of aldehyde fixative and dehydrating fixative in the fixed-cell ELISA. Cannon (1987) found that both methanol containing 0.5% H20 2 and 4% FA were suitable fixatives for HEp-2 cells infected with respiratory syncytial virus (RSV), although methanol/H20 2 was considered to be the fixative of choice. The author also found that where a MAb specific for virus NP was used in liT, staining was only seen following FA fixation but this was not the case for MAb specific for RSV envelope glycoproteins. The latter generally gave good results with either fixative. Similarly, FA fixation seemed to preserve Borna virus-specific antigens better than acetone/methanol or acetone fixation and the labelling was pronounced when using MAb (Pauli et al., 1984) . Smith et al. (1988) also noted that immunoperoxidase staining of BVDV-infected cells fixed with formaldehyde was less intense in cells fixed in acetone.

Our results (Fig. 7) showed that only viral antigens which are present on the membrane of infected cells may be detected after aldehyde fixation and that the accessibility of these antigenic determinants to the appropriate antibodies was not changed after PFA fixation. However, both surface and cytoplasmic antigens were detected after acetone fixation. This is due to the fact that after PFA fixation me cell membrane was still intact but became permeable after acetone fixation. Thus MAbs also bind to cytoplasmic viral antigens (T6 et al, 1991) . It seems that acetone fixation at -20°(: had no effect on the accessibility of the N determinant to the MAb. The same results (data not shown) were also obtained in TGEV-infected cells which were fixed with either act/meth or act/meth/eth using two different fixation procedures. It would seem that the stronger the permeabilizing capacity of the fixative, the more N antigen but less M and S antigens were detected. Consequently, the interpretation of results comparing aldehyde and dehydrating fixatives should take this observation into consideration.

In conclusion 0.1% PFA appears to be a suitable fixative for the detection of TGEV antigens on the membrane of infected cells. For cellular viral antigens; 80% acetone, a mixture of act/meth or act/meth/eth are the fixatives of choice.

Identification of epitopes on respiratory syncytial virus proteins by competitive linking immunoassay

Transmissible gastrnenteritis (TGE) of swine: survivor selection of TGE virus mutant in stomach juice of adult pigs

Antigenic differences between virulent and avirulent strains of Semliki forest viruses detected with mAb

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

Microplate immunoperoxidase detection of infectious respirato~ ~ncytial virus in the lung of infected mice

Comparison of immunofluorescence and enzymelinked immunosorbent assays for the serology of Hantaan virus infections

Virus enzyme-linked cell immunoassay (VELCIA): detection and titration of rotavirus antigen and demonstration of rotavirus neutralizing and total antibodies

Detection of antibody against transmissible gastroenteritis virus of pigs by indirect immunoperoxidase antibody test

Characterization of immunogiobuSn A antibody in serum of swine inoculated with transmissible gastroenteritis virus

1~90) Molecular biology of transmissible gastroenteritis viru!,~

Antigenic ~ structure of transmissible gastroenteritis virus: 1. Properties of mAb directed against virion proteins

Studies on transmissible gastroenteritis of swine. II. Selected characteristics of cytopathogenie virus common to five isolates from transmissible gastrocnteritis

Plaque/focus immunoassays: a simple method for detecting antiviral monoclonal and other antibodies and viral antigens in cells

A mierov~ll immunoperoxidase test for screening hybridomas and for diagnosing Newcastle disease virus and Sendal virus

C (1977) Development and evaluation of an enzyme labelled antibody test for the rapid detection of hog cholera antibodies

Epitope mapping and detection of transmissible gastroenteritis viral proteins in cell culture using biotinylated mAb in a fixed-cen ELISA

Detection of cytopathic and noncythopathic bovine viral diarrhea virus in cell culture with an immunoperoxidase test

Techniques immunoenzymatiques. Soci~t~ Fran~aise d'lmmunologie. Les Editions INSERM

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

Enzyme immunoassay of mumps virus in cell culture with peroxidaselabelled virus specific monoclonal antibodies and its application for determination of antibodies

Transmissible gastroenteritis

We are grateful to Dr. J.M. Aynaud for useful advice, to E. Bottreau and I. Lantier for help with the cell cultures and ELISA. We also wish to 203 acknowledge the Fondation MarceI-M6rieux for providing a scholarship to the leading author.