J. gen. Virol. (1986), 67, 119 130. Printed in Great Britain

Key words: TGEV/coronavirus/monoclonal antibodies

119

Antigenic Structure of Transmissible Gastroenteritis Virus. I. Properties of Monoclonal Antibodies Directed against Virion Proteins

By H U B E R T L A U D E , * J E A N - M I C H E L C H A P S A L , J A C Q U E L I N E G E L F I , S U Z A N N E L A B I A U A N D J E A N N E G R O S C L A U D E

Institut National de la Recherche Agronomique, Station de Recherches de Virologie et d'Immunologie, 78850 Thiverval-Grignon, France

(Accepted 10 September 1985)

S U M M A R Y

Thirty-two hybridoma cell lines producing monoclonal antibodies (MAbs) against the three major structural proteins of transmissible gastroenteritis virus (TGEV) have been isolated. Radioimmunoprecipitation of intracellular viral polypeptides showed that 17 hybridomas recognized both the peplomer protein [E2, 220 × 10 3 tool. wt. (220K)] and a lower mol. wt. species (E'2, 175K), which was characterized as a precursor of E2. Six MAbs selectively immunoprecipitated the E'2 protein. Four hybridomas were directed against the low mol. wt. envelope protein (El, 29K), and three against the nucleoprotein (N, 47K). All major neutralization-mediating determinants were found to be carried by the peplomers. Several anti-E2 MAbs displayed an intrinsic neutralizing activity close to that of the most potent anti-TGEV polyclonal reagents tested (including ascitic fluid of feline infectious peritonitis virus- infected cats). None of the anti-E'2 MAbs induced significant neutralization, although this protein might be incorporated to some extent into the virions. Immunofluores- cence patterns obtained with MAbs directed against either the envelope glycoproteins or the nucleocapsid revealed distinctly different distributions of these antigens within the cells. Comparison of nine TGEV strains using our panel of MAbs confirmed their close antigenic relationship, but revealed the occurrence of distinct antigenic differences.

INTRODUCTION

Transmissible gastroenteritis (TGE), an important disease of swine, is an attractive model for studying neonatal viral enteritis. Absorptive epithelial cells covering the small intestinal villi are the main target of the virus (Pensaert et al., 1970). Their infection leads to a villous atrophy which is responsible for severe intestinal disorders, often fatal for piglets younger than 2 weeks of age (Bohl, 1975).

The causative agent of TGE (TGEV) belongs to the Coronaviridae, a family of enveloped viruses possessing a single-stranded co-linear RNA genome of positive polarity (for review, see Sturman & Holmes, 1983). Like murine or avian coronaviruses, TGE virions are constructed from essentially three polypeptides : a high mol. wt. glycoprotein which forms the characteristic peplomers of the 'corona', a small transmembrane glycoprotein and a phosphorylated nucleoprotein (Garwes & Pocock, 1975; Garwes et al., 1976; Horzinek et al., 1982). The peplomer protein is assumed to be involved in both virus adsorption to the cell and induction of virus-neutralizing antibody. Indeed, a preparation of surface projections from detergent-treated TGEV particles has been shown to induce neutralizing antibodies in animals (Garwes et al., 1978/79). Within the mammalian coronaviruses, TGEV is antigenically related to canine enteric coronavirus (CCV), feline infectious peritonitis virus (FIPV) and human respiratory coronavirus (HCV-229E) (see Horzinek et al., 1982 for references).

We have produced a set of anti-TGEV monoclonal antibodies (MAbs; K6hler & Milstein, 1975) with the aim of obtaining novel information on the antigenicity and function of the virion

0000-6716 © 1986 SGM

120 H. LAUDE AND OTHERS

proteins. This report describes the general properties of 32 hybridomas, most of which were directed against the envelope glycoproteins.

METHODS

Cells and viruses. The American high cell-passaged Purdue-I 15 strain was used as a TGEV source for generating and testing hybridomas. The origin and characteristics of the nine cell-adapted TGEV strains listed in Table 2 have been reported previously (Laude et al., 1981). Infectivity titration as p.f.u, was performed on the ST (swine testis) cell line. Purdue virus stocks destined for purification were produced in the swine kidney cell line PD5. Infected monolayers were maintained in Dulbecco's MEM supplemented with 5 % ultracentrifuged newborn calf serum (CS) and 25 mM-PIPES pH 6.8.

Virus purification and labelling. Twenty h post-infection, harvests of 10 infected cultures (m.o.i. 3 to 5 p.f.u.) in 150 cm 2 flasks (Falcon) were pooled and centrifuged for 30 rain at 8000 g. The supernatants were then frozen and kept at - 7 0 °C until use. Crude suspensions were ultracentrifuged at 35000 r.p.m, in a 45Ti rotor (Beckman) for 90 min. Resulting pellets were resuspended in distilled water and sonicated for 2 min (Bransonic 220, Prosciences, Paris, France). This material was layered onto gradients of 20 to 45 % (w/w) sucrose in distilled water in tubes of a SW27 rotor (Beckman). After centrifugation for 3 h at 25000 r.p.m., the light-scattering band was collected and pelleted in distilled water (35 000 r.p.m., 2 h in a 45Ti rotor). Virus concentration was measured by u.v. absorbance using a calculated mean extinction coefficient of E l~m = 58 at 260 nm. This procedure allowed us to prepare about 1 mg of virus starting from 5 x 108 cells (specific infectivity averaging 3 × 107 p.f.u./~_tg). For some purposes, the virus band was submitted to isopycnic centrifugation (30 to 60% concave sucrose gradients, 14 h at 25000 r.p.m.) before pelleting.

Radiolabelled virus was prepared in the same way using two infected cell cultures maintained in leucine- and lysine-depleted Eagle's MEM with 2 ~ CS, supplemented at 5.30 h post-infection with 3H-amino acids (leucine, lysine, alanine, 15 p.Ci/ml each; CEA, Gif-sur-Yvette, France; 50 to 100 Ci/mmol).

Production o f hybridomas

Immunization procedure. BALB/c mice, initially devoid of anti-TGEV immunity, were injected intraperito- neally with 0.25 ml of a mixture of a 40-fold concentrated crude suspension of Purdue virus (approx. 108 p.f.u.) and the same volume of complete Freund's adjuvant. One to 4 months later the mice were boosted intravenously with 20 to 30 p.g of highly purified virus 3 days prior to spleen removal. The technique of in vitro boosting (Reading, 1982) was also used with one mouse: 5 x 107 splenocytes were cultivated for 2 days preceding fusion in 5 ml hybrid culture medium (see below) containing 1 ~tg/ml purified virus and supplemented with 5 × 10 -5 i -2- mercaptoethanol and 5 × 10 -5 i-hypoxanthine.

Hybridization. Briefly, fusion at a ratio of I : 10 splenocytes with myeloma cells (SP2/0-Ag-14) was carried out with 1 ml 4 5 ~ polyethylene glycol 1000 (Merck, Darmstadt, F.R.G.) following a procedure similar to that described by Nowinski et al. (1979). Then the cells were diluted in RPMI 1640 medium conditioned with 106 thymocytes/ml from 1-month-old mice, supplemented with 20~ foetal CS, and plated at 2-4 × 105 SP2/0-Ag-14 cells/well of 24-well plastic trays (Costar). Selection medium containing 5 × 10 -5 M-hypoxanthine and 10 -5 M- azaserine was added the day after. Twelve days later, azaserine was removed from the medium.

Screening and subcloning. Supernatants of hybrid clones were assayed by immunofluorescence and neutralization tests (see below). TGEV antibody-producing hybridomas were subcloned in thymocyte- conditioned medium by limiting dilution. Their monoclonal nature was checked at a confidence level of 95 °/ according to Poisson's distribution (De Blas et al., 1981).

Isotype determination. Monoclonal isotypes were determined by double diffusion using lysates made from 107 cultured cells in 0 .5~ NP40 (Sigma) in phosphate-buffered saline (PBS). Isotype-specific antisera were purchased from Miles Laboratories.

MAb production and purification. Working preparations of antibody were obtained by production of ascitic fluids in 2-month-old BALB/c mice treated with mayoline (Shell). After centrifugation for 3 min at 13 000 g, ascitic fluids were stored in aliquots at - 2 0 °C. Purified IgG was prepared by gel permeation of 1.5 ml ascitic fluid on a Sephacryl S-300 column (Pharmacia; vol. 450 ml, length 96 cm, flow rate 10 ml/h). Protein concentration in IgG fractions was determined by absorbance measurements at 280 nm using an extinction coefficient of 1.4 per mg protein.

Indirect immunofluorescence test (IIF). One-day-old ST cell monolayers established in microplates (Falcon Microtest II) were infected at a m.o.i, of 2.5 × 10 -z p.f.u./cell in Eagle's MEM. After a 20 h incubation period at 38°C the cultures were rinsed with PBS and fixed with cold acetone/ethanol (1:3, by vol.). This antigen preparation was stored at - 20 °C. For IIF, the plates were incubated for 1 h with serial twofold dilutions of antibodies, and rinsed with PBS containing 0.05~ Tween 20. Anti-mouse Ig ftuorescein isothiocyanate (FITC) conjugate diluted 1 : 100 (Pasteur Production, Paris, France) was added for 1 h. Washed and dried plates were

TGEV monoclonal antibodies 121

examined using an epiftuorescence microscope. For membrane fluorescence examination, cell sheets were treated with 4~o paraformaldehyde for 30 min at 4°C.

Neutralization assays. Two kinds of assay were performed. In the limiting dilution microtest, serial twofold dilutions of antibodies ( 1 : 10 or 1 : 100 pre-diluted heat-inactivated ascitic fluids or purified IgG) were mixed with an equal volume (50 ~tl/well) of medium containing 500 p.f.u, of virus. After 1 h incubation at 37 °C, 4 × 104 trypsinized ST cells in 50 ~1 Eagle's MEM supplemented with 15~ CS were added. Neutralization titres were determined 24 to 48 h later. Alternatively, the neutralization index (N I) was determined using a plaque reduction assay in ST cell monolayers in six-well plates (Costar) as described previously (Laude et al., 1981).

Radiobnmunoprecipitation. Radiolabelled cytosol extracts were prepared as follows: confluent PD5 cell monolayers in Petri dishes (10 cm diam., Costar) were infected (or mock-infected) at a m.o.i, of 50 p.f.u./cell of Purdue virus and maintained at 38 ~C in 10 ml methionine-depleted Eagle's MEM supplemented with 2~ CS. At 3.5 h after infection, 7 ml medium was discarded and 50 ~tCi/ml [35S]methionine was added (Amersham, 1145 Ci/mmol). The dishes were gently agitated until 8.5 h post-infection. The medium was discarded, the cell sheets were rinsed and then scraped into 4 ml chilled lysis buffer (2~ Triton X-100, 0.15 M-NaCI, 0.6 M-KC1, 0.5 mM- MgCI,, 103 units/ml Aprotinin, 10 mM-Tris HCI pH 7.8). Thirty min later, this material was ultracentrifuged for 1 h at 45000 r.p.m, in a 50Ti rotor (Beckman). The supernatant was stored at - 7 0 °C.

For pulse-chase labelling of intracellular polypeptides, monolayers established in 12-well plates (Costar) were infected as above. From 3 h to 5-5 h post-infection, the cells were placed in methionine-deprived medium without CS. Then the same medium containing 500 laCi/ml [3sS]methionine was added (0.1 ml/well). The labelling medium was removed 30 min later and the cells were washed twice with a medium containing CS and supplemented with 20 mM-methionine. At 30 min intervals, cell extracts from three wells were prepared by adding 0.2 ml lysis buffer.

lmmunoprecipitation assays were performed essentially as described by Sheshberadaran et al. (1983). Mixtures of 100 ~tl of cytosot extracts (2 × 106 c.p.m.) and 10 ~tl of undiluted hybridoma ascites were incubated for 1 h at 37 °C. They were then transferred into Eppendorf microfuge tubes containing sedimented Protein A-Sepharose beads (Pharmacia; 80 ~tl of 1:2 slurry per tube) washed with 1 ml 0.05 M-Tris-HC1 pH 8. After 1 h incubation at 37 °C with frequent vortexing, the beads were sedimented at 13 000 g for 2 min (Hettish microfuge) and washed four times with lysis buffer and once with Tris buffer. Beads were finally dried, heated for 2 min at 100 °C in Laemmli's sample buffer, and pelleted. Resulting supernatants were analysed by electrophoresis (PAGE) on

8.75'). 0 or 10°~ polyacrylamide slab gels (Studier, 1973), which were then dried and processed for fluorography (Bonner & Laskey, 1974).

lmmunoblotting. Viral proteins separated by PAGE (40 ~tg per lane) were transferred to nitrocellulose sheets (BA-85, Schleicher & Schfill) essentially as described by Burnette (1981). After saturation with bovine serum albumin, the sheets were incubated with ascitic fluid at a dilution of 1 : 100 and then treated with a sheep anti- mouse Ig-peroxidase conjugate (Pasteur Production), using diaminobenzidine as a substrate (Towbin et al., 1979).

Ascitic fluids jrom FlPV-in[ected cats. Samples derived from three spontaneous cases were employed as polyclonalreagentsagainstTGEV. The most commonly used throughout thisstudy had an anti-TGEV titreof 1 x 100 by neutralization and of 1.6 × 104 by IIF.

RESULTS

Production of hybridomas

The se rum-neut ra l i z ing an t ibody t i tres of T G E V - i m m u n i z e d mice inc reased rapidly, up to 3200 at 2 weeks and decreased slowly dur ing the fol lowing m o n t h s in the absence of boost ing. Th i r ty - two posi t ive viable c lones were es tab l i shed f rom the 212 clones analysed. H o w ev e r , up to 50~o of I IF-pos i t ive clones had been recorded on init ial screening, i nd ica t ing tha t the i m m u n i z a t i o n schedule was adequate .

Polypeptide specificity o f TGEV hybridomas

Immunoprecipitation experiments The results are presented in Fig. 1. The purified preparation of 3H-amino acid-labelled TGEV

used as a standard was resolved into three protein species upon autoradiography after SDS- PAGE: the peplomer protein E2 (220K), the nucleocapsid protein N (47K) and the heterogeneous envelope protein E1 (major species 29K). The mol. wt. values are comparable to those previously published (Garwes & Pocock, 1975; Horzinek et al., 1982).

Heterologous antibody from FIPV-infected cats was preferred as a polyclonal reagent as it was found more potent than anti-TGEV sera derived from hyperimmunized swine or mice. Four viral bands were specifically immunoprecipitated from metabolically labelled celt extracts

E2

=--I

1~

E,2

-==

~

N-,

-~

EI

~

(a)

8 4-

1 2

3 4

5 6

7 8

9 10

Cb)

0 ;>

~

+

l 2

3 4

5 6

7 8

9

~c) 4-

1 2

3 4

5 6

7 8

9 10

Fig.

1.

Pol

ypep

tide

spe

cifi

city

of

21 a

nti

-TG

EV

mon

oclo

nal

anti

bodi

es.

[3sS

]Met

hion

ine-

labe

lled

ext

ract

s fr

om i

nfec

ted

or u

ninf

ecte

d (l

anes

a 6

, b

5, c

1) c

ultu

res

Jcer

e in

cuba

ted

wit

h 10

~tl

of u

ndil

uted

hyb

rido

ma

asci

tic

flui

d. A

scit

ic f

luid

fro

m a

FIP

V-i

nfec

ted

cat

(AF

ip)

dilu

ted

1/10

(la

nes

a 5,

b 4

) or

und

ilut

ed (

lane

c 2

) w

as

ased

as

a po

lycl

onal

rea

gent

. Im

mu

ne

com

plex

es w

ere

rem

oved

usi

ng P

rote

in A

-Sep

haro

se b

eads

and

ana

lyse

d on

8.7

5 ~/

o (a,

b) o

r 1

0~

(c)

SD

S-p

olya

cryl

amid

e ge

ls

wer

e ex

pose

d fo

r 5

days

. P

urif

ied

3H-a

min

o ac

id-l

abel

led

TG

EV

was

inc

lude

d as

a s

tand

ard

(lan

es b

3,

c 7)

.

i'J

,v

Z

-I

7~

TGEV monoclonal antibodies

1 2 3 4 5 6 7 I

!

%=::

:== =

: i / =::

8 9

M:i: -,-E2 - ' , ~ E ' 2

~ :~i ~ ~ ,~

; :::: e •

Fig. 2. Pulse-chase labelling of TGEV-infected cells and immunoprecipitation of viral polypeptides using a heterologous polyclonal reagent (anti-FIPV). Lanes 1 to 7, ST cells pulse-labelled with [35S]methionine for 30 min, then chased with unlabelled methionine at 6 h post-infection; cell extracts were prepared at 30 min intervals. Lane 8, lysate from uninfected control 2.5 h after pulse. Lane 9, lysate of infected cells labelled for 4 h.

123

treated with anti-FIPV: E2, N, E 1 and an additional species designated E'2 (175K). The non- specific 53K and 46.5K bands visible above and below the N protein probably correspond to an artefact at the migrating front of IgG heavy chain, and to actin respectively.

As illustrated in Fig. 1, this technique allowed assessment of the polypeptide specificity of 30 out of the 32 isolated hybridomas. Seventeen hybridomas produced MAbs that recognized the E2 protein and also, without exception, the E'2 species. MAbs of six other hybridomas immunoprecipitated only E'2. In addition, four hybridomas were shown to be directed against the E1 protein and three against the N protein.

Relationship between the E2 and E'2 species

The results of pulse-chase experiments indicated that E'2 is a precursor of E2 (Fig. 2). This polypeptide was synthesized first, then partially converted into E2. It was also noticed that the E2 protein actually appeared to be composed of several discrete bands, probably related to different states of glycosylation.

Since we observed that a faint band with a mol. wt. equivalent to that of E'2 was consistently resolved in highly purified virus analysed on Coomassie Brilliant Blue-stained S D S - P A G E (Fig. 3, lane 1), we attempted an immunoblotting analysis. Whereas this technique was found suitable for confirming the specificity of anti-E1 and anti-N hybridomas, testing of anti-E2/E'2 MAbs was impaired due to an inefficient transfer of the E2 protein to nitrocellulose sheet (Fig. 3, lane 2 and 3). Nevertheless, the minor band weakly reacted on immunostaining with hybridomas

124 H. L A U D E A N D O T H E R S

• I~

165.~ :L: ;~ ~-E'2

2 3

94-~

67~ ~!i

27-~ ~

Fig. 3. Coomassie Brilliant Blue staining of purified TG EV polypeptides (40 ~tg/lane) resolved by 8.75 % SDS-PAGE (lane 1). Coomassie Brilliant Blue staining (lane 2) or immunostaining using the anti-N MAb 5.1 (lane 3) of TGEV polypeptides blotted from a 10~ gel.

25b. 21, 8.8, 5.2, 39.1 and 13.4, whereas the major E2 species was revealed only with the first three of these preparations (not shown).

Reactivities of hybridomas The specificities of the 32 TGEV hybridomas are summarized in Table 1. All subclasses of

IgG were found to be represented, MAbs of IgG2a isotype being the most common. Neutralization studies showed that nine out of 32 MAbs had titres between 6400 and 250000

when tested against 500 p.f.u, o fTGEV; they were directed against E2. The specific activities of every anti-E2/E'2 hybridoma were measured; whereas the last concentration giving positive IF varied from 6 to 0.3 ~tg/ml (not shown), great differences were observed among the limiting neutralizing concentrations, which ranged from 2 ~tg/ml to 15 ng/ml for the neutralizing MAbs (Table 1). Seven anti-E2/E'2 MAbs exhibited low or no neutralizing activity (sp. act. >25 lig/ml), as was the case for the anti-E1 MAbs. Anti-E'2 and anti-N MAbs were all devoid of neutralizing activity.

Immunofluorescence titres of ascitic fluids varied between 400 and 32000. Anti-N MAbs typically exhibited low IIF titres. Moreover, they induced a characteristic pattern of fluorescence: the antigen appeared evenly distributed in the cytoplasm (Fig. 4a). With anti- envelope glycoprotein MAbs, fluorescence appeared as fine granulation essentially limited to the perinuclear area (Fig. 4b). No clear difference could be observed between anti-E2/E'2, anti- E'2 and anti-E1 MAbs. Anti-E2 antibodies were able to induce a bright cell membrane fluorescence (Fig. 4c).

Neutralization kinetics

The neutralization curves of 10 purified anti-E2 MAbs were compared at the same IgG concentration. The following conclusions are drawn from the data summarized in Fig. 5. (i) Some MAbs possessed an intrinsic neutralizing activity close to that of our most potent

TGEV monoclonal antibodies

Table 1. Properties oJ" anti-TGEV monoclonal antibodies

125

Isotype IIF Neutralizing Limiting neutralizing Specificity Designation (Ig) titre titre concentration* (gg/ml)

Anti-E2 + E'2 20.9 G2b 8000 50000 0.2 25b. 21 G2b 25 600 32000 0.2 8.8 G2a 3200 6400 l 3b.5 GI" 3200 6400 0.2

10.4 Gt 3200 6400 0-2 12.18 G2a 25600 64000 1 40. l G2a 12500 18000 0.5 51.13 G2a 32000 256000 0.015 76.2 Gt 800 800 2 48.1 G2a 3200 32000 0.2 11.20 G1 3200 350 25 5.2 G2b 3200 640 6 6. 179 G2a 3200 < 20 i> 25

67.9 G2a 800 < 20 I> 25 69.21 G2a 10000 80 >i 25 78.17 G2a 5000 80 >/25 44.4 G2a 3200 320 25

Anti-E'2 39.1 G 1 5000 < 20 >/25 4.3 G2a 1600 <20 >~25

13.4 G2a 800 < 20 >~ 25 25.8:~ G~ 1600 < 20 >~ 25 31. 191 G2a 3200 <20 >125 61. 142 GI 1600 <20 >125

Anti-E 1 25.22 G 2a 16 000 < 20 ND§ 9.34:~ G 1 7000 160 ND 3.60 G 3 1600 < 20 ND

49.22 G~f 1600 400 ND Anti-N 5.1 G2b 500 < 20 ND

19.1 G2a 400 < 20 ND 22.6 G2a 800 < 20 ND

Undetermined 19.90 M 1600 < 20 ND 16.17 G2a 800 <20 ND

fraction of purified Ig. * Estimated using a 50 ~tg/ml initial concentration of the same t Subtype not determined. :~ Hybridomas obtained by in vitro boosting of splenocytes. § ND, Not determined.

preparations of ant i -TGEV and ant i -FIPV IgGs. (ii) The neutralizing activities as measured by NI or by limiting dilution are more or less in agreement; a major exception, however, concerns the MAbs 3b. 5 and 10.4 which repeatedly displayed a low NI, despite their high titre by the limiting dilution technique (Table l).

Comparison of TGEV strains

The reactivity of MAbs towards nine low and high cell-passaged TGEV strains originating from different countries was measured by IIF. Most antibodies recognized every strain tested. None ofanti-E'2, anti-N and anti-E1 MAbs revealed any antigenic variation. However, several anti-E2 MAbs which failed to recognize certain strain-specific determinants were identified. These were MAbs 3b. 5, 10.4 and 67.9 towards the related SF7 (low passage) and 126 (high passage) British strains, and MAb 76.2 towards the Czechoslovakian M42 strain (Table 2).

Study of the heterologous neutralizing activity corroborated these findings (Table 3). In general, the heterologous NI were found remarkably similar from strain to strain. Again, the British strains were not neutralized by MAbs 3b. 5 and 10.4, nor by MAb 76.2. The latter did not neutralize the M42 strain either. As well as this, some significant hyper- or hyporeactivities could be noticed (e.g. strains SF7 and 126 with MAbs 8.8 and 48.1).

Z

t3

O

,-q

Fig.

4.

Loc

aliz

atio

n of

vir

al a

ntig

en i

n S

T c

ells

at

20 h

aft

er i

nfec

tion

wit

h 0.

01 p

.f.u

./ce

ll o

f P

urdu

e vi

rus.

Im

mun

oftu

ores

cenc

e in

fix

ed c

ells

usi

ng t

he M

Ab

s an

ti-N

5.

1 (a

) an

d an

ti-E

2 25

b.21

(b

); m

emb

ran

e fl

uore

scen

ce w

as o

bser

ved

in p

araf

orm

alde

hyde

-tre

ated

cel

ls s

tain

ed w

ith

MA

b 2

5b.2

1 (c

).

T G E V monoclonal antibodies 127

! i i i i [

r _

O

.~ 4 _

3I, 'i"i" "" ................... - I I

2 5 10 15 30 60

Time at 37 °C (rain)

Fig. 5. Comparative kinetics of TG EV neutralization by monoclonal ( ) or polyclonal (- ) IgG, at 10 gg/ml: -~-, 10.4: ,~, 3b.5; D, 76.2: A, 25b.2l ; O, 12.18; O, 8.8: V , 40.1; ~, anti-TGEV; m, 51.13: IE 48.1 ; A, 20.9; , , anti-FlPV. A stock of Purdue virus plaque-purified twice was used. The neutralization reactions were stopped by serial dilution in ice-cold medium.

DISCUSSION

We have prepared a panel of MAbs to TGEV and identified the virion polypeptide specificity of 30 of them. The peplomer protein, although being the least abundant in the virions, appeared particularly immunogenic since 17 hybridomas produced anti-E2 MAbs. The latter also recognized a well-defined intracellular lower mol. wt. species of 175K, named E'2, which appears to be a precursor according to pulse-chase experiments. The mol. wt. found for E'2 is higher than reported by Hu et al. (1984) for the TGEV peplomer protein component (145K, as compared to 195K for the peplomer). Also, both E2 and E'2 bands were revealed on nitrocellulose blots stained with concanavalin A FITC (not shown). This is in agreement with the fact that no unglycosylated precursor of peplomer polypeptides was detected with murine coronaviruses (Rottier et at., 1981 ; Siddell et al., 198 l). Six MAbs bind preferentially to E'2. The occurrence of anti-precursor hybridomas could imply that TGEV underwent a low-level replication in recipient mice. However, our attempts to retrieve infectious virus by cell culture isolation or mouse passage were unsuccessful. As mentioned above, the neutralizing response was marked, in fact close to that commonly obtained in the hyperimmunized natural host. As shown in Fig. 3, the E'2 species as such might be incorporated into the virions which would explain why specific hybridomas are induced. Preliminary studies have indicated that labelled anti-E'2 MAbs do bind to purified virus. Recently, Wege et al. (1984) reported that five of their MAbs against the murine coronavirus JHM recognized both the gpl70 (peplomer) and the gpl50 (precursor) polypeptides, whereas 15 MAbs immunoprecipitated only the gp 150 species. These authors also showed that anti-gp 150 MAbs bind to purified virus which, however, did not appear to contain gp 150 as revealed by electrophoresis. They concluded that anti-gp 150 MAbs

Ta

ble

2.

Hy

bri

do

ma

An

ti-E

2 t

3b

.5

10.4

76

.2

An

ti-E

2

67

.9

Rea

ctiv

ity

by i

ndir

ect

imm

unof

luor

esce

nce

of

anti

-Pur

due

mon

oclo

nal

anti

bodi

es w

ith

hete

rolo

gous

T

GE

V s

trai

ns*

TG

EV

str

ain

s

Fra

nce

U

.K.

Jap

an

U.S

.A.

r A

~

r--'

~--

',

c-

-~

--

~

Cze

cho

slo

vak

ia

Pu

rdu

e D

52

.5

63

86

.4

Plo

uf.7

S

F7

12

6 S

H 1

5 S

H16

8 M

42

++

++

+

++

+

++

++

+

++

+

--

__

++

+

++

++

+

++

+

++

++

+

++

+

++

++

+

++

+

--

__

++

++

+

++

+

++

++

++

+

++

+

++

+

++

+

++

+

+

++

+

++

+

--

++

+

++

+

++

+

++

+

--

__

++

+

+

++

+

* E

ach

vir

us s

trai

n w

as t

este

d ag

ain

st f

our

dilu

tion

s o

f as

citi

c fl

uid:

1:

20,

1:10

0,

1:50

0 an

d

1:25

00.

Eac

h

+

indi

cate

s p

osi

tiv

e IF

at

the

corr

esp

on

din

g d

ilut

ion.

]"

Neu

tral

izin

g.

Ta

ble

3.

N

eutr

aliz

ing

acti

vity

* o

f an

ti-P

urdu

e st

rain

mon

oclo

nal

antib

odie

s ag

ains

t he

tero

logo

us T

GE

V s

trai

ns

TG

EV

str

ain

f

Hy

bri

do

ma

(dilu

tion)

]"

Pu

rdu

e D

52

.5

6386

.4

Plo

uf.7

S

F7

126

SH

15

SH

168

M

42

20,9

(1

: 100

0)

3"0

3"0

2-9

3"5

~>2"

7 2.

7 3.

2 3"

6 3-

6 25

b.21

(1

:400

) 2.

5 2-

4 2-

8 2.

5 2-

1 2.

6 2.

3 3.

0 3.

4 8.

8 (1

:400

) 2.

0 1.

5 1.

6 1.

8 1.

0 3.

1 2-

2 1,

6 1.

1 3

b.5

(1:

40)

1.9

1.9

1.6

1.1

0.1

0.3

1.9

1.8

1-6

10.4

(1:

40)

1.4

1.1

0.9

0.9

0.1

<<,0.

1 1.

2 1.

4 0.

9 12

.18

(1:4

00)

3.5

3.5

3.2

3.6

~>2.

7 3.

4 3-

8 >~

3.6

3.8

40.1

(1

:400

) 2.

0 1-

9 1.

9 1.

8 1.

3 1.

1 1.

8 1-

7 2.

0 51

.13

(1:2

000)

3.

6 3.

7 3.

9 3-

7 >~

2.7

3.4

3.7

>/3

.6

3.8

76.2

(1:

200)

0.

9 0-

7 0.

7 1.

1 0.

2 0.

1 0.

7 0.

6 <~

0.1

48.

I (1

: 1

000)

1.

8 1.

6 1-

9 2-

3 0.

8 2-

4 1-

6 1.

3 2-

5

* N

eutr

aliz

atio

n i

ndex

(lo

g~0)

for

a

15 r

ain

incu

bat

ion

at

37 °

C.

t F

inal

dil

utio

n o

f as

citi

c fl

uid.

oo

>

> o t-rJ

T G E V monoclonal antibodies 129

probably react with antigenic sites determined by the primary amino acid sequence. Additional experiments are needed to test these two kinds of hypothesis with regard to the anti-E'2 MAbs.

Study of the neutralizing activity associated with each class of MAbs confrmed that neutralization-inducing determinants are mainly located on the peplomers, as reported for murine hepatitis virus type 4, infectious bronchitis virus and bovine enteric coronavirus (Talbot et al., 1984; Mockett et al., 1984, Vautherot et al., 1984); also, in bovine enteric coronavirus, an additional surface 125K glycoprotein appears to be involved in virus neutralization (J. F. Vautherot, personal communication). Our results have also confirmed that MAbs have to bind to critical epitopes of the molecule to reduce virus infectivity markedly, as has been recognized with different enveloped RNA viruses (for review, see Dimmock, 1984). Of the four anti-E1 MAbs, two induced a weak but significant reduction of infectivity, which could not be increased by subsequent incubation with a fivefold concentration of sheep anti-mouse IgG (H + L) antibodies (data not shown). All the six anti-E'2 hybridoma antibodies were devoid of neutralizing activity. The finding that individual anti-E2 MAbs are able to neutralize virus infectivity to an extent equivalent to highly potent polyclonal reagents is at variance with recent studies indicating that cooperative interaction of MAbs directed against distinct spike epitopes is required to achieve substantial neutralization of Newcastle disease and La Crosse viruses (Ioro & Bratt, 1984; Kingsford, 1984).

Although the intracellular localization of TGEV antigens was not investigated in detail, it appeared that the MAbs can be assigned to two groups on the basis of the immunofluorescence pattern they induce. Both envelope glycoproteins were found mainly in the perinuclear area corresponding to the rough endoplasmic reticulum and Golgi apparatus. At the same time (20 h post-infection), about 20~ of cells were expressing E2 antigen at the membrane level. In contrast, anti-nucleoprotein MAbs induced a diffuse cytoplasmic fluorescence; we found the latter particularly suitable for the detection of TGEV strains growing poorly in cell culture.

Studies of the heterologous activity of our panel of MAbs indicated that most antigenic determinants are highly conserved from strain to strain. Distinct antigenic variation, however, was evidenced at the peplomer level of two TGEV isolates which could not be distinguished using conventional antisera. Current studies indicate that these variable epitopes are located outside the main 'neutralizable' domain of E2 (H. Laude et al., unpublished results). Anti-E1 hybridomas were too limited in number to allow conclusions about a possible antigenic variation of this protein. Thus, the situation with TGEV seems to differ from that in the murine coronavirus MHV-4, where each isolate had a unique pattern of reactivity to a set of anti- glycoprotein MAbs was reported (Fleming et al., 1983).

The panel of MAbs may be useful not only for a definition of antigenic determinants on TGEV peplomers in both structural and functional terms, but should also allow insight into the relationships between TGEV and antigenically related coronaviruses, e.g. FIPV and CCV. The fact that these viruses also cause infection of small intestine enterocytes of newborn pigs (Woods et al., 1981) underlines the need for differentiation between these agents.

We are grateful to the following persons for kindly providing us with material: E. H. Bohl, Wooster, Ohio, U.S.A. (ST cell line), G. Buttin, Institut Pasteur, Paris, France (myeloma cells SP2/0-Ag-14) and B. Toma, Ecole V6t6rinaire de Maisons-Alfort, France (ascitic fluids from F1PV-infected cats). The PD5 cell line was a generous gift of Philips-Duphar, Weesp, The Netherlands.

R E F E R E N C E S

BOHL, E. H. (1975). Transmissible gastroenteritis. In Diseases o/Swine, pp. 168 208. Edited by H. W. Dunne & A. D Leman. Ames: Iowa State University Press.

BONNER, W. M. & LASKEY, R. A. (1974). A film detection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels. European Journal of Biochemistry 46, 83 88.

BURNETTE, W. N. (1981). 'Western blotting': electrophoretic transfer of proteins from sodium dodecyl sulfate- polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Analytical Biochemistry 112, 195 203.

DE BLAS, A. t . , RAJNAPARKHI, M. V. & MOSIMANN, J. E. ( 1981 ). Estimation of the number of monoclonal hybridomas in a cell fusion experiment. Effect of post-fusion cell dilution on hybridoma survival. Journal oflmmunological Methods 45, 109-115.

130 H. L A U D E A N D OTHERS

DIMMOCK, N. J. (1984). Mechanisms of neutralization of animal viruses. Journal of General Virology 65, 1015-1022. FLEMING, J. O., STOHLMAN, S. A., HARMON, R. C., LAI, M. M. C., FRELINGER, J. A. & WEINER, L. P. (1983). Antigenic

relationships of murine coronaviruses: analysis using monoclonal antibodies to JHM (MHV-4) virus. Virology 131,296-307.

GARWES, D. J. & POCOCK, D. H. (1975). The polypeptide structure of transmissible gastroenteritis virus. Journal of General Virology, 29, 25-34.

GARWES, D. J., VOCOCK, D. H. & PIKE, B. V. (1976). Isolation of subviral components from transmissible gastroenteritis virus. Journal of General Virology 32, 283 294.

GARWES, D. J., LUCAS, M. H., HIGGINS, D. A., PIKE, B. V. & CARTWRIGHT, S. F. (1978/79). Antigenicity of structural components from porcine transmissible gastroenteritis virus. Veterinary Microbiology 3, 179 190.

HORZlNEK, M., LUTZ, H. & PEDERSEN, N. C. (1982). Antigenic relationships among homologous structural polypeptides of porcine, feline and canine coronaviruses. Injection and Immunity 37, 1148-1155.

HU, s., BRUSZEWSKI, J., 8OONE, Y. & SOUZA, L. (1984). Cloning and expression of the surface glycoprotein gp 195 of porcine transmissible gastroenteritis virus. In Modern Approaches to Vaccines, pp. 219 223. Edited by R. M. Chanock & R. A. Lerner. New York: Cold Spring Harbor Laboratory.

IORO, R. M. & BRATT, M. A. (1984). Neutralization of Newcastle disease virus by monoclonal antibodies to the hemagglut inin-neuraminidase glycoprotein: requirement for antibodies to four sites for complete neutralization. Journal of Virology 51, 445-451.

KtNGSFORD, L. (1984). Enhanced neutralization of La Crosse virus by binding of specific pairs of monoclonal antibodies to the G1 glycoprotein. Virology 136, 265 273.

K{SHLER, G. & MILSTEIN, C. (1975). Continuous cultures of fused cells secreting antibody of predefined specificity. Nature, London 256, 495-497.

LAUDE, H., GELFI, J. & AYNAUD, J. M. (1981). In vitro properties of low- and high-passaged strains of transmissible gastroenteritis coronavirus of swine. American Journal of Veterinary Research 42, 447-449.

MOCKETT, A. P. A., CAVANAGH, D. & BROWN, T. D. K. (1984). Monoclonal antibodies to the S1 spike and membrane proteins of avian infectious bronchitis coronavirus strain Massachuset ts M41. Journal of General Virology 65, 2281-2286.

NOWINSKI, R. C., LOSTROM, M. E., TAM, M. R., STONE, M. R. & BURNETTE, W. N. (1979). The isolation of hybrid cell lines producing monoclonal antibodies against the pl 5(E) protein of ecotropic murine leukemia viruses. Virology 93, 111-126.

PENSAERT, M., HAELTERMAN, E. O. & BURNSTEIN, T. (1970). Transmissible gastroenteritis of swine: virus-intestinal cell-interactions. Archly ,/fir die gesamte Virusforschung 31, 321-334.

READING, ¢. L. (1982). Theory and methods for immunizat ion in culture and monoclonal antibody production. Journal of Immunological Methods 53, 261-291.

ROTTIER, P..r.M., HORZINEK, M. C. & VAN DER ZEIJST, B. A. M. (198l). Viral protein synthesis in mouse hepatitis virus strain A59-infected cells: effect of tunicamycin. Journal of Virology 40, 350 357.

SHESHBERADARAN, H., CHEN, S. N. & NORRBY, E. (1983). Monoclonal antibodies against five structural components of measles virus. I. Characterization of antigenic determinants on nine strains of measles virus. Virology 128, 341 353.

SIDDELL, S., WEGE, H., BARTHEL, A. & TER MEULEN, V. (1981). Coronavirus JHM: intracellular protein synthesis. Journal of General Virology 53, 145-155.

STUDIER, F. W. (1973). Analysis of bacteriophage T7 early R N A s and protein on slab gels. Journal of Molecular Biology 79, 237 248.

STURMAN, L. S. & HOLMES, K. (1983). The molecular biology of coronaviruses. Advances in Virus Research 28, 35-112. TALBOT, P. J., SALMI, A. A., KNOBLER, R. L. & BUCHMEIER, M. J. (1984). Topographical mapping of epitopes on the

glycoproteins of murine hepatitis virus-4 (strain JHM): correlation with biological activities. Virology 132, 250-260.

TOWBIN, H., STAEHELIN, T. & GORDON, J. (1979). Electrophoretic transfer of proteins from gels to nitrocellulose sheets: procedure and some applications. Proceedings ql'the National Academy of Sciences, U.S.A. 76, 4350- 4354.

VAUTHEROT, J. F., LAPORTE, J., MADELAINE, M. F., BOBULESCO, P. & ROSETO, A. (1984). Antigenic and polypeptide structure of bovine enteric coronaviruses as defined by monoclonat antibodies. In Molecular Biology and Pathogenesis oJCoronaviruses, pp. 117-132. Edited by P. J. M. Rottier et al. New York & London: Plenum Press.

WEGE, H., D6RRmS, R. & WEGE, A. (1984). Hybridoma antibodies to the murine coronavirus JHM: characterization of epitopes on the peplomer protein (E2). Journal of General Virology 65, 1931-1942.

WOODS, R. D., CHEVILLE, N. F. & GALLAGHER, J. E. (1981). Lesions in the small intestine of newborn pigs inoculated with porcine, feline and canine coronaviruses. American Journal of Veterinary Research 42, 1163 1169.

(Received 27 May 1985)