key: cord-0004984-kjhzqhbo
authors: Ignjatovic, J.; Galli, L.
title: The S1 glycoprotein but not the N or M proteins of avian infectious bronchitis virus induces protection in vaccinated chickens
date: 1994
journal: Arch Virol
DOI: 10.1007/bf01310043
sha: 896fdda0895470cecd67ecdbcaf4b81aff54b866
doc_id: 4984
cord_uid: kjhzqhbo

The S1, N and M proteins, obtained from the nephropathogenic N1/62 strain of infectious bronchitis virus (IBV) by immunoaffinity purification with monoclonal antibodies, were used for immunization of chickens. For all three antigens multiple immunizations were necessary for induction of an antibody response. Protection of chickens vaccinated with the S1 glycoprotein against virulent challenge was demonstrated by the complete absence of virus in tracheas and kidneys of vaccinated chickens. Following four immunizations with the S1 glycoprotein 71% and 86% of chickens were protected at the level of tracheas and kidneys, respectively. Three immunizations with the S1 glycoprotein protected 70% and 10% of chickens at the level of kidney and trachea, respectively. Neither the N nor the M antigen induced protection to a virulent challenge with the nephropathogenic N1/62 strain of IBV after four immunizations. Virus neutralizing, haemagglutination inhibiting and ELISA antibodies were detected in chickens immunized with the S1 glycoprotein and inactivated N1/62 virus, however there was no correlation between the presence of any of these antibodies and protection.

Infectious bronchitis viruses (IBVs) are members of the Coronaviridae family, enveloped viruses that contain single-stranded RNA of positive polarity. IBVs induce highly contagious respiratory infections or interstitial nephritis [19] in chickens of all ages and the disease is of considerable economic importance. The disease syndrome induced by IBVs is controlled by vaccination with live attenuated vaccines, however "vaccine breaks" due to variant strains have often been reported. Variant strains may arise fiom vaccine viruses by mutation and/or recombination [22, 35] . The scope therefore exists for introduction of a subunit vaccine to replace live vaccination [29] . The antigen which induces protective immunity in IBV-infection has not been identified thus far.

IBV contains four structural proteins. The S1 and $2 glycoproteins with Mr of 92 K and 84 K, respectively form the peplomer (S), the club-shaped oligomer that protrudes from the viral envelope. In the peplomer S, the $2 glycoprotein forms the stalk and act as anchorage for the St [5, 7] . The membrane (M) glycoprotein, heterogeneously glycosylated with Mrs ranging from 27 K to 36 K is embedded in the lipid bilayer and partially (10%) exposed at the outer surface. The nucleocapsid (N) protein is a basic phosphoprotein with Mr of 52 K which is associated with viral RNA [4] . In the virion the structural proteins are present at a molar ratio of I(S):6(M): 15(N) [6] .

The S 1 glycoprotein induces virus neutralizing (VN) and haemagglutination inhibiting (HI) antibodies [8, 9, 20, 27] , enables virus entry into the cell [-9] and virus spread by cell fusion [14, 30] . The $1 glycoprotein of IBV has been considered to be the most likely inducer of protection. Inactivated IBV lacking the S1 glycoprotein did not induce protection in the trachea of immunized chickens whereas whole inactivated IBV protected chickens against virulent challenge [9] . However, chickens vaccinated with the purified peplomer S were not protected against virulent IBV challenge [8] . The peplomer S of murine hepatitis virus (MHV), another coronavirus, was shown to carry determinants for induction of protective immunity [11, 16, 21] . Mice immunized with the purified peplomer S, containing the $1 and $2 glycoproteins, were protected from mortalities following virulent challenge. Immunization with synthetic peptides homologous to two conserved regions in the $2 glycoprotein of the peplomer S of MHV protected mice from virulent challenge, indicating that the protective epitopes might be located on the $2 glycoprotein of the peplomer S [21, 32] . The $2 glycoprotein, which anchors the S1 glycoprotein of the peplomer to the viral membrane, may in IBV, carry also epitopes for crossreactive ELISA and weak VN antibodies [23, 24] . The region in the $2 glycoprotein inducing cross-reactive antibodies was found to be immunodominant and conserved in the majority of IBV strains and the $2 glycoprotein therefore has been suggested as a suitable for development of an IBV vaccine [24] . The role of the N and M antigens in induction of protective immunity is unknown. Hasony and Macnaughton [16] have shown that neither the N nor the M proteins of MHV induce protection in immunized mice. More recent studies have indicated that the N protein of IBV carries T-cell epitopes and that it may have a role to play in immunity and protection [-2, 3] .

The work reported here provides evidence that multiple immunization with purified $1 glycoprotein is able to protect chickens against challenge with virulent IBV.

The nephropathogenic N1/62 ("T") strain of IBV, egg passage 7, was used. Virus titres [expressed as meclian ciliostatic doses (CDso)] and virus infectivity in tissue samples were assayed in tracheal organ cultures [12] . To obtain partially purified virus approximately 104 CDso ofN1/62 was inoculated into the ailantoic sac of 11-day-old embryonated chicken eggs which were incubated for 48 h at 37 °C. After chilling overnight at 4 °C, allantoic fluid was collected, centrifuged at 3000g for 30min and virus pelleted by centrifugation at 48 000g for 1 h in a Beckman ultracentrifuge using SW 28 rotor. The virus pellet was resuspended in PBS pH 7.2 (1:30 of the original volume). For inactivated virus, partially purified N1/62 was incubated with 0.05~ [3-propiolactone (Sigma Chemical Co, St Louis, MO, U.S.A.) at 37 °C for 4 h and then overnight at 4 °C. Absence of infectivity was confirmed by assay in tracheal organ cultures.

Monoclonal antibodies (MAbs) designated 5, 16 and 2, directed against epitopes on the S1, N and M proteins of IBV, respectively, [18] were used to construct affinity columns. The IgG was precipitated from ascitic fluid with (NH4)zSO , (final concentration 40~ w/v). After dialysis against PBS, 10mg of IgG was coupled to 1.5g of CNBr-Sepharose 4B (Pharmacia, Uppsala, Sweden) according to the manufacturer's instructions. Unreactive groups on the gel were blocked using 200mM glycine pH 8.0 and non-bound IgG was removed by three cycles of washing with 100mM NaHCO 3 buffer pH 8.3 containing 500 mM NaC1 followed by 100 mM acetate buffer pH 4.4, containing 500mM NaC1. The MAb-CNBr Sepharose 4B (immunoadsorbent) was stored in PBS containing 0.1~o Nonidet P40 (NP40) and 0.01~ thiomersal (BDH Laboratories, Poole, England). Just before use, the immunoadsorbent was washed twice with 40 ml of PBS and twice with 40 ml of 2~o NP40 in PBS.

The method described by Mockett [26] .was used with some modifications. In brief: partially purified N 1/62 virus (usually 10 ml) was mixed for 2 h at room temperature with 10 ml of 4~o (v/v) N P40 in a Dounce homogenizer. After centrifugation at 17 000 g for 15 min, the supernatant containing soluble proteins was mixed with the MAb-immunoadsorbent overnight at 4 °C using a rotary mixer. The gel (usually 7 ml) was washed twice with 40 ml of2~o NP40 in PBS and twice with 40 ml of 0.1 ~o NP40 in PBS. The MAb-immunoadsorbent was poured into a chromatography column (Bio-Rad Laboratories, Richmond, CA, U.S.A.) and washed with 0.1~o NP40 in PBS (40 ml) and 0.1~o octylglucoside (Sigma Chemical Co, St Louis, MO, U.S.A.) in PBS (20ml). Proteins were eluted using 20ml of 100raM glycine pH 2.5 containing 0.1~ octylglucoside into tubes containing 100 tal of 2 M phosphate buffer pH 9.2. Flow rate was 30 ml/h and fractions of 1.5 ml were collected. A sample from each fraction (usually 25 gl) was used in direct ELISA and SDS-PAGE to determine the titre and purity of the eluted protein. Following elution, fractions were immediately dialyzed against PBS for 24 h. Fractions with peak ELISA activity and optical density were pooled, concentrated using Centricon units (Amicon, Beverly, MA, U.S.A.) and used immediately for immunization of chickens. The amount of S1, M and N used for immunization was estimated by measuring the optical density at 280 nm using a U-2000 spectrophotometer (Hitachi, Japan) and by titration in direct ELISA. Detection limits for the purified S1, N and M proteins in direct ELISA were 10, 5 and 20pg, respectively. The protease inhibitor phenylmethylsulfonyl fluoride (Sigma Chemical Co, MO, St Louis, MO, U.S.A.), 1 mg/ml, was included in all steps during purification of the N protein.

Fractions collected from the MAb-immunoadsorbent and partially purified inactivated N1/62 were diluted in 500 mM carbonate bicarbonate buffer pH 9.5 and 100 gl added to the wells of polystyrene microtiter plates (Disposable Products, Adelaide, Australia). After overnight incubation at 37 °C, wells were washed three times with PBS containing 0.05~o Tween 20 and ascitic fluid of MAbs 2, 5 and 16 (diluted 1:500) added. After incubation for i h at room temperature and washing, goat-anti mouse IgG-HRP conjugate (Bio-Rad Laboratories, Richmond, CA, U.S.A.) was added and incubated for 1 h. After washing, the substrate solution containing 1 mg/ml of 5-aminosalicylic acid (Merck, Darmstadt, Germany) in phosphate buffer pH 5.95, supplemented just prior to use with H20 2 to give a final concentration of 0.005~,, was added. The absorbance at 450 nm was measured in a Titertek Multiskan (ICN, Costa Mesa, CA, U.S.A.) after 1 h.

Microtiter plates were coated with 200 gg in 100 gl of partially purified N1/62 virus diluted in 500mM carbonate bicarbonate buffer pH 9.5, overnight at 37 °C. Plates were washed three times with PBS containing 0.05% Tween 20. Chicken serum, diluted 1:50 and then in log 2 dilutions was added and incubated for 1 h at room temperature. After washing, goat anti-chicken IgG-HRP conjugate (Kirkegaard & Perry Labs, Gaithersburg, MD, U.S.A.) was added and incubated for 1 h at room temperature. Substrate solution containing 5-aminosalicylic acid and H202 was added after washing and the absorbance at 450nm measured after 1 h.

SDS-PAGE was performed in a discontinuous stab gel with a 10~o acrylamide concentration in the separating get. Proteins were detected by Coomassie Brilliant Blue or silver staining (Bio-Rad Laboratories staining kit). For immunoblotting, proteins separated by SDS-PAGE were transferred electrophoretically from the gel to a nitrocellulose membrane for 1 h at 200V. After transfer, the nitrocellulose sheet was incubated in 10mM Tris-HC1, 150 mM NaC1, pH 7.4 (TBS) containing 5~o skim milk powder (blotto). The nitrocellulose sheet was either cut into strips (0.4 cm) or left intact and incubated with ascitic fluid of MAb, diluted 1:500 or 1:4000, or chicken sera diluted 1:20, in blotto. The nitrocellulose sheet or strips were then washed with TBS containing 0.05~o Tween 20 and incubated with goat-anti mouse IgG-HRP or goat-anti chicken IgG-HRP conjugates diluted in blotto. After washing with TBS containing 0.05~o Tween 20, strips were incubated for 1 min with luminol, enchanced chemiluminescence detection kit (Amersham, Buckinghamshire, U.K.) and exposed to an X-ray film (Fuji) or 4-chloro-t-napthot (Sigma Chemical Co, St Louis, MO, U.S.A.), 1 ml of 3 mg/ml solution in methanol to which 5 ml of TBS containing 0.005~ H20 2 was added.

VN test was performed in tracheal organ cultures with 100 CDso of N1/62 virus against log 2 dilutions of chicken sera (strating dilution 1:20) [12] . HI test was performed as described previously [18] .

Fourteen, day-old, specified pathogen-free white leghorn chickens were placed into positive pressure isolation units. At 2 weeks of age 10 chickens in each isolation unit were inoculated in the leg muscle with approximately 50p.g of purified $1, N or M protein, in 0.8ml, or 200 gg of inactivated N1/62 virus, in 0.8ml; all emulsified in complete Freund's adjuvant (1 : 1). Four chickens in each isolation unit were left uninoculated to ascertain that infection with live virus did not occur in the groups for the duration of the experiment. Just before challenge these noninoculated chickens were removed from the inoculated groups and placed together in a separate isolation unit as controls. Ten chickens in one additional group, housed separately, were immunized with 104 CDso of live N1/62 virus in 100 gl, intraocularly. Booster vaccinations of the purified S1, M and N proteins and inactivated N1/62 virus, in Freund's incomplete adjuvant (1:1), in the same amounts as used for the primary immunization, were given at approximately four weekly intervals. Chickens were bled just before each booster vaccination for collection of sera. Three separate experiments were performed. In the first experiment chickens were immunized twice with the purified S1, N and M antigens and live and inactivated N1/62 virus and challenged with 104 CDso of N1/62 virus, in 100 gl, intraocularly three weeks following the second immunization. In the second and third experiment chickens were immunized three and four times, respectively, and challenged three weeks following the last immunization. Five days after challenge, chickens were killed and trachea and kidney tissue collected. Tracheal scrapings, collected in 2ml of Medium 199 (Commonwealth Serum Laboratories, Melbourne, Australia) supplemented with penicillin (100 IU/ml) and streptomycin (100 gg/ml) and kidney homogenates (20~o w/v suspension in Medium 199) were assayed for the presence of infectious virus in tracheal organ cultures. In some groups there were losses during the experiments due to cannibalism. The amount of 200 gg of purified N1/62 antigen used for immunization corresponds to approximately 3 x 108 CDso of infectious virus; 50 gg of purified S l, N and M proteins contained between 10 and 20 times higher titres of the respective antigen than was contained in 200 lag of N1/62 as determined by direct ELISA.

Antisera to synthetic peptides correspondin9 to a defined region in the S1 and $2 91ycoproteins

The nucteotide sequence of the gene coding for the peplomer S (S1 and $2 glycoproteins) of N1/62 strain of IBV has been determined [Sapats S, Wright PJ and Ignjatovic J, in preparation]. From the predicted amino acid sequence of the peplomer S, two peptides homologous to the amino acid residues 188 to 203 (located within the S1 glycoprotein) and 566 to 583 (located within the $2 glycoprotein) were chemically synthesized. For the peplomer S of Massachusetts 4I strain of IBV, corresponding amino acid residues are at positions 187 to 202 and 553 to 570 [22, 23] respectively. These two peptides, referred to as $1 and $2 peptides, respectively, had the following sequence: NH2-188-Lys-Ala-Val-Ser-Ala-Ala-Gly-Val-His-Phe-Lys-Ala-Gly-Gly-Pro-Isol-203-COOH and NH2-566-Asn-Cys-Pro-Tyr-Val-Ser-Tyr-Gly-Lys-Phe-Cys-Ile-Lys-Pro-Asp-Gly-Ser-Ile-583-COOH. Synthetic peptides were conjugated to diphtheria toxoid by the glutaraldehyde procedure. At the C-terminus of the S1 peptide cysteine was included to facilitate conjugation. The N-terminus of the $2 peptide was acetylated and the peptide conjugated to diphtheria toxoid via the lysine side-chain. Peptide synthesis and conjugation was performed by Chiron Mimotopes Pty. Ltd., Clayton, Australia. The peptide: diphtheria toxoid ratio in the conjugate was 1:6.2 and 1 : 15 for the S1 and $2 peptides, respectively. Four-week-old, specified pathogenfree chickens were inoculated four times intramuscularly with peptide-diphtheria toxoid conjugates emulsified in complete (day 0) or incomplete (day 28, day 56, day 86) Freund's adjuvant. Approximately 50lag of synthetic peptide was contained in each inoculum. Chickens were bled 3 weeks following the last immunization and the specificity of anti-peptide sera, at dilution of 1:40, was determined in WB using the N1/62 strain of IBV.

Level of protection and prevalence of seropositive chickens were compared between groups by the non-parametric Fisher's exact test using Tadpole statistical package for personal computers.

The S1, N and M proteins were obtained from partially purified N1/62 virus by immunoaffinity chromatography using MAbs 5, 16 and 2, directed against the S1, N and M proteins respectively. Silver staining and WB of fractions eluted from the MAb-immunoadsorbents ( Fig. 1) showed that the preparations of the S1 and N antigens were not contaminated with each other and were largely free from the other proteins found in allantoic fluid. Preparations of the M protein were also not contaminated with either the N nor the S1 proteins however some preparations of M contained a protein with approximate Mr of 45 K (Fig. la, lane 3) . As this 45 K protein did not react with anti-N and anti-S1 MAbs and with anti-N1/62 chicken sera either in WB (Fig. lb, lane 3) or direct ELISA it is likely to be a protein from allantoic fluid. Purified N and $1 proteins reacted with their respective MAbs in WB (Fig. lb, lanes 1 and 4) and direct ELISA (results not shown) as well as with anti-N1/62 chick sera in both WB and direct ELISA (results not shown) in titres that did not differ markedly Fig. 1 . Purity of the S1, N and M protein preparations following recovery from the immunoaffinity column. Anti-S1, -N and -M monoclonal antibodies were each coupled to the CNBr-Sepharose 4B. Detergent-solubilised partially purified N1/62 virus was mixed with the MAb-immunoadsorbent and bound material eluted using 100 mM glycine pH 2.5. After dialysis against PBS, fractions were subjected to electrophoresis in a 10~o SDS-PAGE slab gel and either a silver stained or b transferred to a nitrocellulose membrane and reacted with a mixture of monoclonal antibodies directed against the S1, N and M proteins. Immune complexes were detected by incubation with anti-mouse IgG-HRP conjugate and 4-chloro-l-naphthol. a Silver-stained SDS-PAGE of the immunoaffinity purified S1, N, and M (1, 2 and 3, respectively), whole N1/62 virus before applied onto the immunoadsorbent (4) and M r markers (5) phosphorylase b (94 K), bovine serum albumin (67 K), ovalbumin (43 K) and carbonic anhydrase (30 K). b Western blotting of the immunoaffinity purified N (1), M (3) and S1 (4) proteins, whole N1/62 virus (5) and PAGE sample buffer only (2) from titres obtained with the same sera on untreated N1/62 antigen. Immunoaffinity purified M glycoprotein did not react with anti-M MAb in WB (Fig. lb,  lane 3) , however it did react with anti-M MAb and anti-N1/62 chick sera in direct ELISA when it was used as a coating antigen (results not shown). By direct ELISA the recovered S1, N and M proteins were not contaminated with each other (results not shown). Purification of the N protein had to be performed in the presence of the proteinase inhibitor phenylmethylsulfonyl fluoride otherwise the complete degradation of the N protein occurred.

Preparations of the N protein recovered from the MAb 16-immunoadsorbent contained usually three protein bands with approximate Mr of 52 K, 43 K and 40 K, the latter two being more readily visible in WB than by silver staining (Fig. 1) . The protein eluted from the MAb 5-immunoadsorbent, the MAb which is was considered to be directed against the St glycoprotein, had only one protein band on gels with approximate Mr of between 94 K and 97 K (Figs. 1 and 2) and this protein stained well with both silver (Fig. la) and Coomassie (results not shown) stains. In WB with anti-N1/62 chick sera the immunoaffinity purified S1 contained only one protein band with approximate Mr of 97K (Fig. 2, lane 7) . The N1/62 virus contained two bands that migrated close to each other, indicative of the presence of S1 and $2 glycoproteins. Immunoaffinity purified S1 reacted in WB only with antisera to the S1 peptide (Fig. 2 , lane 3) but not with antisera to the $2 peptide (Fig. 2, lane 5) . Since the S1 and $2 peptides correspond to amino acid residues 188 to 203 and 566 to 583 in the S 1 and $2 coding domains of the peplomer S respectively, it was therefore presumed that the protein recovered from the MAb-5-immunoadsorbent was the $1 glycoprotein of IBV. Fig. 2 . Electrophoretic mobility and cross-reaction with anti-peptide sera of the S1 protein recovered by immunoafflnity purification. Partially purified N1/62 virus (2, 4 and 6), the protein recovered from the MAb 5-immunoadsorbent (3, 5 and 7) and biotinytated molecular weight markers (I) were subjected to electrophoresis in a 10~o SDS-PAGE slab gel, transferred to a nitrocellulose membrane and reacted with chick anti-Sl-peptide (2 and 3), anti-S2-peptide (4 and 5) and anti-N1/62 sera (6 and 7). After incubation with a mixture of anti-chick IgG-HRP conjugate and avidin, luminol was added followed by autoradiography. Biotinylated molecular weight markers were: rabbit skeletal muscle myosin (200 K), E. coli 13-galactosidase (l16K), rabbit muscle phosphorylase (97K) and bovine serum albumin (66 K)

Ammonium isothiocyanate was used in two studies as elution buffer [11, 26] to recover intact peplomer S. For that reason, 3 M ammonium isothiocyanate in PBS, pH 7.2 was also used for elution of proteins from the MAb-immunoadsorbents, instead of glycine, while all the other conditions were the same. The S1 glycoprotein, and not the peplomer S, was only recovered from the MAb 5-immunoadsorbent with ammonium isothiocyanate (results not shown). The amount and antigenicity of all three proteins, the SI, N and M, recovered with ammonium isothiocyanate was comparable with the amount and antigenicity of the $1, N and M obtained with glycine as determined by their reactions with respective MAbs and polyclonal chicken sera in direct ELISA (results not shown). Elution with glycine was method of choice for this study as it allowed the repeated used of the MAb-immunoadsorbents which was not possible using ammonium isothiocyanate.

In there separate experiments chickens were immunized with 50 ~tg of immunoaffinity purified $1, N and M proteins and inactivated and live N1/62 virus at four weekly intervals. Three weeks after two, three and four immunizations chickens were challenged with live virus to assess the degree of protection afforded by the individual purified viral antigens (Table 1) . Challenge virus was the nephropathogenic N1/62 strain of IBV. Since nephropathogenic IBVs replicate efficiently in both trachea and kidney, protection of both tissues was assessed by virus isolation. As shown in Table 1 , immunization with inactivated N1/62 was less effective than immunization with live N1/62. In chickens immunized twice with inactivated N1/62 virus there was some degree of protection for kidney (33~o) although this was not statistically significant (P ~> 0.05). Three immunizations with inactivated N1/62 virus protected 80~ and 60~ of chickens at the level of kidneys (P ~< 0.001) and tracheas (P ~< 0.01) respectively and this level of protection did not increase with further immunizations (four immunizations, Table 1 ). Kidneys were consistently better protected, 80~o versus 60~o and 75~ versus 65~ following three and four immunizations with inactivated N1/62, respectively although these differences were not statistically significant (P ~> 0.05). Vaccination with live virus totally protected kidney and trachea of all chickens (P ~< 0.001) ( Table 1) .

The $1 glycoprotein induced protection in vaccinated chickens to a virulent challenge only after multiple immunizations (Table 1) . At least three immunizations with the S1 were necessary to induce a significant level of protection (P ~< 0.01), at which time 70~ of chickens were protected at the level of kidneys. At this time there was a significant difference (P ~< 0.01) between protection induced by vaccination with $1 in kidney and tracheal tissues of the same birds as only 10~o of chickens were protected at the level of trachea. The difference in protection between tracheal and kidney tissue was less apparent and not statistically significant (P ~> 0.05) after four immunizations with the S1 glycopro- (Table 1) . Although the N protein was the most immunogenic and induced high titres of ELISA antibodies, it did not provide protection against virulent challenge and 100~ of chickens had virus present in both trachea and kidneys 5 days after challenge with virulent N1/62.

Antibody responses following each immunization were determined by VN, ELISA, WB and HI test. Results obtained in experiment in which chickens were immunized four times with the purified S1, N and M antigens and live and inactivated N1/62 virus, are shown in Table 2 and Fig.3 . N1/62 9/9 9/9 10/10 10/10 NT d NT NT NT N1/62 inactivated 0/5 0/5 2/10 8/10"** 6/10"*'* 10/10"** 8/8*** 8/8*** N 0/5 0/5 0/10 10/10"** 0/10 10/10"** 0/7 7/7*** M 0/5 0/5 0/10 0/10 0/10 0/10 0/8 6/8** S1 0/5 0/5 0/10 0/10 1/10 10/10"** 5/7** 7/7*** Controls e 0/12 0/12 0/12 0/12 0/12 0/12 0/t0 0/10 "Antibodies measured four weeks following the first (1°), second (2°), third (3 °) and fourth (4 °) immunization bTwo-week-old chickens were immunized at four weekly intervals with live (104 CDs0 ) and inactivated N1/62 virus (200 lag), and the N, M and $1 antigens (50 lag) obtained from N1/62 virus by immunoatfinity purification. Live virus given intraocularly (100 lal), other antigens intramuscularly in Freund's adjuvant (1:1, 0.8 ml/chick) ~VN antibodies against N1/62 strain determined in tracheal organ cultures; for ELISA microtiter plates were coated with purified N 1/62 virus and sera tested diluted 1:50

aNT Not tested ° Uninoculated controls kept in contact with groups inoculated with inactivated N1/62 virus and the N, M and S1 proteins ***'**'*Values were statistically different (Fisher's exact test); ***P ~< 0.001, **P ~ 0.01 when compared with the control group; * P ~< 0.05 when compared with the S1 samples positive for VN antibodies in the group with three immunizations; All other comparisons for the N, M, S1 and inactivated N1/62 groups were not statistically significant (P/> 0.05) Fig. 3 . Specificity of antibodies in chickens immunized four times, at four weekly intervals, with a inactivated N 1/62 strain of IBV, b S 1 glycoprotein, e N protein and d M glycoprotein obtained from N1/62 strain by immunoaffinity purification; e noninoculated controls. Purified N1/62 virus was subjected to electrophoresis in a 10~ SDS-PAGE slab gel, transferred to nitrocellulose and strips incubated with representative chicken sera (1:20) . After incubation with anti-chick IgG-HRP conjugate immune complexes were detected by incubation with luminol, followed by autoradiography Two immunizations with inactivated N1/62 induced ELISA antibodies in the majority of chickens (8/10) (P ~< 0.001) however only 2/10 chickens (P >~ 0.05) had VN antibodies (Table 2) . Following three immunizations with inactivated virus, VN, ELISA and HI antibodies were detected in the majority of chickens (P ~< 0.01 and P ~<0.001 for VN and ELISA antibodies respectively) (results for HI antibodies not shown).

Of the three purified proteins used for immunization, the N protein was the most immunogenic. The ELISA antibodies were detected in all chickens immunized with the N antigen four weeks following secondary immunization (P ~< 0.001) ( Table 2) . The M glycoprotein, given in the same amount as the N protein, was the least immunogenic. ELISA antibodies were detected in significant number of chickens (P ~< 0.01) only after four immunizations. Neither the N nor the M antigens induced VN (Table 2) or HI antibodies (result not shown). The purified S1 glycoprotein induced ELISA antibodies in all chickens after three immunizations (P ~< 0.001) ( Table 2) , however only one chick had VN antibodies at that time (P >~0.05). Following the fourth immunization with $1, the majority of chickens (5/7) (P ~< 0.01) developed VN antibodies ( Table 2 ) and four of those had also HI antibodies (results not shown). The geometric mean titre of VN and HI antibodies in the group vaccinated with the S1 (1/40 and 1/16, respectively) was lower in comparison to the geometric mean VN and HI titres in chickens vaccinated with inactivated NI/62 (1/256 and 1/64 respectively).

As shown in Fig. 3 chickens immunized four times with the purified S1, N and M proteins developed only antibodies directed against the S1, N and M proteins, respectively, confirming that the immunoaffinity purified antigens were free from contamination with each other. Antibodies elicited by four immunizations with inactivated N1/62 were directed to the $1, $2, N and M proteins (Fig. 3a) . Non-inoculated chickens kept in contact with chickens inoculated with either the S1, N and M proteins, or inactivated N1/62 virus, remained antibody negative throughout the immunization experiments (Table 2, Fig. 3e) . 

As s h o w n in T a b l e s 1 a n d 2, c h i c k e n s i m m u n i z e d twice with the S1 a n t i g e n w e r e all a n t i b o d y n e g a t i v e a n d 

aChickens inoculated at four weekly intervals with inactivated N1/62 (200 lag in 0.8ml) strain of IBV or S1 glycoprotein (50 lag in 0.8 ml) obtained from N1/62 by immunoaffinity purification bChickens were immunized three or four times, respectively and challenged three weeks after the last immunization with virulent N1/62 c Chickens were from experiments detailed in Tables 1 and 2 dChallenge virus in trachea and kidney five days following challenge. Absence of challenge virus was taken as indication of protection eVNAb Virus neutralizing antibodies against N1/62 determined in tracheal organ cultures at serum dilution of 1:20. All sera were positive for ELISA antibodies f +, -Positive, negative g NA Not applicable (2/35) or trachea (4/35) or both (3/35) and were not protected. Only in 11/35 chickens did the presence of VN antibodies correlated with the absence of challenge virus in trachea and kidneys. Also there was no correlation between the presence of ELISA antibodies and protection as all chickens in Table 3 were positive for ELISA antibodies.

Our results showed that the S 1 glycoprotein of IBV induces protective immunity against IBV-infection whereas neither the N protein nor the M glycoprotein induced any level of protection. The immunoattiinity purified protein that induced protection in this study was a single protein with Mr of between 94 K and 97 K which induced VN and HI antibodies, properties attributed to the S1 glycoprotein of IBV [8, 27] . The result confirmed the previous finding of Cavanagh and co-workers who have indirectly indicated that the S 1 glycoprotein might be the inducer of protective immunity in IBV-infection [9] . In MHV the $2 glycoprotein has been shown to carry epitopes which induce protective immunity [21, 32] although the same studies did not evaluate whether the S1 glycoprotein also carries such determinants. The possible involvement of the $2 glycoprotein of IBV in protection was not evaluated in this study because of the lack of MAbs directed against this glycoprotein. The immunoatfinity purified S1 was able to induce protection in 70 to 80% of vaccinated chickens, protection comparable to that induced by inactivated N1/62 virus having all four structural antigens. Protection induced by the purified S1 glycoprotein was complete and prevented replication of the challenge virus in trachea and kidney tissues which are the target organs for IBV replication. In studies performed with MHV protection from mortalities, but not prevention of virus replication, which still occurred, was taken as a measure of protection [11, 16, 21, 32] . In the development of an effective vaccine for control of IBV-infections in commercial poultry it is however necessary to completely prevent replication of a challenge virus in the trachea of chickens as IBV infection predisposes the trachea to invasion with opportunistic organisms such as E. coli resulting in debilitating respiratory disease with significant economic losses [19] .

Although the N protein did not have a role in prevention of virus replication in trachea and kidneys it was the most immunogenic protein and induced significant titres of antibodies in all vaccinated chickens. The N proteins of MHV and feline infectious peritonitis virus were also shown not to be involved in protection [16, 34] although in one study the recombinantly expressed N protein of MHV protected mice against virulent challenge [36] . It has been previously suggested that the N protein of IBV might have a role in immunity and protection, although the direct evidence for its involvement in protection has not been provided [3] .

The M glycoprotein was the least immunogenic and did not induce protection in vaccinated chickens. It could not be discounted however that the J. Ignjatovic and L. Gatli purification procedure had adversely affected the immunogenicity of the M glycoprotein. At least four immunizations with the purified M glycoprotein were necessary to induce a detectable antibody response. Contrary results have been obtained thus far regarding protection induced by the M glycoprotein. The M glycoprotein of MHV has been indirectly implicated in protection [15] and disease progression [25] however purified and recombinantly expressed M glycoprotein did not induce protection in vaccinated mice [16, 36] . On the other hand, the recombinantly expressed M glycoprotein of feline infectious peritonitis virus induced some level of protection in vaccinated kittens [34] .

It is not clear to what extent the purification procedure may have affected the immunogenicity of the S1, N and M proteins in chickens. The antigenic properties of the purified S1 and N proteins were comparable with properties of the same antigens in the purified untreated virus. Both purified proteins retained the ability to react with their respective MAbs and polyclonal sera in both native and denaturated state in a similar fashion as did the S1 and N proteins in untreated virus. Changes however were detected in the properties of the purified M glycoprotein as it did not react in WB with MAb although it reacted with the same MAb in ELISA. Elution at acid pH with glycine, which was used in this study, did not alter the antigenicity of either the $1, N or M in comparison to elution at neutral pH with ammonium isothiocynate which was used in studies performed by others [11, 26] . The N and M antigens were the most and the least immunogenic, inducing ELISA antibodies after two and four immunizations, respectively. At least three immunizations with the S1 glycoprotein were necessary to induce ELISA antibodies however VN antibodies were not present at that time. In studies of Hasony and Macnaughton [16] with MHV, the peplomer S was more immunogenic than either the N or M proteins, whereas in studies of Mockett [26] the M protein of IBV, purified using ammonium isothiocyanate, was more immunogenic in rabbits than the peplomer S. The S1 glycoprotein obtained by immunoaffinity purification was less effective in inducing VN and HI antibodies than the S1 glycoprotein present in inactivated virus although a greater amount was used for immunization than that present in inactivated N1/62 virus. Three immunizations with the inactivated N1/62 induced VN and HI antibodies in 60~ of chickens whereas only 10~ of chickens immunized three times with the purified S1 glycoprotein had VN antibodies. Possibly the purification in the presence of detergents has affected the immunogenicity of the $1. Another more likely reason for the poor immunogenicity of the purified St might be the lack of association with the $2 glycoprotein, a conformation which enables a correct presentation'of antigenic sites in the S1 glycoprotein to the immune system and which is present in inactivated virus [8, 9] . The purified peplomer S of MHV, containing both the $1 and $2 glycoproteins, induced a considerable level of VN antibodies in all mice following two immunizations [11] . Only the $1 glycoprotein of the peplomer S was obtained using MAb 5-immunoadsorbent and this was the case regardless whether elution was carried out at acid or neutral pH. Contrary to that, the peplomer S of IBV and MHV containing the $1 and $2 glycoproteins, were obtained using elution at neutral pH with ammonium isothiocyanate [11, 26] . The reason for this discrepancy is not clear.

Immunization of chickens with inactivated N1/62 virus and purified $1 antigen was less effective than immunization with live virus in spite of using a considerably larger amount of antigen for immunization. Inactivated IBV vaccines are poor inducers of protection if administered to unprimed chickens and live attenuated IBV have been used for control of IBV-infections [19] . The protective antigen of IBV must therefore be delivered via a replicating agent if currently used live vaccines are to be replaced by subunit vaccines.

In this study protection induced by the purified S1 glycoprotein and inactivated IBV did not correlate with the presence of circulating antibodies. Chickens were protected in the absence of VN and HI antibodies and conversely some chickens with circulating VN, HI and ELISA antibodies were not protected. Only in one third of chickens did the presence of VN antibodies correlate with protection. The mechanism of protection in coronavirus infection is not clear. Antibodies were implicated in protection in a number of coronavirus induced diseases. Suckling mice and piglets have been protected from development of acute disease if nursed by immunized dams [13, 17, 28] and antibody levels correlated with protection in mice immunized with the peplomer S [11, 16, 32] . Abrogation of B cells in chickens leads to more severe and longer lasting infection [10] suggestive of the role of antibodies in protection. In other studies, however, antibodies such as VN antibodies, did not correlate with protection [9, 2t] . Cell mediated immunity is generally believed to be of importance in protection against coronavirus infections. Adoptive transfer of CD4 + and CD8 + cells and virus specific T-cells prevented development of acute disease in mice [31, 37] . The N antigen of IBV, as well as the S and M antigens of murine hepatitis virus, were shown to carry T-cell epitopes [2, 25] suggestive of the involvement of cell mediated immunity in coronavirus infections.

There was an indication that the protection of tracheal and kidney tissues in IBV-infection, may differ. Intramuscular immunization with the S1 glycoprotein, and inactivated N1/62 to a lasser degree, provided better protection for kidneys than tracheas. Whether this difference in protection resulted from the route of antigen delivery or reflects separate pathways by which trachea and kidney tissue become infected, remains to be determined. Differential protection of target tissues has also been observed in some instances with MHV [1] .

Virus strain specificity of challenge immunity to coronavirus

Localization of a T-cell epitope within the nucleocapsid protein of avian coronavirus

Induction of anti-viral immune responses by immunization with recombinant-DNA encoded avian coronavirus nucleocapsid protein

Sequences of the nucleocapsid genes from two strains of avian infectious bronchitis virus

Coronavirus IBV glycopolypeptides: size of their polypeptide moieties and nature of their oligosaccharides

Coronavirus IBV: further evidence that the surface projections are associated with two glycopolypeptides

Coronavirus IBV: structural characterization of the spike protein

Induction of humoral neutralising and haemagglutination-inhibiting antibody by the spike protein of avian infectious bronchitis virus

Coronavirus IBV: virus retaining spike glycopolypeptide $2 but not S 1 is unable to induce virus-neutralizing or haemagglutination-inhibiting antibody, or induce chicken tracheal protection

Effect of in ovo bursectomy on the course of an infectious bronchitis virus infection in line C White Leghorn chickens

Protection from lethal coronavirus infection by affinitypurified spike glycoprotein ofmurine hepatitis virus, strain A59

Taxonomic studies on strains of avian infectious bronchitis virus using neutralisation tests in tracheal organ cultures

Epitope specificity of protective lactogenic immunity against swine transmissible gastroenteritis virus

Stably expressed FIPV peplomer protein induces cell fusion and elicits neutralizing antibodies in mice

Monoclonal antibodies to the matrix (El) glycoprotein of mouse hepatitis virus protect mice from encephalitis

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

PassiVely aquired challenge immunity to enterotropic coronavirus in mice

Monoclonal antibodies to three structural proteins of avian infectious bronchitis virus: characterization of epitopes and antigenic differentiation of Australian strains

Infectious bronchitis

Diseases of poultry

Antigenic domains on the peplomer protein of avian infectious bronchitis virus: correlation with biological functions

Immunogenic peptide comprising a mouse hepatitis virus A59 B-cell epitope and an influenza virus T-cell epitope protects against lethal infection

Van Der Zeijst BAM (t990) Sequence evidence of RNA recombination in field isolates of avian coronavirus infectious bronchitis virus

Analysis of an immunodominant region of infectious bronchitis virus

Antigenicity of the peplomer protein of infectious bronchitis virus

Immune response to a murine coronavirus: identification of a homing receptor-negative CD4 ÷ T cell subset that responds to viral glycoproteins

Envelope proteins of avian infectious bronchitis virus: purification and biological properties

Monoclonal antibodies to the S1 spike and membrane proteins of avian infectious bronchitis coronavirus strain Massachusetts M41

Maternal antibody-modulated MHV-JHM infection in C57BL/6 and Balb/c mice

Progress towards a coronavirus recombinant DNA vaccine

Proteolytic cleavage of the E2 glycoprotein of murine coronavirus: activation of cell-fusing activity of virions by trypsin and separation of two different 90 K cleavage fragments

T-cell-mediated clearance of mouse hepatitis virus strain JHM from the central nervous system

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

Early death after feline infectious peritonitis virus challenge due to recombinant vaccinia virus immunization

Primary structure of the membrane and nucleocapsid protein genes of feline infectious peritonitis virus and immunogenicity of recombinant vaccinia viruses in kittens

Evidence of natural recombination within the S1 gene of infectious bronchitis virus

Mouse hepatitis virus spike and nucleocapsid proteins expressed by adenovirus vectors protect mice against a lethal infection

Protection of mice from a lethal coronavirus infection in the central nervous system by adoptive transfer of virus-specific T cell clones

The authors thank Fiona Ashton for excellent assistance. Special thanks are due to Ms. Sue Federizzi and Ms. Carmel Rogan for their care of chickens used in this study. This work was supported in part by the Australian Chicken Meat and Egg Industry Research and Development Councils. 

Received January 6, 1994