key: cord-0004955-1bzc0j69
authors: Wang, L.; Parr, R. L.; King, D. J.; Collisson, E. W.
title: A highly conserved epitope on the spike protein of infectious bronchitis virus
date: 1995
journal: Arch Virol
DOI: 10.1007/bf01323240
sha: 6a7739601d17ef1dcc1d0918bbca7db23a13d6bd
doc_id: 4955
cord_uid: 1bzc0j69

The predicted amino acid sequence and secondary structures of S1 of the spike protein (S) of infectious bronchitis viral (IBV) strains from Europe, the U.S.A., and Japan were compared. An antigenic determinant that was highly conserved in both the primary amino acid sequence and secondary structure of all strains was identified between amino acid positions 240 to 255. A synthesized peptide corresponding to this region was found to react with all polyclonal antisera examined from various IBV strains and with one monoclonal antibody (MAb), 9B1B6, out of nine known to react with the S of Gray. The specificity of the interaction with MAb 9B1B6 was confirmed by competitive ELISA using bound and unbound peptide. Interestingly, the previously described epitope for 9B1B6 had been characterized as cross-reactive with several strains of IBV, as conformation-independent but reacting only with intact whole S, and as associated with the functional integrity of other epitopes, including neutralizing epitopes on the S protein. The apparent critical functional and structural nature of this highly immunogenic determinant suggests a potential contribution in developing protective, cross-reactive subunit vaccines to IBV.

Infectious bronchitis virus (IBV), a highly contagious respiratory pathogen of poultry, represents a serotypically diverse group of viruses [10] . The potential of IBV for genetic variation undoubtedly plays an important role in the occurence of antigenicatly distinct, virulent viruses that are often responsible for outbreaks of bronchitis in vaccinated flocks. It has also been shown that this remarkable evolution of IBV depends on both point mutations and recombination events and that the latter have commonly involved Mass vaccine-like strains [7, 20, 36, 37] . Although recombination between genes of murine hepatitis viruses had been experimentally produced under laboratory conditions [24, 27, 28] , IBV was the first coronavirus in which recombination has been suggested to generate new naturally occurring strains [20, 25, 36, 37] .

The IBV particle has three major and one minor structural proteins [2, 10] . Antibodies are known to be readily induced to the structural proteins; the spike (S), membrane (M) and nucleocapsid (N) [32] . Whereas antibodies to N are strongly cross-reactive among strains, epitopes on the S and M proteins have been shown to be more variable [32] . The S protein contains determinants that dictate serotype and that induce neutralizing antibody and protection [4, 5, 18, 19, 29, 31] . This protein is synthesized as a large protein that is post-translationally processed to the S 1 and $2 subunits. The primary neutralization epitopes of S are found on the outer globular-like S 1, whereas minor neutralizing epitopes are found on the $2 which anchors the complex into the viral envelope, that is the bilipid membrane [5, 18] . A hypervariable region (HVR) within the S 1 is probably associated with serotype determination of a strain [6, 10, 26, 30] .

Monoclonal antibodies (MAb) specific for S have been used to identify both conformation-dependent epitopes, including S 1 neutralizing epitopes, and conformation-independent epitopes [19, 29, 31] . However, certain non-neutralizing monoclonal antibodies react with S determinants that are both stable under denaturing conditions and conserved on various serologically distinct strains. The cross-reactive nature of these amino acid sequences could make them valuable if they should contribute or enhance protective immunity.

The purpose of this study was to characterize the amino acid sequence of the S 1 in order to identify highly conserved, potentially immunogenic regions. The antigenic nature of a highly conserved determinant lying downstream of the HVR was examined using serotype-specific polyclonal antibody and S-specific MAb.

The S 1 predicted amino acid sequences were determined from the previously determined nucleotide sequences. The secondary structures were predicted and compared by computer analysis based on the algorithms of Chou and Fasman [9] . The antigenicity indices were determined from the values of hydrophilicity [17] , flexibility [23] , surface probability [13] , and glycosylation (GCG system 7.3).

The chicken polyclonal antisera were collected after a single, intranasal/ocular inoculation [16] . The preparation ofmonoctonal antibodies specific for the spike protein of the IBV Gray strain and both positive and negative mouse ascites have been previously described [31] .

Modifications of the western blot procedure described by Parr and Collisson [31] were used for immunoblot assays. Briefly, immobilon-P membranes (Millipore Corp, Bedford, MA) were soaked in methanol then TBS (Tris buffered saline solution, pH 7.5) before placing on the dot blot or slot blot apparatus (BioRad Laboratories, Richmond, CA). The Gray virus prepared as described [32] or peptide (synthesized and purified by Biosynthesis, Inc. Denton, TX) were added to the wells at concentrations of 1 gg and 5 ng per milliliter, respectively. Two hundred microliters of the virus or peptide were added to each well and filtered through the membrane by vacuum. The membranes were treated with diluent containing 3% bovine serum albumin (Kirkegard & Perry Labs, Gaithersburg, MA) and reacted with either polyclonal or monoclonal primary antibody, or normal serum diluted 1:10 in diluent. The membranes were removed from the apparatus, washed with TTBS (Tween-20 and TBS) before blocking again with diluent and adding secondary antibody consisting of either alkaline phosphatase-conjugated goat anti-chicken or anti-mouse (heavy and light chains) antibodies. The substrate, NBT/BCIP (Kirkegaard and Perry Laboratories Inc., Gaithersburg, MA), was used to detect antigen-antibody complexes.

Similarly prepared antigens were used to coat wells overnight at 4 °C in 96-wetl plates with concentrations used above in immunoblot assays for virus and, unless otherwise specified, for peptide [31] . After blocking for nonspecific binding with diluent (Kirkegaard and Perry Laboratories, Inc., Gaithersburg, MA) and incubating at 37 °C, specific viral antibodies or normal antibody controls were added (150 ng/ml and 100 gl/well) for one additional hour at 37 °C. The plates were then washed five times with PBS containing 0.05% Tween 20. The secondary antibodies, either goat anti-mouse-HRPO labeled antibody or goat anti-chicken-HRPO labeled antibody (0.5 mg/mt), were added at a 1:500 dilution for one hour at 37 °C. After thorough washing, the antigen-antibody complexes were detected with a mixture of the substrate ABTS (Kirkegaard and Perry laboratories, Inc., Gaithersburg, MA) and hydrogen peroxide. The color intensity was read with an ELISA reader at OD 490 (Dynatech Laboratories Incorp, Alexandria, VA; Manual, Kirkegard and Perry laboratories, Inc., Gaithersburg, MA). Those ratios of sample to background greater than or equal to 2 were considered positive [31] .

The competitive binding assay was similar to the ELISA, except that the concentration of bound peptide was varied and the primary antibody was reacted with free or unbound peptide for 1 h at room temperature, before adding to the 96-well microtiter plate containing bound peptide [31] . The background OD in the absence of antibody were substracted from all values before determining the percentage binding with or without unbound peptide inhibitor. The maximum or 100% binding was the OD minus background in the absence of unbound peptide and the percent binding was calculated from the OD resulting from maximum binding.

The origins of the S1 gene of 24 strains of IBV isolated from Europe, Japan, and the United States that have been sequenced in our laboratory and other laboratories throughout the world (Genbank, NCBI, Bethesda, MD) are shown in Table 1 . All strains represent distinct isolates. Mass42, also included in the following but not shown in the table, is a laboratory derivative of Beaudette [30] . The S 1 gene of these sequences were compared and analyzed in order to identify secondary structures and determined putative conserved features of the protein.

Using Ark99 as the reference strain, the conserved and variable regions, derived Sutou et al. [33] Japan "The years during which these strains were isolated are shown in parenthesis from the predicted amino acid sequences, were superimposed onto a schematic of the computer generated secondary structure that included a-helices, 13-pleated sheets, random coils and 13-turn regions (Fig. 1 ). Based on the amino acid comparisons of these IBV strains, the S l protein was divided into highly conserved (C1 to C4), with more than 80% identity among groups, and variable (V1 to V3), tolerating more than 20% variation. The HVR, also referred to as V1, extended from residue 50 to 150 and demonstrated less than 50% identity among groups. The V1 could also be further subdivided into 2 hypervariable domains (HV 1 and HV2), separated by several internal conserved amino acids residues. The overall conservation of the secondary structure is illustrated in Fig. 2 in which the structures of S 1 from 14 strains are compared. A highly conserved 13turn occupying a prominent position was predicted in C2 of the S1 from all 14 strains (Fig. 3) . A region which included this t3 -turn between two [3-pleated sheets was also predicted to have several highly antigenic sites.

Group-specific peptide on the S1

The amino acid sequence of residues 201 to 300 from the 24 strains was compared in detail in order to determine the degree of conservation of this common secondary structure lying within C2 (Fig. 4a) . The primary amino acid sequence of a site that incorporated the conserved [3-turn was found to be nearly totally conserved. A region within this site from 240 to 255, GlnTyrAsnTyrGlyAsnPhe-SerAspGlyPheTryProPheThrAsn (underscored in Fig. 4a ) was also found to have a high antigenicity index (Fig. 4b) . The only variations among the 24 strains were at position 248 and 251 where the serine was substituted by threonine and phenylalanine was substituted by leucine, respectively, in the Dutch strains, and at 255 where the threonine was substituted for by a isoleucine in the Mass 41 strain.

The conservation of the primary and secondary structures suggested that this site might serve a critical role in either maintaining conformational integrity or biological function. In addition, the high antigenicity index suggested that the peptide could be valuable in inducing broad based immunity to IBV. The P240 peptide corresponding to the above amino acid sequence was synthesized in order to evaluate actual antigenicity of the site. The peptide was used to determine potential interactions with polyclonal antibody specific for eight strains of IBV (Table 2 ). This peptide reacted in a slot blot assay with primary chick antibody specific for Mass41, Ho1152, JMK, SE17, Ark99, F188, Gray and Holte strains of IBV as determined by viral neutralization assays ( [16] , unpubl, obs., D. J. King). Antibody specific for ILTV (infectious laryngotracheitis virus) or NDV (Newcastle disease virus) did not react with the peptide nor did uninoculated control chicken sera. The polyclonal antisera specific for IBV strains, included PP14, also reacted with the P240 in an ELISA assay. Therefore, the sequence was not only conserved among strains in its primary sequence and predicted secondary structure but also in its functional immunogenicity. 

The binding properties of several MAbs generated to the Gray S had been previously examined [31] . The relative binding avidities, neutralization potentials and competitive interactions of these MAbs, and therefore, their corresponding epitopes, had been well characterized. The interactions of nine of these MAbs with P240 were examined. In a slot blot assay, only the MAb 9B1B6 consistently bound to the peptide (Fig. 5 ). The reaction with 9B1B6 produced a more intense "Reaction with 100 gg of peptide UAmount of peptide in gg that was added to each well band than the polyclonal anti-Gray antibody. The specific interaction with P240 was confirmed in an ELISA and a dot blot assay (Table 3 ). With the exception of a weak reaction with 5C5B5 in the dot blot assay, the peptide reacted with only 9B1B6 and the polyclonal mouse antibody and the intensity of the observed bands decreased with decreasing concentrations of 9B 1B6. Therefore, the peptide generated to this highly conserved region appeared to correspond to the epitope, or a part of the epitope, that interacted with the MAB 9B 1B6. Consistent with the above results, 9B1B6 had been shown to identify a conformation-independent and group-specific epitope [31] . In order to confirm the specific interaction of P240 with 9B1B6, unbound peptide was used to competitively inhibit the reaction of bound peptide with 9B 1 B6. Varying concentrations of 9B 1B6, maximum or 100% binding was determined in the absence of unbound peptide. The unbound peptide did inhibit the interaction between the MAb and the bound P240, and the inhibition occurred in a dose-dependent manner (Fig. 6) . The inhibition profile was very similar with the four concentrations of bound peptide shown. Lower concentrations of bound peptide used were below the level of sensitivity of the assay. Therefore, the peptide corresponding to a highly conserved region, consisting mostly of a B-turn common to most, if not all, S 1 proteins, specifically reacted with 9B 1B 16, a MAb that defined a conserved and conformation-independent epitope.

Because S, especially the S1, has been implicated in the binding of the IBV particle to the host cell membrane, it is a logical target antigen for developing p. [37] . The apparent continuous evolution of IBV, implicated with emerging virulent strains distinct from vaccine strains, make the generation of relevant protective vaccines difficult. An ideal vaccine for a virus that is as variable as IBV would include highly conserved antigenic determinants that could contribute to the induction of responses that inhibit viral replication or the spread of virus. Antigenic determinants not associated with neutralization of virus could contribute to the control of viral infection; for example, through the induction of cellular immunity. Although epitopes that stimulate IBV neutralizing antibody appear to lie in highly variable regions, the effects of conserved antigenic regions would be universal. Computer generated comparisons of the amino acid sequences of 24 IBV strains, derived from isolations made throughout the world, identified in S 1 four conserved regions separated by three variable regions, the first of which is the HVR. A region of -turns, occupying a prominent position about 100 residues downstream of the HVR (V1) was conserved among all strains examined. This region, also conserved in amino acid sequence, was predicted to be a strongly antigenic region. Interestingly, two residues in this region were also found to be conserved in other coronavirus strains, such as MHV, bovine coronavirus, transmissible gastroenteritis virus, feline infectious peritonitis virus and human coronavirus [35] . At about position 500, a second conserved, highly antigenic determinant of 13 residues was identified in which five amino acids were common in a similar region throughout the coronavirus genus. These apparently functionally critical determinants on the protein could be useful targets for the induction of immunity that might contribute to protection.

Conserved epitope on the IBV S

The synthetic P240, corresponding to only 16 residues of 540 amino acids in the S 1, reacted only with the one out of nine MAb examined. The P240 region was found to serve as the epitope of a well characterized MAb that had been generated against the Gray S [31] . Although, according to the present studies, the epitope was obviously within the S 1 subunit, in western blot assays this MAb had been found to be conformation-independent but only reacted with the whole S and not with either subunit alone [31] . This had suggested that either the epitope lay between the two subunits and was destroyed following post-translational cleavage, or the maintenance of the intact S1 and $2 correlated with the integrity of the epitope. Another unusual property of gB 1 B6 was that it had competitively inhibited the binding of most of the S-specific MAbs, where such broad competition was not common among the other MAbs [31] . The differences in competition could not be related to differences in binding constants of the MAb. It is difficult to explain the interactions of the 9B1B6 epitope with other determinants on S and the inability to recognize the S 1 subunit alone except that it would appear to occupy a critical position in the S l protein and functionally contribute to the integrity of the whole protein.

It would be of interest to determine if there is any association of this conserved region in C2 with cell attachment. Cleavage in the S could result in alterations in protein conformation that may be biologically important, for example, for viral entry into the host cell. In spite of serologic and pathogenic variations, the receptor-binding epitopes of IBV would be expected to be relatively conserved because most of the IBV strains do infect the respiratory system, as well as cultured chicken embryo kidney cells. In fact, we have identified a tissue receptor for the Gray strain from lung and kidney cells with a single molecular weight (unpubl. obs.), that could be the basis for future studies defining any direct or indirect interactions with the P240 site. Whether or not the corresponding determinant, conserved in both primary and secondary structure, is associated with cell membrane attachment, the eventual definition of the viral function that necessitates its evolutionary conservation should provide valuable insite into the overall structure and the impact of structure on S 1 function.

Cultivation of the virus of infectious bronchitis

Structural polypeptides of coronavirus IBV

Sequencing approach to IBV antigenic variation and epizootiology

Induction of humoral neutralizing 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

Amino acids within hypervariable region 1 of avian coronavirus IBV (Massachusetts serotype) spike glycoprotein are associated with neutralization epitopes

Infectious bronchitis virus: evidence for recombination within the Massachusetts serotype

Isolation of a new serotype of infectious bronchitis-like virus from chickens in England

Prediction of protein conformation

An overview of the molecular characteristics of avian infectious bronchitis virus

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

Occurrence and significance of infectious bronchitis virus variant strains in egg and broiler production in the Netherlands

Induction of hepatitis A virusneutralizing antibody by a virus-specific synthetic peptide

Antigenic differences among isolates of avian infectious bronchitis virus

Serologic and immunologic properties of a recent isolate of infectious bronchitis virus

Serological comparisons of strains of infectious bronchitis using plaque purified isolates

Prediction of protein antigenic determinants from amino acid sequences

The S1 glycoprotein but not the N or M proteins of avian infectious bronchitis virus induces protection in vaccinated chickens

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

A novel variant of avian infectious bronchitis virus resulting from recombination among three different strains

A new serotype of infectious bronchitis virus responsible for respiratory disease in Arkansas broiler flocks

Immunologic differences in strains of infectious bronchitis virus

Prediction of chain flexibility in proteins

Multiple recombination sites at the 5' end of murine coronavirus RNA

Sequence evidence for RNA recombination in field isolates of avian coronavirus infectious bronchitis virus

Phylogeny of antigenic variants of avian coronavirus IBV

The recombination between nonsegrnented RNA genomes of routine coronaviruses

High-frequency RNA recombination of murine coronaviruses

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

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

Epitopes on the spike protein of a nephropathogenic strain of infectious bronchitis virus

Comparison of the structural proteins of avian infectious bronchitis virus as determined by western blot analysis

Cloning and sequencing of genes encoding structural proteins of avian infectious bronchitis virus

Sperling FG (1951) Infectious bronchitis

Molecular studies of the infectious bronchitis virus genome

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

Evolutionary implications of genetic variation in the S1 gene of infectious bronchitis virus

Etiology of an infectious nephritis-nephrosis syndrome of chickens

We thank Dr. Yuan Xu for her excellent technical assistance. This research was supported by Southeast Poultry and Egg Association No. 29