IBV抗原肽鉴定
Abstract The spike (S)protein,containing two subunits,S1and S2,is the major immunity-eliciting antigen of avian infectious bronchitis virus (IBV),a highly contagious disease of chickens.Several immunogenic regions,mainly located within the S1subunit,have been identified.Nonetheless,these immune-dominant regions were defined using selected mono-clonal antibodies or using a short peptide approach that involves only certain limited regions of the S protein.In addition,some immune-dominant regions are located in hypervariable regions (HVRs)which are not present in all serotypes.Hence,the aim of this study was to determine a
broader range of antigenic regions that have strong antibody eliciting ability;these could then be applied for development of an IBV-diagnostic tool.Initially,the S1and part of the S2subunit protein (24–567amino acids)were expressed as five fragments in prokaryotic system.The antigenicity was con-firmed using IBV immunized sera.Performance of the S subfragments was evaluated by ELISA using a panel of field chicken sera with known IBV titres determined by a commer-cial kit.This indicated that,among the five antigenic recom-binant proteins,the region S-E showed the highest specificity and sensitivity,namely 95.38%and 96.29%,respectively.The κvalue for the in-house ELISA using the S-E fragment compared to a commercial kit was 0.9172,indicating a high agreement between these two methods.As region S-E harbors strong immunogenicity within the spike protein,it has the potential to be exploited as an antigen when developing a cost-effective ELISA-based diagnosis tool.
Keywords Infectious bronchitis virus .Spike protein .Antigenic regions .ELISA
Introduction
Infectious bronchitis virus (IBV),belonging to gammacor-onavirus within the Coronaviridae family,is an economi-cally important etiological disease agent in chickens (Carstens 2010).It affects chickens of all ages and causes a range of clinical signs depending on the age of the chicken,including respiratory disease and nephritis in young chicks,and a reduction in egg production in adult birds.Control of such a highly contiguous disease relies on vaccination.Hence,early diagnosis and a determination of the immune status of chickens are important approaches for controlling IBV
K.-H.Lin :A.-P.Hsu :T.-J.Chang :J.-H.Shien (*)
Department of Veterinary Medicine,College of Veterinary Medicine,National Chung Hsing University,250Kuo Kuang Road,Taichung 402,Taiwan
e-mail:jhshien@https://www.sodocs.net/doc/498024167.html,.tw K.-H.Lin
lin.guanshiun@https://www.sodocs.net/doc/498024167.html,
C.-F.Lin
Department of Medical Laboratory Science and Biotechnology,Central Taiwan University of Science and Technology,Taichung,Taiwan
S.-S.Chiou :C.-C.Chang :W.-L.Hsu (*)
Graduate Institute of Microbiology and Public Health,College of Veterinary Medicine,National Chung Hsing University,250Kuo Kuang Road,Taichung 402,Taiwan
e-mail:wlhsu@https://www.sodocs.net/doc/498024167.html,.tw
A.-P.Hsu :M.-S.Lee
Council of Agriculture,Animal Health Research Institute,376Chung-Cheng Road,Tamsui,Taipei 251,Taiwan
METHODS AND PROTOCOLS
Application of purified recombinant antigenic spike
fragments to the diagnosis of avian infectious bronchitis virus infection
Kuan-Hsun Lin &Chuen-Fu Lin &Shyan-Song Chiou &Ai-Ping Hsu &Min-Shiuh Lee &Chao-Chin Chang &Tien-Jye Chang &Jui-Hung Shien &Wei-Li Hsu
Received:14March 2012/Revised:23April 2012/Accepted:25April 2012/Published online:25May 2012#Springer-Verlag 2012
Appl Microbiol Biotechnol (2012)95:233–242DOI 10.1007/s00253-012-4143-8
outbreaks in large flocks.Since first reported in1981, enzyme-linked immunosorbent assays(ELISA)utilizing inac-tivated IBV particles as the coating antigen have been used worldwide to measure the level of IBV specific antibodies in birds(Marquardt et al.1981).Purified recombinant viral proteins have been exploited as antigens for the serological detection of IBV(Chen et al.2003;Gibertoni et al.2005; Wang et al.1995;Wang and Khan2000).
IBV encodes four envelope glycoproteins;of these gly-coproteins,spike(S)protein,containing1165amino acid residues,is a highly glycosylated homotrimer and consists of the N-terminal region of the S1protein and C-terminal region of the S2,which are generated by post-translational cleavage.The peplomer consists of the S1subunit,which forms a bulb structure,and the S2subunit,which forms the stalk structure that anchors the S protein to viral envelope (Cavanagh1983).The S1glycoprotein is the major immu-nogen,and it is able to induce neutralization antibodies that provide protection against virulent challenge(Cavanagh et al.1992;Cavanagh et al.1988;Ignjatovic and Galli1994). Many of antigenic epitopes able to induce neutralization antibodies are located in the hypervariable regions(HVRs) of the S1protein(Cavanagh et al.1992;Niesters et al. 1986).These HVRs has been used to determine the relevant differences between various vaccine strains and a range of newly emerging strains;these epitopes have the potential to be developed into serotype-specific diagnosis tool(Kusters et al.1989a,b;Wang and Huang2000).In contrast,the S2 glycoprotein induces cross-reactive antibodies and T cell responses that are important for recovery from the disease (Ignjatovic and Sapats2005).Moreover,using monoclonal antibody,one neutralization epitope was mapped to N-terminal of S2region(Koch et al.1990).
At present,inactivated whole virus is the most commonly used coating antigen in the commercially available diagnosis kits that are used for IBV diagnosis.Recombinant antigenic protein expressed using prokaryotic,yeast,or baculovirus systems has become a new trend when preparing specific coating antigen for ELISA kits(Chan et al.2009;Chen et al. 2003;Gibertoni et al.2005;Niesters et al.1986).Due to the protein being a strong immunogen,ELISA kits for the detec-tion of IBV based on the S protein have been reported in the literature(Chen et al.2011;Wang et al.2002).
An ideal serological reagent requires a strong immune eliciting ability.Kant et al.(1992)predicted antigen epitopes by analyzing the sequences of mutant viruses that escaped from neutralization using eleven monoclonal antibodies.A comparison of the nucleotide sequences of the antibody-resistant mutant viruses with that of the IBV parental strain gave five conformation dependent antigenic regions within the S1subunit;these were S1D(24–61amino acid;a.a), S1E(132–149a.a),and S1C/A/B(291–398a.a).Other research group identified residues304–386on the S1glycoprotein that are involved in a serotype-specific,con-formationally dependent epitope by analysis of the S1gene of monoclonal antibody-neutralization-resistant mutants (Moore et al.1997).More recently,Ignjatovic and Sapats (2005)mapped the antigenic and protective epitopes of S1 using seven short synthetic peptides.They identified the additional epitopes SP4(294–316a.a.),SP6(532–537 a.a.),and SP7(566–584a.a.)that may contribute to eliciting protective immune responses(Ignjatovic and Sapats2005). Up to the present,a systemic screening of entire S1subunit has not been carried out;although the location of these antigenic regions have been determined,many of them are located in the hypervariable regions of S1and very likely the residue of these epitopes might not exist in IBV isolates from Taiwan(Cavanagh et al.1992;Kant et al.1992; Kusters et al.1989a,b;Niesters et al.1986).The IBV spike is a large protein with intensive hydrophobicity that limits its yield when heterologously expressed in various systems (unpublished data).To develop a cost-effective and sensitive serological diagnosis assay,subunit S proteins capable of eliciting strong immune responses and able to be expressed at high level are needed.In these circumstances,the aim of the present study is to determine which S1subunit frag-ments show strong immunogenicity and can therefore be used to develop ELISA detection systems.To this end,five recombinant S fragments that cover S1region and part of the S2subunit were expressed and purified;the antigenicity of these fragments was then characterized.
Materials and methods
Amplification of the IBV S gene(nucleotides1–1764) Total nucleic acid was extracted from105EID50of genotype I IBV(strain CK/TW/T15/2006),given by Animal Health Research Institute,Council of Agriculture,Taiwan,using a MagNA Pure Compact Nucleic acid isolation kit I and the automated MagNA Pure Compact System(Roche Applied Science).The concentration of nucleic acid was determined by spectrophotometer.The S gene was amplified from the RNA samples by One-step RT-PCR.The primer set was designed based on the IBV sequence reported previously (GenBank accession number:DQ646405)and consisted of forward primer(S1F)5′-ATTGTGCATGGTGGACAATG-3′and reverse primer(S1R)5′-TCTGTAACATTAAG TAAAGG-3′.First-strand synthesis of cDNA was followed by S gene amplification.The amplification was carried out using a One-step RT-PCR kit(Bertec;Taiwan)according to the manufacturer's descriptions.The thermocycling condi-tions for amplification were42°C(50min),95°C(2min) followed by35cycles of denaturation(95°C,40s),anneal-ing(50°C,50s),and extension(72°C,1min20s);this was
followed by a final extension(72°C,7min).The amplified fragment was then cloned into the TA vector PCR2.1 (Invitrogen)and the resulting plasmid designated IBV-S-pTA.The authenticity of the PCR products was determined by automatic sequencing(Mission Biotech,Taipei,Taiwan). The sequences of the partial S gene(nucleotides1–1764, which consists of the full length of S1and part of S2)have been submitted to GenBank and have the accession number HQ185567.
Sequence alignment
The sequence similarity was established between the local IBV genotypes,such as Taiwan local genotypes(genotype I CK/TW/T15/2006;accession number:HQ185567)obtained in this study,and genotype II TW22296/95(DQ646404), and various other IBV strains,namely strain D207(acces-sion number:M21969),Mass H120(accession number: FJ888351),Beaudette42(accession number:DQ830981) and M41(accession number:DQ834384),has also been reported to GenBank.These were then analyzed using Meg-align from the DNAstar software package(DNASTAR,Inc. Madison,Wisconsin,USA).
Construction of the plasmids expressing the five antigenic regions
The study was designed to investigate the antigenic regions of the IBV spike proteins,and this involved assessment using IBV(strain CK/TW/T15/2006)immunized chicken sera.Five fragments(A to E)are located in the immunodo-minant region of the IBV spike(consisting of entire S1 subunit and part of S2)as defined in previous studies (Ignjatovic and Sapats2005;Kant et al.1992).These regions were amplified and cloned into an expression vector. Briefly,region A(24–149a.a),and region C(235–302a.a) were amplified from pET24a plasmid(Novagen)harbours the S gene of genotype I CK/TW/T15/2006using the uni-versal forward primer RBS-F:5′-TAAGAAGGAGATATC CATATGGCTA-3′together with either IBV-S-A-R5′-GTACTGTCGACAAGCTTCTGTTAAATT-3′or IBV-S-C-R(5′-AAAATGTCGACAAGCTTTCTGAGCTG-3′) for region A(24–149a.a),and C(235–302a.a),respective-ly.For the ease of subsequent cloning,restriction enzymes cutting sites,as indicated by underline,were introduced into each oligonucleotide.The PCR products were digested with Eco RV/Hin dIII,and then subcloned to pET32a(Novagen). The resulted expression plasmids were named IBV-S-A-pET32a and IBV-S-C-pET32a.
The regions B and D of the S gene were amplified from IBV-S-pTA,which contains the IBV Spike protein(nucleo-tides1–1764),by PCR using two primer set,IBV-S-B-F(5′-ATTCGGATCCGTTAGTGTATCTAAGTAC-3′)/IBV-S-B-R(5′-AGATCTCGAGACATGCTAGCAAACCTCT-3′) and IBV-S-D-F(5′-ATTCGGATCCAGTGGTTATTA TAATTTT-3′)/IBV-S-D-R(5′-AGATCTCGAGCACTA CAAACTGCTGATT-3′)for region B(150–234a.a),and D(303–503a.a),respectively.Subsequently,the PCR prod-ucts were cut with Bam HI/Xho I and cloned into pET32a. The resulting expression plasmids were named IBV-S-B-pET32a,and IBV-S-D-pET32a,respectively.
Due to a lack of restriction enzyme sites,a two-step cloning procedure was used to construct the plasmid expressing the E region(504–567a.a).Initially,region E fragment was amplified from IBV-S-pTA by PCR using the primer set IBV-S-E-F(5′-AGTTTGCTAGCTCAGGCGG TAAG-3′)/IBV-S-E-R(5′-CCGCAAGCTTGTTGTT TAAGTGAAC-3′).The PCR product was digested with Nhe I/Hin dIII and subcloned to pET24a.Next,the region E was amplified from this plasmid by PCR using the primer set RBS-F(5′-TAAGAAGGAGATATCCATATGGCTA-3′) and IBV-S-E-R(5′-CCGCAAGCTTGTTGTTTAAGT GAAC-3′).The PCR product was then digested with Eco RV/Hin dIII and subcloned to pET32a.The resulted ex-pression plasmid was named IBV-S-E-pET32a.
Amplification of all of these fragments used the same conditions:initial denaturation95°C(3min),followed by 35thermal cycles of denaturation(95°C,30s),annealing (55°C,40s),and extension(72°C,40s),followed by a final extension(72°C,5min).All amplification cycles were performed in a DNA thermal cycler(GeneAmp PCR system 9700,PerkinElmer)and the resulting products were re-solved on a2%agarose gel.The authenticity of all the plasmids was initially checked by restriction enzyme diges-tion and this was then followed by automated DNA se-quence analysis(Mission-Biotech,Taipei).
Recombinant protein expression and purification
The various plasmids were transformed into Escherichia coli BL21(DE3)competent cells.Protein expression was induced by0.8mM IPTG at16°C for24h.Bacterial cell pellets were initially dissolved in1/10volume of binding buffer without denaturing agent(0.01M imidazole,0.5M NaCl,0.05M Tris–HCl,200μg/ml lysozyme)followed by three cycles of freeze-thawing and then sonication.The soluble and insoluble protein fractions were separated by centrifugation at13,000rpm for15min at4°C.The insoluble fraction(pellet),which contained the majority of recombinant S protein,was then resuspended in denatur-ation buffer(0.01M imidazole,0.5M NaCl,0.05M Tris–HCl,6M urea)and this was followed by further sonication. As all the recombinant proteins were expressed with6-histidine tag at C-terminus that allows for further purifica-tion by metal affinity chromatography according manufac-turer protocol(Pharmacia).In brief,the bacteria crude
extract was mixed with Ni-NTA sepharose packed in col-umn for2h at room temperature with rotation,followed by five times of washes with buffer(0.02M imidazole,0.5M NaCl,0.05M Tris–HCl,6M urea),and then eluted using a buffer containing a high concentration of imidazole(100,or 400mM).After step-wise dialysis against PBS(0.02M phosphate,0.15M NaCl)containing gradually decreased concentration of urea at4°C to remove the urea and imid-azole,the protein concentration was determined by Bradford method(Bio-Rad).
Western blot analysis
Recombinant proteins were separated by10%or12.5% SDS-PAGE and electrotransferred to nitrocellulose mem-brane using Mini Proten III equipment(BioRad).The mem-branes were blocked in PBS-T buffer(0.02M phosphate, 0.15M NaCl,0.05%Tween-20)containing5%fat free dried skimmed milk powder and probed with mouse anti-his tag antibody(1:10,000)or1:50diluted high immune serum that was obtained from chicken initially immunized with106 EID50of genotype I IBV(homologous strain)followed by two boosters with2-week interval for overnight at4°C. After several washes with PBS-T buffer,the filter was next incubated with the secondary antibody,either goat anti-mouse IgG conjugated with horseradish peroxidase(HRP) diluted1:10,000or goat anti-chicken IgG conjugated HRP antibody diluted1:2,000for1h at room temperature.After another PBS wash to remove the unbound antibodies,the signal was detected using an AEC(Red)Substrate Kit (Invitrogen).
Enzyme-linked immunosorbent assay
The enzyme-linked immunosorbent assay(ELISA)protocol was initially optimized as follows:a checkerboard titration was conducted using various concentrations of the recom-binant proteins(at concentrations of2.48,7.4,22.3,67,200, and600μg/ml)in combination with serially diluted high immunized chicken serum(dilutions of1:1,000,1:2,000, 1:4,000,1:8,000,and1:16,000).All combinations were repeated in triplicate.The recombinant S proteins(regions A-E)were diluted in coating buffer at the various concen-trations,and100μl were added to each well of the ELISA plate(Nunc).After overnight incubation at4°C,the plates were washed three times with PBS-T buffer(0.02M phos-phate,0.15M NaCl,0.05%Tween-20)and then blocked with PBS-T containing5%fat free dried skimmed milk powder at room temperature for1h.After washing three times with PBS-T,the serially diluted test chicken serum samples were added for1h of incubation at37°C.The plates were then washed three times with PBS-T.This was followed by the addition of the secondary antibody(goat anti-chicken IgY conjugated with HRP diluted1:10,000, Thermo).The plates were again washed three times with PBS-T and then the substrate3,3′5,5′-tetramethylbenzi-dine(TMB Microwell substrate system)was added.Finally, the plates were incubated for10min in the dark,and the reaction was terminated by adding2M H2SO4.The plates were read at450nm using a microtitration plate ELISA reader(Tecan Sunrise).
Following the optimized protocol,the performance of the in-house recombinant IBV-S ELISA was then evaluated using a panel of field sera(81samples IBV-positive and 65samples IBV negative as diagnosed by a commercial IDEXX ELISA kit)collected from the Animal Disease Diagnostic Center(ADDC),NCHU.
Statistical analysis to obtain a receiver operating characteristic curve
The cutoff values of the IBV-S epitope ELISA were deter-mined by receiving operating characteristics(ROC)analysis using the OD450nm values obtained from the81IBV-positive and65IBV-negative chicken sera.The ROC curves were generated by the GraphPad statistical package analysis tool(GraphPad Software,San Diego,CA).This was used to determine the cutoff OD450nm values that provided the optimal sensitivity and specificity for each ELISA.The area under the ROC curve(AUC)was used to evaluate the performance of a diagnostic test to determine the evidence of infection.
Agreement between the results of the commercial and S subfragment ELISA assays was evaluated by calculating the κstatistic(Sackett et al.1991).Theκvalue was calculated as the agreement beyond chance divided by the amount of agreement possible beyond chance(observed agreement percent/expected agreement percent)/(1?expected agree-ment percent).Theκvalue can be interpreted in the follow-ing qualitative manner:0.0–0.2,slight agreement;0.2–0.4, fair agreement;0.4–0.6,moderate agreement0.6–0.8,sub-stantial agreement;and0.8–0.1,almost perfect agreement. Results
Sequence analysis to identify the immunogenic regions
To determine the subunit regions that need to be expressed, we initially analyzed the sequences of the spike proteins of various local IBV strains.As shown in Fig.1,a sequence alignment of1–149amino acids(HVR regions)of the S protein of Taiwan local strains,namely genotype I,and II, with those isolated from other countries,demonstrated a high level of variation.This region contains antigenic sites S1D(24–61a.a),and S1E(132–149a.a)as defined by Kant
et al.(1992).The sequence divergence between genotype I,the major genotype in Taiwan and other genotypes,includ-ing genotype II,was up to 41.6%(Table 1).Hence,it was worth exploring the antigenicity of the corresponding region designated region S-A in this study (Fig.2a ).In addition,region S-C (235–302a.a.),also located in S1,was selected due to its high level of sequence conservation and the fact that it contained previously defined non-conformational de-pendent epitopes (i.e.,SP3245–260a.a.and SP4294–316a.a.)(Ignjatovic and Sapats 2005;Wang et al.1995).The region S-E (504–567a.a.),which is located at the junction of S1and S2subunits,consists of SP6epitopes and includes residues of the non-conformational dependent epitope;this is known to eliciting cross-protection antibodies and was
chosen for comparative purposes in this study (Ignjatovic and Sapats 2005).To evaluate the entire S1,two other regions,S-B (150–234a.a)and S-D (303–503),which are located between S-A/S-C and S-C/S-E were also selected (Fig.2a ).
Plasmid construction,recombinant protein expression,and protein purification
In order to prepare the recombinant proteins containing the antigenic epitopes,the recombinant proteins were expressed and purified from E.coli (Fig.2b )after IPTG induction.All constructs successfully produced antigenic proteins at vari-ous levels as indicated by asterisks compared
with
Fig.1Alignment of the deduced amino acid sequence s of the IBV spike protein.Amino acid residues 1to 149of the S1subunit of Taiwan local genotypes genotype I CK/TW/T15/2006(HQ185567),and genotype II TWII-2296/95(DQ646404)were aligned with those of other IBV strains,namely Mass H120(FJ888351),M41(DQ834384),Beaudette 42(DQ830981),and D207(M21969).The dots indicate regions where the sequences are identical to the consensus sequences.Deletions within the sequences are shown as dashes
uninduced cell lysate.After Ni-NTA chromatography,pro-teins of the predicted size were purified (Fig.2c ).Recognition of recombinant antigenic proteins by polyclonal chicken IBV antisera
The identity of the recombinant protein was verified by Western blot analysis.As expected,the five recombinant proteins were recognized by antibody targeting the proteins'his-tags (Fig.3a ).Nevertheless,it is of importance to deter-mine whether the recombinant proteins showed appropriate antigenicity with respect to IBV.To do so,the serum of chickens infected with IBV was used to test the antigenic properties of the purified recombinant proteins.Each recom-binant protein (500μg)was resolved by SDS-PAGE and by analyzed by Western blotting (Fig.3b ).Noticeably,IBV-immunized serum reacted with all of the five recombinant antigenic proteins (as indicated with asterisks,Fig.3b ),but not with thioredoxin,the fusion peptide at N-terminal of the recombinant proteins.Interestingly,higher intensity of sig-nals were observed for the S-A and S-E recombinant pro-teins (lane 1and 5,Fig.3b ),which suggests a stronger interaction between IBV serum and these two recombinant proteins.
Table 1Sequence alignment of the various IBV strains showing percent similarity
CK/TW/T15/2006TW2-2296/95
Mass H120M41
Beaudette 42D207
Percent Identity
D i v e r g e n c e
1
234561
75.5
72.669.274.068.01229.768.3
68.369.766.42334.141.291.8
95.267.63439.641.28.791.8
64.14532.038.8 5.08.766.2
56
41.644.342.348.544.76
1
2
3
4
5
6
Fig.2Identification,expression,and purification of the recombinant partial spike proteins.a Schematic illustration of the location and length of the five antigenic fragments.Recombinant proteins (indicated
with asterisks )were expressed under IPTG treatment (+)(b ),and then purified by Ni-NTA affinity chromatography (c )
Detection by ELISA using field sera and the recombinant antigenic proteins
Using the checkerboard titrations described in “Materials and methods ”section,the optimized ELISA conditions for coating antigen and detection antibody were found to be 0.74μg of recombinant proteins per well and a 1:1,000dilution of the field serum.The performance of the in-house recombinant IBV-S ELISA was then evaluated using a panel of field sera collected from the Animal Disease Diagnostic Center (ADDC),NCHU.Among this panel,81samples were defined as IBV-positive and 65samples were defined as IBV negative using a commercially available ELISA kit (IDEXX).
Antigen performance was statistically analyzed by ROC curves as shown in Fig.4a .Overall,the results indicated that S subfragment ELISA using the S-E antigen (AUC 00.986)more accurately discriminated between presumptive IBV-positive and -negative serum specimens than did the antigen S-A,S-B,S-C,and S-D (AUC 00.9413,0.9436,0.9320,0.9502,respectively).
On the basis of the ROC analysis,the overall cutoff value for the IBV-S subfragment ELISAwas https://www.sodocs.net/doc/498024167.html,pared
with
Fig.3Recognition of the recombinant antigenic proteins by polyclon-al chicken IBV antisera.Each recombinant protein (approximately 500μg),namely including S-A (lane 1),S-B (lane 2),S-C (lane 3),S-D (lane 4),and S-E (lane 5),together with a fusion tag peptide thioredoxin as a negative control (lane 6),were separated by SDS-
PAGE and then analyzed by Western blot assay using antibody against His tag (a ),and chicken serum immunized with IBV (b ).All five recombinant proteins containing partial spike sequences were recog-nized by the IBV antisera (as indicated by asterisks
)
1-Specificity %
S e n s i t i v i t y %
a
b c
IBV-S-subfragment protein
IBV-S-subfragment protein
Fig.4Detection of field specimens by ELISA using recombinant the S subfragments.Panels (a )of field chicken serum samples that had been previously diagnosed by a commercial IBV ELSIA kit were used to evaluate the performance of the various S subfragment ELISAs.A plot of the sensitivity versus the false-positive rate (1-specificity),the receiver operator characteristic (ROC)curve analysis,was used to determine the discriminatory accuracies of the tests.The area under
the ROC curve (AUC)was used to evaluate the performance of a diagnostic test (a ).Based on ROC curve,the overall cutoff value derived from the ELISA with five S subfragments was 0.8965.The OD450nm values of individual samples from low titer (IBV negetive controls,n 065),and high titer IBVantisera (IBV positive controls,n 081)groups are shown in panels b and c ,respectively
the negative samples(Fig.4b),the IBV-positive sera reacted with the recombinant S subfragments to a much higher extent (Fig.4c).The specificity and sensitivity of each IBV subfrag-ment ELISA was calculated based on the cutoff value.The specificity of the IBV-S protein fragment ELISA assays using the various antigens were:S-A—89.23%,S-B—100.00%,S-C —100.00%,S-D—75.38%,and S-E—95.38%.The sensitiv-ities of the IBV-S protein fragment ELISA assays were S-A—88.88%,S-B—62.96%,S-C—38.27%,S-D—95.06%,and S-E—96.29%(Table2).Among the five S subfragment pro-teins,the recombinant S-E antigen gave the highest sensitivity and specificity.Theκvalue was0.9172(almost perfect agree-ment),indicating excellent coordination between the commer-cial and in-house IBV S-E region ELISA.
Confirmation of the IBV-S-based ELISA using Western blot analysis
When tested,the S-A antigen detected IBV antibody in7/65 of the presumptively negative samples.This disagreement between the commercial and in-house ELISA results might be due to the high frequency of variation in the region S-A, as indicated in Fig.1.
To confirm the accuracy of these diagnoses,the samples
were further tested by Western blotting.As shown in Fig.5, four out of seven presumptive IBV-negative sera reacted positively with S-A antigen,indicating that antibodies spe-cific to the S-A antigenic region exist in those specimens. Discussion
An ideal antigen for a serological test needs readily available preparations of pure antigen.Due to the complexity,expen-sive and laborious procedures needed to produce pure virus, the use of recombinant viral proteins expressed in eukaryotic systems,such as yeast and insect cell,have been developed as serological antigens(Gibertoni et al.2005;Niesters et al. 1986).To establish a cost-effective and specific method for IBV diagnosis,the robust prokaryotic host E.coli was selected for protein production.As E.coli is a common commensal microorganism of all animals,it is reasonable to suspect that most field specimens might contain antibodies against E.coli and hence the purity of the coating antigen expressed in such a host might affect diagnosis accuracy.Contamination with E. coli proteins of the coating antigen has been proposed to lead to false positives when serodiagnosing animal diseases (Zanoni et al.1991).Ni-NTA affinity chromatography in combination with ion-exchange chromatography or pre-absorption of the tested serum specimens with E.coli crude extract have been suggested as ways of eliminating the false positive problem(Wu et al.2008;Zanoni et al.1991).In the present study,despite the insolubility of all five recombinant S fragments,rather than using denaturing buffer,the bacteria pellet was initially lysed in buffer without urea.After centri-fugation,supernatant containing the soluble protein was re-moved and the enriched insoluble protein fraction(pellet)was then dissolved in denaturing buffer.This was followed by the
Table2A comparison of the assay results by a commercial kit and S subfragment ELISA using 146field sera
a81chicken serum samples
that were diagnosed as
IBV-positive by IDEXX kit
b65chicken serum samples that were diagnosed as
IBV-negative by IDEXX kit
Recombinant IBV spike proteins
IBV-S-A IBV-S-B IBV-S-C IBV-S-D IBV-S-E
+?+?+?+?+?
IBV positive chicken sera(n081)a72951303150774783 IBV negative chicken sera(n065)b7580650651649362 Sensitivity(%)88.8862.9638.2795.0696.29 Specificity(%)89.23100.00100.0075.38
95.38
Fig.5Comparison analysis of the field samples by Western blot
analysis.Seven IBV negative field samples that were defined as IBV
negative by commercial ELISA kit but recognized by ELISA with S-A
antigen were examined.Further testing by Western blot analysis
showed that four out of the seven presumptive IBV-negative sera
reacted positively with S-A antigen
further steps of the purification procedure for insoluble pro-teins.Such an addition step is useful to increase protein purity by depleting the endogenous soluble fraction of bacterial proteins prior to affinity chromatography;as shown in Fig.2c,one Ni-NTA affinity chromatography run was suffi-cient to purify the recombinant proteins to high homogeneity, even when the protein had a relatively low yield.
The immunogenicity of the coating antigen is one of the crucial factors when performing serological tests such as ELISA for antibody detection.The high hydrophobicity of the full-length recombinant spike protein,as predicted by Kyte-Doolittle hydropathy plots(data not shown),as well as its high molecular weight(180kDa),makes expression of the full-length recombinant spike protein in E.coli difficult(Cav-anagh1983).As a full screen of immunogenic regions on a broad scale has not yet been done,to clearly map the immu-nogenic regions of the entire S1and partial S2subunits,five subfragment proteins harboring distinct regions of S protein were expressed and the antigenicity of these individual spike fragments was determined.In addition,to simplify the inter-pretation,immunogenic property of alternate fragments over-lapping with these five fragments was not considered.
In our study,among the five fragments,S-A and S-E were strongly recognized by IBV immunized sera(Fig.3b).The S-A antigen harbors the conformational epitopes S1D(24–61 a.a),and S1E(132–149a.a),which were previously defined based on the sequence of IBV strain D207(Kant et al.1992). It is worth noting that these two epitopes are located in hyper variable regions.The sequence alignment indicates that in epitopes S1D and S1E,9out of38and4out of18amino acids were different between D207and TW1,respectively (Fig.1).This suggests that either the amino acid variations do not alter the conformation of the epitopes,or the S-A region of TW-1strain harbors epitopes independent of S1D and S1E. Noticeably,the specificity of the ELISA using S-A antigen was89.23%indicating that some of IBV-negative samples examined by commercial kit were detected as positive by S-A https://www.sodocs.net/doc/498024167.html,ing Western blot analysis,the S-A antigen was able to recognize by four out of seven IBV-negative samples, indicating the existence of IBV antibodies in these serum specimens(Fig.5).The inconsistent results between the com-mercial and the S-A ELISA might be due to the genotype of the IBV virus used as coating antigen in commercial ELISA kit.This is likely to be significantly different from the Taiwan local IBV,and therefore field serum samples may react less efficiently with foreign IBV than that with the recombinant S-A antigen.Similarly,it has been suggested that sequence variations in the S1subunit might be a major cause of poor cross-protection between different serotypes of IBV(Lin et al. 2005;Wang and Huang2000).Hence,the antigenic region S-A may be useful as a reagent for type-specific diagnosis. Among the five fragments,ELISA with the S-E antigen shows the highest consistency with the commercial kit(κvalue of 0.9172).This region covers the previously defined S1-F and SP6-Sp6epitopes that have been described as non-conformation epitopes(Ignjatovic and Sapats2005;Kant et al. 1992).The immunogenicity of this region has also been con-firmed by another research group.Wang et al.established an S-fg ELISA using a coating protein(amino acid381–555)that partially overlapped with the S-E fragment(amino acid504–567)expressed in current study(Wang et al.2002).Notwith-standing the fact that a larger region of S1subunit was included in their antigen fragment,the antibody titers of the tested sera identified by commercial ELISA were significantly correlated with that optical values indicated by the S-fg ELSIA.
Despite the fact that serological methods for IBV detec-tion have been described by other research groups,an anal-ysis of the antigenic regions within the entire S1subunit has not been done before.In the present study,two strong immunogenic regions,S-A and S-E,were identified by both Western blot analysis and ELISA.S-A,which is located in the HVR,might be useful as a type-specific serological diagnosis reagent.Furthermore,due to the strong immuno-genicity and high cross-reactivity between the different gen-otypes,the S-E region could serve as an ideal antigen for serosurveillance.
Acknowledgments This study was partly supported by the Bureau of Animal and Plant Health Inspection and Quarantine,the Council of Agriculture,grant number:96–1.1.5-B1(5).
References
Carstens EB(2010)Ratification vote on taxonomic proposals to the international committee on taxonomy of viruses(2009).Arch Virol155:133–146.doi:10.1007/s00705-009-0547-x Cavanagh D(1983)Coronavirus IBV:structural characterization of the spike protein.J Gen Virol64(Pt12):2577–2583.doi:10.1099/ 0022-1317-64-12-2577
Cavanagh D,Davis PJ,Mockett AP(1988)Amino acids within hyper-variable region1of avian coronavirus IBV(Massachusetts sero-type)spike glycoprotein are associated with neutralization epitopes.Virus Res11:141–150
Cavanagh D,Davis PJ,Cook JK,Li D,Kant A,Koch G(1992) Location of the amino acid differences in the S1spike glycopro-tein subunit of closely related serotypes of infectious bronchitis virus.Avian Pathol21:33–43.doi:10.1080/03079459208418816 Chan KW,Hsieh HH,Wang HC,Lee YJ,Sung MH,Wong ML,Hsu WL(2009)Identification,expression and antigenic analysis of recombinant hemagglutinin proteins of canine distemper virus.J Virol Methods155:18–24.doi:10.1016/j.jviromet.2008.09.024 Chen H,Coote B,Attree S,Hiscox JA(2003)Evaluation of a nucleoprotein-based enzyme-linked immunosorbent assay for the detection of antibodies against infectious bronchitis virus.
Avian Pathol32:519–526.doi:10.1080/0307945031000154125 Chen HW,Wang CH,Cheng IC(2011)A type-specific blocking ELISA for the detection of infectious bronchitis virus antibody.
J Virol Methods173:7–12.doi:10.1016/j.jviromet.2010.12.016 Gibertoni AM,Montassier Mde F,Sena JA,Givisiez PE,Furuyama CR,Montassier HJ(2005)Development and application of a Saccharomyces cerevisiae-expressed nucleocapsid protein-based
enzyme-linked immunosorbent assay for detection of antibodies against infectious bronchitis virus.J Clin Microbiol43:1982–1984.doi:10.1128/JCM.43.4.1982-1984.2005
Ignjatovic J,Galli L(1994)The S1glycoprotein but not the N or M proteins of avian infectious bronchitis virus induces protection in vaccinated chickens.Arch Virol138:117–134
Ignjatovic J,Sapats S(2005)Identification of previously unknown anti-genic epitopes on the S and N proteins of avian infectious bronchitis virus.Arch Virol150:1813–1831.doi:10.1007/s00705-005-0541-x Kant A,Koch G,van Roozelaar DJ,Kusters JG,Poelwijk FA,van der Zeijst BA(1992)Location of antigenic sites defined by neutral-izing monoclonal antibodies on the S1avian infectious bronchitis virus glycopolypeptide.J Gen Virol73(Pt3):591–596.
doi:10.1099/0022-1317-73-3-591
Koch G,Hartog L,Kant A,van Roozelaar DJ(1990)Antigenic domains on the peplomer protein of avian infectious bronchitis virus:correlation with biological functions.J Gen Virol71(Pt
9):1929–1935.doi:10.1099/0022-1317-71-9-1929
Kusters JG,Jager EJ,Lenstra JA,Koch G,Posthumus WP,Meloen RH,van der Zeijst BA(1989a)Analysis of an immunodominant region of infectious bronchitis virus.J Immunol143:2692–2698 Kusters JG,Niesters HG,Lenstra JA,Horzinek MC,van der Zeijst BA (1989b)Phylogeny of antigenic variants of avian coronavirus IBV.Virology169:217–221
Lin KY,Wang HC,Wang CH(2005)Protective effect of vaccination in chicks with local infectious bronchitis viruses against field virus challenge.J Microbiol Immunol Infect38:25–30
Marquardt WW,Snyder DB,Schlotthober BA(1981)Detection and quantification of antibodies to infectious bronchitis virus by enzyme-linked immunosorbent assay.Avian Dis25:713–722 Moore KM,Jackwood MW,Hilt DA(1997)Identification of amino acids involved in a serotype and neutralization specific epitope
within the S1subunit of avian infectious bronchitis virus.Arch Virol142:2249–2256
Niesters HG,Lenstra JA,Spaan WJ,Zijderveld AJ,Bleumink-Pluym NM,Hong F,van Scharrenburg GJ,Horzinek MC,van der Zeijst BA(1986)The peplomer protein sequence of the M41strain of coronavirus IBV and its comparison with Beaudette strains.Virus Res5:253–263
Sackett DL,Tugwell P,Guyatt GH(1991)Clinical epidemiology:a basic science for clinical medicine,2nd edn.Little,Brown and Co,Boston
Wang CH,Huang YC(2000)Relationship between serotypes and genotypes based on the hypervariable region of the S1gene of infectious bronchitis virus.Arch Virol145:291–300
Wang X,Khan MI(2000)Molecular characterization of an infectious bronchitis virus strain isolated from an outbreak in vaccinated layers.Avian Dis44:1000–1006
Wang L,Parr RL,King DJ,Collisson EW(1995)A highly conserved epitope on the spike protein of infectious bronchitis virus.Arch Virol140:2201–2213
Wang CH,Hong CC,Seak JC(2002)An ELISA for antibodies against infectious bronchitis virus using an S1spike polypeptide.Vet Microbiol85:333–342.doi:S0*******(01)00525-9
Wu PC,Chien MS,Tseng YY,Lin J,Lin WL,Yang CY,Huang C (2008)Expression of the porcine circovirus type2capsid protein subunits and application to an indirect ELISA.J Biotechnol 133:58–64.doi:10.1016/j.jbiotec.2007.09.015
Zanoni RG,Nauta IM,Pauli U,Peterhans E(1991)Expression in Escherichia coli and sequencing of the coding region for the capsid protein of Dutch maedi-visna virus strain ZZV1050: application of recombinant protein in enzyme-linked immunosor-bent assay for the detection of caprine and ovine lentiviruses.J Clin Microbiol29:1290–1294