key: cord-0885634-bbn16uf1
authors: Lai, Szu-Chia; Chong, Pele Choi-Sing; Yeh, Chia-Tsui; Liu, Levent Shih-Jen; Jan, Jia-Tsrong; Chi, Hsiang-Yun; Liu, Hwan-Wun; Chen, Ann; Wang, Yeau-Ching
title: Characterization of neutralizing monoclonal antibodies recognizing a 15-residues epitope on the spike protein HR2 region of severe acute respiratory syndrome coronavirus (SARS-CoV)
date: 2005-08-19
journal: J Biomed Sci
DOI: 10.1007/s11373-005-9004-3
sha: 5214e22c132b0127b8783e4cd14ee8378f0bee71
doc_id: 885634
cord_uid: bbn16uf1

The spike (S) glycoprotein is thought to play a complex and central role in the biology and pathogenesis of SARS coronavirus infection. In this study, a recombinant protein (rS268, corresponding to residues 268–1255 of SARS-CoV S protein) was expressed in Escherichia coli and was purified to near homogeneity. After immunization with rS268, S protein-specific BALB/c antisera and mAbs were induced and confirmed using ELISA, Western blot and IFA. Several BALB/c mAbs were found to be effectively to neutralize the infection of Vero E6 cells by SARS-CoV in a dose-dependent manner. Systematic epitope mapping showed that all these neutralizing mAbs recognized a 15-residues peptide (CB-119) corresponding to residues 1143–1157 (SPDVDLGDISGINAS) that was located to the second heptad repeat (HR2) region of the SARS-CoV spike protein. The peptide CB-119 could specifically inhibit the interaction of neutralizing mAbs and spike protein in a dose-dependent manner. Further, neutralizing mAbs, but not control mAbs, could specifically interact with CB-119 in a dose-dependent manner. Results implicated that the second heptad repeat region of spike protein could be a good target for vaccine development against SARS-CoV.

It is now well-documented and strong evidence supporting that a new emerging coronavirus (CoV) is etiologically linked to the outbreak of Severe Acute Respiratory Syndrome (SARS) in Asia and Canada last year [1] [2] [3] . SARS associated CoV has several major structural proteins: membrane (M) and spike (S) glycoprotein, and nucleocapsid (N) and envelop (E) proteins [4] . Of these structural proteins, it is thought that the S protein is a multifunctional protein that plays complex and central role in the biology and pathogenesis of SARS infection [4] . Major known properties and functions of the coroavirus S protein had been found to include: (a) binding to specific receptor glycoproteins on the surface of host cells; (b) after binding to receptor the S protein possible conformational changes causing induction of fusion of the viral envelop with the cell membrane; (c) cell fusion facilitating viral RNA entry into host; (d) a target for induction of the neutralizing antibody and (e) elicitation of cell-mediated immunity [4] . It had been demonstrated the N-terminal 600 amino acids (S1 domain) associated with altered antigenicity and virulence [5, 6] . Actually, there were considerable diversity in both the lengths and nucleotide sequences of the S1 derived from different groups of coronaviruses, and sometime even within different strains of a single family of coronavirus [7] [8] [9] [10] . Alignment of the amino acid sequences of the SARS-CoV S (GI 29836496) and HCoV-229E spike proteins (GI 13604334) had revealed only 29% identity. This indicated that there was very considerable diversity of S proteins between SARS-CoV and other human coronavirus (HCoV-229E and HCoV-OC43).

A rapid treatment of SARS patient is urgently needed. Therefore, it is very necessary to develop an effective and safe vaccine to prevent the infection of SARS-CoV. In this study, we evaluated the neutralization effect of mAbs to the spike protein of SARS-CoV, and mapped the interaction region of spike protein with neutralizing mAbs. It is anticipative to find out conserved neutralizing epitope(s) of spike protein that can be used as a vaccine target or a therapeutic agent against SARS-CoV.

Escherichia coli TOPO10 was the plasmid used in the cloning SARS spike protein, and Escherichia coli BL21 Star (DE3) was used for over expression of proteins under the control of phage T7 lac promoter. The plasmid vector pET101/D-TOPO (QIAGEN, K101-01) was used to express the histidine-tagged fused at the carboxyl-terminus of S protein to generate recombinant plasmid, rS 268-1255 .

The mRNA was extracted from Vero E6 cells infected with SARS-CoV isolated from suspected-SARS patients using the QIAamp Viral RNA Mini Kit (Qiagen) according to the instruction of manual. Extracted mRNA was resuspended into TE buffer (10 mM Tris-Cl and 1 mM EDTA, pH 8.0) and used as a template in the RT-PCR for amplification of amino acids 268-1255 of the SARS coronavirus. Oligo-dT-18 (5¢-TTTTTTTTTTTTTTTTTT-3¢) and one pair of primers corresponding to nucleotides 25241-25215 (22278-22295 and 25241-25215 (CoV268pET: 5¢-CACC-ATggAAAATggTA CAATCACA-3¢; and CoV25241pET: 5¢-TgTgTAA TgTAATTTgACACCCTTgAg-3¢) of the SARS-CoV spike protein-encoding sequences were designed based on the published sequences (GenBank accession no. NC_004718). The PCRs were carried out with an initial denaturation step of 94°C for 5 min followed by 30 cycles of denaturation (94°C for 1 min), annealing (50°C for 1 min), and extension (68°C for 4 min), with a final prolonged extension step (68°C for 10 min). The amplified coding sequences were inserted into the pET101/ D-TOPO vectors to generate the plasmid, pET101/ D-TOPO-S 268 . Recombinant plasmid DNA was sequenced and correct coding sequence was confirmed.

The ability of antisera from immunized mice to inhibit SARS-CoV virus infection of Vero E6 cells was assessed by virus neutralization assay. To prepare virus stock solution for the neutralization assays, Vero E6 cells were infected with SARS-CoV (GenBank accession no. GI 29836496) and incubated at 37°C in 5% of CO 2 for 3 days. The infectious virus stock with TCID50, was calculated using the method of Reed and Muench (1938) [11] , and 1Â10 7 TCID50 of the virus stock solution was aliquoted into individual tubes and stored at )70°C. For virus neutralization testing, 2Â10 5 Vero E6 cells/ml was inoculated onto a 12-well tissue culture plate (Falcon #3043, 96-well flexible plate) at 37°C in 5% of CO 2 overnight. Preimmune serum and antiserum of BALB/c raised against the recombinant SARS-CoV spike protein were pre-treated at 56°C for 30 min to destroy heat-labile, non-specific viral inhibitory substances, and diluted to the beginning dilution 1/ 20 with DMEM maintenance medium, then added into a well containing 2Â10 4 TCID50 of the virus in a volume of 0.15 ml. MAbs were diluted to 50, 20, 5, and 1 lg per 0.15 ml. Equal volumes of the serum or mAb solution and the test-virus dilutions were mixed and incubated at 37°C for 1 h. Then, the serum-virus mixtures and virus controls (no sera) were inoculated into Vero E6-containing culture plates, which had been pre-washed with DMEM maintenance medium and emptied just prior to the addition of serum-virus mixtures into culture plates. After absorption at 37°C (or at 4°C for virus entry-inhibition assay) for 2 h, the wells were washed with DMEM maintenance medium and then 2% of fetal calf serum/DMEM buffer was added. After 24 h incubation at 37°C, the wells were washed twice with PBS (pH 7.4), lyses buffer was added. The microplates were stored at )70°C until SDS-PAGE/Western blot analysis for the presence of virus replication. The 90% virus neutralization titer is calculated when the SARS-CoV viral protein were not detected in SDS-PAGE/Western blot analysis.

For the identification of purified rS 268 protein, equal amounts of lysated from Escherichia coli BL21 containing mock or induced recombinant spike protein, and purificated protein were boiled and loaded onto an SDS-PAGE as described below. Purified fusion proteins were detected with mouse antihistidine antibody (1/1000; Amersham Pharmacia Biotech) by Western immunoblot analysis. For confirmation of the neutralization effect of antiserum against the recombinant SARS-CoV spike protein, equal amounts of virus-infected cell lysates (1/100 of total lysates, 10 ll) were boiled in a sample buffer (125 mM Tris-HCl, [pH 6.8], 100 mM DTT, 2% SDS, 20% glycerol, 0.005% brophenol blue) for 5 min, and then loaded onto an 8% SDS-polyacrylamide gel. After electrophoresis of SDS-PAGE, specific proteins were detected with BALB/c anti-rS268 polyclonal antibody (our laboratory) or mouse anti-b actin monoclonal antibody (Sigma, cat. no. SI-A5441) by Western blotting described previously [12] .

Recombinant proteins were expressed and purified using the denature method, as described previously with some modification [12] . In brief, Escherichia coli BL21DE3 bacteria, containing the plasmid pET-S 268-1255 , were used for prokaryotic expression and purification of histidine-tagged proteins. After dialysis and brief centrifugation in a Sigma 3K12 centrifuge 5402 (14,000 rpm for 5 min) at 4 C, the supernatant was quickly frozen in liquid nitrogen and stored at )80°C.

Normal healthy BALB/c and rat sera were obtained from the Animal Center of the Institute of Preventive Medicine of the National Defense Medical Center, Taiwan. All animals were confirmed healthy by a licensed veterinarian. Antisera against spike protein were prepared by three subcutaneous inoculations (15, and 5 lg of recombinant proteins mixed with Freund's adjuvant for rat, and BALB/c, respectively; complete adjuvant and incomplete adjuvants for the first, and for the second and third inoculations, respectively, at an interval of 1 month). One month after the last immunization, the animals were bled and the sera centrifuged at 3000 Â g for 10 min at 4°C in a Sigma 3K12 centrifuge with a Nr. 12154 rotor after 2 h of agglutination at room temperature. Antisera were collected and mixed with 50% glycerol and stored at )20°C.

For immunization, BALB/c mice received an intraperitoneal (i.p.) injection of 5 lg of recombinant spike protein (rS 268 ) in 100 ll of PBS emulsified with an equal volume of complete Freund's adjuvant. After an interval of 2 weeks and 4 weeks, booster injections were given as above, except that we used incomplete Freund's adjuvant instead. Three weeks after the third injection, final boosters containing 5 lg of antigen were administered via i.p. injection. Fusion was performed 5 days after the last injection with the spleen cells of the donor mouse. Hybridomas secreting anti-spike antibodies were generated according to the standard procedure [13] . Hybridoma colonies were screened by ELISA, and selected clones were subcloned by the limiting dilution method. Immunoglobulin classes and subclasses were determined using subtyping kit (Roche Diagnostics, Penzberg, Germany). Ascitic fluids were produced in pristane-primed BALB/c mice. Monoclonal antibodies were purified using protein A affinity chromatography, and stored in 1 lg mAb per ml at )80°C.

To check the SARS-CoV spike protein-recognizing capabilities of mAbs against recombinant spike protein, an IFA [14] was performed. Briefly, the Vero E6 monolayer cells infected with SARS-CoV were washed three times with PBS and then fixed in acetone-Methanol mixture (1:1) for 3 min at room temperature. After blocking with 3% skim milk in PBS for 1 h at room temperature, an IFA assay was performed. Each stained monolayer was viewed with an immunofluorescent microscope (LEICA, DMIRB).

The 96-well microtiter plates (Falcon, #3912) were coated with antigens (0.2 lg/ml purified recombinant rS268-truncated protein [268-1255 amino acid]) in 0.05 M carbonate buffer, pH 9.6) at 4°C overnight. Bound mAbs were detected using secondary antibody (goat anti-mouse IgG-HRP, 37°C for 1 h). After washing, ABT (Boehringer) or (TMB or OPD (o-phenylediamine, Sigma, P-6787)) substrate was added and stayed for 30 min (ABT, 15 min for TMB, and 30 min for OPD) at room temperature. Reaction was stopped with 1N H 2 SO 4 , and specific SARS-CoV IgG was detected by OD 405nm (ABT, OD 405nm for TMB) endpoint reading. Experiments were performed as described previously [12] . Sera were always assayed in duplicate; each plate also included an air blank, as well as negative control and positive controls.

For the competitive-inhibition assay, different concentrations of synthetic peptides (CB119, CB119IA, SP-SGNCD, SP-SGIAA, SP-DLG, and SP2; as illustrated in Table 1 ) were mixed with mAbs before being transferred to the rS268 recombinant protein-coated plates and incubated for 1 h. After washing, the plates were incubated with HRP-conjugated goat anti-mouse Ig antibodies (1:2000 in PBST) (ICN Biomedicals) at 37°C for 30 min, and the procedures described above were followed.

Synthetic peptide (SP 1143-1172 , 100 lg) was labeled with Digoxigenin (Roche, Digoxigenin-3-O-methylcarbonyl-e-aminocaproic acid-N-hydroxysuccinimide ester, cat. no. 1 333 054) according to the instruction of manual. Labeled peptide was purified by passage through a NucTrap purification column (Strategen, cat. no. 400702) according to the manufacture's instructions. It was then concentrated to 100 ll (1 lg/ll) using CENTRICON centrifugal filture devices (Millipore, cat. no. YM-3000) according the instruction of manual.

For gel migration shift assay, 1 ll of Digoxigeninlabeled probe (SP 1143-1172 -Dig) was incubated with mAbs (1 ll of Mab 5, 1/25 and 1/100; or Mab 6, 1/ 25) in the presence or the absence of cell lysates (Vero cell lysates, 1/100 of total lysates; or SARS-CoV virus-infected cell lysates, 1/500, 1/200, and 1/ 100 of total lysates), in the reaction buffer (20 mM Hepes, pH 8.0, 25 mM KCl, 10% glycerol, 2 mM MgCl 2 , 0.5 mM dithiothreitol) for 30 min at room temperature. Then 2 ll of 10Â sample dye (0.5 Â TBE, 10% glycerol, 0.25% bromophenol blue, 0.25% xylene cyanol) was added in and probe-protein complexes were separated from the free probe by electrophoresis at 4°on a 5% native polyacrylamide gel. After electrophoresis, the gel was transferred to Hybond-C extra membrane. The membrane was blocking and further incubated with Anti-Digoxigenin-POD, Fab fragment (Roche, cat. no. 1 207 733). After finally washing with blocking buffer, the membrane was developed with ECL Western Blotting Reagent as described above in Western Immunoblotting Analysis.

Microtiter plates coated with different fusion proteins or synthetic peptides (D2-TM, S2-Fc, S3, rRBD2, D1 and D2 long peptide, CB116-CB123, CB124-131, CB132-138, SP1, SP2, PEP508, PEP509, and NP; 1 lg/well in carbonate buffer at 4°C for overnight) were incubated with diluted mAbs (1-1, 3-2, 5-1, and 8-1; 1:5000 in PBS) at 37°C for 1 h after blocking. Bound mAbs were detected using secondary antibody (goat antimouse IgG-HRP, 37°C for 1 h). After washing, ABT (Boehringer) or TMB substrate was added and stayed for 30 min (ABT, 15 min for TMB) at room temperature. Reaction was stopped with 1 N H 2 SO 4 , and specific SARS-CoV IgG was detected by OD 405nm (ABT, OD 405nm for TMB) endpoint reading. Each plate also included blanks and negative controls.

Background reactivity and possible cross-reactivity were assessed by analyzing pre-immune serum specimens from healthy rabbits, mice, and rats. The cutoff values were set at: ODn+3 SD, where ODn is the mean of ODs recorded or the preimmune serum or mock specimens. Those ODs greater than the calculated threshold ODs regarded as positive sera and all others regarded as negative. Synthetic peptides and recombinant S fragments used in monoclonal antibodies mapping

The recombinant S fragments are: D1-TM (D1 fragment is residues 74-253 then linked with 8 Gly as linker to TM fragment residues 1130-1255) expressed in Escherichia coli; D2-TM (D2 fragment is residues 294-739 then linked with 8 Gly as linker to TM fragment residues 1130-1255) expressed in Escherichia coli; S1-Fc (S1 fragment is residues 1-333 linked to human Fc fragment of IgG1) expressed in baculovirus system; S3 (residues 667-999) expressed in Escherichia coli; rRBD2 (residues 294-739) expressed in Escherichia coli. CB116-CB138 peptide sequences are from residues 1128-1255, 15-mer overlapping by 10 residues (Kelowna, Taipei, R.O.C.) ( Table 1) .

The coding sequence, corresponding to amino acids 268-1255 of the S protein of SARS coronavirus, was amplified using RT-PCR and cloned into the prokaryotic expressional vector containing a 6-histidine tag coding sequence, pET101/D-TOPO, to produce pET101/D-TOPO-S 268 plasmid. The expression of recombinant protein was induced by IPTG and the recombinant protein was purified to near homogeneity using Ni 2+ -NTA agarose affinity chromatography (Figure 1a , lane 5 for rS 268 ). Purified recombinant protein was confirmed with mouse antihistidine antibody as described in Materials and Methods. As shown in Figure 1b , purified rS 268 protein indeed to be a histidine-tagged fusion protein.

To determine whether Escherichia coli expressed recombinant spike protein could conserve its immunological and biological properties, SARS CoV-specific antibodies from SARS patients were used to detect the purified recombinant with ELISAs. Using rS 268 as coating antigen, ELISAs were performed to measure SARS-specific IgG in the sera of controls and suspected SARS patients. Recombinant protein rS 268 was recognized by the IgG of the SARS patients (S4, #0612, S45, S25, S18, Pt-Shi, and Pt-Yea), but not by control sera (S3, S31 and #284) (Figure 1c ). In short, the recombinant protein could be detected by SARS-specific IgG in SARS-CoV infected patients implicated that rS 268 might conserve its antigenic characterization and epitopes.

The polyclonal antibodies from rat, and BALB/c sera raised against purified rS 268 protein, were tested and confirmed to recognize SARS coronavirus spike proteins using Western blot, and indirect immunofluorescent assay (IFA) (Figure 2 , panel a, and b). As illustrated in Figure 2a , mouse anti-rS 268 antiserum recognized the spike glycoprotein in SARS-CoV-infected Vero E6 cell lysate (lane 2). The antibody reactivity titer against SARS-CoV spike protein by ELISA was found to be above 25,600 (for rat anti-rS 268 IgG), and 68,260 (for BALB/c anti-rS 268 IgG) after the last immunization (data not shown). When mAbs raised against rS 268 were assayed for their capabilities to recognize SARS-CoV spike protein using an IFA assay. As illustrated in Figure 2b , mAbs (Mab 1, 3 and 5) specifically recognized SARS-CoV in virus-infected Vero E6 cells (panels E, D, and F). To characterize monoclonal antibodies, the isotype of each mAb was analyzed using IsoStrip mouse mAb isotyping kit (Roche). As shown in Table 2 , most of mAbs were IgG1.

SARS patient's serum also recognized SARS-CoV infected Vero cells in the IFA assay ( Figure 2b , panel H), while sera from mouse preimmune or control patient did not (panels A, G). Taken all results together, the antisera against rS268 could recognize and bind to the spike protein of SARS-CoV.

In order to characterize the neutralization of antisera raised against recombinant SARS spike proteins, virus-neutralization assays were performed to test the efficacy of BALB/c mice mAbs raised against rS 268 . As shown in Figure 3a Table 2 . Results revealed that recombinant rS 268 protein maintained some conserved antigenic epitopes of SARS-CoV spike glycoprotein, so that antisera and mAbs raised from this recombinant protein were capable of neutralizing the infection of Vero E6 cells by SARS-CoV.

To elucidate the mechanism of neutralization as well as to map the neutralizing epitope of spike protein, a systematic epitope-mapping assay was performed using synthetic peptides derived from S protein and mAbs by ELISA. As shown in Figure 4a , neutralizing mAbs (Mab 1, 3, 5, and 8) bound to recombinant S protein fragments common with the trans-membrane domain (D1-TM and D2-TM). Further epitope mapping shown that all these neutralizing mAbs recognized peptides covering residues 1128-1168 (CB 116-123), as illustrated in Figure 4b . Actually, all these neutralizing mAbs specifically bound to an individual synthetic peptide, CB119, which only comprised of a 15-amino acid (SPDVDLGDISGINAS) and was located at the tip of one heptad repeat region (HR2) of the SARS-CoV spike protein (Figure 4c ). The binding of antibodies to this heptad region was specific and sufficiently to neutralize the infection of VERO E6 cells by SARS-CoV, since all strong neutralizing mAbs (Mab 1, 2, 3, 5, 7, and 8) bound to CB119 peptide but not for the weak or non-neutralizing mAbs (4, 6, 9, 12, and 13) (Figure 5a ). Results from this study revealed that mAbs binding to the heptad repeat region of spike protein could sufficiently neutralize the infection of SARS-CoV. 

In order to reveal the neutralization mechanism of mAbs against spike protein of SARS-CoV, different synthetic peptides mimicking the neutralization epitope (CB119, corresponding to 1143-1157 amino acid of spike protein), its mutant, and other controls were used to test their competition abilities for inhibiting the interaction of mAbs and spike protein. As illustrated in Figure 5b , CB119 peptide, but not control peptide SP2, inhibited neutralizing mAbs (Mab 1 and 2) binding to spike protein in a dose-dependent manner. Meanwhile, CB119 peptide could not efficiently inhibit non-neutralizing mAb (Mab 6) binding to spike protein. The competitive inhibition of mAbsspike protein interaction by CB119 was specific, since control peptides (SP2, SP-SGNCD, SP-SGIAA, SP-DLG) had no inhibition effect (Figure 5c ). Further characterization of mAb-HR2 interaction by ELISA, as illustrated in Figure 5d , neutralizing mAbs have differential features for their binding to synthetic peptides mimicking various amino acid residues of heptad repeat region of spike protein. For example, some mAbs (Mab 1, 5, 7, and 8) bound to CB119IA peptide almost equally to wild type peptide (CB119); some did not (Mab 2 and 3) . Results from these data implicated that neutralizing mAbs inhibiting the infection of SARS-CoV through specially binding to the amino acid residues 1143-1157, located on the HR2 region, of spike protein.

Neutralizing monoclonal antibodies bind to synthetic peptides mimicking amino acid residues of HR2 region of spike protein in a dose-dependent manner Aforementioned results implicated that mAbs might neutralize the infection of cells by SARS-CoV through specially binding to HR2 region of spike protein. To test this hypothesis, an EMSA assay was performed to examine the binding characterization of mAbs to synthetic peptides. This experiment used a Digoxigenin-labeled synthetic peptide (SP 1143-1172 -Dig) as a probe. When there is no binding protein, it should migrate faster in polyacrylamide gel. Once there is binding protein (or antibody), the peptide-antibody complex should migrate slower than free peptide. Both free peptide and peptide-antibody complex should bind to POD-conjugated Anti-Digoxigenin secondary antibody and be detected by ECL substrate. As illustrated in Figure 6 , synthetic peptide probe (SP 1143-1172 -Dig) representing HR2 heptad repeat region specially bound to neutralizing mAb (Mab 5) in a dose-dependent manner (lanes 8-9), but not to non-neutralizing mAb (Mab 6) (lane 10). When lysates of SARS-CoV infected cells was added in, the peptide-mAb complex was competed by SARS-CoV virus-infected cell lysates in a dosedependent manner (lanes 4-6). Result from this study showed that mAbs specifically recognizing the heptad repeat region might neutralize the infection of cells by SARS-CoV in a dose-dependent manner with differential features. Thus, it implicates that it might be practicable to prevent human infection of SARS-CoV by recombinant subunit vaccine or synthetic peptides mimicking amino acid residues of heptad repeat region of spike protein.

The SARS pandemic had great impacted on the health and economic integrity of countries all over the world associated with transmission of this novel pathogen. Effective vaccines and drugs are important research candidates for the prevention and therapy of this novel viral disease. Therefore, the infection and the viral entry mechanism is important target for these works.

In the present studies, we examined the biological properties of Escherichia coli expressed recombinant spike protein and the immunological prop-erties of small animal antisera raised against recombinant rS 268 proteins compared to those obtained from SARS patients. Mouse mAbs generated from immunizing recombinant spike proteins were evaluated for their biological activities in virus neutralization against SARS-CoV, and then mapped out and characterized the virus neutralization epitope(s).

Since SARS-CoV spike protein was a glycoprotein, it could be argued that lack of protein modification in prokaryotic cell expression systems would not retain the native conformation and folding of S protein. However, as shown in Figure 1c , the purified recombinant proteins appeared to be recognized by antibodies from the sera of SARS-CoV infected patients. Further, the antisera induced by recombinant rS 268 proteins did Table 1 ) were incubated with diluted mAbs (1/1000 in PBST) at 37°C for 1 h after blocking. All bound mAbs [in (a) (b) (c) (d)] were detected by ELISA, as described in Materials and Methods. Error bars indicate standard deviations of duplicate tests. Each plate also included blanks, as well as negative controls. recognize and captured SARS-CoV spike glycoprotein in Western blot, IFA, and capture sandwich ELISA (Figure 2a , b, and data not shown). Actually, several lines of evidence demonstrated that spike protein of SARS-CoV interacted with human ACE2 receptor in a glycosylation-independent manner [15] . A recombinant spike protein expressed in Escherichia coli could induce protective immunity against SARS-CoV [15] . Therefore, the recombinant rS 268 proteins were valuable and their induced antisera actually recognized the major characteristics and biological functions of spike protein (as shown in Figures 1, 2) , even though they might lose some minor conformationdependent antigenic epitopes related to glycosylation.

Small-animal antisera raised against recombinant spike proteins were evaluated for their neutralization effect in Vero cells infection by SARS-CoV. Rabbit antisera generated from recombinant S full immunization had potent virus neutralization activity comparable to SARS patients' sera (data not shown). Meanwhile, mouse immunized with rS 268 did generate virus neutralizing antibodies against SARS-CoV. In the present studies, some of mAbs against recombinant spike protein had strong neutralization effect as seen in the virus neutralization assay; but some of mAbs had little or no neutralization effect (Table 2, Figure 3a was further supported that different mAbs have various binding activity for synthetic peptides mimicking spike protein (Figure 5c and d) . As 10 lg per 10 4 pfu/10 5 cells monoclonal antibodies [1, 5, 8, 16, 17] recognized the epitope (residues 1143-1157), could sufficiently neutralize SARS-CoV infection through blocking the virus entry into Vero E6 cells. The neutralization epitope (residues 1143-1157) mapped in this study was located at the tip of the heptad repeat region of spike protein, the HR-C domain. Thus, these mAbs likely neutralized the infection of SARS-CoV through inhibiting virus entry into Vero E6 cells with a mechanism that was different from virus-ACE2 receptor binding blocking mechanism reported by others [15] . Because viral-cellular membrane fusion mediated by several type I viral fusion proteins (for example, HIV gp160) could be inhibited by peptide mimicking either the HR-N or HR-C regions [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] . Author suggest that the HR1 and HR2 regions are crucial in the entry of SARS-CoV, and support the assumption that SARS-CoV may employ the same fusion mechanism as HIV-1 and other viruses with class I envelope proteins [28] . Studies indicate that after SARS-CoV binding to the target cell, the transmembrane spike protein might change conformation by association between the HR1 (residues 902-952; 915-950; 889-926) and HR2 (1145-1184; 1151-1185; 1161-1178) regions to form an oligomeric structure, leading to fusion between the viral and target-cell membranes [28] [29] [30] [31] . Therefore, HR2 has been suggested to use as a potent inhibitor peptides or a target for the development of therapeutic drug for SARS-CoV [28] [29] [30] [31] [32] [33] . In this study, mAbs bound to HR-C regions and might inhibit the HR-N and HR-C forming coiled-coil structure that was required for viral-cellular membrane fusion, so neutralized the infection of SARS-CoV by blocking virus entry into Vero cells. However, our data showed rabbit antisera immunized with recombinant spike protein did not react with the CB119 peptide (data not shown). These rabbit antisera could have different virus neutralization effects. Recently, angiotensin-converting enzyme 2 (ACE2) was reported as a functional receptor for the SARS coronavirus in Vero E6 cells by Li et al. [34] . ACE2 receptorassociated epitope on SARS-CoV spike protein had been identified to be within residues 310-527 6-7) . The probe-antibody complex was separated from the free probe by electrophoresis on a 5% non-denaturing polyacrylamide gel. After this, Western blotting was finished as described in the Materials and Methods. Position of the probe-antibody complexes are indicated on the right. [15] . Further studies should be investigated to map the neutralization epitopes recognized by the rabbit antisera.

In summary, mouse mAbs raised against recombinant spike protein were effective and sufficient for neutralization of SARS-CoV infection against Vero E6 cells. Result from this study showed that mAbs, specifically recognizing the heptad repeat region of spike protein, might neutralize the infection of cells by SARS-CoV through binding to the neutralizing epitope located on the HR-C of spike protein. Further, monoclonal antibodies differently bind to synthetic peptides mimicking HR2 region implicated presence of several potential neutralizing epitopes. Thus, it might be practicable to prevent human infection of SARS-CoV by recombinant subunit vaccine or synthetic peptides mimicking optimal amino acid residues, such as virus neutralization epitope SPDVDLGDISGINAS, of heptad repeat region of spike protein.

Identification of a novel coronavirus in patients with severe acute respiratory syndrome

A novel coronavirus associated with severe acute respiratory syndrome

Coronavirus as a possible cause of severe acute respiratory syndrome

Coronaviridae: The Viruses and their Replication

The V5A13.1 envelope glycoprotein deletion mutant of mouse hepatitis virus type-4 is neuroattenuated by its reduced rate of spread in the central nervous system

Porcine respiratory coronavirus: molecular features and virus-host interactions

A clustering of RNA recombination sites adjacent to a hypervariable region of the peplomer gene of murine coronavirus

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

Identification of an immunodominant linear neutralization domain on the S2 portion of the murine coronavirus spike glycoprotein and evidence that it forms part of complex tridimensional structure

Neutralization-resistant variants of a neurotropic coronavirus are generated by deletions within the amino-terminal half of the spike glycoprotein

Improvements in methods for calculating virus titer estimates from TCID50 and plaque assays

Development of a recombinant protein-based enzyme-linked immunosorbent assay and its applications in field surveillance of rodent mice for presence of immunoglobulin G against Orientia tsutsugamushi

Continuous cultures of fused cells secreting antibody of predefined specificity

Orientia tsutsugamushi suppresses the production of inflammatory cytokines induced by its own heat-stable component in murine macrophages

Potent neutralization of severe acute respiratory syndrome (SARS) coronavirus by a human mAb to S1 protein that blocks receptor association

Short Protocols in Molecular Biology 2

Design of a novel peptide inhibitor of HIV fusion that disrupts the internal trimeric coiled-coil of gp41

The coronavirus spike protein is a class I virus fusion protein: structural and functional characterization of the fusion core complex

Structure-function study of a heptad repeat positioned near the transmembrane domain of Sendai virus fusion protein which blocks virus-cell fusion

C-Terminal gp40 peptide analogs inhibit feline immunodeficiency virus: cell fusion and virus spread

An antiviral peptide targets a coiled-coil domain of the human T-cell leukemia virus envelope glycoprotein

A synthetic all D-amino acid peptide corresponding to the Nterminal sequence of HIV-1 gp41 recognizes the wild-type fusion peptide in the membrane and inhibits HIV-1 envelope glycoprotein-mediated cell fusion

A synthetic peptide corresponding to a conserved heptad repeat domain is a potent inhibitor of Sendai virus-cell fusion: an emerging similarity with functional domains of other viruses

Peptides derived from the heptad repeat region near the C-terminal of Sendai virus F protein bind the hemagglutinin-neuraminidase ectodomain

Analysis of the murine leukemia virus R peptide: delineation of the molecular determinants which are important for its fusion inhibition activity

Analysis of a peptide inhibitor of paramyxovirus (NDV) fusion using biological assays NMR, and molecular modeling

Structural characterization of the human respiratory syncytial virus fusion protein core

Suppression of SARS-CoV entry by peptides corresponding to heptad regions on spike glycoprotein

Interaction between heptad repeat 1 and 2 regions in spike protein of SARS-associated coronavirus: implications for virus fusogenic mechanism and identification of fusion inhibitors

Structural Characterization of the SARS-Coronavirus Spike S Fusion Protein Core

Characterization of the heptad repeat regions HR1 and HR2, and design of a fusion core structure model of the spike protein from severe acute respiratory syndrome (SARS) coronavirus

Severe acute respiratory syndrome coronavirus (SARS-CoV) infection inhibition using spike protein heptad repeat-derived peptides

Following the rule: formation of the 6-helix bundle of the fusion core from severe acute respiratory syndrome coronavirus spike protein and identification of potent peptide inhibitors

Angiotensinconverting enzyme 2 is a functional receptor for the SARS coronavirus

This work was supported by a grant from the Institute of Preventive Medicine, National Defense Medical Center, Taiwan, R.O.C. awarded to Y.-C. Wang. We are very grateful to Mr. Lih-Jeng Tarn and Ms. Yu-Ying Huang for helping to take care of small animals and maintain the cell cultures, respectively. We also thank Dr. Ching-Len Liao for some material assistance. Part of this research was also supported by the National Science Council, Taiwan, Grants #SVAC12-06 (Chong, P).

Figure A.1. Neutralizing monoclonal antibodies bind to synthetic peptides mimicking aminoacid residues of HR2 region of spike protein in a dose-dependent manner. Neutralizing monoclonal antibodies specially bind to HR2 region of spike protein in a dosedependent manner. A Digoxigenin-labeled synthetic peptide (SP 1143-72 -Dig) was used as a probe in the gel migration shift assay (EMSA). The presence (+) or absence ()) of synthetic peptide probe (SP 1143-72 -Dig) and mAbs (Mab 5 and 6) are indicated above each lane of the photograph. For the interaction, 1 ll of SP 1143-72 -Dig probe and mAbs (Mab 5, 1/100, 1/10, and 1/1 for lane 2-4; or Mab 6, 1/1 for lane 5) were incubated at room temperature for 20 min. The probe-antibody complex was separated from the probe by electrophoresis on a 5 nondenaturing polyacrylamide gel. After this, Western blotting was finished as described in the Materials and Methods. Position of the probe-antibody complexes are indicated on the right.