key: cord-0024565-yrslx0my
authors: Li, Hua; Ding, Jian; Chen, Ying-Hua
title: Recombinant protein comprising multi-neutralizing epitopes induced high titer of antibodies against Influenza A virus
date: 2004-11-10
journal: Immunobiology
DOI: 10.1078/0171-2985-00244
sha: 4f85e618b81aea607f458c81c74c7b80eb740825
doc_id: 24565
cord_uid: yrslx0my

In previous studies, we suggested that epitope-vaccine might be a new strategy against virus infection. Based on this hypothesis, we designed and expressed a recombinant immunogen (multi-epitope-peptide) comprising repeats of three neutralizing-epitopes (neutralizing epitopes: aa92 – 105, 127 – 133 and 183 – 195) of hemagglutininin (HA) of influenza virus (H3N2) in E. coli. After vaccination, the recombinant multi-epitope protein could induce a high level of antibodies with predefined multi-epitope-specificity in mice and rabbits. The epitope-specific antibodies in sera were tested using three different epitope-peptides (synthetic peptides) in ELISA assay, and the serum dilutions from 1 : 6400 to 1 : 25600 were confirmed. In western blot analysis, both the antiserum and the antibodies purified by synthetic epitope-peptide coupled sepharose columns could recognize natural HA from influenza virus particles (strain A/Wuhan/359/95 H3N2). In hemagglutination inhibition (HI) tests, these three antisera at the dilutions from 1 : 20 to 1 : 80 showed inhibitory activity. Interestingly, antisera and purified antibodies induced by the epitope-vaccine could partially inhibit plaque-formation of influenza virus (strain A/Wuhan/359/95) on MDCK cell monolayers. These results suggest that the recombinant multi-epitope vaccine can simultaneously induce multi-antiviral activities against influenza virus, which may provide a new way to develop effective vaccines against influenza virus.

Influenza is a mild febrile illness in general, but it causes severe symptoms and sometimes death in those who have underlying diseases or in the aged. Large influenza epidemics, even at the present time, become serious public health problems all over the world. It has been generally considered that neutra lizing antibodies are cardinal for protection against influenza A infection (Virelizier, 1975; Ada & Jones, 1986 ). However, the effect of currently available vaccines against influenza is extremely variable (Cox & Subbarao, 1999) . To solve the problem, intrana sal immunization (Corrigan & Clancy, 1999) , new adjuvants (Martin, 1997) and increasing the dose of hemagglutinin (HA) and neuraminidase (NA) anti gens with purified protein have been used (Couch et al., 1997) . In previous studies, we suggested epitopevaccine as a new strategy to induce multi-antiviral activities (Chen et al., 1999; Xiao et al., 2001) .

The membrane protein HA of influenza A viruses, mediating the attachment of the virus to target cells and the release of viral content into attached cells, is the major target of neutralizing antibody responses (Wilson & Cox, 1990; Skehel & Wiley, 2002) . After decades of research, several neutralizing epitopes on HA have been identified. The peptide (HA aa127-133) could induce neutralizing antibodies and T cell response after introduction into an MHC II binding component (Naruse et al., 1994; Ogasawara et al., 1992) . The peptide , when con jugated to tetanus toxin (TT), can induce both humoral and Tcell responses and protect Balb/c mice from lethal virus infection (Simeckova-Rosenberg et al., 1995) . The peptide (HA aa91 -108) can also elicit immune responses and lead to partial protec tion (Muller et al., 1982; Shapira et al., 1984) . Intersubunit region of HA (aa333-357) is immu nogenic in Balb/c mice without the administration of any carrier (Rajnavolgyi, 1992) . Hybrid flagella containing one B, TH and CTL (cytotoxic T lym phocyte) epitope of influenza virus showed signifi cant protection against sub-lethal viral challenge (Simeckova-Rosenberg et al., 1995; Jeon et al., 2002) .

Although epitope-vaccines prepared by synthetic epitope-peptides have shown some distinct proper ties and advantages in comparison with the tradi tional vaccine (Lu et al., 2002) , it is necessary to develop a new method together with syntheticpeptide-based epitope-vaccine to fulfill the multi epitope-vaccine strategy. In this study, we designed and expressed a recombinant immunogen contain ing three neutralizing epitopes on [127] [128] [129] [130] [131] [132] [133] [183] [184] [185] [186] [187] [188] [189] [190] [191] [192] [193] [194] [195] of influenza virus (H3N2), and wanted to induce high levels of antibodies with predefined multi-epitope-specificity against influen za virus.

Three epitope-peptides (P1, P2 and P3) containing neutralizing epitopes on HA of influenza A virus (strain A/Wuhan/359/95, H3N2) and a control peptide (CP) were commercially synthesised in Genemed Synthesis Inc. (USA). p 1: C-(KAYSNCY-PYDVPDYG)2, HA a a 9 2 -105; P2: C-(GHHPITDSDQTRLY)2, HA aa1 8 3 -195; P3: C-(GWTGVAQD)2, HA aa12 7-133; CP: (KGGG)7-K. Peroxidase-conjugated rabbit anti-mouse and goat anti-rabbit immunoglobulin were obtained from Dako (Denmark); mAb His • Tag was obtained from Novagen (USA); control mAb 10E3 was prepared in our laboratory (Tian et al., 2001) . Three epitope-specific antibodies against P1, P2 and P3 were purified from the rabbit antisera against the peptides P1, P2 and P3 using peptide-coupled NHSactivated sepharose columns.

Based on the nucleotide sequence of influenza virus A/Wuhan/359/95, three oligonucleotide frag ments encoding repeats of HA peptides a a 9 2 -105, 183-195 and 127-133 respectively, and five pri mers were synthesized (Sangon, China) ( Table 1) . Primers F1/R1 were used to amplify the O1 fragment while primers F2/R2 were used to amplify the O2 fragment. 100 ng of each product were mixed together and used as templates for overlap extension PCR using primers F1/R2. After two steps of PCR, four restriction enzyme sites (BamHI, SalI, BglII, XhoI) were introduced into the linked DNA frag ment O 1-O 2. The fragment was inserted into pGEM T-easy vector (Promega, USA) and repeated once using BamHI, BglII and XhoI. The primers F3/ R2 were used to amplify the fragment O3. A fragment of four repeats of O3 was generated as mentioned above and linked downstream to the fragment (O1 -O2)2 . The target gene sequence was determined by the didoxynucleotide chain termina tion method.

After cutting from the recombinant T-easy vector using BamHI/XhoI, the multi-epitope gene was ligated into similarly treated expression vector pET-28a (Invitrogen, USA) to yield recombinant plasmid pET-epitope. Competent E. coli strain BL21 cells were then transformed and grown in 2 x YT agar medium containing 60 pg/ml kanamycin sul fate. Overnight cultures of the recombinant bacteria having plasmid pET-epitope were diluted 1 : 50 in 200 ml 2 x YT mediums and grown at 37 °C for about two hours. 1 mM isopropylthiogalactoside (IPTG) was added to induce protein expression. After a further 4 hours incubation under vigorous agitation, E. coli cells were pelleted for further purification.

The cell paste was resuspended in 20 ml ice-cold PBS and sonicated twice for 2 minutes with 10 minutes rest on ice. The pellet was collected by centrifugation at 7800g, resuspended in PBS containing 0.1% Triton X-100, sonicated briefly, and centrifuged at Table 1 . Oligonucleotides and primers used for generation of the multi epitope antigen gene containing three different neutralizing epitopes. Primers F1/R1 or F2/R2 were used for amplifying DNA fragments O1 and O2 corresponding to ( KAYSNCYPYDVPDYG)2 and (GHHPITDSDQTRLY)2 separate ly. Primers F3 and R2 were used to amplify DNA fragment O3 corresponding to ( <5WTGVAQD)2

2000g. The inclusion bodies were then solubilized in 50 mM sodium phosphate buffer (PBS), pH 7.5, containing 0.5 M sodium chloride, 5 mM imida zole, and 6 M urea (buffer A). Insoluble material was removed by centrifugation at 17,000g. The supernatant was loaded onto a Ni2+-NTA column pre-equilibrated with buffer A. After further equili bration, recombinant protein was eluted with the same buffer containing 250 mM imidazole. The refolding procedure on the Ni2+-NTA column was used. After dialyzing in buffer A, purified protein was reloaded onto the column and refolded by applying a 1000 ml reverse gradient of urea (6 to 0 M) in the same buffer at a flow rate of 2 ml/minute at room temperature. The column was then washed with 80 ml of 50 mM PBS, pH 7.5, containing 0.5 M sodium chloride, 5 mM imidazole. Refolded protein was eluted in the same buffer with 250 mM imidazole. After dialyzing with 20 mM Tris-HCl, pH 7.8, the purified recombinant fusion protein was tested by western blot analysis as described in Ding et al. (2000) .

After mixing with complete Freund's adjuvant, each of four mice was immunized with 10 pg recombi nant protein intraperitoneally while each of three rabbits was immunized with 50 pg antigen subcuta neously at axillea and groin. Boosters with the same dose were given in incomplete Freund's adjuvant on days 11, 22 and/or 33. Antisera were separated on the 5th day after the last booster. Pre-immune sera were collected before immunization. Epitope-speci fic antibodies were detected in the enzyme linked immunosorbent assay (ELISA) as described in Tian et al. (2001) .

Antiserum of rabbit was loaded onto peptide coupled sepharose-4B. After equilibration with PBS (pH 7.5), epitope specific antibodies were eluted with 0.1 M glycine (pH 2.7). Influenza virus particles (A/Wuhan/359/95) were subjected to elec trophoresis and electroblotted onto NC membrane. Binding of epitope specific antibodies to HA were detected by western blot analysis as described in Ding et al. (2000) .

Hemagglutination unit (HU) of the influenza virus was titrated using chicken red blood cells (RBCs). Allantoic fluid was serially diluted with PBS (pH 7.2) in V-shaped 96-well microtiter plates, 25 pl in each well. Another 25 pl PBS (pH 7.2) and 50 pl 0.5% chicken RBCs suspension in PBS were added to each well separately. A 4 HU/25 pl virus suspension was prepared according to the results of the HA titration and back titrated. Sera were treated with trypsin and NaIO4 to remove nonspecific agglutination inhibitors and nonspecific agglutinins. 25 pl of serially diluted serum was mixed with 25 pl 308 H. Li et al. of virus suspension and incubated for 1 hour at room temperature. 50 pl 0.5% chicken RBCs in PBS (pH 7.2) was added. The HI endpoint was taken as the highest dilution clearly demonstrating hemag glutination of chicken RBCs.

Madin-Darby canine kidney (MDCK) cells (ATCC) were cultured to monolayers in 24-well plates (Costar, USA). Influenza virus A/Wuhan/359/95, grown in the allantoic cavity of 9-day-old embryonated eggs, was serially diluted in DMEM (high glucose) (Gibco, USA) (40 x 10~3 to 4~5 x 10~3). After one hour incubation at 34 °C, the mixtures (100 pl per well) along with fresh medium (400 pl per well) were added to the plate and incubated for 2 hours at 34 °C. After removal of the medium, 1% agarose in DMEM was laid into each well. Non serum DMEM was added after the agarose solidi fied. The plate was stained by 1% crystal violet after 72 hours incubation at 34 °C. Plaque number was counted and plaque formation unit (PFU) in the viral stock was calculated. Plaque formation inhibition was performed by the same procedure. Serial dilu tions of antisera or antibodies were mixed with the same volume of medium containing 60 PFU of virus.

In this study, a gene of a recombinant multi-epitope peptide (immunogen) bearing three epitopes on influenza HA protein was designed and constructed.

The O1 and O2 fragments were joined together (O1 -O2) by using overlap-extension PCR, and then the O1 -O2 fragment was ligated into T-easy vector. Employing the enzymes of BamHI, BglII and XhoI, the O 1-O 2 fragment was repeated again and the O3 fragment was repeated four times after the repeats of O 1-O 2. The final composition of the gene fragment becomes (O 1-O 2)2-(O 3)4 (Fig. 1) . The sequence analysis of the recombinant DNA demonstrated that no alteration had occurred during the synthesis process, PCR amplification or cloning. After sub-cloning into plasmid pET-28a and trans formation into competent BL21 cells, the gene of the recombinant protein antigen was expressed in E. coli when induced with IPTG. Purified by Ni2+-NTA affinity column, the recombinant protein (34 kDa) containing His • Tag and three different epitopes of HA were identified in western blot assay using mAb His • Tag (Fig. 2, lane B) . Three epitope-specific antibodies purified from the rabbit antisera against the HA peptides P1, P2 and P3 could recognize the recombinant protein (34 kDa) respectively (Fig. 2 , lanes D, E and F), while the control antibody (negative control) did not (Fig. 2, lane C) . These results indicate that the recombinant protein (34 kDa) contains His • Tag and three different epitopes of HA.

After vaccination, the recombinant multi-epitope protein could induce a high level of antibodies with predefined multi-epitope-specificity in mice and rabbits. The epitope-specific antibodies in antisera were tested using three different epitope-peptides (synthetic peptides P1, P2 and P3) in ELISA assay, mouse antisera dilution up to 1:6400 -1 : 25600 for P1,1: 6400 for P2, while P3-specific antibodies were (20 pg) were subjected to electrophoresis in SDS-PAGE and electro-blotted onto nitro-cellulose membrane. The binding of specific antibodies to the recombinant protein was detected with peroxidase conjugated rabbit anti-mouse immunoglobulin and goat anti-rabbit immunoglobulin. Lane A, molecular weight marker; Lane B, binding of mouse mAb (His-Tag) to recombinant protein; Lane C, binding of control antibody (mouse antibody as negative control) to recombinant protein; Lane D, binding of peptide C-(KAYSNCYPYDVPDYG)2 specific rabbit antibodies to recombinant protein; Lane E, binding of peptide C-(GHHPITDSDQTRLY)2 specific rabbit antibodies to recombinant protein; Lane F, binding of peptide C-(GWTGVA QD)2 specific rabbit antibodies to recombinant protein; Lane G, binding of IgG purified from normal rabbit serum by protein-G column as negative control to recombinant protein.

undetectable (Fig. 3A) . Rabbit antisera were diluted up to 1:25600 for P1 and P3, 1:6400 for P2 (Fig. 3B) . In western blot analysis, rabbit antisera and three epitope-specific antibodies purified from the antisera by three sepharose columns coupled with the synthetic epitope-peptides P1, P2 and P3 could recognize natural HA from influenza virus particles (strain A/Wuhan/359/95) (Fig. 4, lanes B , C, D and E). In hemagglutination inhibition (HI) tests, these three antisera at the dilutions from 1 : 20 to 1:80 respectively showed inhibitory activity against influenza virus A/Wuhan/359/95 (Fig. 5) . Interestingly, the purified P1 and P2 specific-anti bodies could partly inhibit plaque-formation of influenza virus (strain A/Wuhan/359/95) on mono layer MDCK cells, and more than 50% reduction was shown at 10 pg/ml of these antibodies, while the P3 specific antibodies showed a lower inhibitory effect (Fig. 6 ). Besides these purified antibodies, these antisera could also inhibit plaque-formation of this influenza virus, at dilutions from 1: 100 up to 1:1000 (Fig. 7) .

Antigenic variation of HA is one of the most important characteristics of influenza virus (Wilson & Cox, 1990) . Only one amino acid mutation in each antigenic site of HA may generate a new epidemic strain (Wiley et al., 1981) . Currently used vaccines must be amended annually to follow the rapid change of epidemic strains. The main difficulty in design of a new influenza vaccine is to develop one that is capable of inducing a broad antibody response protective against infection by heterolo gous viruses. Many synthetic vaccines have been studied for induction of the neutralizing capacity, but the main trend is to develop a better component or more efficient adjuvant (Levi et al., 1995; Levi & Arnon, 1996) . We suggested epitope vaccine as a new strategy against viral infection (Chen et al., 1999; Xiao et al., 2001 ). An epitope vaccine com prising repeats of linear neutralizing B cell epitope had improved the immunogenicity of this epitope and induced a potent neutralizing antibody response (Lu et al., 2002) . We have succeeded in synchro nously inducing high levels of antibodies with predefined multi-epitope-specificity by a recombi nant multi-epitope-vaccine (Li et al., 2002) . In this study, the recombinant immunogen bearing three predefined epitopes of HA was identified by three 1: 400 1: 1600 1: 6400 1 25601 epitope-specific antibodies in western blot analysis (Fig. 2) . The recombinant immunogen elicited a strong humoral immune response with multi-speci ficities in mice and rabbits, the serum dilutions from 1:6400 to 1: 25600 were confirmed in ELISA assay. Binding of antiserum and three epitope-specific antibodies to natural HA in western blot analysis. Three epitope-specific antibodies were purified from rabbit antisera against the recombi nant immunogen by three sepharose columns coupled with the synthetic epitope-peptides P1, P2 and P3. Virus particles of strain A/ Wuhan/359/95 were subjected to electrophoresis and electroblotted onto NC membranes. Lane A, protein molecular weight maker; Lane B, rabbit antiserum induced by multi-epitope protein bind to HA; Lane C, P1-epitope specific antibodies bind to HA; Lane D, P2epitope specific antibodies bind to HA; Lane E, P3-epitope specific antibodies bind to HA; lane F, rabbit pre-immune sera did not bind to HA; Lane G, IgG purified from pre-immunized sera by protein-G column did not bind to influenza proteins.

These antisera and the purified epitope-specific antibodies could recognize natural HA in blotting tests. In a functional study, these three antisera at the 6 . Inhibition of plaque formation by epitope specific anti bodies. Three epitope-specific antibodies (Ab1, Ab2 and Ab3) were purified from rabbit antisera against the recombinant immunogen by three sepharose columns coupled with the synthetic epitopepeptides P1, P2 and P3. IgG as negative control was isolated by protein-G column from rabbit pre-immune sera. The figure shows the data from one of three separate experiments. 5 . Hemagglutination inhibition by rabbit antiserum induced by multi-epitope vaccine. All the antisera and the pre-immune sera were treated with trypsin and NaIO4 to reduce non-specific agglutinins and non-specific agglutination inhibitors respectively before use in the HI assay. Viral suspensions were prepared to contain 4 HUs per 25 pl PBS. Sera were serially diluted with PBS. Chicken RBCs in PBS were used to clarify the HI titres of the antisera and the pre-immune sera. Triple wells were repeated at each dilution. The figure shows the data from one of three separate experiments. Fig. 7 . Inhibition of plaque formation by antiserum induced by multi-epitope vaccine. Antisera of three rabbits and pre-immune serum were treated with trypsin and NaIO4 to reduce non-specific agglutinins and non-specific agglutination inhibitors before use. The figure shows the data from one of three separate experiments.

dilutions from 1:20 to 1:80 showed inhibitory activity in the hemagglutination inhibition test. These antisera and the purified epitope-specific antibodies could partially inhibit plaque-formation of influenza virus. These results suggest that the recombinant multi-epitope vaccine can simulta neously induce multi-antiviral activities against influenza virus, which may provide a new way to develop an effective vaccine against influenza virus. Experimental data from animal models showed that antibody response induced by whole influenza virion immunization might focus on only one epitope of HA (Lambkin & Dimmock, 1995 . This bias was more obvious after second and third immunization and did not broaden. The human immune system behaved in a similar way. By this token, the recombinant multi-epitope vaccine showed predominance by synchronously inducing a high level of antibodies with predefined multi epitope-specificity. Wang et al. (1986) found a relatively restricted response in human sera taken between 1969 and 1971 but not in sera taken in 1978. This indicated that mutations in one antigenic site might lead to virus escape in a subpopulation. A vaccine that can induce neutralizing antibodies against two or more different antigenic sites may dramatically decrease the possibility of escape. Sequence analysis within the H3N2 subtype had revealed the high conservation rate of all the sites of the HA1 epitope aa92 -104 with only one restricted variable residue (Lu et al., 2002) . Five more variable sites with restricted mutations were found in the other two HA1 epitopes aa127-133 and aa183-195 while the others are highly conserved (data not shown). As the candidate multi-epitope vaccine had shown an advantage in synchronously inducing antibodies with multi-epitope-specificities, broad protection against infection of heterogeneous viru ses might be expected by the multi-epitope-specific immune response induced by a candidate vaccine with more epitopes including these three epitopes.

The immune response to influenza infection

Epitope-vaccine: A new strategy against HIV-1

Is there a role for a mucosal influenza vaccine in the elderly?

Improvement of inactivated influenza virus vaccines

Candidate multi epitope vaccines in aluminum adjuvant induce high levels of antibodies with predefined multi-epitope specificity against HIV-1

Intranasal immunization with synthetic recombinant vaccine containing multiple epitopes of influenza virus

All rabbits immunized with type A influenza virions have a serum haemagglutination-inhibition antibody re sponse biased to a single epitope in antigenic site B

Longitudinal study of an epitope-biased serum haemagglutination-inhibition antibody response in rabbits immunized with type A influenza virions

Intranasal immunization of mice against influenza with synthetic peptides anchored to proteosomes

Synthetic recombinant influenza vaccine induces efficient long-term immu nity and cross-strain protection

Recombinant multi-epitope vaccine induces predefined epitopespecific antibodies against HIV-1

A candidate vaccine against influenza virus intensively improved the immunogenicity of a neutralizing epitope

Development of an adjuvant to enhance the immune response to influenza vaccine in the elderly

Anti influenza response achieved by immunization with a synthetic conjugate

Analysis of epitopic residues introduced into the hybrid peptide vaccines prepared according to the cassette theory

A strategy for making synthetic peptide vaccines

Synthetic peptides in the search for T-and B-cell epitopes

Immunity and protection against influenza virus by synthetic peptide corresponding to antigenic sites of hemagglutinin

Protection of mice against lethal viral infection by synthetic peptides corresponding to B and T-cell recognition sites of influenza A hemag glutinin

Influenza haemagglutinin

Induction of monoclonal antibodies with predefined epitope-specificity by epitope-vaccines

Host defenses against influenza: the role of anti-hemagglutinin antibody

Comparative analyses of the specificities of anti-influenza hemag glutinin antibodies in human sera

Structural identification of the antibody-binding sites of Hong Kong influenza haemagglutinin and their involve ment in antigenic variation

Structural basis of immune recognition of influenza virus hemagglutinin

Epitope-vaccine as a new strategy against HIV-1 mutation

Acknowledgements. This work was supported by the Fund of the S863-Program (2001AA215181) and the National Science Foundation for Outstanding Young Scientist of China (No. 30025038). Jian Ding and Hua Li contributed equally to this paper.