Mimotopes selected with neutralizing antibodies against multiple subtypes of influenza A

C179 monoclonal antibody (A/H2N2 subtype) was purchased from Takara Bio Inc (Otsu, Shiga, Japan); Mouse monoclonal antibody (IV.C102) against influenza virus A strain H1N1 was from Santa Cruz (Santa Cruz, CA, USA); Purified H3N2 goat polyclonal IgG specific to influenza A/Texas 1/77 was from Virostat (Portland, ME, USA); SIV sera were prepared from patients hospitalized by swine-origin influenza virus A/2009 and their binding activities were tested by ELISA. Endotoxin was removed by purification with polymyxin B chromatography. Endotoxin levels were < 0.1 unit/μg of protein by the Limulus Amebocyte Lysate QCL-1000 pyrogen test (Cambrex).

Ph.D.-7, Ph.D.-12 and Ph.D.-C7C were produced by New England Biolabs, Inc (Ipswich, MA, USA), with random linear 7-mer, 12-mer or cyclic 7-mer peptides fused to minor coat proteins (pIII) of M13 filamentous phages.

C179 (0.2 ml, 10 μg/ml) was coated on three wells of 24-wells microplate at 4°C overnight. The coated wells were blocked with 2% bovine serum albumin (BSA) at 37°C for 1 h, then washed with Tris•HCl buffer solution (TBS) containing 0.1% Tween-20 (TBST) for 6 times. Ten microliter Ph.D.-7 (2 × 10¹¹ pfu), Ph.D.-12 (1.5 × 10¹¹ pfu) and Ph.D.-C7C (2 × 10¹¹ pfu) libraries diluted with 0.2 ml TBST were dropped into the coated wells respectively. The incubation wells were rocked gently for 30 min, followed by discarding nonbinding phages. The binding phages were eluted with 0.2 M Glycine-HCl (pH 2.2), 1 mg/ml BSA and neutralized with 1 M Tris•HCl (pH 9.1). The eluted phages were used to infect log-phase bacteria ER2738, concentrated by PEG precipitation and submitted to the second round of selection. The following selections were performed as above except that the Tween-20 concentration was raised from 0.1% to 0.5% in the wash steps. After 3 rounds of selection, the eluted phages were used for plaque isolation. 32 plaque clones were amplified for ELISA test and single-strand DNA preparation.

IV.C102 (0.2 ml, 10 μg/ml) was coated on three wells of 24-wells microplate at 4°C overnight. The blocking and bio-spanning procedures were carried as did in C179 antibody, except that the binding phages were eluted with IV.C102 (0.2 ml, 10 μg/ml). After 3 rounds of selection, the eluted phages were used for plaque isolation.

H3N2 polyclonal antibody (30 μg/ml) and goat IgG (100 μg/ml) were coated respectively. After blocking with BSA and washing with TBST for 6 times, Ph.D.-7, Ph.D.-12 and Ph.D.-C7C libraries were added into the goat IgG-coated wells for 30 min incubation. Then, the nonbinding phages were transferred to H3N2 polyclonal antibody-coated wells for additional 30 min incubation. After that, the non-binding phages were discarded, and the binding phages were eluted with H3N2 polyclonal antibody. The eluted phages were amplified in bacteria ER2738 and used in the following selections as above.

Goat anti-human IgG (100 μg/ml) was coated on six wells. After blocking, three wells were added with diluted SIV sera (1:20), the others were added with human IgG (100 μg/ml). Then they were placed at 4°C overnight and washed with TBST for 6 times. Peptide phage display libraries were added into the IgG wells for 30 min incubation. Subsequently, the nonbinding phages solutions were transferred to the sera wells for 30 min incubation. Then, the binding phages at the sera wells were eluted with goat anti-human IgG. The eluted phages were used for plaque isolation at the end of the 3rd selection.

Ninety-six-well plates were coated with mAb and BSA (10 μg/ml, 100 μl) by incubation at 4°C overnight, and blocked with 5% BSA in TBS. Affinity-selected phage were added to the wells and allowed to bind at 37°C for 1 h. After the unbounded phages were removed with 5% TBST, the bound phages were detected by incubation with peroxidase-labelled murine anti-M13 antibodies (Pharmacia). The bound peroxidase was determined by incubation with Opheny lenediamine dihydrochloride (Pierce Chemicals) in buffer (30 mM citrate, 70 mM Na2HPO4, and 0.02% H2O2, pH 5.5). When the reaction was stopped by the addition of 3 N HCl, A450 nm was determined with an ELISA reader (BioRad). All the assays were carried out in triplicate.

The phage from the 3rd biopanning eluate was cloned for immune-analysis. The nucleotide sequence of the gene III insert was determined as the instruction manual. The amino acid sequence of the insert was deduced from the nucleotide sequence and was compared with native influenza A. The sequences that appeared > 3 times among the selected phage clones were classified as the consensus sequence. The aligned amino acid sequences shared by three or more identical amino acids within the dodecapeptides (heptapeptides) were determined as the mimotopes of the matched protein sequences.

The 7- and 12-mer mimotopes of H1N1, H2N2, H3N2 and SIV were linked by GSGGS with the mimotope sequences of SIV7-SIV12-H1N17-H1N112-H3N27- H3N212-H2N27-H2N212. Each mimotope represents the peptide with the highest frequency on phage surface. The codon usage was optimized by species preference and GC content. The gene was synthesized with Eco R I/Bam H I enzyme site by Sangon (Shanghai, China).

The multi-mimotope gene was cut with Eco R I and Bam H I endonucleases, and ligated separately into precut pGEX-2 T-1, pGEX-4 T-1 and pET43a (+) plasmids. The ligation products were used to transform competent Trans 109 E. coli cells, which were selected on LB-AMP agar plates at 37°C for 12 h. Three AMP-resisitant clones were picked randomly for plasmids extraction, Eco R I/Bam H I digestion and gene sequencing. The confirmed plasmids with correct insert were transformed into competent BL21 (DE3) E. coli cells for protein expression. Four transformed bacteria BL21 (DE3) clones were picked from LB-AMP agar plates for culturing overnight in LB-AMP medium. The resultant bacteria were inoculated into 5 ml fresh medium, cultured to mid-log growth phase for protein expression induction with 1 mM IPTG. After 12 h induction, the bacteria sample was aliquot for SDS-PAGE analysis.

One hundred milliliters of BL21 (DE3) E. coli cells containing pET43a (+)-multi-mimotope plasmids were induced with IPTG for 12 h. The resultant medium was centrifuged with 8000× g for 10 min and the pellet was resuspended into 10 ml of 20 mM TBS (pH 7.9). The cells were broken by ultrasounding with 100 W for 100 s, followed by centrifuging to remove the cell debris. The supernatant was filtered through 0.22 μm membrane and then loaded onto pre-equilibrated Ni²⁺-NTA-resin. Then, the resin was rinsed by TBS containing 5 mM imidazole; the binding protein was eluted by TBS containing variable imidazole.

The recombinant multi-mimotope was coated on 96-well microplate with concentration of 10 μg/ml at 4°C overnight, followed by blocking with 2% BSA at 37°C for 2 h. The bait antibodies including C179, H1N1 monoclonal antibody, H3N2 polyclonal antibody and SIV sera were added into wells separately to incubate with coated protein at 37°C for 2 h. The wells were washed with PBST for 6 times. Then, HRP-conjugated secondary antibodies were added for binding at 37°C for 2 h. The TMB solution was then added into the wells for color development, which was stopped with 3 N HCl. The unrelated protein BSA was coated as control protein to determine the binding specificity of multi-mimotope to the antibodies.

To evaluate the potential of the selected mimotopes as experimental vaccine candidates, purified phage mimotopes were used to immunise female inbred specific-pathogen-free BALB/c mice through intraperitoneal administration.

The multi-mimotope protein was concentrated to 1 mg/ml and injected intraperitoneally (50 μg, 100 μg, 200 μg per mouse) or subcutaneously (100 μg per mouse) into BALB/c mice (9 per group) as emulsion (1:1) with complete Freund's adjuvant (CFA) for the first immunization and with incomplete Freund's adjuvant (IFA) for the booster injection at 14 days later. The control group were injected with PBS. Ten days after the booster injection, except for control group, other 5 groups were challenged with 2 × 10³ f.f.u A/Puerto Rico/8/1934 (H1N1) by intranasal inoculation of 50 μl per mouse. Mice were weighed on the day of virus challenge and then every three days for two weeks. Two weeks after the challenge, lungs were removed for pathological examination. Blood samples were taken to measure serum Ab titers by ELISA. Animals were conducted and approved by the Institutional Animal Care and Use Committee of the Academy of Military Medical Science, under protocol number 0054921. All experiments were performed according to institutional guidelines.

The concentrations of IgG Abs against H1N1 influenza virus were measured by ELISA. Purified antigen was coated on the microtitre plates (100 μl/well, 5 μg/ml in coating solution, 0.1 M sodium bicarbonate, pH 9.6) (Corning, Corning, NY, USA) and incubated at 4°C overnight. Serial 2-fold dilutions of sera (100 μl/well) from each group of unimmunized or immunized and immunized infected were incubated for 1 h at 37°C. Goat anti-mouse IgG HRP (1:10,000 dilution with washing buffer) was used to detect IgG Abs and O-phenylenediamine dihydrochloride (Pierce Chemicals) was used as substrate for HRP and the reaction was monitored at an absorption of 492 nm using an ELISA reader (Labsystems Multiskan, Finland).

Lung tissue samples were fixed in 10% formalin and embedded with paraffin,sections were cut at 5 μm thickness and were stained with hematoxylin eosin (HE).

The data from test groups were evaluated by Student'st-test. The survival rates of mice in test and control groups were compared by using Fisher's exact test. All differences were considered significant at P values0.05.

With H2N2 monoclonal antibody C179 as bait protein, binding phages from Ph.D.-7, Ph.D.-12 and Ph.D.-C7C peptide phage-display libraries were enriched by three rounds of binding-elution-amplification. Thirty-two binding phage clones were picked up randomly from every library for ELISA testing with C179 coated and uncoated wells (BSA blocked). Twelve phage clones with higher ELISA signal ratio (C179 to BSA) were chosen for single-strand DNA preparation and DNA sequencing. As shown in Figure 1, the overall binding affinity of linear 7-mer and cyclic 7-mer peptide-displayed phages to C179 was much lower than that of linear 12-mer group. The clone H2-12-22 had the highest ELISA signal ratio in linear 12-mer group, which was above 10. In linear 7-mer group, the clone H2-7-18 had the highest ELISA signal ratio of 3.89. In cyclic 7-mer group, the clone H2-C7-29 had the highest ELISA signal ratio of 2.55 (Figure 1). The selected phages were amplified in bacteria ER2738 and their single-strand DNA was extracted for gene sequencing. The peptide sequences were summarized in Table 1. The dominant sequences were considered as the mimotopes of the influenza A virus A/Okuda/57 strain, according to C179 monoclonal antibody. The linear heptopeptide WHWRLPS, linear dodecapeptide WHTHKWSLSAKA and cyclic cysteine-restricted heptopeptide NLSSSWI had low similarity (Table 1).

The dominant sequence displayed on linear heptopeptide phages binding to IV.C102 H1N1 monoclonal antibody was QWTWTQY, whereas the linear dodecapeptide was DCWQMDRKTCPL, cyclic heptopeptide was PLHARLP. All sequences and their frequencies were listed in Table 2.

The dominant sequence displayed on linear heptopeptide phages binding to swine-origin influenza virus A sera was ETKAWWL, whereas the linear dodecapeptide was QAHNWYNHKPLP, cyclic heptopeptide was PLHARLP. All sequences and their frequencies were listed in Table 3.

The dominant sequence displayed on linear heptopeptide phages binding to influenza A virus A/H3N2 polyclonal IgG was WPWHNHR, whereas the linear dodecapeptide was VWSTPPHADGPA, cyclic heptopeptide was LGALSHT. All sequences and their frequencies were listed in Table 4.

The multi-mimotope SIV7-SIV12-H1N17-H1N112-H3N27-H3N212-H2N27- H2N212 of influenza A covered the mimotopes of H2N2, H1N1, H3N2 and SIV subtypes. Each subtype contained heptopeptide and dodecapeptide mimotopes. The whole amino acids sequence and nucleotide sequence were shown in Table 5

The multi-mimotope gene was subcloned into pGEX-2 T-1, pGEX-4 T-1 and pET43a (+) expression plasmids, respectively. Their reading frames were confirmed by EcoR I/BamH I digestion and gene sequencing. After transforming into BL21 (DE3) E. coli cells and inducing with IPTG, the multi-mimotope genes were expressed as GST-fused protein with pGEX-2 T-1, pGEX-4 T-1 plasmids or as Nus/His6-fused protein with pET43a (+) plasmid. The proteins were of 42KD and 73KD in respective fused forms.

Large portion of Nus/His6-fusion multi-mimotope was produced as soluble form; in contrast, the GST-fusion proteins were expressed as insoluble form. So, BL21 (DE3) E. coli cells containing pET43a (+)-multi-mimotope plasmids were chosen for induction with IPTG. The Nus/His6-fusion multi-mimotope with endogenous His6 tag was bound by Ni²⁺-NTA-resin and eluted with TBS containing 60 mM imidazole (Figure 2a). The imidazole gradient was further investigated with the consequence that fusion protein could be eluted completely between 10 to 30 mM imidazole in TBS (Figure 2b). The eluted protein was dialyzed against TBS containing 5 mM imidazole and loaded on pre-equilibrated Ni²⁺-NTA-resin for further affinity purification. This time, the eluted protein was of high purity (Figure 2c).

The results (Figure 3) showed that the recombinant protein could be bound by H2N2 monoclonal antibody (C179), H1N1 monoclonal antibody (IV.C102), goat H3N2 polyclonal antibody and human SIV sera, respectively. But the control protein BSA could not.

Fourteen days after A/Puerto Rico/8/1934 (H1N1) virus challenge, 9 of the 9 mice in PBS group, 7 of the 9 mice in 50 μg multi-mimotope i.p. group, 5 of the 9 mice in 100 μg multi-mimotope i.p. group and 8 of the 9 mice in 100 μg multi-mimotope s.c. group died. However, only 2 of the 9 mice in 200 μg multi-mimotope i.p. group died (Figure 4a). Six days after virus challenge, mice in PBS group lost 32% body weight, compared with 19% in 50 μg multi-mimotope i.p. group, 16% in 100 μg multi-mimotope i.p. group, 8% in 200 μg multi-mimotope i.p. group and 23% in 100 μg multi-mimotope s.c. group (Figure 4b). These findings suggested that multi-mimotope effectively reduced infection of influenza virus A.

Sera were collected and pooled from mice infected with A/PR8 (H1N1) influenza virus. The ELISA method was used to detect IgG Ab titers against H1N1 influenza viruses. As shown in Figure 5, the immune responses to PBS and unimmunized control mice were very low. For the mice, immunization with 200 μg multi-mimotope induced antigen-specific Abs. The specific IgG Abs in immunized mice after infected with H1N1 influenza virus were significantly increased.

Gross lesions were observed in the lungs of mice of the unimmunized infected groups,including pulmonary hyperemia,hemorrhage and consolidation. Importantly, the lung inflammation levels of the 200 μg multi-mimotope immunized infected group were significantly decreased compared with those of the matched unimmunized infected groups. Necrotizing interstitial pneumonia was found with light microscope in all cases of death in the unimmunized infected groups. The pulmonary tissue exhibited hyperemia, emorrhage and inflammatory exudation, leading to consolidation. The lumina of alveoli and bronchioles were variably filled with protein-rich edema fluid, fibrin, erythrocytes and cell debris, admixed with many neutrophils and lymphocytes in the unimmunized infected groups (Figure 6).