Since the two identified human coronaviruses (HCoV-OC43 and HCoV-229E) only triggered mild upper respiratory infection [19], the low-homologous motifs of their relative SARS-CoV, designated as S_1–217, S_425–513 and M_1–20 were first screened by comparing sequence between 40 different Chinese SARS-CoV isolates and 36 various coronaviruses. The SARS-CoV protein sequences in the NCBI GenBank database (Genbank AY274119) representing a Canadian Tor2 isolate were analyzed to identify potential B cell epitopes according to the algorithms concerning the hydrophilicity, surface probability, hydrophobicity, antigenic value, flexibility and secondary structure using DNASTAR software. Four segments designated as S_174–195, S_437–459, S_556–568 and M_1–20 were selected for construction of multi-epitope DNA vaccine.

The four selected epitopes from S and M protein were engineered into a DNA vaccine and separated one from another with AAY spacers to enhance appropriate epitope processing (Fig. 1 ). The multi-epitope gene was constructed using overlapping oligonucleotides in a PCR-based synthesis with the sequence as “AGATCT GCC ACC GAGAAGTCCGGCAACTTTAAGCACTTACGCGAGTTTGTGTTTAAGAACAAGGACGGCTTTCTGTACGCCGCCTACAACTACAAGTACAGGTACCTGAGACACGGCAAGCTGAGGCCCTTTGAGAGAGACATCTCCAACGTGCCCGCCGCCTACTCCGACTTCACTGACTCCGTTCGCGACCCCAAGACCTCCGCCGCCTACATGGCCGACAACGGCACCATCACCGTGGAGGAGCTGAAGCAGCTGCTGGAGCAGTGGAAC GAATTC”. The construct included a Kozak sequence (bold) at the N-terminus and the gene was optimized for mammalian codon usage according to the previous study [20]. The synthetic nucleotide sequence was then incorporated into expression vector pVAX1 (Invitrogen, USA), pcDNA4-his/myc (Invitrogen, USA) eukaryotic plasmid vector as well as pET-32a (Novagen, Germany) prokaryotic vector. The final plasmids identified as pVAX-epis, pcDNA4-his/myc-epis and pET-32a-epis were purified using Qiagen MegaPrep columns (Valencia, CA) according to manufacturer's instructions and were adjusted to 1 mg/ml in PBS buffer, pH 7.0.Fig. 1

Schematic diagram of the design of a multi-epitope DNA vaccine. The four selected B cell epitopes from S and M protein (S_174–195, S_437–459, S_556–568 and M_1–20) were screened from the SARS-CoV spike protein and M protein and engineered into a DNA vaccine separated one from another with alanine–alanine–tyrosine (AAY) spacers. Evaluation of the hydrophilicity, surface probability, hydrophobicity, antigenic value, flexibility and secondary structure of the multi-epitope chimera protein was performed by DNASTAR software. The synthetic nucleotide sequence was then incorporated into a pVAX1 vector. Codon optimization was employed in constructing this plasmid.

The pET-32a-epis plasmid was transformed into Escherichia coli strain BL21. The expressed chimeric protein with a 6xHis tag was purified by a Ni²⁺ affinity chromatography column according to the manufacturer's instructions (Novagen, Germany) and was eluted with PBS for future analysis.

293T cell was transfected with pcDNA4-his/myc-epis using Lipofectamine 2000 reagent. Cell lysates and supernatants were collected 48 h post-transfection. After separation by 10% SDS-PAGE, samples were transferred onto a nitrocellulose membrane by electroblotting. The membrane was incubated with monoclonal mouse anti-myc antibody (Santa Cruz, USA) at 4 °C overnight then with HRP-conjugated goat anti-mouse IgG (Santa Cruz, USA) at 37 °C for 1 h. After washing the membrane was developed with enhanced chemiluminescence Kit (Piece Corp., USA).

BALB/c mice were intramuscularly injected with 100 μg pcDNA4-his/myc-epis or mock plasmid, 36 h later; quadriceps were surgically removed and made frozen sections. The section was dried at 37 °C for 45 min, fixed with acetone then treated with 0.3% H₂O₂ in methanol to quench endogenous peroxidase. After blocking by normal serum, it was incubated with monoclonal mouse anti-myc tag antibody; biotinylated secondary antibody and ABC reagent were used following the protocol of ABC Kit (R.T.U. VECTASTAIN^® UNIVERSAL, Vector Lab) before DAB staining.

BALB/c mice at 6–8-week-age were obtained from Shanghai Laboratory Animal Center, Chinese Academy of Sciences. All animals were housed in the pathogen-free mouse colony and all animal experiments were performed according to the guidelines for the Care and Use of Medical Laboratory animals (Ministry of Health PR China, 1998) and the guidelines of the Laboratory Animal Ethical Commission of Fudan University. Mice were injected with 100 μg pVAX-epis or mock plasmid in 100 μl PBS into the tibialis anterior muscle by two legs on week 0, 3 and 6. Then 20 μg chimeric proteins in 50 μl PBS was administered subcutaneously (s.c.) on week 10 or 18 as a boosting immunization. To optimize the prime–boost strategy, 100 μg pVAX-epis DNA i.m. immunization was followed by three 20 μg chimeric proteins boosting on week 3, 5 and 7. Blood samples were collected by retro-orbital bleeding.

ELISA plates were coated with 10 μg/ml individual synthesized epitope peptide, mixed synthesized peptides or purified prokaryotic expressed SARS-CoV S protein fragment (125–683aa) (Shukang Biotechnology, Shanghai) at 4 °C overnight and blocked with 10% goat serum in PBST (PBS with 0.05% Tween 20) for 1 h. After washing serum sample were added in duplicate (1:100) and incubated at 37 °C for 1 h. After another three times of washing, HRP-coupled goat anti-mouse IgG (Sigma–Aldrich, USA) was added before color development with OPD. The reaction was stopped with 2 M H₂SO₄ and absorption at 490 nm was measured by a microplated reader (BioLab, USA).

Anti-sera (week 10) derived from mice receiving DNA priming 1× (week 0) and protein boosting 3× (week 3, 5 and 7) was used for Western blotting analysis and neutralization assay. Inactivated SARS-CoV(strain Z-1 (6) IV, China) lysates were provided by the biosafety level 3 laboratory, Wuhan institute of Biological Products, China and mixed with equal volume of loading buffer (50 mM Tris–HCl, pH 6.8, 100 mM DTT, 2% SDS, 0.1% bromophenol blue, and 10% glycerol). After separation by 12% SDS-polyacrylamide gel electrophoresis (PAGE), the protein band was then transferred onto a nitrocellulose membrane. After blocking with 5% skimmed milk in PBS for 2 h, the membrane was incubated with 1:100 diluted serum overnight at 4 °C and then with an HRP-conjugated goat anti-mouse IgG for 2 h at 37 °C before color development with enhanced chemiluminescence kit (Piece Corp., USA).

Virus neutralization assay was performed in the biosafety level 3 laboratory. Vero E6 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 1% fetal bovine serum, 100 U/ml of penicillin G per ml, and 100 μg/ml of streptomycin. Viral infection was performed by adding 100 μl medium containing 50 × TCID₅₀ of SARS-CoV (strain Z-1 (6) IV, China) or 50 μl serial twofold diluted serum premixed with 50 μl 50 × TCID₅₀ of SARS-CoV for 1 h to plate containing 4 × 10⁴ cells per well. Cytopathic effects (CPE) were observed 96 h afterwards by microscopy and confirmed by crystal violet staining. CPE of cells was blindly evaluated by two pathologists on a scale of 0 to 4, in which − represents 100% cell viability; + represents 0–25% cell death or apoptosis; + + represents 25–50% cell death or apoptosis; + + + represents 50–75% cell death or apoptosis; + + + + represents 75–100% cell death or apoptosis.

Data were presented as means and standard deviation. One-way ANOVA test was used to compare difference of antibody titers between all the groups of mice. Difference between two groups of mice was compared using wilcoxon-matched pairs test. A probability of less than 0.05 was taken as significant. Statistical calculation was performed using the GraphPad Prism (version 4.0) statistical program.

S and M protein were screened for potential B cell epitopes using epitope prediction software online and four candidate B cell epitopes designated as S_174–195 (EKSGNFKHLREFVFKNKDGFLY), S_437–459 (NYKYRYLRHGKLRPFERDISNVP), S_556–568 (SDFTDSVRDPKTS) and M_1–20 (MADNGTITVEELKQLLEQWN) were selected for construction of a multi-epitope DNA vaccine (Fig. 1).

In vitro expression of the multi-epitope gene was first demonstrated using pET-32a prokaryotic expression system. The corresponding multi-epitope protein fused with a trxA fragment and a 6xHis tag expressed in BL21 cells was purified by a Ni²⁺ affinity chromatography. It was shown in Fig. 2A that the multi-epitope construct was well expressed in vitro with a molecular weight about 28 kDa. The immunogenicity of this prokaryotic expressing protein was proved by s.c. injection of 20 μg chimera protein into mice emulsified with complete Freund's adjuvant (CFA) which resulted in induction of peptide-specific serum IgG 3 weeks post-immunization (Fig. 2B).Fig. 2

Prokaryotic expression and the immunogenicity of the chimeric multi-epitope protein. (A) Expression of the chimeric multi-epitope protein with pET-32a system. Purified protein by Ni²⁺ affinity chromatography (lane 1), BL21 lysates without IPTG induction (lane 2) and with 4 h induction (lane 3) were separated by 12% Gel and stained overnight with Coomassie Brilliant Blue G-250. (B) BALB/c mice were s.c. immunized with 20 μg chimera protein emulsified with complete Freund's adjuvant (CFA). 1:50 diluted sera samples were analyzed their reactivity to the 4-epitope mixtures by ELISA assay (*p < 0.01 compared with control group).

The multi-epitope chimera gene was also cloned into a pcDNA4-his/myc eukaryotic expression vector. The in vitro transfection efficiency was evaluated in 293T cell using Lipofectamine 2000 reagent. Multi-epitope chimera protein expressed in the cell lysates was confirmed by Western blotting (Fig. 3A). To assess the expression and distribution of the chimera gene in vivo, tibialis anterior muscles of mice were stained by anti-myc Ab 36 h after 100 μg DNA administration. As shown in Fig. 3B, the eukaryotic encoding multi-epitope gene was well expressed in vivo.Fig. 3

Eukaryotic expression of the chimeric multi-epitope protein in vitro and in vivo. (A) Western blotting analysis of the chimeric multi-epitope protein expressed in 293T cells. Presence of the chimeric protein in the cell lysates of 293T cells transfected with pcDNA4-his/myc-epis (1 and 2) or mock plasmid (3 and 4) was determined by an anti-myc monoclonal antibody (1 and 3) or an isoform antibody (2 and 4). (B) Immunohistochemistry analysis of the chimeric multi-epitope protein in vivo. Frozen sections prepared from the quadriceps of mice injected with mock plasmid (a) or pcDNA4-his/myc-epis (b) were reacted with an anti-myc monoclonal antibody and then stained with DAB. Magnification: 100×.

To determine the immunogenicity of the multi-epitope SARS-CoV DNA vaccine, BALB/c mice received 3 injections of 100 μg pVAX-epis or mock DNA by 3 weeks intervals. As shown in Fig. 4A, significant increase of IgG level was only observed in multi-epitope DNA immunized mice. The specific serum IgG titer increased with time and reached its peak at week 8 amounting to 512 then declined. A single 20 μg prokaryotic chimera protein boosting on week 10 or 18 quickly restored the humoral response to a higher extent (IgG titer more than 4000) indicating that the multi-epitope DNA vaccine elicited both a prime and memory immune response which lasted at least 3 months. However, priming with a mock DNA followed by protein boosting could only achieve a relatively weak immune response. Reactivity of the anti-sera to the individual epitope peptide showed that S_437–459 and M_1–20 peptides were highly immunogenic for the serum antibodies induction while S_174–195 and S_556–568 epitopes were relatively weak immunogenic (Fig. 4B). The immune sera could also recognize prokaryotically expressed S and M protein (data not shown).Fig. 4

Humoral immune response induced by the chimeric multi-epitope DNA vaccine. (A) Mice (n = 12) were injected with 100 μg pVAX-epis plasmid into the tibialis anterior muscle on week 0, 3 and 6. 20 μg prokaryotic expressed chimeric multi-epitope protein was s.c. injected on week 10 (n = 6) or week 18 (n = 6) as a boost immunization. Specific IgG level in the 1:100 diluted sera reactive to the mixed epitopes was measured by ELISA assay. Arrows indicate the time points of the 3 DNA injections (**p < 0.01 compared with other 2 groups; *p < 0.05, ^##p < 0.01 compared with mock DNA priming group). (B) Specific reactivity of the week 8 serum to the four individual predicted epitope peptides was investigated by ELISA assay (*p < 0.01).

Various DNA prime–protein boost strategies were investigated to promote the vaccine efficiency. Mice were primed with one DNA/protein injection (week 0) and boosted with three prokaryotic chimera protein s.c. immunization (week 3, 5 and 7). After the last protein boosting, extremely higher level of serum IgG (titer amounted to near 10⁶) was only seen in DNA priming/protein boosting immunized mice (Fig. 5 , p  < 0.01) indicating the multi-epitope SARS-CoV DNA vaccine elicited a potent SARS-specific priming antibody response which could be synergistically amplified by the following protein antigen immunization.Fig. 5

SARS-CoV–specific humoral immune response was greatly augmented using modified DNA prime–protein boost strategies. Four groups of mice were primed with 100 μg pVAX-epis DNA i.m. immunization or 20 μg chimeric proteins s.c. injection followed by three chimeric protein (20 μg) boosting on week 3, 5 and 7. The specific serum IgG titer was determined by ELISA assay (**p < 0.01 compared with protein priming (with adjuvant)).

The week 10 immune sera from mice treated with DNA prime–protein boost 3× (Fig. 5) were first investigated their binding capacity to SARS-CoV proteins by Western blotting assay. The 1:100 diluted serum was shown to bind to the protein in the lysates of viral infected Vero E6 cells with MW of 180 kDa, equivalent to MW of reported S protein (Fig. 6 ) and to prokaryotic expressed S protein (data not shown), indicating that the anti-sera could recognize the native S proteins expressed by SARS-CoV.Fig. 6

Western blot analysis of the reactivity of the immune-sera to SARS-CoV lysates. Cell lysates of Vero E6 cells without (lanes 2 and 4) or with SARS-CoV infection (lane 3 and 5) were separated on SDS-PAGE, lane 1 represents the pre-stained marker. Then lane 2 and 3 were hybridized with the anti-sera (week 10) derived from mice receiving DNA priming–protein boosting 3× immunization; while serum from mock DNA-immunized mice were reacted with lane 4 and 5 as negative control.