Induction of SARS-nucleoprotein-specific immune response by use of DNA vaccine

2.1. Mice

Female C3H/He mice, 4–5 weeks, were purchased from Shanghai Experimental Animal Center (Shanghai, China). Female adult New Zealand rabbits, 5–6 kg, were purchased from Nanjing Experimental Animal Center (Nanjing, China).

2.2. Cell line and transfection

C2C12 is a myoblast cell line derived from C3H mice. The cell line was cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco BRL) supplemented with 2 mM l-glutamine, 25 mM HEPES, 100 μg/ml streptomycin and 100 u/ml penicillin and 10% heat-inactivated fetal bovine serum (FBS) (Gibco BRL). Electroporation transfection was performed exactly as described by the manufacturer’s manual for Cell Porator II (Gibco BRL).

2.3. Construction of expression vectors

SARS-Cov nucleocapsid protein (also nucleoprotein) gene (n gene was mentioned in following text) was synthesized chemically based on sequence reported (AY304495, GeneBank). At the 5′(ATG) and 3′(TAA) ends of full coding region, EcoRI and BamHI restriction sites were introduced respectively for subcloning into pGEM3Z. The resultant plasmid (pGEM-fn) was digested with EcoRI and BamHI and a 1.2 kb released fragment was then ligated to pcDNA3.1 vector digested with the same enzymes. The recombinant plasmid (pcDN-fn) could express full length N protein in mammalian cells and was used as a DNA vaccine. pcDNA3.1 vector was used for vaccination control. To express recombinant SARS N protein in Escherichia coli for polyclonal antibody preparation, we used pfu DNA polymearase chain reaction with primer pair P1 (5′-GAATTCATATGTCTGATAATGGACCCC-3′) and P2 (5′-AAGGATCCTTATGCCTGAGTTGAATCAG-3′), and modified the SARS n gene in pGEM-fn by introducing NdeI and BamHI site to the 5′ end and 3′ end, respectively. pET-fn expression plasmid was made by ligating the NdeI/BamHI n gene fragment into pET24a vector digested with NdeI and BamHI. To make an expression plasmid of truncated N protein, pET-fn was cut with HindIII and BamHI to remove the region from 121 to 422 aa and then self ligated after blunting the ends by Klenow large fragment. The resultant plasmid (pET-pn) was able to express truncated N protein in E. coli. All constructs are summarized in Fig. 1 .

2.4. Polyclonal antibody preparation

After transformation of BL21(DE3) with pET-fn and pET-pn, the recombinants were induced by 0.2 mM IPTG for 5 h. Cell pellets were collected by centrifugation and lysed by 8 M urea and ultrasonic. The lysates were loaded onto DEAE-A50 column pre-equilibrated by Tris buffer (20 mM Tris–HCl pH 8.3, 5 mM 2-ME, 2 mM EDTA), and the fraction passing through was collected for further purification. Proteins in this fraction were resolved by SDS-PAGE gel electrophoresis and recovered from the gel by KCl staining method [16].

For making anti-N protein polyclonal antibodies, purified N proteins were mixed with Freund’s adjuvant and immunized sequentially in rabbits three to four times. The sera were collected a week after final immunization and the antibody titres were detected with enzyme-linked immunosorbent assay (ELISA).

2.5. Determination of serum IgG by ELISA

The mice were bled and their sera were assayed for antibody activity by standard quantitative ELISA. Briefly, 10 μg purified protein was coated in each well of ELISA plate by which captured reactive antibody from serum. The binding IgG was detected by HRP-labeled secondary antibody. A serum pool from several non-immune mice was used as normal serum control.

2.6. DNA vaccination of mice

Plasmid DNA for vaccination was prepared with a Qiagen plasmid purification kit. The purified DNA was dissolved in 20% sucrose at a concentration of 1 mg/ml. Female C3H/He mice were immunized by intramuscular injections with 0.05 ml of plasmid DNA in a separate site on each side of the quadriceps muscle, followed by one or two booster vaccination at 2 week intervals [17]. In all cases, a total volume of 50 μl was injected per muscle, using an insulin syringe with 5 G needle.

2.7. Splenocyte re-stimulation in vitro

For cytotoxicity assays, mice splenocytes were harvested 2 weeks after both two and three injection protocols. In vitro re-stimulation of primed splenocytes was achieved by C2C12-N, a C2C12 cell line expressing N protein made by transfection of pcDN-fn plasmid into C2C12 cells as described in the Section 2.2. The single-cell suspension of splenocytes was prepared from individual mouse spleen in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM l-glutamine, 50 μM 2-mercaptoethanol and 20 iu/ml of interleukin 2. A total of 3×10⁷ splenocytes were then re-stimulated with 6×10⁵ Mitomycin C-treated C2C12-N target cells for 5 days at 37 °C in 5% CO₂ prior to CTL assay. The target cells were been treated with 10 μg/ml of Mitomycin C for 5 h and washed 3 times.

2.8. Cytotoxic T-cell assay

Target cells were washed three times and added in triplicate at the rate of 1×10⁴ cells per well to 96-well round-bottom plates containing different amounts of re-stimulated effector cells. The effector:target ratios were adjusted from 100:1 to 25:1. After incubation at 37 °C in 5% CO₂ for 4 h, LDH activity released into the supernatants was determined according to the method published previously [18]. Maximal LDH release was made by treating cells with PBS containing 1% Triton-X100; spontaneous LDH release was determined from the wells containing the target cells and medium only. The percent specific lysis was calculated as follows: 100 × (experimental LDH release – spontaneous LDH release)/maximum LDH release – spontaneous LDH release).

3.1. Expression and purification of SARS-N protein in E. coli

To detect SARS-n gene expression in C2C12-N target cells and measure N protein-specific antibody produced in vaccinated mice, recombinant N protein and polyclonal antibody were prepared first. Full length and truncated N genes were inserted into pET24a expression vector, respectively. After induction with IPTG, as we expected, both pET-fn and pET-pn cell lysates showed obvious extra expression bands of ∼ 48 and 15 kDa, respectively, as shown in Fig. 2A and B . It was also found that both proteins were expressed in soluble form other than in inclusion body form. The results of Western blot assay showed that these two proteins could also react with the serum antibody of mice immunized with pcDN-fn DNA vaccine as shown in Fig. 2C. We purified the recombinant proteins by DEAE-A50 absorption and SDS-PAGE gel extraction method. The resultant pure proteins were injected into New Zealand rabbits for polyclonal antibody preparation and two kinds of antibodies (FN-Ab, anti-full length N protein, and PN-Ab, anti-partial N protein) were obtained. The titration of both antibodies was over 1:30,000 averagely.

3.2. Stable expression of N protein in C2C12 cells and as target cells for measuring CTL response

To determine expression of pcDN-fn in myoblast cells, C2C12 cells were tansfected with pcDN-fn plasmid by electroporation. G418 resistant stable lines were screened by Western blot assay and an N protein-expressible stable clone was amplified for target cell identification. Initially, we used PN-Ab as the first antibody and detected a unique band of size 48 kDa as shown in Fig. 3 . This suggests no post-translational modification of SARS-Cov nucleocapsid gene in C2C12 cells.

3.3. Induction of SARS-N protein-specific antibody

C3H/He mice were injected intramuscularly with 100 μg of plasmid DNA for two times at 0 and 2nd week or for three times at 0, 2nd and 4th week. Anti-N protein antibody was measured at the 6th week after the first injection. Sera were collected from the mice injected with the pcDN-fn construct or pcDNA3.1 control and the titers were quantified by ELISA with full length or truncated recombinant N proteins. As shown in Fig. 4 , anti-N protein antibodies were detected in mice vaccinated with pcDN-fn DNA. When the mice received two injections, specific antibody level was much lower than three injections. Antibody titre was also calculated and defined as dilution folds at which OD490 nm reading value was three times over control serum level (P/N>3). The result showed that the antibody titres of the two injections group were 1:200–1:3200, and averaged 1:985. For groups receiving three injections, titres were 1:3200–1:6400, and the average was 1:4580. The anti-N protein antibody raised by DNA vaccination also reacted specifically with recombinant N protein expressed in E. coli as shown in Fig. 2C.

3.4. Induction of CTL in vivo by pcDN-fn DNA

To examine CTL level in DNA-immunized mice, splenocytes were collected from individual mice 2 weeks after two or three injections with pcDN-fn and pcDNA3.1. Splenocytes cultures were re-stimulated in vitro for 5 days with Mitomycin C-treated C2C12-N cells. Cytolysis activity of the splenocytes was measured in a LDH release assay. When C3H/He mice were immunized by three injections, as expected, the group of vaccinated mice that received pcDN-fn developed a much higher level of target lytic activities: 67.4±8.4% (E:T=25:1), 69.6±6.7% (E:T=50:1) and 71.8±6.2% (E:T=100:1). In the case of two injections, CTL activities were also at a high level: 56.6±12.7% (E:T=25:1), 57.4±11.7% (E:T=50:1) and 63.0±6.3% (E:T=100:1) as shown in Fig. 5 (left part and right upper part). There was no significant difference between groups the received two or three injections. CTL activities of both groups were much higher than that of pcDNA3.1 control as shown in Fig. 5 (right lower part).