Virus-like particles of SARS-like coronavirus formed by membrane proteins from different origins demonstrate stimulating activity in human dendritic cells

Abstract

The pathogenesis of SARS coronavirus (CoV) remains poorly understood. In the current study, two recombinant baculovirus were generated to express the spike (S) protein of SARS-like coronavirus (SL-CoV) isolated from bats (vAcBS) and the envelope (E) and membrane (M) proteins of SARS-CoV, respectively. Co-infection of insect cells with these two recombinant baculoviruses led to self-assembly of virus-like particles (BVLPs) as demonstrated by electron microscopy. Incorporation of S protein of vAcBS (BS) into VLPs was confirmed by western blot and immunogold labeling. Such BVLPs up-regulated the level of CD40, CD80, CD86, CD83, and enhanced the secretion of IL-6, IL-10 and TNF-alpha in immature dendritic cells (DCs). Immune responses were compared in immature DCs inoculated with BVLPs or with VLPs formed by S, E and M proteins of human SARS-CoV. BVLPs showed a stronger ability to stimulate DCs in terms of cytokine induction as evidenced by 2 to 6 fold higher production of IL-6 and TNF-alpha. Further study indicated that IFN-gamma+ and IL-4+ populations in CD4+ T cells increased upon co-cultivation with DCs pre-exposed with BVLPs or SARS-CoV VLPs. The observed difference in DC-stimulating activity between BVLPs and SARS CoV VLPs was very likely due to the S protein. In agreement, SL-CoV S DNA vaccine evoked a more vigorous antibody response and a stronger T cell response than SARS-CoV S DNA in mice. Our data have demonstrated for the first time that SL-CoV VLPs formed by membrane proteins of different origins, one from SL-CoV isolated from bats (BS) and the other two from human SARS-CoV (E and M), activated immature DCs and enhanced the expression of co-stimulatory molecules and the secretion of cytokines. Finding in this study may provide important information for vaccine development as well as for understanding the pathogenesis of SARS-like CoV.

Figure 1. Confirmation of the incorporation of BS protein in BVLPs.

(A) Budding of BVLPs from insect cells. Sf21 cells were co-infected with recombinant baculovirus vAcBS and vAcME at a moi of 5. At 72 h post-infection, cells were harvested and fixed with 2% glutaraldehyde and then with 1% osmium tetroxide. Thin-section samples were prepared and stained with 1% uranyl acetate followed by examination under a transmission electron microscope. Arrows indicates BVLPs. Bar = 250 nm. (B) Western blot analysis of BVLPs. 5 µg purified BVLPs were analyzed by western blot using rabbit-derived anti-SARS-CoV antibody. An equivalent amount of purified human SARS CoV VLPs was run in parallel. Preimmune rabbit antiserum was used as negative control. Left part, band corresponds to S protein. Lane 1, negative control; lane 2, human SARS CoV VLPs; lane 3, BVLPs. Right part, band corresponds to E or M protein. Lane 1, human SARS CoV VLPs; lane 2, negative control; lane 3, BVLPs. The protein marker was Fermentas #SM0671. (C) Detection of BVLPs by immunogold labeling. The collodion-coated EM grids were loaded with purified BVLPs (left), SARS CoV VLPs (middle) or VLPs without BS or S protein (right) for 5 min. After removal of the excess sample solution, grids were incubated with antibody specific against BS2 protein for 1 h and then incubated with 15 nm gold conjugated anti-rabbit IgG.. The samples were stained with 2% PTA for 1 min, drained and examined under the EM. Arrow indicates the gold particles and triangle indicates the VLPs. Bar = 100 nm.

Figure 2. BVLPs induced phenotype changes in DCs.

Immature DCs (10⁶cells/ml) were incubated with 10 µg/ml of BVLPs, 10 µg/ml of LPS, 10 µg/ml of heated BVLPs, Ac (culture supernatant of sf21cells transfected with baculovirus expression vector pFastBac DUAL with gfp under the control of p10 promoter) or PBS. 16 h post incubation, cells were collected, washed, stained with indicated antibodies, and analyzed by flow cytometry. One representative experiment out of 3 is shown.

Figure 3. BVLPs induced secretion of cytokines in DCs.

DCs were incubated with BVLPs (10 µg/ml), heated BVLPs (10 µg/ml), LPS (10 µg/ml), Ac (culture supernatant of sf21 cells transfected with baculovirus expression vector pFastBac DUAL with gfp under the control of p10 promoter), or PBS. BVLPs + PMB or LPS + PMB: 10 µg/ml BVLPs or 10 µg/ml LPS was treated with 10 µg/ml polymyxin B at room temperature for 1 h before incubation with DCs. Following 16 h of treatment, supernatants were collected, and IL-6, IL-10 and TNF-α were analyzed using ELISA. Data are expressed as the mean±SD of triplicate samples.

Figure 4. Phenotypic characterization of DCs treated with BVLPs or with SARS CoV VLPs.

DCs were incubated with SL-CoV BVLPs or SARS CoV VLPs at a concentration of 10 µg/ml. DCs treated with LPS (10 µg/ml) and PBS were used as positive and negative control, respectively. The cells were harvested at 16 h post-inoculation and the expression of CD40, CD86, CD83 and CD80 were analyzed by flow cytometry. Isotype-matched controls were shown.

Figure 5. Cytokines secretions in DCs treated with BVLPs or with SARS CoV VLPs.

DCs were incubated with SL-CoV BVLPs or SARS CoV VLPs at a concentration of 10 µg/ml. DCs treated with LPS (10 µg/ml) and PBS were used as positive and negative control, respectively. After 16 h incubation, supernatants were collected, and IL-6, IL10 and TNF-α were analyzed using ELISA. Data are expressed as the mean±SD of triplicate samples.

Figure 6. Evaluation of CD4 T cell types stimulated by BVLPs-exposed DCs.

Immature DCs (1×10⁴) were exposed to BVLPs (10 µg/ml), Ac or PBS for 24 h and then co-cultured with CD4+T cells (1×10⁵) in 96-well plates in triplicate for another 72 h. The DCs/T mixtures were then collected for intracellular IFN-γ or IL-4 staining followed by flow cytometry analysis. IFN-γ and IL-4 positive populations in CD4 T cells were analyzed. Data are expressed as the mean±SD of triplicate samples.

Figure 7. Detection of SARS-CoV S-specific IgG and the subclasses in vaccinated mice.

Mouse sera (8 per group) were collected 10 days after the final immunization and assayed for IgG1 and IgG2a against the S2 protein of SARS-CoV. ELISA was used to detect the level of S-specific IgG antibodies. Recombinant S2 protein expressed in E. Coli was purified and used as the detection antigen. Data are presented as means±SD.

Figure 8. SARS-CoV S protein-specific IFN-γ and IL-4 ELISPOT.

The number of INF-γ or IL-4 -secreting cells was assayed by ELISPOT using splenocytes harvested from mouse spleens 10 days after the final immunization and stimulated in vitro with purified S2 protein. The results represent the averages of triplicate wells of 3 mice and are expressed as means±SD. (A) SARS-CoV S protein-specific IFN-γ-secreting cells. (B) SARS-CoV S protein-specific IL-4-secreting cells.