Severe acute respiratory syndrome coronaviruses with mutations in the E protein are attenuated and promising vaccine candidates

Abstract

Severe acute respiratory syndrome coronavirus (SARS-CoV) causes a respiratory disease with a mortality rate of 10%. A mouse-adapted SARS-CoV (SARS-CoV-MA15) lacking the envelope (E) protein (rSARS-CoV-MA15-ΔE) is attenuated in vivo. To identify E protein regions and host responses that contribute to rSARS-CoV-MA15-ΔE attenuation, several mutants (rSARS-CoV-MA15-E*) containing point mutations or deletions in the amino-terminal or the carboxy-terminal regions of the E protein were generated. Amino acid substitutions in the amino terminus, or deletion of regions in the internal carboxy-terminal region of E protein, led to virus attenuation. Attenuated viruses induced minimal lung injury, diminished limited neutrophil influx, and increased CD4(+) and CD8(+) T cell counts in the lungs of BALB/c mice, compared to mice infected with the wild-type virus. To analyze the host responses leading to rSARS-CoV-MA15-E* attenuation, differences in gene expression elicited by the native and mutant viruses in the lungs of infected mice were determined. Expression levels of a large number of proinflammatory cytokines associated with lung injury were reduced in the lungs of rSARS-CoV-MA15-E*-infected mice, whereas the levels of anti-inflammatory cytokines were increased, both at the mRNA and protein levels. These results suggested that the reduction in lung inflammation together with a more robust antiviral T cell response contributed to rSARS-CoV-MA15-E* attenuation. The attenuated viruses completely protected mice against challenge with the lethal parental virus, indicating that these viruses are promising vaccine candidates.

Importance: Human coronaviruses are important zoonotic pathogens. SARS-CoV caused a worldwide epidemic infecting more than 8,000 people with a mortality of around 10%. Therefore, understanding the virulence mechanisms of this pathogen and developing efficacious vaccines are of high importance to prevent epidemics from this and other human coronaviruses. Previously, we demonstrated that a SARS-CoV lacking the E protein was attenuated in vivo. Here, we show that small deletions and modifications within the E protein led to virus attenuation, manifested by minimal lung injury, limited neutrophil influx to the lungs, reduced expression of proinflammatory cytokines, increased anti-inflammatory cytokine levels, and enhanced CD4(+) and CD8(+) T cell counts in vivo, suggesting that these phenomena contribute to virus attenuation. The attenuated mutants fully protected mice from challenge with virulent virus. These studies show that mutations in the E protein are not well tolerated and indicate that this protein is an excellent target for vaccine development.

FIG 1

Schematic of mutations and deletions introduced within SARS-CoV E protein and growth kinetics of the mutant viruses (rSARS-CoV-MA15-E*). (A) The SARS-CoV genome is shown in the top, and the expanded region shows the E protein sequence and its different regions. White boxes represent the amino acids deleted within the E protein in each virus. Gray letters indicate the amino acids mutated to change the amino-terminal region of the protein. (B) Mutant virus growth kinetics. Subconfluent monolayers of Vero E6 and Huh7.5.1 cells were infected with wt, ΔE, and rSARS-CoV-MA15-E* viruses at an MOI of 0.001 on Vero E6 cells. Error bars represent standard deviations of the mean using data from three independent experiments. Hexagon, WT; diamond, ΔE; pentagon, Mut 1; triangle, Δ2; star, Δ3; inverted triangle, Δ4; square, Δ5; and circle, Δ6.

FIG 2

Subcellular localization of mutant E proteins. (A) Vero E6 cells were infected with either rSARS-CoV-MA15-ΔE, -Δ3, or -Δ5 or wt recombinant viruses, at an MOI of 0.3, and fixed at 24 hpi. E protein (green) and ERGIC (red) were labeled with specific antibodies. Nuclei were stained with DAPI (blue). Merge indicates superposition of both labels. Original magnification was ×126. (B) The panel represents the percentage of overlap coefficient between E protein and ERGIC, calculated with Leica LAS AF v2.6.0 software.

FIG 3

Stability of rSARS-CoV-MA15-E* after serial infections. The stability of rSARS-CoV-MA15-E* virus deletion mutants was examined after 8 passages in Vero E6 cells by sequence analysis. Asterisks denote the presence of nucleotide substitutions.

FIG 4

Weight loss and survival rate of mice inoculated with rSARS-CoV-MA15-E* mutants. BALB/c mice were intranasally infected with 1 × 10⁵ PFU of each virus (n = 5 mice). Animals were monitored daily for weight loss (A) and survival (B). Animals that lost more than 30% of their initial body weight were euthanized. Differences in weight loss between attenuated and virulent viruses were statistically significant (*, P < 0.05).

FIG 5

rSARS-CoV-MA15-E* growth in the lungs of infected mice. BALB/c mice were intranasally inoculated with 1 × 10⁵ PFU of the indicated viruses. At 2 and 4 days p.i., lung tissue was harvested and viral titers were analyzed in Vero E6 cell monolayers. Means and standard deviations are shown (n = 3 mice).

FIG 6

Lung pathology caused by infection with rSARS-CoV-MA15-E* mutants. BALB/c mice were intranasally inoculated with 1 × 10⁵ PFU of the indicated SARS-CoV deletion mutants and sacrificed at days 2 and 4 p.i. (A) Lungs were removed and sections were prepared and stained with hematoxylin and eosin. Asterisks indicate edema accumulation in both bronchiolar and alveolar airways. Original magnification was ×20. Representative images are shown. (B) Macroscopic lung pathology in rSARS-CoV-MA15-E*-infected mice. Representative images of lung gross pathology. (C) Lung weight. Three lungs were evaluated in each case (n = 3 mice). Statistically significant data compared to mice infected with the wt virus are indicated (**, P < 0.01).

FIG 7

Specific infectivity of rSARS-CoV-MA15-E* mutants. Vero E6 cells were independently infected with each of the rSARS-CoV-MA15-ΔE, -Δ3, and -Δ5 and wt viruses, at an MOI of 0.3. At 8 hpi, the culture medium was harvested and replaced with fresh medium supplemented with 2% FBS. Cell supernatants were collected at 11 hpi (“nascent virus”), and the levels of genomic RNA and infectious virus titer were analyzed by RT-qPCR and plaque assay, respectively. The ratio of infectious virus titer (PFU/ml) to genomic RNA is depicted in the graph as percentage of specific infectivity. Means and standard deviations are shown (n = 3 mice).

FIG 8

E protein stability of rSARS-CoV-MA15* mutants. Stability of E protein deletion mutants was evaluated in relation to that of the full-length E protein after infection of Vero E6 cells with rSARS-CoV-MA15-E* mutants. At 12 hpi, the cells were treated with cycloheximide. Cells were harvested at the indicated times, and the amount of E protein mutants was determined by Western blotting. (A) Membranes were probed with anti-E protein and with β-actin-specific antibodies as a loading control. (B) The graph represents the values obtained after densitometry analysis. The percentage of protein remaining after cycloheximide addition is represented. Bars represent standard deviations of the mean (n = 3 mice).

FIG 9

Interaction of SARS-CoV E protein deletion mutants with M protein. Vero E6 cells were cotransfected with a plasmid pcDNA3 encoding the N-terminal HA-tagged M protein, combined with plasmids expressing the E-wt protein or the E protein with a small deletion, E-Δ3. Cotransfections of a plasmid encoding HA-M protein and of a plasmid without E protein were used as controls. Cells were lysed and analyzed by Western blotting with specific antibodies for the E protein and HA (left) or subjected to immunoprecipitation with the monoclonal HA-specific antibody. The presence of E and M proteins was analyzed in the precipitated fractions using E- or HA-specific antibodies (right).

FIG 10

Leukocyte infiltrates present in the lungs of rSARS-CoV-MA15-E*-infected mice. BALB/c mice were intranasally infected with 1 × 10⁵ PFU of the indicated SARS-CoV deletion mutants and sacrificed at 4 days p.i. The total number of leukocytes (A) and total numbers and percentages of macrophages (B), neutrophils (C), CD4⁺ T cells (D), and CD8⁺ T cells (E) were determined. Error bars represent the standard deviations of the means. Data are representative of three independent experiments (n = 3 mice). Statistically significant differences compared to infection with the wt virus are indicated (*, P < 0.05).

FIG 11

Differential gene expression in rSARS-CoV-MA15-E*-infected mice. Total lung RNA was extracted from mice infected with the indicated SARS-CoV deletion mutants at 2 days p.i. (n = 3 mice). (A) Differential gene expression was measured using microarrays. Only genes with an FDR less than 0.05 were considered candidate genes. Points with a differential expression higher or lower than 2-fold are represented as darker or lighter dots, respectively. (B) Candidate genes that were up- and downregulated in ΔE, Δ3, and Δ5 viruses were grouped according to Gene Ontology terms. Numbers on the x axis indicate DAVID FDR values. (C) Genes differentially expressed in ΔE-, Δ3-, and Δ5-infected mice compared to in wt-infected mice were classified according to their main biological functions. Black and gray lettering is used to indicate up- and downregulated genes, respectively. Asterisks indicates genes whose expression was confirmed by RT-qPCR. The numbers indicate the fold change for each gene in ΔE-, Δ3-, and Δ5-infected mice compared to in wt-infected mice. For those genes detected with more than one probe, the value corresponding to the highest upregulation or downregulation is represented.

FIG 12

Effect of rSARS-CoV-MA15-E* deletion mutants on the expression of cytokine mRNAs in BALB/c mouse lung. Mice were infected with 1 × 10⁵ PFU of rSARS-CoV-MA15-Δ3, -Δ5, and -ΔE and wt viruses, and total RNA was extracted at 2 and 4 days p.i. The expression of mRNAs encoding inflammatory genes (A) and immune response and IFN response genes encoding TGFβ and 18S rRNA (B) was measured by qRT-PCR. In each case, levels of expression in infected lungs were compared to those in mock-infected ones. Bars represent standard deviations of the mean (n = 3 mice). Statistically significant data compared to mice infected with the wt virus are indicated (*, P < 0.05; **, P < 0.01).

FIG 13

Expression of cytokines in rSARS-CoV-MA15-E*-infected mice. Mice were infected with 1 × 10⁵ PFU of rSARS-CoV-MA15-wt, -Δ3, -Δ5, or -ΔE virus or were mock infected (mock). Lung proteins were extracted at 2 days p.i., and the accumulation of several cytokine proteins was measured. Protein concentration is expressed as picograms per milliliter of lung tissue extract. Means and standard deviations are shown (n = 3 mice). Statistically significant differences compared to mice infected with wt virus are indicated (*, P < 0.05).

FIG 14

Protection conferred by immunization with the rSARS-CoV-MA15-E* mutants. Six-week-old BALB/c mice were mock immunized or immunized with 6,000 PFU of the SARS-CoV-MA15-E* mutants and challenged at day 21 postimmunization with 1 × 10⁵ PFU of MA15 virus (n = 5 mice). Weight loss (A) and survival (B) were recorded daily.