Vaccine efficacy in senescent mice challenged with recombinant SARS-CoV bearing epidemic and zoonotic spike variants - PubMed (original) (raw)

doi: 10.1371/journal.pmed.0030525.

Timothy Sheahan, Mark Heise, Boyd Yount, Nancy Davis, Amy Sims, Mehul Suthar, Jack Harkema, Alan Whitmore, Raymond Pickles, Ande West, Eric Donaldson, Kristopher Curtis, Robert Johnston, Ralph Baric

Affiliations

Vaccine efficacy in senescent mice challenged with recombinant SARS-CoV bearing epidemic and zoonotic spike variants

Damon Deming et al. PLoS Med. 2006 Dec.

Erratum in

Abstract

Background: In 2003, severe acute respiratory syndrome coronavirus (SARS-CoV) was identified as the etiological agent of severe acute respiratory syndrome, a disease characterized by severe pneumonia that sometimes results in death. SARS-CoV is a zoonotic virus that crossed the species barrier, most likely originating from bats or from other species including civets, raccoon dogs, domestic cats, swine, and rodents. A SARS-CoV vaccine should confer long-term protection, especially in vulnerable senescent populations, against both the 2003 epidemic strains and zoonotic strains that may yet emerge from animal reservoirs. We report the comprehensive investigation of SARS vaccine efficacy in young and senescent mice following homologous and heterologous challenge.

Methods and findings: Using Venezuelan equine encephalitis virus replicon particles (VRP) expressing the 2003 epidemic Urbani SARS-CoV strain spike (S) glycoprotein (VRP-S) or the nucleocapsid (N) protein from the same strain (VRP-N), we demonstrate that VRP-S, but not VRP-N vaccines provide complete short- and long-term protection against homologous strain challenge in young and senescent mice. To test VRP vaccine efficacy against a heterologous SARS-CoV, we used phylogenetic analyses, synthetic biology, and reverse genetics to construct a chimeric virus (icGDO3-S) encoding a synthetic S glycoprotein gene of the most genetically divergent human strain, GDO3, which clusters among the zoonotic SARS-CoV. icGD03-S replicated efficiently in human airway epithelial cells and in the lungs of young and senescent mice, and was highly resistant to neutralization with antisera directed against the Urbani strain. Although VRP-S vaccines provided complete short-term protection against heterologous icGD03-S challenge in young mice, only limited protection was seen in vaccinated senescent animals. VRP-N vaccines not only failed to protect from homologous or heterologous challenge, but resulted in enhanced immunopathology with eosinophilic infiltrates within the lungs of SARS-CoV-challenged mice. VRP-N-induced pathology presented at day 4, peaked around day 7, and persisted through day 14, and was likely mediated by cellular immune responses.

Conclusions: This study identifies gaps and challenges in vaccine design for controlling future SARS-CoV zoonosis, especially in vulnerable elderly populations. The availability of a SARS-CoV virus bearing heterologous S glycoproteins provides a robust challenge inoculum for evaluating vaccine efficacy against zoonotic strains, the most likely source of future outbreaks.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: Experiments described in this manuscript use technology that ND helped develop at the University of North Carolina, and that has been licensed to AlphaVax, Inc., for commercial use. ND is a shareholder in AlphaVax, Inc.

Figures

Figure 1

Figure 1. VRP Expression of SARS S and N and VRP-S Induction of Anti-SARS S Antibody

(A) Western blot of cell lysates infected with VRP-S or VRP-N and probed with human serum collected from a convalescent SARS patient, 1128. (B–D) Western blot of lysates from cells infected with the SARS-CoV strains Urbani, Tor2, icSARS, or icGDO3-S and probed with human convalescent serum, 1128 (B), mouse anti–SARS-S serum from VRP-S–vaccinated animals (C), or mouse serum from a mouse vaccinated with VRP-N (D).

Figure 2

Figure 2. VRP-S Induces Short- and Long-Term Protection against icSARS-CoV Challenge

icSARS titers are expressed as the log10 plaque-forming units per gram (pfu/g) of lung. Tissues were homogenized in PBS to form a 20% suspension and titered on Vero monolayers. The titers for individual mice are shown as a filled circle, and the mean titer for the group is represented by a solid bar. Limit of detection (lod) is 2.4 log10 pfu/g. (A) Lung titers of16-wk-old BALB/c mice harvested 2 d after being i.n. infected with 105 PFU of icSARS-CoV (n = 6). (B) Lung titers of BALB/c mice vaccinated and boosted with 106 infectious units (IU) of VRP expressing the influenza HA (VRP-HA), SARS-S glycoprotein (VRP-S), SARS-N protein (VRP-N), or a combination of VRP-S and VRP-N (VRP-S+N). Mice (n = 7 VRP-HA, n = 8 for other groups) were vaccinated at 5 wk of age, boosted 5 wk later, then i.n. challenged with 105 pfu of icSARS-CoV 54-wk post-boost. Lungs were harvested 4 d later and titered. (C) Plaque assay results were confirmed by in situ hybridization to sectioned lungs of five mice from each vaccinated group with a radiolabeled riboprobe complementary to the SARS CoV N gene. In senescent mice challenged with the icSARS-CoV, representative lung sections from VRP-HA– (unpublished data) and VRP-N–vaccinated (a) animals exhibited extensive in situ signal (black arrows), whereas only one of five sections from VRP-S–vaccinated (b) and zero of five sections from VRP-S+N–vaccinated (c) mice exhibited SARS-CoV–specific signal above background levels.

Figure 3

Figure 3. Synthetic Reconstruction of icGD03-S

(A) Unrooted phylogenetic gene tree of 35 SARS isolates ranging from early, middle, and late phases of the 2002–2003 epidemic to 2003–2004 animal isolates. Branch confidence values are shown as posterior probabilities. The three human isolates that fall within the cluster otherwise isolated from animals (shown in boxes), GZ0402, GD03, and GZ0401, may represent infections in which a human acquired the virus from a Himalayan palm civet. (B) The GDO3-S glycoprotein. Amino acid changes unique to the GDO3-S with the GDO3-S amino acid listed on the left and the corresponding Urbani to the right. The GDO3-S amino acid changes are shown in relation to the S1 and S2 subunits, the receptor binding domain (RBD), heptad repeats one (HR1) and two (HR2), the transmembrane domain (TM), and known neutralizing epitopes. Two mutations that arose during tissue culture passage of the chimeric icGDO3-S are shown in red. (C) Growth curves of the Urbani strain of SARS-CoV (diamond, solid line), the recombinant Urbani icSARS (squares, dashed line), and the recombinant chimeric virus icGDO3-S (triangles, dotted line) in human airway epithelial cells. (D) Comparing growth of icSARS-CoV to icGDO3-S in the lungs of mice. Six-week-old female BALB/C mice were infected with icSARS-CoV or icGDO3-S (n = 5 per group). The individual titer of each mouse is represented by a filled circle, and the mean titer of the group is represented as a solid bar.

Figure 4

Figure 4. VRP-S Induces Short-Term Protection against icGDO3-S in Young and Partial Protection in Old Mice

(A) Lung titers of BALB/c mice vaccinated and boosted with 106 IU of VRP-S, VRP-N, a combination of VRP-S plus VRP-N (VRP-S+N), or mock vaccinated with PBS, then challenged with 105 pfu of icGDO3-S challenge (n = 8 per group). Lungs were harvested 2 d post-challenge. (B) Lung titers of aged BALB/c mice vaccinated at greater than 26 wk of age, boosted 4 wk later, then challenged 12 wk post-boost with icGDO3-S (n = 7 VRP-N, n = 8 for other groups). Tissue was harvested 4 d post-challenge. (C) SARS CoV specific in situ signal (black arrows) was observed in the lungs of senescent mice that were vaccinated with PBS (unpublished data) or VRP-N (a) and challenged with icGDO3-S, although overall, the signal appeared to be less intense than that observed in the icSARS challenge animals (Figure 2C). Vaccination with VRP-S (b) or VRP-S+N (c) failed to induce complete protection from icGD03-S challenge, as sections from two of five S- and three of four S+N-vaccinated animals exhibited signal above that of uninfected controls.

Figure 5

Figure 5. An 80% Plaque Reduction Neutralization Titers (PRNT80) for VRP-S and VRP-S+N Hyperimmune Serum

(A) Mice vaccinated young: icSARS-CoV (left) and icGDO3-S (right) PRNT80 for VRP-S immune serum (experiment 2) collected at 5 wk post-boost (n = 5) and 53 wk post-boost (n = 8). (B) Mice vaccinated old: icSARS (left) and icGDO3-S (right) PRNT80 values for VRP-S and VRP-S+N immune serum (experiment 4) at 12 and 29 wk post-boost (n = 6 for icSARS; n = 5 for icGDO3-S). The PRNT80 values for individual animals are show as black circles, and the mean value is shown as a solid bar. The limits of detection (1:1,600 upper and 1:100 lower) are represented by horizontal dotted lines.

Figure 6

Figure 6. ELISA Titers for Anti-S and Anti-HA igG in Vaccinated Animals

(A) Log10 half-maximum ELISA titers for anti-S IgG antibody in aged mice vaccinated with VRP-S or VRP-S+N when young (Table 1, experiment 2) or senescent (Table 1, experiment 4). Values represent mean values, and error bars indicate standard deviation. (B) Log10 half-maximum ELISA titers for anti-HA IgG antibody in aged mice vaccinated when young (experiment 2) or senescent (experiment 4).

Figure 7

Figure 7. Pathogenic Findings Following Homologous Challenge

Light photomicrographs of representative histologic lung sections (Table 1, experiment 2) taken from an untreated control mouse (A), a VRP-N–vaccinated mouse (B) and (D), a VRP-HA–vaccinated mouse (C), a VRP-S–vaccinated mouse (E), and a VRP-S– and VRP-N–treated mouse (F). No histopathology was evident in (A). A marked mixed inflammatory infiltrate composed mainly of mononuclear leukocytes (lymphocytes and plasma cells) and widely scattered eosinophils are evident in the perivascular and peribronchiolar interstitium (asterisk) in (B). Similar inflammatory cells are also present in bronchiolar (br) airways and alveolar airspaces along with enlarged and vacuolated alveolar macrophages (arrows). The box in (B) denotes the site of the light photomicrograph (D) that was taken at a higher magnification to better illustrate the lymphoplasmacytic inflammatory cell infiltrate with lesser numbers of eosinophils (arrows). Similar, but slightly less severe, perivascular inflammatory infiltrates (asterisk) are also present in (F), but without accompanying alveolitis. Minimal lymphoplasmacytic cell accumulations around the pulmonary arteriole (a) are evident in (C) and (E). All tissues were stained with hematoxylin and eosin. Bars denote the scale of the magnification. a, pulmonary arteriole; ap, alveolar parenchyma; br, bronchiolar lumen; e, surface epithelium of the bronchiole.

Figure 8

Figure 8. Pathogenic Findings Following Heterologous Challenge

Light photomicrographs of representative histologic lung sections (Table 1, experiment 4) taken from a mock PBS–vaccinated mouse (A) and (B), a VRP-N–vaccinated mouse (C) and (D), a VRP-S–vaccinated mouse (E) and (F), or a VRP-S+N–vaccinated mouse (G) and (H). The boxes in (A), (C), (E), and (G) (200× magnification) denote the site of the light photomicrograph that was taken at a higher magnification (400×) to better illustrate the lymphoplasmacytic inflammatory cell infiltrates including eosinophils (yellow arrows). All tissues were stained with hematoxylin and eosin.

Figure 9

Figure 9. Kinetics of VRP-N–Associated Inflammation

Light photomicrographs of lung sections taken from VRP-HA– and VRP-N–vaccinated mice harvested at days 2, 4, 7, and 14 post–icSARS-CoV challenge (Table 1, experiment 5). Representative lung sections (200× magnification) comparing pulmonary inflammation between VRP-HA–vaccinated (A), (C), (E), and (G) and VRP-N–vaccinated (B), (D), (F), and (H) mice. Enhanced inflammation was evident by day 2 (A) and (B) in some VRP-N–vaccinated animals relative to lung sections of VRP-HA–inoculated mice. By day 4 post-infection (C) and (D), increased inflammation in VRP-N–vaccinated animals was widely apparent and was maintained through days 7 (E) and (F) and 14 (G) and (H).

Figure 10

Figure 10. Identifying Eosinophils among Inflammatory Infiltrates

The 400× magnification comparing eosinophil infiltration within the lung sections of VRP-HA–vaccinated (A), (C), (E), and (G) and VRP-N–vaccinated (B), (D), (F), and (H) mice (Table 1, Experiment 5). At day 2 post-infection (A) and (B), eosinophils are rarely evident in the lungs of either VRP-HA (A) or VRP-N (B) mice. Day 4 post-infection (C) and (D), extensive eosinophils (yellow arrows) are present within the lungs of VRP-N–vaccinated mice. Widespread eosinophils are seen at day 7 post-challenge in VRP-N–vaccinated (F), but not VRP-HA–vaccinated (E) mice. By day 14 (G) and (H), eosinophils are rarely found among inflammatory cells of VRP-N–vaccinated mice. An identical experiment in old animals was performed simultaneously (Table 1, experiment 6), showed results indistinguishable from those of young mice (unpublished data). All tissues were stained with hematoxylin and eosin.

Similar articles

Cited by

References

    1. Drosten C, Gunther S, Preiser W, van der Werf S, Brodt HR, et al. Identification of a novel coronavirus in patients with severe acute respiratory syndrome. N Engl J Med. 2003;348:1967–1976. - PubMed
    1. Ksiazek TG, Erdman D, Goldsmith CS, Zaki SR, Peret T, et al. A novel coronavirus associated with severe acute respiratory syndrome. N Engl J Med. 2003;348:1953–1966. - PubMed
    1. Han Y, Geng H, Feng W, Tang X, Ou A, et al. A follow-up study of 69 discharged SARS patients. J Tradit Chin Med. 2003;23:214–217. - PubMed
    1. Lau SK, Woo PC, Li KS, Huang Y, Tsoi HW, et al. Severe acute respiratory syndrome coronavirus-like virus in Chinese horseshoe bats. Proc Natl Acad Sci U S A. 2005;102:14040–14045. - PMC - PubMed
    1. Poon LL, Chu DK, Chan KH, Wong OK, Ellis TM, et al. Identification of a novel coronavirus in bats. J Virol. 2005;79:2001–2009. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources