Enterovirus D-68 Infection, Prophylaxis, and Vaccination in a Novel Permissive Animal Model, the Cotton Rat (Sigmodon hispidus)

Abstract

In recent years, there has been a significant increase in detection of Enterovirus D-68 (EV-D68) among patients with severe respiratory infections worldwide. EV-D68 is now recognized as a re-emerging pathogen; however, due to lack of a permissive animal model for EV-D68, a comprehensive understanding of the pathogenesis and immune response against EV-D68 has been hampered. Recently, it was shown that EV-D68 has a strong affinity for α2,6-linked sialic acids (SAs) and we have shown previously that α2,6-linked SAs are abundantly present in the respiratory tract of cotton rats (Sigmodon hispidus). Thus, we hypothesized that cotton rats could be a potential model for EV-D68 infection. Here, we evaluated the ability of two recently isolated EV-D68 strains (VANBT/1 and MO/14/49), along with the historical prototype Fermon strain (ATCC), to infect cotton rats. We found that cotton rats are permissive to EV-D68 infection without virus adaptation. The different strains of EV-D68 showed variable infection profiles and the ability to produce neutralizing antibody (NA) upon intranasal infection or intramuscular immunization. Infection with the VANBT/1 resulted in significant induction of pulmonary cytokine gene expression and lung pathology. Intramuscular immunization with live VANBT/1 or MO/14/49 induced strong homologous antibody responses, but a moderate heterologous NA response. We showed that passive prophylactic administration of serum with high content of NA against VANBT/1 resulted in an efficient antiviral therapy. VANBT/1-immunized animals showed complete protection from VANBT/1 challenge, but induced strong pulmonary Th1 and Th2 cytokine responses and enhanced lung pathology, indicating the generation of exacerbated immune response by immunization. In conclusion, our data illustrate that the cotton rat is a powerful animal model that provides an experimental platform to investigate pathogenesis, immune response, anti-viral therapies and vaccines against EV-D68 infection.

Fig 1. A cotton rat model of EV-D68 infection and replication.

Cotton rats were infected i.n. with 10⁶ TCID₅₀ per rat with 3 different strains of EV-D68, ATCC (prototype Fermon strain), VANBT (2012 pre-outbreak strain from Nashville, TN) and MO/49 (2014 outbreak strain from Kansas, MO). (A) Evolutionary tree showing three major clades of EV-D68, A, B and C, distributed worldwide. Strains used in this study are shown in red. The tree is rooted by the oldest EV-D68 sequence in GenBank, the Fermon strain (referred to as ATCC), collected in 1962 in California, USA. (B, C) Quantification of infectious virus titers of each EV-D68 strain in nose and lung homogenates from infected animals at 10 h (B) and 24 h (C) p.i. Groups of 5 animals were euthanized at each time point. Each bar corresponds to an individual animal. UN = uninfected. (D, E, F) Time course of VANBT replication in cotton rats; groups of 4 animals were euthanized at the indicated time to measure infectious virus titers in nose (D) and lung (E) homogenates. Results are representative of 2 independent experiments. *p<0.05 for 4 h nose titer compared with 0.5 h p.i. and #p<0.05 for 10 h nose titer compared with 4 h p.i. Animals inoculated with UV-VANBT and sacrificed at 6 h p.i. (UV-6) were shown as control. (F) Quantification of VANBT (-) vRNA by qRT-PCR in lung tissue. Insert is a blowup of the 24, 48 and 96 h time points from VANBT-infected and UV-VANBT-inoculated rat sacrificed at 6 h p.i. (G) Homologous NA titers measured in serum samples collected 3 weeks p.i. with the indicated EV-D68.

Fig 2. VANBT infection in cotton rats induces pulmonary chemokine, IFN, and proinflammatory cytokine mRNA expression.

Cotton rats were infected i.n. with 10⁶ TCID₅₀ of VANBT or the same amount of UV-VANBT. Groups of 4 animals were euthanized at each indicated time point. Results are representative of two independent experiments. Relative gene expression profile of chemokines (GRO, MCP-1, IP-10 and RANTES), IFN-β, two IFN-inducible genes (Mx-1 and Mx-2), and pro-inflammatory cytokines (IL-6 and IFN-γ) in the lung tissues were measured by qRT-PCR. Results are calculated as fold induction over uninfected (naïve) animals and expressed as geometric means ± SE, *p<0.05 for VANBT compared with UV-VANBT at each time point.

Fig 3. VANBT infection in cotton rats causes lung pathology.

(A) Histopathology scores obtained from lungs of uninfected (UN) or UV-VANBT or VANBT infected animals and euthanized at the indicated time p.i. Combined scores represent extent of peribronchiolitis, perivasculitis, interstitial inflammation and alveolitis. n = 4–8, *p<0.05 for VANBT compared to UV-VANBT at respective time point. (B) Representative H&E-stained lung sections in naïve rats (a) or lungs taken after 48 h after i.n. inoculation with UV-VANBT (b), or VANBT (c). Peribronchial inflammation (black arrow) and alveolitis (black arrow head) are indicated. Scale bars, 500 μm. Images shown are representative of four cotton rats.

Fig 4. Intramuscular immunization with live EV-D68 strains induces virus-specific homologous NA titers.

(A) Female cotton rats were immunized i.m. with 10⁶ TCID₅₀/100 μl of indicated virus on day 0 and boosted 3 weeks after the first immunization. Serum samples were obtained at 3 week (before boosting) and 7 weeks after the first immunization and homologous serum NA titers were determined. The sera were assayed in duplicate and NA titer is expressed as Log₂ geometric mean ± SE. n = 10–15 from two independent experiments, ***p< 0.001 for 3 weeks NA titer compared with 7 weeks NA. (B) Passive transfer of VANBT immune sera protects animals from VANBT challenge. Animals were treated intraperitoneally with 0.5 ml 1 x PBS (mock) or serum from animals immunized i.m. with either UV-VANBT, VANBT, or MO/49. The following day, animals were challenged i.n. with 10⁶ TCID₅₀ of VANBT and euthanized 10 h later to determine nose and lung viral titers. n = 4–5 per group. * p<0.05 for each group compared with mock group.

Fig 5. Intramuscular immunization with EV-D68 protects against VANBT challenge.

(A) Schematic representation of infection and immunization regimen. Female cotton rats were mock-immunized i.m. on day 0 and at 3 weeks, infected with VANBT on day 0, or immunized i.m. on day 0 and at 3 weeks with UV-VANBT, ATCC, VANBT, or MO/49 using 10⁶ TCID₅₀/100 μl of virus. At 7 weeks, all animals were challenged i.n. with 10⁶ TCID₅₀ of VANBT and sacrificed at 10 h or 48 h p.i. (B) Viral titers in the nose and lung of animals sacrificed at 10 h p.i. n = 5 per group. Data are representative of two independent experiments. * p<0.05 where each group is compared with mock-immunized group.

Fig 6. Intramuscular immunization with VANBT induces both Th1 and Th2 cytokines upon virus challenge.

Female cotton rats were mock-immunized i.m. on day 0 and at 3 weeks, infected with VANBT on day 0, or immunized i.m. on day 0 and at 3 weeks with UV-VANBT, or VANBT, using 10⁶ TCID₅₀/100 μl of virus. At 7 weeks, all animals were challenged i.n. with 10⁶ TCID₅₀ of VANBT and sacrificed at 10 h or 48 h p.i. Relative mRNA expression profiles of IFN-β (A), Mx-2 (B), IP-10 (C), IFN-γ (D), IL-6 (E), IL-4 (F), IL-5 (G), and IL-13 (H) at either 10 or 48 h p.i. in the lung tissues were measured by qRT-PCR. Results were calculated as fold-induction over uninfected (naïve) animals and expressed as geometric means ± SE. Results are representative of two independent experiments, n = 5. * p<0.05 for each group compared with mock-immunized group, #p<0.05 when VANBT/i.m. compared to VANBT/i.n. and ⦁p<0.05 when VANBT/i.m. compared with UV-VANBT/i.m.

Fig 7. Lung histopathology scores of VANBT-immunized cotton rats upon virus challenge.

Groups of female cotton rats were infected or immunized as described above for Fig 6. At 7 weeks, all groups were challenged with VANBT i.n. and sacrificed at 48 h p.i. Graphs represent the average scores for bronchiolitis, vasculitis, interstitial pneumonia, and alveolitis. Results are representative of two independent experiments, n = 5. * p<0.05 for each group compared with mock immunized group or VANBT/i.n. group.