Adverse events post smallpox-vaccination: insights from tail scarification infection in mice with Vaccinia virus

Abstract

Adverse events upon smallpox vaccination with fully-replicative strains of Vaccinia virus (VACV) comprise an array of clinical manifestations that occur primarily in immunocompromised patients leading to significant host morbidity/mortality. The expansion of immune-suppressed populations and the possible release of Variola virus as a bioterrorist act have given rise to concerns over vaccination complications should more widespread vaccination be reinitiated. Our goal was to evaluate the components of the host immune system that are sufficient to prevent morbidity/mortality in a murine model of tail scarification, which mimics immunological and clinical features of smallpox vaccination in humans. Infection of C57BL/6 wild-type mice led to a strictly localized infection, with complete viral clearance by day 28 p.i. On the other hand, infection of T and B-cell deficient mice (Rag1(-/-)) produced a severe disease, with uncontrolled viral replication at the inoculation site and dissemination to internal organs. Infection of B-cell deficient animals (µMT) produced no mortality. However, viral clearance in µMT animals was delayed compared to WT animals, with detectable viral titers in tail and internal organs late in infection. Treatment of Rag1(-/-) with rabbit hyperimmune anti-vaccinia serum had a subtle effect on the morbidity/mortality of this strain, but it was effective in reduce viral titers in ovaries. Finally, NUDE athymic mice showed a similar outcome of infection as Rag1(-/-), and passive transfer of WT T cells to Rag1(-/-) animals proved fully effective in preventing morbidity/mortality. These results strongly suggest that both T and B cells are important in the immune response to primary VACV infection in mice, and that T-cells are required to control the infection at the inoculation site and providing help for B-cells to produce antibodies, which help to prevent viral dissemination. These insights might prove helpful to better identify individuals with higher risk of complications after infection with poxvirus.

Figure 1. VACV infection in single-cytokine deficient mice did not alter the outcome of infection.

C57BL/6 (n = 6), IL-12/23^−/− (n = 4), TNFR^−/− mice (n = 7), IFNγ^−/− (n = 6) and iNOS^−/− (n = 6) animals were infected with 10⁷ p.f.u. of VACV-WR by tail scarification. (A) Weight curve of C57BL/6, IL-12/23^−/−, TNFR^−/− mice. (B) Weight curve of IFNγ^−/− and iNOS^−/− mice. (C) WT, IFNγ^−/− and iNOS^−/− animals were euthanized at days 3, 5 and 7 post-infection (p.i.) and viral titers in tails were assessed by titration in BSC-40 cells. Bars represent the mean and standard error (SEM). The results shown are representative of two separate experiments.

Figure 2. T and B-cell deficient mice had severe disease upon tail scarification infection with VACV-WR.

Rag1 ^−/− animals (n = 6) were infected with 10⁷ p.f.u. of VACV-WR by tail scarification. (A) Weight curve of infected or mock-infected Rag1 ^−/− mice (**p<0.01, unpaired t test). (B) Animals were followed during 60 days post-infection (p.i.) to compute mean survival (**p<0.01, log-rank test). (C) WT and Rag1 ^−/− animals were euthanized at the times indicated and the tail, spleen and ovaries were removed and titrated in BSC-40 cells. Bars represent the mean and standard error (SEM). The results are representative of three experiments.

Figure 3. Clinical signs and lesion kinetic in T and B-cell deficient mice.

Rag1 ^−/− animals (n = 6) were infected with 10⁷ p.f.u. of VACV-WR by tail scarification. (A) Clinical signs of disseminated disease in Rag1 ^−/−. Starting on the second week post-infection, Rag1 ^−/− mice had signs of generalized disease, such as ruffled fur, arched back and disseminated lesions, primarily at the face, forepaw and tail. The animal shown is representative of the group. (B) Rag1 ^−/− presented a delayed lesion healing compared to WT mice. Note the ulcerative, purulent nature of the lesion in Rag1 ^−/− compared to the scab formed in C57BL/6 animals. Photos are representative of the groups indicated. Black bar represents 1 cm.

Figure 4. Antibody response seems not crucial to prevent mortality, but it is important to block viral dissemination.

(A) WT (C57BL/6) mice were infected with 10⁷ p.f.u. of VACV-WR by tail scarification and euthanized as described. Sera were collected at the times indicated. Indirect IgG ELISA was performed using purified VACV-WR as antigen (10⁶ p.f.u./well). Bars represent mean and standard deviation (SD). ND: not detected. (B) Survival curve of Rag1 ^−/− animals inoculated with anti-VACV rabbit serum or rabbit naïve serum intraperitoneally every week after infection. At days 10 and 15 p.i., the tails (C) and the ovaries (D) were removed and titrated in BSC-40 cells. Bars represent mean and standard error (SEM). *p<0.05, unpaired t test.

Figure 5. B-cell deficient mice did not show any signs of disseminated disease, but had delayed viral clearance.

µMT mice (n = 6) were infected with 10⁷ p.f.u. of VACV-WR by tail scarification. (A) Weight curve of µMT mice infected with VACV-WR or mock-infected. (B) The same animals were followed for 60 days to compute mean survival. (C) Viral titers were determined in tails of C57BL/6 and µMT mice at days 10, 15, 28 and 60 p.i. (*p<0.05 and **p<0.01, t test). (D) Viral titers were determined in ovaries of µMT mice at days 10, 15, 28 and 60 p.i. and of Rag1 ^−/− mice at days 10 and 15 p.i.(*p<0.05 and **p<0.01, t test). Bars represent mean and standard error (SEM). Results shown are representative of two experiments.

Figure 6. Mice lacking T cells are as susceptible to primary VACV infection as Rag1 ^−/− mice.

NUDE athymic mice (nu/nu) and their wild type littermates (nu/+) were infected with 10⁷ p.f.u. of VACV-WR by tail scarification. (A) Weight curve of NUDE mice (***p<0.001, unpaired t test). (B) Survival curve of nu/nu and nu/+ after infection (***p<0.001, log-rank test). (C) Viral titers were determined in tail of C57BL/6 and nu/nu mice (**p<0.01 day 10 p.i. and *p<0.05, t test). Data of viral titer in tails of nu/+ mice are missing. (D) Spleens of nu/nu and nu/+ mice were removed at the times indicated and virus titer was determined in BSC-40 cells. (E) VACV-specific IgG antibodies were determined by ELISA in serum of nu/nu and nu/+ mice (**p<0.01 and ***p<0.001, t test). Bars represent the mean and standard error (SEM), except for 6E, where bars represent mean and standard deviation (SD). The results shown are representative of two experiments. ND: not detected.

Figure 7. Rag1 ^−/− mice passively transferred with both T cells were fully protected from VACV infection.

Spleens of naïve C57BL/6 mice were removed and made into single-cell suspension and this population was enriched in CD4⁺ and CD8⁺ T cells using a negative selection system. (A) Rag1 ^−/− were inoculated intravenously with either CD4⁺ and CD8⁺ T cells or either type alone (n = 6 per group). As a control, Rag1 ^−/− mice received splenocytes prior to isolation intravenously (10⁷ viable cells). Four days after inoculation, all animals were infected with VACV-WR by tail scarification and weighted daily (***p<0.001, paired t test). Bars represent mean and SEM. (B) Survival curve of Rag1 ^−/− passively transferred with both CD4⁺ and CD8⁺ T cells or with either type alone, compared to the mortality in the same strain without previous treatment (**p<0.01, log-rank test). Viral titers in tail (C) and ovaries (D) of mice passively transferred with WT T cells and infected with VACV-WR. Bars represent mean and standard error (SEM) (*p<0.05, **p<0.01 and ***p<0.001 compared to Rag1 ^−/− non-treated group, t test). The results shown are representative of two experiments (CD4⁺ plus CD8⁺ T cells group). For animals that received only CD4⁺ or CD8⁺ T cells, only one experiment was performed.