Innate Immune Basis for Rift Valley Fever Susceptibility in Mouse Models

Abstract

Rift Valley fever virus (RVFV) leads to varied clinical manifestations in animals and in humans that range from moderate fever to fatal illness, suggesting that host immune responses are important determinants of the disease severity. We investigated the immune basis for the extreme susceptibility of MBT/Pas mice that die with mild to acute hepatitis by day 3 post-infection compared to more resistant BALB/cByJ mice that survive up to a week longer. Lower levels of neutrophils observed in the bone marrow and blood of infected MBT/Pas mice are unlikely to be causative of increased RVFV susceptibility as constitutive neutropenia in specific mutant mice did not change survival outcome. However, whereas MBT/Pas mice mounted an earlier inflammatory response accompanied by higher amounts of interferon (IFN)-α in the serum compared to BALB/cByJ mice, they failed to prevent high viral antigen load. Several immunological alterations were uncovered in infected MBT/Pas mice compared to BALB/cByJ mice, including low levels of leukocytes that expressed type I IFN receptor subunit 1 (IFNAR1) in the blood, spleen and liver, delayed leukocyte activation and decreased percentage of IFN-γ-producing leukocytes in the blood. These observations are consistent with the complex mode of inheritance of RVFV susceptibility in genetic studies.

Figure 1

Host genetics contributes to susceptibility and symptoms of RVF disease. (a) Survival curves of BALB/c (n = 28) and MBT (n = 30) mice infected intraperitoneally with 10² PFU RVFV strain ZH548. (b) Trajectory of symptoms of these mice as observed daily, preceding death. Mantel-Cox’s Logrank test was performed to assess survival curve differences. ***P < 0.001.

Figure 2

Histopathological analysis from non-infected and RVFV-infected BALB/c and MBT mice at day 3 post infection. (a,b,d,e) Hematoxylin and eosin (HE) staining of liver: arrowheads indicate lesions with necrotic and apoptotic hepatocytes, and neutrophil infiltrate (insets: intra-nuclear inclusion bodies with margination of nuclear chromatin, arrows). (c,f) Immunohistochemical detection of the N protein of RVFV.

Figure 3

Higher viral antigen load correlates with higher cell death in the spleen of MBT mice. (a–c) Flow cytometric analysis for the N viral nucleoprotein (N-antigen, N-ag) within the blood (a), spleen (b), and liver (c) cells from BALB/c (n = 5/day) and MBT (n = 5/day) mice on days 1–3 post RVFV infection. (d) Percentages of viable cells in the spleen of RVFV-infected BALB/c and MBT mice using flow cytometry (n = 5 mice/strain/timepoint). Data are represented as mean ± SEM. Two-way ANOVA was used to test differences between means. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4

Changes in the percentages of innate immune cells in RVFV-infected BALB/c and MBT mice. (a) A comprehensive flow cytometric profiling of leukocytes defined as positive for CD45 in the blood, spleen, liver and bone marrow of RVFV-infected BALB/c and MBT mice on days 0–3 post infection. (b) Percentages of key innate immune cell populations, DCs (CD45+, CD11c+), natural killer (NK) cells (CD45+, CD335+), and neutrophils (CD45+, Ly6C/Ly6G+), after infection with RVFV in the spleen of BALB/c and MBT mice. (c) Linear discriminant (LDA) analyses ranks neutrophils, as having the greatest impact on the dissimilarity between BALB/c and MBT spleen at day 3 post infection. Cytometry experiments were performed twice for each time point (n = 5 mice/strain/timepoint). Representative experiments are displayed. Data are represented as mean ± SEM. Two-way ANOVA was used to test differences between means. *P values < 0.05, **P < 0.01, and ***P < 0.001.

Figure 5

Faster recruitment of neutrophils, and higher PSGL-1 levels in neutrophils at day 3 post infection in MBT mice. (a) Flow cytometric profile of neutrophil (CD45+, Ly6C/Ly6G+) populations in the blood, liver, and bone marrow of RVFV-infected BALB/c and MBT mice on days 0–3 post infection. Two-way ANOVA was used to test differences between means (b) Percentages of PSGL-1-positive circulating neutrophils at day 3 post infection. Cytometry experiments were performed twice (n = 5 mice/strain/timepoint, unless otherwise indicated). Data are represented as mean ± SEM. Mann- Whitney U test was used to compare differences between means. *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 6

Greater IFN-α expression, but weaker IFNAR1 expression in MBT mice. (a) Serum IFN-α concentrations in BALB/c and MBT mice on days 0–3 post infection (n = 8 per strain, infected, n = 3 per strain, sham-infected). (b) Percentages of total IFNAR1-positive leukocytes in the blood, spleen, and liver of BALB/c and MBT mice as measured by flow cytometry. Cytometry experiments were performed twice (n = 5 mice/strain/timepoint). Representative experiments are displayed. Data are represented as mean ± SEM. Two-way ANOVA was used to test differences between means. *P values < 0.05, **P < 0.01, and ***P < 0.001.

Figure 7

Delayed activation of circulating leukocytes in MBT mice. (a) Percentages of circulating leukocytes expressing IFN-γ in BALB/c and MBT mice on days 1–3 post infection as evaluated by intracellular flow cytometry. (b) Percentages of PSGL-1-positive circulating leukocytes in BALB/c and MBT mice on days 0–3 post infection. Cytometry experiments were performed twice (n = 5 mice/strain/timepoint). Data are represented as mean ± SEM. Representative experiments are displayed. Two-way ANOVA was used to test differences between means. *P < 0.05, **P < 0.01, and ***P < 0.001.