Impact of caspase-1/11, -3, -7, or IL-1 β/IL-18 deficiency on rabies virus-induced macrophage cell death and onset of disease

Abstract

Rabies virus is a highly neurovirulent RNA virus, which causes about 59000 deaths in humans each year. Previously, we described macrophage cytotoxicity upon infection with rabies virus. Here we examined the type of cell death and the role of specific caspases in cell death and disease development upon infection with two laboratory strains of rabies virus: Challenge Virus Standard strain-11 (CVS-11) is highly neurotropic and lethal for mice, while the attenuated Evelyn-Rotnycki-Abelseth (ERA) strain has a broader cell tropism, is non-lethal and has been used as an oral vaccine for animals. Infection of Mf4/4 macrophages with both strains led to caspase-1 activation and IL-1β and IL-18 production, as well as activation of caspases-3, -7, -8, and -9. Moreover, absence of caspase-3, but not of caspase-1 and -11 or -7, partially inhibited virus-induced cell death of bone marrow-derived macrophages. Intranasal inoculation with CVS-11 of mice deficient for either caspase-1 and -11 or -7 or both IL-1β and IL-18 led to general brain infection and lethal disease similar to wild-type mice. Deficiency of caspase-3, on the other hand, significantly delayed the onset of disease, but did not prevent final lethal outcome. Interestingly, deficiency of caspase-1/11, the key executioner of pyroptosis, aggravated disease severity caused by ERA virus, whereas wild-type mice or mice deficient for either caspase-3, -7, or both IL-1β and IL-18 presented the typical mild symptoms associated with ERA virus. In conclusion, rabies virus infection of macrophages induces caspase-1- and caspase-3-dependent cell death. In vivo caspase-1/11 and caspase-3 differently affect disease development in response to infection with the attenuated ERA strain or the virulent CVS-11 strain, respectively. Inflammatory caspases seem to control attenuated rabies virus infection, while caspase-3 aggravates virulent rabies virus infection.

Figure 1

Comparison between viral load and cell death in Mf4/4 cells following infection with ERA or CVS-11 rabies virus strains. (a) Virus titer was measured at 48 hpi using titration in BHK-21 cells. (b) Cellular viability was determined using the crystal violet staining method to measure the viable adherent cells as a percentage compared to uninfected control cells. Data are representative of at least five independent experiments and are presented as mean values±S.E.M. (c) Fluorescence microscopy imaging of viral N protein to measure the number of infected Mf4/4. Cells were stained with Hoechst (left panel) and FITC-labeled anti-rabies virus nucleocapsid antibody (right panel). (d) The percentage of antigen-positive cells is expressed by: (number of FITC-anti-N cells/numbers of Hoechst-positive cells)×100. (e) Analysis of caspase activity in infected Mf4/4 cells at 48 h post virus inoculation. Cells were stained with Hoechst (left panel) and with the green fluorescent caspases inhibitor (FLICA) that covalently binds to the active caspases (right panel). (f) The percentage of cells with active caspases is expressed by: (number of FLICA-positive cells/number of Hoechst-positive cells)×100.

Figure 2

Activation of signaling pathways of apoptosis and pyroptosis in rabies virus-infected Mf4/4. Analysis of the proteolytic activation of caspases-1, -3, -7, -8, and -9, IL-1β and Bid in MF4/4 cells inoculated by ERA. Western blots were performed using antibodies specific for the active and inactive forms of these proteins. Upon activation, the protein is cleaved and appears as a smaller band representing the active protein on the blot. Total cellular lysates were prepared 24 and 48 h after virus inoculation. Uninfected Mf4/4 were prepared in parallel to each condition as negative controls (CTL). An anti-β-actin antibody was used to verify that equal amounts of protein were loaded. Data are representative of at least three independent experiments.

Figure 3

Cellular viability and viral load of caspases-1/11, -3, or -7-deficient primary BMDMs inoculated by CVS-11 (a, c) or ERA (b, d) virus. (a, b) Cellular viability was determined at 48 hpi using the crystal violet staining method to measure the percentage of viable adherent cells. P-value was calculated using an unpaired t-test (each group of deficient BMDM compared to the WT group) and was indicated as follows: P-value<0.01 (**). A protector effect on the viability of the cells was observed in caspase-3^−/− macrophages inoculated with ERA or CVS-11. Data were pooled and are representative of at least four independent experiments. They are presented as mean values±S.E.M. (c and d) Viral load was determined 48 hpi by RT-qPCR following CVS-11 or ERA infection. Results are presented as a ratio compared to the WT cells. Each point represents one independent experiment.

Figure 4

Impact of a deficiency in caspase-1/11, -3, or -7 or IL-1β/IL-18 in mice on rabies disease progression, mortality, brain viral load, and neutralizing antibody response upon infection with a virulent strain of rabies. WT (n=6), caspase-1/11^−/− mice (n=7), caspase-3^−/− mice (n=4), caspase-7^−/− mice (n=6), and IL-1β/IL-18^−/− mice (n=3) were inoculated intranasally with the virulent CVS-11 virus (10^2.5 TCID₅₀/mouse). Disease progression and mortality within each group was evaluated by mean clinical score per mouse (b) and survival curves (a), respectively. P-value was calculated using a log-rank test (survival curves) and an ANOVA test followed by Bonferroni multiple comparison post-test (mean score per mouse) and was indicated as follows: P-value <0.05 (*); P-value <0.01 (**). (a) Survival curves of WT mice and caspase-3-deficient mice were significantly different (**). (b) The mean clinical score per mouse was significantly different between WT and caspase-3^−/− mice at 8 DPI. Caspase-3 deficiency delayed CVS-11-related disease, but had no effect on final lethal outcome. (c) Brain tissue was sampled upon peak clinical score and viral load measured by RT-qPCR. P-value was calculated with an ANOVA test followed by Bonferroni multiple comparison post-test. No significant difference could be observed between WT mice and deficient mice. (d) The neutralizing antibody titer was evaluated by rapid fluorescent focus inhibition test and expressed in IU/ml. P-value was calculated by an ANOVA test followed by Bonferroni multiple comparison post-test. Most mice were unable to mount an antibody response (>0.5 IU/ml) prior to killing. No significant difference was observed between WT and deficient mice. Two out of four caspase-3^−/− mice had developed antibody titers >0.5 IU/ml, which might be because they survived long enough to allow an antibody response.

Figure 5

Impact of a deficiency in caspase-1/11, -3, or -7 or IL-1β/IL-18 in mice on rabies disease progression, mortality, brain viral load, and neutralizing antibody response upon infection with an attenuated strain of rabies. WT (n=6), caspase-1/11^−/− mice (n=6), caspase-3^−/− mice (n=4), caspase-7^−/− mice (n=5), and IL-1β/IL-18^−/− mice (n=4) were inoculated intranasally with the attenuated ERA virus (10⁵ TCID₅₀/mouse). Disease progression and mortality within each group was evaluated by mean clinical score per mouse (b) and survival curves (a), respectively. P-value was calculated using a log-rank test (survival curves) and an ANOVA test followed by Bonferroni multiple comparison post-test (mean score per mouse) and was indicated as follows: P-value <0.05 (*); P-value <0.01 (**). (a) All mice survived the infection except for one caspase-7^−/− mouse. (b) The mean clinical score per mouse was significantly different between WT and caspase-1^−/− inoculated with ERA virus at 21 and 20 DPI (c). Brain tissue was sampled at 35 DPI (except for one mouse) and viral load measured by RT-qPCR. P-value was calculated with an ANOVA test followed by Bonferroni multiple comparison post-test. No significant difference could be observed between WT mice and deficient mice. (d) The neutralizing antibody titer was evaluated by rapid fluorescent focus inhibition test and expressed in IU/ml. P-value was calculated by an ANOVA test followed by Bonferroni multiple comparison post-test. No significant difference was observed between WT and deficient mice, but antibody titers seemed higher in caspase-3^−/− mice.