Impaired immune responses following spinal cord injury lead to reduced ability to control viral infection

Abstract

Spinal cord injuries disrupt central autonomic pathways that regulate immune function, and increasing evidence suggests that this may cause deficiencies in immune responses in people with spinal cord injuries. Here we analyze the consequences of spinal cord injury (SCI) on immune responses following experimental viral infection of mice. Female C57BL/6 mice received complete crush injuries at either thoracic level 3 (T3) or 9 (T9), and 1 week post-injury, injured mice and un-injured controls were infected with different dosages of mouse hepatitis virus (MHV, a positive-strand RNA virus). Following MHV infection, T3- and T9-injured mice exhibited increased mortality in comparison to un-injured and laminectomy controls. Infection at all dosages resulted in significantly higher viral titer in both T3- and T9-injured mice compared to un-injured controls. Investigation of anti-viral immune responses revealed impairment of cellular infiltration and effector functions in mice with SCI. Specifically, cell-mediated responses were diminished in T3-injured mice, as seen by reduction in virus-specific CD4(+) T lymphocyte proliferation and IFN-γ production and decreased numbers of activated antigen presenting cells compared to infected un-injured mice. Collectively, these data indicate that the inability to control viral replication following SCI is not level dependent and that increased susceptibility to infection is due to suppression of both innate and adaptive immune responses.

Fig. 1

Spinal cord injured mice exhibit increased mortality and higher viral titer following viral infection compared to un-injured infected mice. At 1 week post-SCI, mice were infected with increasing dosages of MHV (5 × 10²–5 × 10⁵ PFU) and mortality recorded. A) The survival of T3-injured mice following 5 × 10⁵ PFU infection resulted in mortality, yet 100% survival was observed in un-injured mice and laminectomy surgery-control mice. Survival of T3-injured mice was prolonged following infection at lower dosages. Viral titers were recorded following infection at 1 week (B) and 4 weeks (C) post-SCI in T3- and T9-injured mice. Injured mice showed higher viral titers compared to un-injured mice at all infection dosages, independent of infection time post-SCI. Survival studies began with 10–8 mice in each infection group. Viral titers are presented as logarithmic means of PFU per gram of liver, as shown in columns in B and C bar graphs, with each data point representing one mouse. The limit of detection was ~200 PFU/g liver; BD=below level of detection. Significant differences between T3-injured and un-injured controls (*p ≤ 0.05), T9-injured and un-injured controls (**p ≤ 0.02), and T3-injured and T9-injured (***p ≤ 0.003) groups are shown in B and C.

Fig. 2

T3-injured mice have reduced spleen size and decreased number of B lymphocytes prior to MHV infection. Spleen weight and lymphocyte populations were examined in injured and un-injured mice prior to infection. A) T3-injured and not T9-injured mice showed significant reduction in spleen weight compared to un-injured controls at 1 week post-SCI (*p ≤ 0.004). The average mean weight of injured mice did not increase over time like un-injured mice, thus both T3-injured (*p ≤ 0.001) and T9-injured (**p ≤ 0.001) mice had significantly reduced spleen weights at 4 weeks post-SCI compared to un-injured controls. Following 1 week post-SCI, spleens were collected and processed for flow cytometric analysis of B and T lymphocytes. B–D) Representative dot plots show the frequency (mean ± SEM) positive cells within gated population, with the corresponding number in adjacent bar graphs. The frequencies of CD45R⁺ B cells in T9-injured and un-injured mice were similar (B, dot plots). However the frequency in T3-injured mice was reduced (B, dot plots), and there was a significant decrease in the number of B cells compared to un-injured controls (*p ≤ 0.002) (B, graph). The number of B cells in T9-injured mice was reduced compared to un-injured control, but was not significantly different (B, graph). Injured mice showed similar increases in the frequency of CD8⁺ T cells, however only T3-injured mice had significant increase compared to un-injured controls (*p ≤ 0.004) (C, dot plots). Comparison of the numbers of CD8⁺ T cells between all experimental groups revealed no significant differences (C, graph). There was no significant difference in the frequency of CD4⁺ T cells from T9-injured and un-injured mice (D, dot plots). However, T3-injured mice showed significantly increased frequency of CD4⁺ T cells compared to un-injured mice (*p ≤ 0.006) and T9-injured mice (***p ≤ 0.001) (D, dot plots). Although injured mice show reduced numbers of CD4⁺ T cells compared to un-injured mice, comparison between all experimental groups revealed no significant differences (D, graph). Each data point represents one mouse, and the group mean is presented as column in bar graphs.

Fig. 3

Spleen size and the generation of an adaptive immune response are reduced following infection of T3–injured mice. Spleen size increases following infection, thus on day 5 p.i. with 1 × 10⁴ PFU MHV spleen weight and cellular composition were assessed. A) Following infection at 1 week and 4 weeks post-SCI, T3-injured mice showed significantly reduced spleen weight compared to un-injured mice (*p ≤ 0.03). B–D) At 1 week post-SCI, T3-SCI mice were infected with 1 × 10⁴ PFU and the generation of an adaptive immune response in the spleen evaluated on day 5 p.i. using flow cytometric analysis. Virus-specific T cells were determine by ex vivo stimulation with CD4 Ag-specific peptide (epitope M133–147), CD8 Ag-specific peptide (epitope S598–605), or non-specific OVA control peptide, prior to flow cytometric assessment. T3-injured mice showed significant increase (*p ≤ 0.002) in the frequency of CD8⁺ T cells compared to un-injured mice (B, first panel), however there was no difference in the number of total CD8⁺ T cells nor the number of virus-specific CD8⁺ IFN-γ⁺ T cells (B, second and right-hand panels). The frequency of CD4⁺ T lymphocytes was similar between infected T3-injured and un-injured mice (C, first panel), however the numbers of total CD4⁺ and virus-specific CD4⁺ IFN-γ⁺ T cells were significantly reduced in injured mice (*p ≤ 0.04) (C, second and right-hand panels). The frequency of CD11b⁺ macrophages was significantly increased in T3-injured mice (*p ≤ 0.04) (D, first panel), however the frequency of CD86⁺ activated CD11b⁺ macrophages within the total macrophage population was significantly decreased compared to un-injured mice (*p ≤ 0.01) (D, second panel). The mean number of cells is presented in bar graphs, with each data point representing one mouse from two separate experiments.

Fig. 4

SCI suppresses CD4⁺ T cell effector functions following infection. T3-injured mice were infected with 1 × 10⁴ PFU MHV 1 week post-injury, and splenocytes harvested at day 5 p.i. Cells were stimulated with CD4 Ag-specific peptide (epitope M133–147), CD8 Ag-specific peptide (epitope S598–605), or non-specific OVA control peptide to determine proliferation and IFN-γ production. A) The frequency of proliferating CFSE-labeled CD8⁺ responding to CD8 Ag-specific peptide was similar in T3-injured and un-injured mice. B) There was no difference in proliferation of Ag-specific CD8⁺ T cells, indicated by similar total number of CFSE⁺ CD8⁺ T cells between injured and un-injured mice. C) T3-injured mice showed a decrease in the frequency of proliferating CFSE-labeled CD4⁺ responding to CD4 Ag-specific peptide compared un-injured mice, but the difference was not significant. D) T3-injured mice, however, showed a significant decrease in proliferation of Ag-specific CD4⁺ T cells (*p ≤ 0.007), compared to un-injured mice, as indicated by the total number of CFSE⁺ CD4⁺ T cells. E) ELISA assay revealed splenocytes from infected mice that were stimulated with CD8 Ag-specific peptide induced similar IFN-γ protein expression between T3-injured and un-injured mice, while (F) CD4 Ag-specific peptide stimulation resulted in significantly reduced IFN-γ expression in T3-injured mice compared to un-injured mice (*p ≤ 0.001). CD8 or CD4-Ag stimulation of splenocytes from uninfected mice induced very low, yet comparable levels of IFN-γ between T3-injured and un-injured mice (E and F). Representative dot plots are shown in A and C, and the frequency (mean ± SEM) of dual-positive cells indicated above the gate. The corresponding number of proliferating CFSE-labeled virus-specific T cells shown in B and D, is relative to the number of CD8⁺ and CD4⁺ virus-specific T cells, respectively. Cytokine expression in E and F is reported as fold-change to OVA control peptide stimulation. The mean number of cells is presented in bar graphs, with each data point representing one mouse from two separate experiments.

Fig. 5

Infection of T3-injured mice results in decreased activated macrophages in the liver. 1 week following SCI, mice were infected with 1 × 10⁴ PFU MHV and livers collected and processed for flow cytometric analysis of activated macrophages. A) At day 5 p.i., the numbers of macrophages in the liver were similar between T3-injured and un-injured mice. However, the frequencies of gated MHC II⁺ CD11b^hi activated macrophages (B) and the corresponding numbers (C) were significantly reduced in T3-injured mice, compared to uninjured mice following infection (*p ≤ 0.003). Prior to infection, the total numbers of CD11b^hi macrophages and CD11b^hi MHC II⁺ activated macrophages were minimal, yet similar between T3-injured and un-injured uninfected mice (A and C). The mean number of cells is presented in bar graphs (A and C) with each data point representing one mouse from two separate experiments. Representative dot plots from infected mice are shown in B, and the frequency (mean ± SEM) of dual-positive cells is indicated in the upper right quadrant.

Fig. 6

MHV-specific CD4⁺ T cell infiltration to the liver is altered following SCI. 1 week following SCI, mice were infected with 1 × 10⁴ PFU MHV, and livers collected at day 5 p.i. Tissues were processed for ex vivo peptide stimulation and subsequent flow cytometric analysis to determine the infiltration of adaptive T lymphocytes to the liver. Following MHV infection, the frequency (A, lower gate) and numbers (C) of CD8⁺ T cells in the liver were comparable between T3-injured and un-injured mice. T3-injured mice showed significant decrease in frequency (B, lower gate) and numbers (E) of CD4⁺ T cells in the liver compared to un-injured mice (*p ≤ 0.006). In addition, there was significant decrease in the infiltration of Ag-specific CD4⁺ T cells, indicated by decreased frequency (B, upper gate) and total numbers (F) of virus-specific CD4⁺ T cells expressing IFN-γ in T3-injured mice compared to un-injured mice, as determined by intracellular cytokine staining following stimulation with CD4 epitope M133–147 (*p ≤ 0.01). There were similar frequencies (A, upper gate) and numbers (D) of virus-specific CD8⁺ IFN-γ⁺ T cells between T3-injured and un-injured mice as determined by intracellular cytokine staining following stimulation with CD8 epitope S598–605. Prior to infection, the total numbers of assessed T lymphocyte populations were minimal, yet similar between T3-injured and un-injured mice (C–F). Representative dot plots are shown in A and B, with the frequency (mean ± SEM) of CD8 and CD4 single-positive cells, indicated in lower gate, and the corresponding number in C and E bar graphs. Representative dot plots also show the frequency (mean ± SEM) of CD8/IFN-γ or CD4/IFN-γ dual-positive cells, indicated in upper gate, and the corresponding number in D and F bar graphs. The mean number of cells is presented in bar graphs (C–F) with each data point representing one mouse from two separate experiments.