Targeted blockade in lethal West Nile virus encephalitis indicates a crucial role for very late antigen (VLA)-4-dependent recruitment of nitric oxide-producing macrophages

Abstract

Infiltration of Ly6C(hi) monocytes from the blood is a hallmark of viral encephalitis. In mice with lethal encephalitis caused by West Nile virus (WNV), an emerging neurotropic flavivirus, inhibition of Ly6C(hi) monocyte trafficking into the brain by anti-very late antigen (VLA)-4 integrin antibody blockade at the time of first weight loss and leukocyte influx resulted in long-term survival of up to 60% of infected mice, with subsequent sterilizing immunity. This treatment had no effect on viral titers but appeared to be due to inhibition of Ly6C(hi) macrophage immigration. Although macrophages isolated from the infected brain induced WNV-specific CD4(+) T-cell proliferation, T cells did not directly contribute to pathology, but are likely to be important in viral control, as antibody-mediated T-cell depletion could not reproduce the therapeutic benefit of anti-VLA-4. Instead, 70% of infiltrating inflammatory monocyte-derived macrophages were found to be making nitric oxide (NO). Furthermore, aminoguanidine-mediated inhibition of induced NO synthase activity in infiltrating macrophages significantly prolonged survival, indicating involvement of NO in the immunopathology. These data show for the first time the therapeutic effects of temporally targeting pathogenic NO-producing macrophages during neurotropic viral encephalitis.

Figure 1

VLA-4 is the primary integrin for inflammatory monocyte derived macrophages. Analysis at day 7 p.i. showed (A) CD4⁺ and CD8⁺ T cells, Ly6G⁺ neutrophils (~6%, 7% and 3%, respectively, of all infiltrating leukocytes), and large numbers of Ly6C⁺ macrophages (~50% of infiltrating leukocytes, of which >90% were Ly6C^hi) infiltrating into the brain from day 5 p.i., and (B) lymphocyte function-associated antigen (LFA)-1 and very late antigen (VLA)-4 was expressed on Ly6C^lo and Ly6C^hi blood monocytes. (C) Immunohistochemical staining for intercellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1 performed on day 7 p.i. on the sham-infected and WNV-infected brains of C57BL/6 mice showed significant widespread upregulation of both molecules in WNV-infected mice and co-localized with CD31. Treatment with 100 μg anti-VLA-4 antibody from day 6 p.i. onwards resulted in reduced neutrophil and T cell recruitment into day 7 p.i. WNV-infected brains. A 66% reduction was seen in the CD45⁺/CD11b⁺/Ly6G⁻/Ly6C^hi macrophage compartment. (D) By contrast, LFA-1 blockade reduced Ly6C^hi macrophage infiltration by 33%. (E) VLA-4 blockade resulted in increased survival of 10 to 60%, depending on the inoculation dose (P < 0.001), whereas (F) LFA-1 blockade did not increase survival. (G) Viral titers in the brain at day 7 p.i. were similar in all treatment groups. Data are representative of at least three independent experiments with at least four mice per group. Antibody-mediated effects on the disease course were determined on four separate occasions with at least ten mice per group. Data shown are the mean ± standard deviation. Statistical analysis was conducted using one-way ANOVA with a Tukey-Kramer post hoc test. Statistical analysis of survival was conducted using the log-rank (Mantel-Cox) test. * P < 0.05; **P < 0.001; ***P < 0.001. The inoculation dose used was 6 × 10³ plaque-forming units, unless otherwise noted.

Figure 2

Assessment of macrophages isolated from the West Nile virus (WNV) infected brain. (A, B) Macrophages isolated from day 7 p.i. brain tissue infected with West Nile virus expressed major histocompatability complex (MHC)-II and CD86. (C) More than 70% of Ly6C^hi macrophages in the WNV-infected brain were positive for fluorescent agent (ProSense; VisEn Medical, Bedford, MA, USA), indicative of cathepsin activity. (D) The ability of macrophages (CD45⁺/CD11b⁺/Ly6G⁻/GFP⁺/CD11c⁻/Ly6C^hi) and resident microglia (CD45^lo/int/CD11b⁺/GFP⁻/CD11c⁻/Ly6C^hi), isolated from WNV-infected colony-stimulating factor 1 receptor (cFMS)-enhanced green fluorescent protein (EGFP) chimeric mice [2] to stimulate naïve and activated CD3⁺/CD4⁺ T cells from sham-infected and WNV-infected mice, respectively, was measured. Except at the highest antigen-presenting cell (APC):CD4 ratios, neither population was capable of stimulating naive T-cell proliferation, as measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. (E) Notwithstanding the apparent APC capabilities, depletion of CD8⁺ or CD4⁺ or CD4⁺ and CD8⁺ had no effect on overall survival in this model. All experiments were performed at least twice with four to five mice per group. Cells were sorted from 10 pooled cFMS-EGFP chimeric mice, and data represent the mean of triplicate wells and the standard deviation of absorbance. T-cell depletion was performed at least twice with eight to ten mice per group. Statistical analysis was conducted using one-way ANOVA with a Tukey-Kramer post hoc test. * Indicates significance of comparison of CD4⁺ T-cell proliferation between wells with macrophages and wells with CD4⁺ T cells only. * P < 0.05; **P < 0.001; ***P < 0.001. Mice infected with 6 ×10⁴ plaque-forming units were used to obtain APC populations.

Figure 3

Macrophages produce NO, which is pathogenic during WNV infection. (A-E) Analysis of sections from sham-infected and West Nile virus (WNV)-infected mice at day 7 p.i. showed significant upregulation of nitric oxide synthase (NOS)-2 message in the WNV-infected brains. (C, F) Furthermore, immunohistochemical staining for NOS-2 protein (red) and lectin (green) showed that most cells expressing NOS-2 in the WNV-infected brain parenchyma at day 7 p.i. were lectin-positive. (G) Using the dye 4-amino-5-methylamino-2',7'-difluorofluorescein diacetate (DAF-FM), which, in the presence of NO becomes fluorescent, flow cytometry showed that more than 70% of macrophages (CD45^hi/CD11b⁺/CD11c⁻/Ly6C⁺) expressed NO compared with approximately 25% of resident microglia (CD45^lo/int/CD11b⁺/CD11c⁻/Ly6C⁻), and treatment of mice with aminoguanidine resulted in a significant reduction in macrophages expressing NO. (H) Inhibition of NO with aminoguanidine did not increase survival when used over the entire infectious period, but it had a demonstrably protective effect given at day 6 p.i. (P < 0.05). (I) A significant reduction in DAF-FM-positive macrophages was seen from day 7 p.i. in the brains of WNV-infected mice that had been treated with anti-VLA-4 neutralizing antibodies on day 6 p.i. In situ hybridization was performed on sagittal brain sections from at least six mice per group. DAF-FM data are representative of three experiments with at least three mice per group. Aminoguanidine therapy was performed four times with at least ten mice per group. Values shown are means ± SD. Statistical analysis was conducted using one-way ANOVA with Tukey-Kramer post hoc test. * P < 0.05; **P < 0.001; ***P < 0.001. Statistical analysis of survival was determined using the log-rank (Mantel-Cox) test. *P < 0.05 was considered significant.