Persistence of West Nile virus in the central nervous system and periphery of mice

Abstract

Most acute infections with RNA viruses are transient and subsequently cleared from the host. Recent evidence, however, suggests that the RNA virus, West Nile virus (WNV), not only causes acute disease, but can persist long term in humans and animal models. Our goal in this study was to develop a mouse model of WNV persistence. We inoculated immunocompetent mice subcutaneously (s.c.) with WNV and examined their tissues for infectious virus and WNV RNA for 16 months (mo) post-inoculation (p.i.). Infectious WNV persisted for 1 mo p.i. in all mice and for 4 mo p.i. in 12% of mice, and WNV RNA persisted for up to 6 mo p.i. in 12% of mice. The frequency of persistence was tissue dependent and was in the following order: skin, spinal cord, brain, lymphoid tissues, kidney, and heart. Viral persistence occurred in the face of a robust antibody response and in the presence of inflammation in the brain. Furthermore, persistence in the central nervous system (CNS) and encephalitis were observed even in mice with subclinical infections. Mice were treated at 1 mo p.i. with cyclophosphamide, and active viral replication resulted, suggesting that lymphocytes are functional during viral persistence. In summary, WNV persisted in the CNS and periphery of mice for up to 6 mo p.i. in mice with subclinical infections. These results have implications for WNV-infected humans. In particular, immunosuppressed patients, organ transplantation, and long term sequelae may be impacted by WNV persistence.

Figure 1. Persistence and levels of WNV RNA varies for different tissues.

Adult, female B6 mice were inoculated s.c. with diluent alone (mock) or 10³ PFU of WNV in the left rear footpad, and tissues were harvested from 8 to 9 WNV-inoculated mice and one mock-inoculated mouse at various times p.i.. Levels of WNV genome copies per µg of GAPDH RNA were determined by real-time RT-PCR in (A) skin-inoculation site (left rear footpad), (B) skin-distal sites (right rear footpad and both front footpads), (C) spinal cord, (D) brain, (E) lymph nodes (both popliteal and inguinal lymph nodes), (F) spleen, (G) kidney, and (H) heart. Axis titles are the same for A–H. Each data point represents an individual animal, and the horizontal solid line is the geometric mean. Data points on the x-axis are negative. Data for mock-inoculated mice were negative for WNV RNA at each time point and are not shown on the graphs. Similar results were obtained in three independent studies performed at 1 mo p.i. (n = 8 to 10) and two independent studies performed at 2 mo p.i. (n = 4 or 8).

Figure 2. Mice maintain a robust antibody response during WNV persistence.

Adult, female B6 mice were inoculated s.c. with diluent alone (mock) or 10³ PFU of WNV in the left rear footpad, and sera were harvested by serial tail bleeds or at time of sacrifice (Table S1 for details). Sera were tested for WNV-specific antibodies by (A and B) MIA for anti-WNV E, (C and D) MIA for anti-WNV NS5, and (E) PRNT with 90% endpoint titers. Axis titles are the same for A–D. The number of mice tested at each time point in A, C and E are placed above the whisker-box plots. In B and D, the same mice (n = 15) were bled at each time point, the results for an individual mouse over time is represented by a solid line, and the number of mice positive at each time point is listed at the top of the graphs. The horizontal, dotted lines correspond to the LOD, and data points below the dotted line are negative. Data for mock-inoculated mice were negative for WNV-specific antibodies in all assays and are not shown on the graphs.

Figure 3. Histopathologic changes persist in the brains of mice during WNV persistence.

Adult, female B6 mice were inoculated s.c. with diluent alone (mock) or 10³ PFU of WNV in the left rear footpad, and brains were harvested for histopathology from 8 WNV-inoculated mice and one mock-inoculated mouse at 1, 2, 4, 6 and 9 mo p.i.. Representative photomicrographs of brains are shown at (A and B) 1 mo p.i. in the thalamus, (C and D) 2 mo p.i. in the brainstem, and (E and F) 4 mo p.i. in the midbrain for WNV-inoculated mice (left panels) and mock-inoculated mice (right panels). Black arrows point to areas of inflammation. Sections were stained with hematoxylin and eosin, and photomicrographs are shown at a magnification of 200× (A, B, C, D, and F) and 400× (E).