Hantaan virus infection causes an acute neurological disease that is fatal in adult laboratory mice

Abstract

Hantaan virus, the etiological agent of Korean hemorrhagic fever, is transmitted to humans from persistently infected mice (Apodemus agrarius), which serve as the primary reservoir. Here we demonstrate that several strains of adult Mus musculus domesticus (C57BL/6, BALB/c, AKR/J, and SJL/J) were susceptible to Hantaan virus infection when infected intraperitoneally. First clinical signs were loss of weight, ruffled fur, and reduced activity, which were followed by neurological symptoms, such as paralyses and convulsions. Within 2 days of disease onset, the animals died of acute encephalitis. PCR analysis indicated a systemic infection with viral RNA present in all major organs. Immunohistochemical and in situ hybridization analyses of postmortem material detected viral antigen and RNA in the central nervous system (predominantly brain), liver, and spleen. In the central nervous system, viral antigen and RNA colocalized with perivascular infiltrations, the predominant pathological finding. To investigate the involvement of the interferon system in Hantaan virus pathogenesis, we infected alpha/beta interferon receptor knockout mice. These animals were more susceptible to Hantaan virus infection, indicating an important role of interferon-induced antiviral defense mechanisms in Hantaan virus pathogenesis. The present model may help to overcome shortcomings in the development of therapeutic and prophylactic measurements against hantavirus infections.

FIG. 1.

Laboratory mice are susceptible to HTNV infection. (A) Different strains of laboratory mice were intraperitoneally infected with 10⁵ PFU of HTNV strain 76-118. Except for AKR/J mice, all other animals were susceptible to infection and showed similar but sometimes delayed disease courses. (B) C57BL/6 mice were used to determine the LD₅₀ by intraperitoneal infection with 10-fold dilutions of the virus stock. The LD₅₀ was calculated to be approximately 60 PFU. (C) Knockout mice lacking a functional receptor for IFN-α/β (IFNAR-1^−/− mice) were infected intraperitoneally with 100 LD₅₀ of HTNV. Weight loss was registered earlier for IFNAR-1^−/− mice than for C57BL/6 mice. The greater susceptibility to infection may indicate a role of IFN-α/β in antiviral defense. p.i., postinfection.

FIG. 1.

Laboratory mice are susceptible to HTNV infection. (A) Different strains of laboratory mice were intraperitoneally infected with 10⁵ PFU of HTNV strain 76-118. Except for AKR/J mice, all other animals were susceptible to infection and showed similar but sometimes delayed disease courses. (B) C57BL/6 mice were used to determine the LD₅₀ by intraperitoneal infection with 10-fold dilutions of the virus stock. The LD₅₀ was calculated to be approximately 60 PFU. (C) Knockout mice lacking a functional receptor for IFN-α/β (IFNAR-1^−/− mice) were infected intraperitoneally with 100 LD₅₀ of HTNV. Weight loss was registered earlier for IFNAR-1^−/− mice than for C57BL/6 mice. The greater susceptibility to infection may indicate a role of IFN-α/β in antiviral defense. p.i., postinfection.

FIG. 2.

HTNV causes a systemic infection in laboratory mice. RNA was isolated from different organs, and RT-PCR was performed with L segment-specific oligonucleotides (see Material and Methods). The amplicons were analyzed on 1% agarose gels and visualized with ethidium bromide. HTNV-specific amplions (1,080 bp) were detected in all organs tested. Lanes 1 and 10, DNA standards; lanes 2 and 9, negative PCR control (H₂O, no template); lane 3, negative control for RNA isolation (no template); lanes 4 to 8, samples from the liver, spleen, lungs, kidneys, and brain, respectively.

FIG. 3.

HTNV infection causes encephalitis in laboratory mice. Sections of different organs were HE stained. The brain sections show perivascular edema (arrow in upper panel) with massive mononuclear cell infiltration in and around vessel walls and leukocytes which adhere to endothelial cells. Strongly basophilic nuclear fragments indicate apoptosis (arrows in lower panel). The liver sections show focal large necrotic areas (arrows in upper panel). A higher magnification of the sections shows giant cells with prominent nucleoli (arrow in lower panel) and a mononuclear cell infiltrate consisting of lymphocytes and monocytes.

FIG. 4.

HTNV antigen can be detected in the brain, spleen, and liver. Sections from the brain, spleen, and liver were prepared for immunohistochemical analysis (see Material and Methods). To detect viral antigen, the sections were incubated with a mouse polyclonal antiserum against HTNV (1:100 dilution) followed by rabbit anti-mouse IgG conjugated to horseradish peroxidase (1:6,000). Viral antigen was detected in neurons (brain), endothelial cells (brain and liver), and lymphoid cells (spleen).

FIG. 5.

HTNV nucleic acid can be detected in the brain and spleen. Sections from the brain and spleen were prepared for in situ hybridization with an S segment-specific, ³²P-labeled RNA probe that was 350 nucleotides long (see Material and Methods). Sections were analyzed by bright- and dark-field microscopy. Viral nucleic acid was detected in neuronal cell layers of the brain and hyperplastic germ centers of the white pulp of the spleen.

FIG. 6.

Immunized mice are partially protected against lethal HTNV challenge. Adult C57BL/6 mice were intraperitoneally and subcutaneously immunized three times with 10⁵ PFU of recombinant vaccinia viruses expressing either the glycoprotein precursor of HTNV (VVHTNV/GPC) or the T7 RNA polymerase (VVT7). Prior to challenge with HTNV (100 LD₅₀), antibody titers were determined by using an HTNV-specific ELISA. In contrast to the control animals (VVT7), all immunized animals (VVHTNV/GPC) developed an antibody response, indicating successful immunization (A). While all control animals died, protective immunity was achieved in 60% of the immunized animals (three of five) (B). OD, optical density; p.i., postinfection.