High genomic stability of Andes virus following successive passage in vivo in Syrian hamsters

Abstract

Andes virus (ANDV) is a South American hantavirus that causes hantavirus cardiopulmonary syndrome (HCPS), a severe respiratory infection with a case fatality rate as high as 40%. A critical model for studying HCPS is the Syrian hamster model, which faithfully recapitulates key aspects of the disease. Recent studies have shown that strains of ANDV from different origins, i.e., from rodent or human hosts, can produce widely different outcomes in the hamster model. CHI-7913, a human isolate, does not cause lethal disease and replicates poorly in hamster tissues compared with the classical rodent isolate Chile-9717869. Here, we passaged CHI-7913 in vivo in Syrian hamsters 25 times and infected groups of hamsters with virus from passages 2, 12, and 24 to assess whether viral adaptation occurs that might ultimately drive the phenotype of the virus during infection in this model. Very few mutations occurred during passage in hamsters, with only a single coding mutation discovered after 25 passages. Nearly all mutations occurring in hamsters occurred early, within two passages, and none of the passaged virus caused overt disease or pathology in infected hamsters, consistent with minimal changes in the viral genome. Early in infection, CHI-7913 viral RNA levels were still well below those seen in lethal Chile-9717869 infection. Overall, our data offer insights into the lack of selective pressure on hantaviruses in certain hosts and further evidence of distinct outcomes between rodent and human-derived viral isolates that should be studied further in additional viral species.IMPORTANCEInfection of Syrian hamsters with Andes virus (ANDV) can result in disparate outcomes, depending on the source of the virus used for the infection. The ANDV strain CHI-7913 does not cause lethal disease in hamsters but is able to replicate in hamster tissues. We successively passaged CHI-7913 in vivo to study how continued infection influences adaptation in hamsters and whether passaging would lead to the development of a lethal model, as sometimes occurs with other viruses. Surprisingly, even after 25 passages, minimal mutations occurred in ANDV CHI-7913 genomes, with only one coding mutation present above consensus by passage 25, in the viral glycoprotein. Our data suggest that depending on both viral origin and the host, hantaviruses may face minimal selective pressure to mutate toward a disease phenotype. Studies with additional ANDV strains and other hantavirus species are warranted to further study this phenomenon.

Keywords: Andes virus; disease modeling; hantavirus; hantavirus pulmonary syndrome; orthohantavirus; pathogenesis; virulence.

Fig 1

Frequency of non-synonymous and non-coding mutations in the ANDV CHI-7913 genome throughout the hamster adaptation passaging. Mutations for the ANDV CHI-7913 (A) S segment, (B) M segment, and (C) L segment are plotted across the passages. Solid lines represent mutations that reached consensus, and dashed lines represent mutations that did not.

Fig 2

Survival and replication of Andes virus in Syrian hamsters. (A) Survival of hamsters infected with either ANDV strain Chile-9717869 or in vivo P2, P12, or P24 of ANDV CHI-7913. (B) Viral RNA levels in the tissues or serum of infected hamsters on 3, 7, and 12 DPI. (C) ANDV N-specific IgG endpoint titers in hamsters infected with ANDV strain Chile-9717869 or in vivo P2, P12, or P24 of ANDV CHI-7913. n = 16 hamsters per virus, with 4 euthanized on 3, 7, and 12 DPI and 4 kept for survival analysis (28 DPI). In B and C, medians and individual data points are shown. Statistical significance assessed by Log rank test in A, and by one-way ANOVA in B. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig 3

Hematological parameters in ANDV-infected hamsters. Hamsters were infected with either ANDV strain Chile-9717869 or in vivo P2, P12, or P24 of ANDV CHI-7913, and complete blood counts were performed with whole blood on 3, 7, and 12 DPI. Medians and individual data points are shown. Significance was assessed by one-way ANOVA. *P < 0.05, **P < 0.01.

Fig 4

Serum biochemical parameters in ANDV-infected hamsters. Hamsters were infected with either ANDV strain Chile-9717869 or in vivo P2, P12, or P24 of ANDV CHI-7913, and serum biochemical analysis was performed with fresh serum on 3, 7, and 12 DPI. Medians and individual data points are shown. Significance was assessed by one-way ANOVA. *P < 0.05, **P < 0.01.

Fig 5

Histopathology in ANDV-infected hamsters. Hamsters were infected with ANDV CHI-7913 (P2, P12, P24) or Chile-9717869 via the intranasal route. (A) Hematoxylin and eosin staining was performed on the lungs from ANDV-infected hamsters taken at 7, 12, and 28 DPI (n  =  2/group). (B and C) Lungs from ANDV-infected hamsters taken at 7, 12, and 28 DPI were stained with rabbit hyperimmune serum and counterstained with hematoxylin (n  =  2/group). Representative sections are shown.

Fig 6

Cytokine production in ANDV-infected hamsters. Hamsters were infected with either ANDV strain Chile-9717869 or in vivo P2, P12, or P24 of ANDV CHI-7913, and cytokine levels in serum were assessed by Luminex assay. Mean + SEM are shown for each group at each time point. n = 4 per group per day post-infection.