The First Nonmammalian Pegivirus Demonstrates Efficient In Vitro Replication and High Lymphotropism

Abstract

Members of the Pegivirus genus, family Flaviviridae, widely infect humans and other mammals, including nonhuman primates, bats, horses, pigs, and rodents, but are not associated with disease. Here, we report a new, genetically distinct pegivirus in goose (Anser cygnoides), the first identified in a nonmammalian host species. Goose pegivirus (GPgV) can be propagated in goslings, embryonated goose eggs, and primary goose embryo fibroblasts, and is thus the first pegivirus that can be efficiently cultured in vitro Experimental infection of GPgV in goslings via intravenous injection revealed robust replication and high lymphotropism. Analysis of the tissue tropism of GPgV revealed that the spleen and thymus were the organs bearing the highest viral loads. Importantly, GPgV could promote clinical manifestations of goose parvovirus infection, including reduced weight gain and 7% mortality. This finding contrasts with the lack of pathogenicity that is characteristic of previously reported pegiviruses.IMPORTANCE Members of the Pegivirus genus, family Flaviviridae, widely infect humans and other mammals, but are described as causing persistent infection and lacking pathogenicity. The efficiency of in vitro replication systems for pegivirus is poor, thus limiting investigation into viral replication steps. Because of that, the pathogenesis, cellular tropism, route of transmission, biology, and epidemiology of pegiviruses remain largely uncovered. Here, we report a phylogenetically distinct goose pegivirus (GPgV) that should be classified as a new species. GPgV proliferated in cell culture in a species- and cell-type-specific manner. Animal experiments show GPgV lymphotropism and promote goose parvovirus clinical manifestations. This study provides the first cell culture model for pegivirus, opening new possibilities for studies of pegivirus molecular biology. More importantly, our findings stand in contrast to the lack of identified pathogenicity of previously reported pegiviruses, which sheds lights on the pathobiology of pegivirus.

Keywords: goose; pegivirus.

FIG 1

Schematic presentation of the GpgV genome and location of PCR fragments used for its sequencing and predicted RNA secondary structures in the 5′ region of GPgV genome. (A) Two-way arrows indicate the relative sizes and locations of the overlapping PCR fragments. (B) Sequence corresponding to the 5′ UTR and 14 nt located downstream of the translation initiation codon (indicated in blue) was analyzed using Mfold. IRES-associated sequences (GNRA motif and PPT) are indicated.

FIG 2

Characterization of the GPgV genome and polyprotein. (A) GPgV-1 genome and polyprotein organization. Putative cleavage sites within the polyprotein are indicated by a red rhombus (structural proteins) or an orange rhombus (nonstructural proteins). Sequence alignment shows a comparison of the predicted cleavage sites for cellular signalase and viral proteases in the GPgV-1 polyprotein with other pegiviruses. (B) Locations and sequence comparisons of the conserved NS2 protease motif, NS3 protease catalytic triad, zinc-binding residues, Walker A, Walker B, and NS5B RNA-dependent RNA polymerase motifs.

FIG 3

Phylogenetic analysis of GPgV relative to other pegiviruses. Phylogenetic analysis was carried out using MEGA 7 software. (A) Phylogenetic trees of GPgV and representative strains from the Pegivirus genus built using the neighbor-joining method based on the sequences of NS3 (upper panel) and NS5B (lower panel) proteins. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (200 replicates) is shown next to the branches. The evolutionary distances were computed using the JTT matrix-based method and are presented in units of the number of amino acid substitutions per site. (B) Amino acid sequence divergence between polyproteins of GPgV, TDAV, EPgV, and SPgV. The sliding window size is 50 amino acid residues.

FIG 4

GPgV reduces weight gain and results in changes in the thymus of infected goslings. (A) Schematic representation of the study design. Three-day-old goslings from the experimental group (n = 24) were infected intravenously with 200 μl of allantoic fluid from GPgV-infected goose embryos (1.16 × 10¹⁰ RNA copies/ml), while goslings from the control group (n = 20) were mock infected with 200 μl PBS. Every gosling was weighed at PID 1, 4, 7, 10, 13, 16, 19, 22, 25, and 28. At 30 PID, surviving birds were sacrificed and spleen, thymus, bursa of Fabricius, duodenum, rectum, pancreas, and liver were sampled for HE staining, ICH, and tissue viral load detection. (B) Goslings were weighed at the indicated days postinfection. Mean values are shown, and error bars indicate the standard deviation: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant (Student’s t test). (C) The viral RNA copy numbers in the spleen (red dots) and the body weights (black column) of the infected goslings at PID 30. (D) Upper: representative HE staining of the thymus in the control group (left) and experimental group (right) at PID 30. Bars = 20 μm. Lower: the gross findings of the thymus from mock-infected (left) and GPgV-infected geese (right) at PID 30. (E) The detection of GPgV in the thymus and spleen of infected goslings at PID 30 using an anti-GPgV E2 monoclonal antibody. Bars = 20 μm.

FIG 5

GPgV coinfection reduces weight gain and survival of GPV-carrying goslings. (A) Schematic representation of the study design. Two-day-old GPV-positive goslings (n = 54) were intravenously infected with 200 μl of allantoic fluid containing 1.16 × 10¹⁰ GPgV copies/ml, while the control group (n = 46) was mock infected with 200 μl of PBS. (B) Three randomly selected live goslings from both groups were weighed at the indicated days postinfection. Mean values are shown, and the error bars indicate the standard deviation: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant (Student’s t test). (C) Kaplan-Meier survival plot of GPV-infected and GPgV+GPV-coinfected goslings; birds were monitored up to 21 days postinfection.

FIG 6

Growth characteristics of GPgV in 9-day-old embryonated goose eggs. (A) GPgV stock (10th generation) was passaged in 9-day-old embryonated goose eggs four consecutive times; for each passage, allantoic fluid was harvested at 168 h p.i. The virus genome copy number in the 11th, 12th, 13th, and 14th generation stocks was determined by qRT-PCR. Each symbol represents an individual egg (n = 5, 5, 5, 4 for each generation, respectively); the line depicts the mean of each group, and the error bars indicate the standard deviation. (B) GPgV from the 14th passage (200 μl, 4.55 × 10⁹ RNA copies/ml) was inoculated into embryonated eggs (n = 4 for each time point). At the indicated time points, virus genome copy numbers in the allantoic fluid were determined by qRT-PCR; data are presented as described for panel A.

FIG 7

Growth characteristics of GPgV in GEFs. (A) GPgV derived from allantoic fluid was passaged 8 times in GEFs. The virus genome copy number in the cell culture supernatant was measured for every passage. (B) The GEFs were infected with cell-culture-adapted GPgV (7th passage, 1.15 × 10⁸ RNA copies/ml) and, at the indicated times, culture supernatants and cells lysates, which were obtained using three freeze-thaw cycles, were collected. The virus genome copy number was determined by qRT-PCR. The cell lysate indicates amount of viral RNAs in 1 ml of lysate. (C) GPgV negative-strand RNA assay. GPgV-spreverse primers were used for a reverse transcription reaction of GPgV negative-strand RNA assay. H2O (no primer) used for a reverse transcription reaction functioned as the blank control. Nested RT-PCR using in vitro-transcribed RNA for a template functioned as the negative control. Lanes: 1, marker; 2, GEF-1; 3, GEF-4; 4, GEF-8; 5, thymus; 6, spleen; 7, small intestine. 8, marker; 9, GEF-1; 10, GEF-4; 11, GEF-8; 12, thymus; 13, spleen; 14, small intestine. 15, marker; 16, positive-sense RNA. (D) Detection of the GPgV E2 protein in GEFs at 96 h p.i. Control cells (Mock) were mock infected with PBS. The samples were treated with a mouse anti-GPgV E2 monoclonal antibody as the primary antibody and FITC goat anti-mouse IgG as the secondary antibody (green); nuclei were counterstained with DAPI (blue). Fluorescence was detected using a Nikon 80i microscope. Scale bars indicate 20 μm. (E) The viability of GPgV-infected and mock-infected GEFs was tested by a CCK8 assay. (F) GEFs grown in a 6-well plate were infected with GPgV (7th passage, 1.15 × 10⁸ RNA copies/ml) and/or GPV (6.46 × 10⁶ DNA copies/ml). The numbers of GPgV genome copies (left) and GPV DNA copies (right) were determined at 168 h p.i. All the experiment was repeated three times. Mean values are shown, error bars indicate the standard deviation: *, P < 0.05; ns, not significant (Student’s t test).