Characterization of a fatal feline panleukopenia virus derived from giant panda with broad cell tropism and zoonotic potential

Abstract

Represented by feline panleukopenia virus (FPV) and canine parvovirus (CPV), the species carnivore protoparvovirus 1 has a worldwide distribution through continuous ci13rculation in companion animals such as cats and dogs. Subsequently, both FPV and CPV had engaged in host-to-host transfer to other wild animal hosts of the order Carnivora. In the present study, we emphasized the significance of cross-species transmission of parvoviruses with the isolation and characterization of an FPV from giant panda displaying severe and fatal symptoms. The isolated virus, designated pFPV-sc, displayed similar morphology as FPV, while phylogenetic analysis indicated that the nucleotide sequence of pFPV-sc clades with Chinese FPV isolates. Despite pFPV-sc is seemingly an outcome of a spillover infection event from domestic cats to giant pandas, our study also provided serological evidence that FPV or other parvoviruses closely related to FPV could be already prevalent in giant pandas in 2011. Initiation of host transfer of pFPV-sc is likely with association to giant panda transferrin receptor (TfR), as TfR of giant panda shares high homology with feline TfR. Strikingly, our data also indicate that pFPV-sc can infect cell lines of other mammal species, including humans. To sum up, observations from this study shall promote future research of cross-host transmission and antiviral intervention of Carnivore protoparvovirus 1, and necessitate surveillance studies in thus far unacknowledged potential reservoirs.

Keywords: TfR; cross-species transmission; fatal; feline panleukopenia virus; giant panda.

Figure 1

Characterization of the infectivity and morphology of pFPV-sc. (A) Cytopathic effect induced by pFPV-sc (clonal virus population) in F81 cells. (B) Morphology of pFPV-sc particles exhibited with negative-stained transmission electron microscopy. (C) Growth curve of pFPV-sc as measured by end-point dilution. Each datapoint shows averages of three independent replicates, and standard deviations are indicated as error bars.

Figure 2

VP2-based phylogenetic analysis indicates pFPV-sc clades with recent FPV strains. The tree was generated based on full-length VP2 sequence of different FPVs using the neighbor-joining method with the Jukes-Cantor algorithm of distance correction, with bootstrapping up to 1000 replicates. GenBank accession numbers are specified for each reference strain. The three FPV clades, namely clade A, B and C, were highlighted with different colors, with the branch of pFPV-sc highlighted in red. CPV was used as out group.

Figure 3

pFPV-sc has specific hemagglutination ability towards porcine erythrocytes. (A) Hemagglutination assay (HAA) of pFPV-sc performed with giant panda, human, rat, pig and rabbit erythrocytes. Twofold serial dilutions of pFPV-sc, starting at 10⁴ TCID₅₀ per well, were mixed 1:1 with 0.5% erythrocytes diluted in PBS. Hemagglutination was assessed after 2 hour incubation on ice. HAAs were repeated at least three times and representative experiments are shown. (B) Hemagglutination of pFPV-sc is not sensitive to neuraminidase (NA) treatment. Untreated and NA (from Arthrobacter ureafaciens) treated porcine erythrocytes were compared via HAA as in (A), and NA treatment does not affect the outcome. (C) Inhibition of pFPV-sc hemagglutination with hyperimmune serum. Hemagglutination inhibition assay was twofold serial dilutions of unimmunized serum and immunized hyperimmune serum mixed with 8 hemagglutination units of pFPV-sc.

Figure 4

Seroprevalence of parvovirus antibodies in giant panda samples from 2010 to 2018. Reactivity of giant panda serum samples (n=14) against pFPV-sc were measured by virus neutralization assay (A) and hemagglutination inhibition (HI) assay (B). Reactivity profiles of all serum samples are displayed as distribution dot plots, with each data point represents the virus neutralization titer (VNT) or HI titer of a particular sample.

Figure 5

pFPV-sc can infect cells other than feline origin. Immunofluorescent staining was performed with hyperimmune serum upon pFPV-sc infected human, feline, porcine and African green monkey cell lines. Green fluorescence indicates infection; scale bars, 250 μm. Note that three out of four human cell lines tested (HRT-18, HEK-293T and Hela) are susceptible to pFPV-sc. This panel shows representative micrographs from at least three repeats.

Figure 6

Bioinformatical and structural analysis indicates giant panda transferrin receptor (TfR) as a putative receptor for pFPV-sc. (A) Giant panda TfR is akin to the feline TfR at the virus-binding motif. Multiple sequence alignment indicated that the majorities of the crucial amino acid residues that forms the virus binding motif (βI-1, βII-1, βII-2, βII-3, βI-6, βII-7 and βII-8 loops) are conserved between feline and giant panda TfRs. Differences between giant panda TfR and feline TfR are marked red. (B) TfR topology are relatively conserved between different host species, with the virus binding motif well exposed. Side view of the dimeric human TfR (PDB accession number: 7ZQS), feline TfR and giant panda TfR (modelled using homology modelling) are shown in surface representation and colored by one monomer gray, and another monomer blue. Virus binding motifs of feline and giant panda TfRs, specified in (A), were depicted in red in the gray-colored monomer.