Identification of chicken CAR homology as a cellular receptor for the emerging highly pathogenic fowl adenovirus 4 via unique binding mechanism

Abstract

Since 2015, the prevalence of severe hepatitis-hydropericardium syndrome, which is caused by the novel genotype fowl adenovirus serotype 4 (FAdV-4), has increased in China and led to considerable economic losses. The replication cycle of FAdV-4, especially the emerging highly pathogenic novel genotype FAdV-4, remains largely unknown. The adenovirus fibre interacts with the cellular receptor as the initial step in adenovirus (AdV) infection. In our previous studies, the complete genome sequence showed that the fibre patterns of FAdV-4 were distinct from all other AdVs. Here, protein-blockage and antibody-neutralization assays were performed to confirm that the novel FAdV-4 short fibre was critical for binding to susceptible leghorn male hepatocellular (LMH) cells. Subsequently, fibre 1 was used as bait to investigate the receptor on LMH cells via mass spectrometry. The chicken coxsackie and adenovirus receptor (CAR) protein was confirmed as the novel FAdV-4 receptor in competition assays. We further identified the D2 domain of CAR (D2-CAR) as the active domain responsible for binding to the short fibre of the novel FAdV-4. Taken together, these findings demonstrate for the first time that the chicken CAR homolog is a cellular receptor for the novel FAdV-4, which facilitates viral entry by interacting with the viral short fibre through the D2 domain. Collectively, these findings provide an in-depth understanding of the mechanisms of the emerging novel genotype FAdV-4 invasion and pathogenesis.

Keywords: D2 domain; Emerging FAdV-4; cell receptor; chicken CAR; novel binding mechanism; short fibre.

Figure 1.

Fibre 1 is the short fibre protein of the novel FAdV-4. (A) The mode pattern of distinct fibres in different AdVs. (B) LMH cells were infected with FAdV-4 at a MOI of 0.1 and a cell lysate was prepared for western blot analysis, using pAbs against fibre 1 and fibre 2. The fibre 2 protein (70 kDa) was clearly larger than fibre 1 (45 kDa).

Figure 2.

Fibre 1 determined the infectivity of FAdV-4. SDS-PAGE analysis of purified soluble fibre 1 (A) and fibre 2 (B) proteins. Different concentrations (3.125, 12.5, or 50 ng/μl) of the fibre 1 or fibre 2 protein were pre-incubated with LMH cells at 37°C for 1 h. Next, the cells were inoculated with FAdV-4 (MOI = 0.1) and further incubated for 1 h at 37°C. After washing the cells three times with PBS, they were cultured for 24 h under normal culture conditions. Finally, the samples were collected for qPCR analysis. The fibre 1 protein significantly blocked the infectivity of FAdV-4 (P < 0.05) (C), but the fibre 2 protein did not (P ≥ 0.05) (D). Different dilutions of antibodies against fibre 1 or fibre 2 (or mouse IgG as a negative control) were incubated with 0.1 MOI FAdV-4 at 37°C for 1 h and then washed twice with PBS. Thereafter, the mixture was added to LMH cells and incubated for another 1 h at 37°C. After washing the cells three times, they were maintained under normal culture conditions and collected at 24 hpi for qPCR analysis. The antibody against fiber1 (E), but not fibre 2 (F), blocked FAdV-4 infection in a concentration-dependent manner. The data are presented as the mean ± SD of at least three independent experiments.

Figure 3.

The chicken CAR homolog was identified as a protein that interacts with FAdV-4 fibre-1. Fibre 1 interaction with cellular proteins were analyzed by LC/MS and confirmed by western blotting. (A) LMH cell lysates were incubated with Fc-tagged fibre 1 or Fc protein (direct bound to protein A/G) for 6 h and then subjected to SDS-PAGE. An additional band was observed by silver staining in the fibre 1 sample, compared to the negative control sample. The band was identified as the chicken CAR homolog (red box) by LC/MS. (B) The band was further confirmed to be CAR by performing western blot analysis with a CAR-specific antibody.

Figure 4.

CAR facilitated FAdV-4 infection. To determine whether CAR influenced the infectivity of FAdV-4, RNAi and overexpression assays was performed. (A) All of the CAR siRNAs tested effectively downregulate the mRNA transcription of CAR. (B) The tested siRNAs effectively depleted CAR protein expression. (C) After pre-treatment with CAR siRNA for 24 h, LMH cells were inoculated with FAdV-4 (MOI = 0.1) and the relative infection levels were measured after 24 h. FAdV-4 infection was significantly inhibited by RNAi (p < 0.05). (D) CAR overexpression promoted the infectivity of FAdV-4. FAdV-4 (MOI = 1) was incubated with CAR-overexpressing LMH cells for 1 h and then the distribution of the virus was analyzed. FAdV-4 (green fluorescence) was located on the membrane of mock-treated LMH cells. However, following CAR overexpression (red fluorescence), FAdV-4 migrated into the cytoplasm and co-localized with the CAR protein. The data shown are presented as the mean ± SD of at least three independent experiments.

Figure 5.

Both soluble CAR and a CAR-specific antibody blocked the infection of FAdV-4. The proper expression and purification of Fc-tagged CAR were identified by SDS-PAGE (A) and western blotting (B). (C) Different concentrations of the soluble CAR protein, ranging from 3.125 to 12.5 ng/μl (with the Fc protein serving as a negative control), were used in the protein-based blockade assay. All detected concentrations of soluble CAR protein could block FAdV-4 infection by 76% to 80% (p < 0.05). (D) In the antibody-based blockade assay, we applied different dilutions of the CAR pAb, ranging from 1:2 to 1:32, with mouse IgG serving as a negative control. The CAR antibody could block the infectivity of FAdV-4 in a concentration-dependent manner (1:2 dilution, 73% inhibition; 1:8 dilution, 20% inhibition; 1:32 dilution, 0% inhibition) (p < 0.05). The data shown are presented as the mean ± SD of at least three independent experiments.

Figure 6.

CAR conferred FAdV-4 entry into non-permissive cells. (A) Non-permissive 293 T cells were transfected for 24 h with a pCAGGS vector driving HA-tagged CAR overexpression (empty pCAGGS vector, negative control), followed by FAdV-4 infection (MOI = 1) for 24 h. The samples were analyzed by indirect immunofluorescence using rabbit antibodies against the HA tag and the FAdV-4 Hexon protein. FAdV-4 (green fluorescence) co-localized with CAR (red fluorescence) in the cytoplasm of 293 T cells, whereas virus and CAR were not detected in the negative-control cells. CAR overexpressing 293 T cells (B) and PK-15 cells (C) were infected with FAdV-4 (MOI = 1) and collected for qPCR detection at 24 hpi. The results showed that non-permissive cells could confer viral entry into cells, with the viral load increased by 4.79-fold in 293 T cells and 3.78-fold in PK-15 cells (P < 0.05). The data shown are presented as the mean ± SD of at least three independent experiments.

Figure 7.

The D2 domain of CAR enabled FAdV-4 infection. (A) The mode pattern of CAR functional domains in different species. The Gallus CAR Ig V-set domain (D1) and the Ig domain (D2) are shown in the red box. (B) SDS-PAGE detection of soluble His-tagged variants of the D1, D2, and ECD (D1 + D2) domain (red box). (C) Western blot-based identification of the D1, D2, and ECD domains of CAR. Soluble CAR D1, D2, and ECD proteins were used in the protein-based blockade assay. CAR D2 at concentrations of 12.5 ng/μl (D) and 25 ng/μl (E) block the infectivity with efficacies of 78% and 99.9% (P < 0.05), respectively, in contrast to the D1 domain (which did not block infection). The data shown are presented as the mean ± SD of at least three independent experiments.