Role of Small Hydrophobic Protein of J Paramyxovirus in Virulence

Paramyxoviruses are associated with many devastating diseases in animals and humans. J paramyxovirus (JPV) was isolated from moribund mice in Australia in 1972. Newly isolated viruses, such as Beilong virus (BeiPV) and Tailam virus (TlmPV), have genome structures similar to that of JPV. A new paramyxovirus genus, Jeilongvirus, which contains JPV, BeiPV, and TlmPV, has been proposed. Small hydrophobic (SH) protein is present in many paramyxoviruses. Our present study investigates the role of SH protein of JPV in pathogenesis in its natural host. Understanding the pathogenic mechanism of Jeilongvirus is important to control and prevent potential diseases that may emerge from this group of viruses.

ABSTRACT

J paramyxovirus (JPV) was first isolated from moribund mice with hemorrhagic lung lesions in Australia in 1972. It is a paramyxovirus classified under the newly proposed genus Jeilongvirus. JPV has a genome of 18,954 nucleotides, consisting of eight genes in the order 3′-N-P/V/C-M-F-SH-TM-G-L-5′. JPV causes little cytopathic effect (CPE) in tissue culture cells but severe disease in mice. The small hydrophobic (SH) protein is an integral membrane protein encoded by many paramyxoviruses, such as mumps virus (MuV) and respiratory syncytial virus (RSV). However, the function of SH has not been defined in a suitable animal model. In this work, the functions of SH of JPV, MuV, and RSV have been examined by generating recombinant JPV lacking the SH protein (rJPV-ΔSH) or replacing SH of JPV with MuV SH (rJPV-MuVSH) or RSV SH (rJPV-RSVSH). rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH were viable and had no growth defect in tissue culture cells. However, more tumor necrosis factor alpha (TNF-α) was produced during rJPV-ΔSH infection, confirming the role of SH in inhibiting TNF-α production. rJPV-ΔSH induced more apoptosis in tissue culture cells than rJPV, rJPV-MuVSH, and rJPV-RSVSH, suggesting that SH plays a role in blocking apoptosis. Furthermore, rJPV-ΔSH was attenuated in mice compared to rJPV, rJPV-MuVSH, and rJPV-RSVSH, indicating that the SH protein plays an essential role in virulence. The results indicate that the functions of MuV SH and RSV SH are similar to that of JPV SH even though they have no sequence homology.

IMPORTANCE Paramyxoviruses are associated with many devastating diseases in animals and humans. J paramyxovirus (JPV) was isolated from moribund mice in Australia in 1972. Newly isolated viruses, such as Beilong virus (BeiPV) and Tailam virus (TlmPV), have genome structures similar to that of JPV. A new paramyxovirus genus, Jeilongvirus, which contains JPV, BeiPV, and TlmPV, has been proposed. Small hydrophobic (SH) protein is present in many paramyxoviruses. Our present study investigates the role of SH protein of JPV in pathogenesis in its natural host. Understanding the pathogenic mechanism of Jeilongvirus is important to control and prevent potential diseases that may emerge from this group of viruses.

INTRODUCTION

J paramyxovirus (JPV) is a member of the genus Jeilongvirus of the Paramyxoviridae family. Jeilongvirus is a newly proposed genus that includes paramyxoviruses isolated from rodents or as a contaminant from rodent cell culture (1, 2). JPV was first isolated in northern Queensland, Australia, in 1972. Extensive hemorrhagic lung lesions were reported in mice from which the virus was isolated. JPV produced characteristic syncytia in kidney autoculture monolayers, and electron microscopy of the virus revealed the herringbone-shaped nucleocapsid typical of paramyxoviruses (3). The full-length genome of JPV was sequenced and contains 18,954 nucleotides and eight genes in the order 3′-N-P/V/C-M-F-SH-TM-G-L-5′. The SH gene and TM gene encode integral membrane proteins, the small hydrophobic (SH) and transmembrane (TM) proteins, which are 69 and 258 amino acids (aa), respectively (4). TM is a type II glycosylated integral membrane protein, which promotes cell-to-cell fusion (5). JPV has a fusion (F) protein, which is predicted to be a type I membrane protein. JPV G is the largest paramyxovirus attachment protein sequenced to date. The G gene encodes a putative 709-aa residue attachment protein and a distal second open reading frame (ORF), termed ORF-X, which has not yet been detected in infected cells. Nucleotide probes specific for both the G-coding and ORF-X regions identified a mRNA species matching the G gene (4, 6). Beilong virus (BeiPV) and Tailam virus (TlmPV) are also included in the same proposed genus due to their identical genome organizations and isolation from a rodent source. BeiPV was isolated from rat and human mesangial cell lines. TlmPV was isolated from Sikkim rats (Rattus andamanensis). Genomic and phylogenetic evidence supports the grouping of these viruses within the new genus Jeilongvirus (1, 7, 8). Jeilongviruses are isolated from bats (9), demonstrating their zoonotic potential, as bats are the natural reservoirs of zoonotic paramyxoviruses like Nipah and Hendra viruses.

There are two different strains of JPV: JPV-LW and JPV-BH. JPV-LW is not pathogenic in mice, but JPV-BH is highly pathogenic in mice. It is thought that JPV-LW is a laboratory-adapted strain of JPV-BH. Replacing the L gene of JPV-BH with the L gene of JPV-LW resulted in attenuation in mice, confirming the role of the L gene in viral pathogenesis (10). These findings demonstrated that JPV-BH can be used as a model to study the pathogenic mechanisms of Jeilongviruses.

The SH protein is expressed by some paramyxoviruses during infection. Parainfluenza virus 5 (PIV5), mumps virus (MuV), metapneumoviruses, and respiratory syncytial virus (RSV) contain the SH gene (11–14). PIV5 SH is a type II membrane protein of 44 aa and is located between the F and HN genes (13, 15). A recombinant PIV5 lacking the coding region of SH (rPIV5ΔSH) had no growth defect in tissue culture cells, but it induces more apoptosis in both MDBK and L929 cells through a tumor necrosis factor alpha (TNF-α)-mediated extrinsic apoptotic pathway (16, 17). MuV SH protein is a type I membrane protein of 57 aa, and SH is not essential for the in vitro growth of MuV (18). Although there is no sequence homology between PIV5 SH and MuV SH, MuV SH was able to functionally replace PIV5 in cell culture (14). RSV, a member of the family Pneumoviridae, also encodes an SH protein. RSV lacking SH (RSVΔSH) was viable and exhibited a growth pattern similar to that of wild-type RSV (19, 20). RSVΔSH infection caused more apoptosis in A549 cells and L929 cells (12). JPV SH is located between the F and TM genes. It is a type I membrane protein with 69 amino acid residues, with a predicted N-terminal ectodomain of 5 residues, a transmembrane domain of 23 residues, and a C-terminal cytoplasmic domain of 41 residues. JPV SH has no sequence similarity with any other paramyxovirus SH but has a hydrophilicity profile similar to those of the SH proteins of MuV and RSV (6). The function of JPV SH protein was studied previously using JPV-LW and revealed the apoptosis-blocking function of SH. The JPV-LW strain with its SH ORF replaced with the ORF of Renilla luciferase (RLuc) had no growth defect in Vero cells (21). Due to the lack of pathogenicity of JPV-LW in mice, no differences in terms of mortality or morbidity were seen between mice infected with JPV-LW and those infected with recombinant JPV-LW lacking SH. Thus, definitive functions of JPV SH in an infection model have not been explored.

Recombinant RSV lacking the expression of SH was attenuated in vivo (22–24). RSV is a human virus, and the ideal animal model to study RSV pathogenesis is the chimpanzee, so the study of RSV SH in a suitable animal model is difficult. Deletion of SH reduced the neurovirulence of MuV in a newborn rat intracerebral infection model (25), but MuV poorly replicates in this animal model and does not cause disease. The lack of an ideal animal disease model simulating the mode of natural infection prevented studies to elucidate the role of SH in viral pathogenesis.

Since JPV-BH is pathogenic in its natural host, we used laboratory mice to compare the pathogenicities of JPV mutant viruses to study the role of JPV genes in pathogenesis. In this work, we replaced the ORF of the SH gene of JPV-BH with enhanced green fluorescent protein (EGFP) without changing the gene start (GS) and gene end (GE) regions of the transcriptional unit. Similarly, we made recombinant chimera viruses, rJPV-MuVSH and rJPV-RSVSH, by replacing SH of JPV-BH with SH of MuV or RSV. The role of the SH gene in pathogenesis was studied for the first time in the natural host of a virus. In this work, JPV-BH is referred to as JPV unless indicated otherwise.

RESULTS

Recovery of recombinant virus rJPV-ΔSH.

To study the function of SH, we replaced the SH coding sequence in a full-length JPV-BH plasmid with EGFP (Fig. 1A). This plasmid together with three other helper plasmids encoding the N, P, and L proteins and a plasmid encoding T7 RNA polymerase were cotransfected into HEK293T cells and cocultured with Vero cells as described previously (10). After obtaining the rescued virus, PCR amplification of cDNA with primers MA12F and MA09R was used to identify rJPV-ΔSH (Fig. 1B). Expression of EGFP in the place of the SH ORF was visualized with a fluorescence microscope (Fig. 1C). In addition, the full-length genome sequence of plaque-purified rJPV-ΔSH was confirmed by reverse transcription-PCR (RT-PCR) and Sanger sequencing. Recombinant virus lacking SH was further confirmed by an immunofluorescence assay (IFA) with antibodies against JPV SH and F proteins (Fig. 1D).

FIG 1.

Recovery of recombinant virus rJPV-ΔSH. (A) Schematics of rJPV and rJPV-ΔSH indicating the location where the ORF of SH was replaced with EGFP. (B) To confirm the deletion of SH and the presence of EGFP in rJPV-ΔSH, RT-PCR was performed using primers MA12F and MA09R to amplify the SH region. (C) EGFP expression of rJPV-ΔSH. Vero cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. At 2 dpi, fluorescence was examined using a Nikon FXA fluorescence microscope (magnification, ×10). (D) Immunofluorescent staining of Vero cells infected with rJPV or rJPV-ΔSH. Vero cells were mock infected or infected with rJPV or JPV-ΔSH. At 2 dpi, cells were washed with PBS and fixed with 0.5% formaldehyde. The cells were permeabilized with a 0.1% PBS–saponin solution, incubated for 30 min with polyclonal anti-F or -SH rabbit serum at a 1:100 dilution, and PE labeled with goat anti-rabbit antibody. The cells were incubated for 30 min and examined and photographed using a Nikon FXA fluorescence microscope.

Virus morphology and analysis of growth kinetics in vitro.

To compare the growth kinetics of rJPV and rJPV-ΔSH, growth rates at a high multiplicity of infection (MOI) (Fig. 2A) and a low MOI (Fig. 2B) were determined in Vero cells. Vero cells were infected with rJPV and rJPV-ΔSH at MOIs of 5 and 0.1. The medium was harvested at different time points, and titers of virus in medium were determined by a plaque assay. Similar growth patterns were observed for rJPV and rJPV-ΔSH. There was no difference in the plaque morphologies of rJPV and rJPV-ΔSH (Fig. 2C). Examination of virions by electron microscopy was not able to detect any difference in the structures of rJPV and rJPV-ΔSH (Fig. 2D). Infection of L929 cells with rJPV and rJPV-ΔSH produced similar levels of N and F proteins (Fig. 2E and F).

FIG 2.

Comparison of rJPV and rJPV-ΔSH in vitro. (A) High-MOI growth curves of rJPV and rJPV-ΔSH. Vero cells in a 6-well plate were infected, in triplicates, with rJPV or rJPV-ΔSH at an MOI of 5, and the medium was harvested at 24-h intervals. Plaque assays were performed on Vero cells to determine the virus titer. (B) Low-MOI growth curves of rJPV and rJPV-ΔSH. Vero cells in a 6-well plate were infected, in triplicates, with rJPV or rJPV-ΔSH at an MOI of 0.1, and the medium was harvested at 24-h intervals. Plaque assays were performed on Vero cells to determine the virus titer. (C) Plaques formed by rJPV and rJPV-ΔSH in Vero cells. Plaques were stained with 0.5% crystal violet. (D) Morphology of rJPV and rJPV-ΔSH. Viruses were grown in Vero cells and concentrated in a 20% sucrose gradient. Concentrated viruses dissolved in PBS were adsorbed onto electron microscopy grids and negatively stained with phosphotungstic acid. (E and F) Expression of N and F in virus-infected cells. L929 cells in six-well plates were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. The cells were collected at 2 dpi and fixed with 0.5% formaldehyde for 1 h. The fixed cells were resuspended in FBS-DMEM (50:50) and permeabilized with 70% ethanol overnight. The cells were washed once with PBS and then incubated with mouse anti-N monoclonal antibody and mouse anti-F monoclonal antibody. Secondary staining was performed using APC-goat anti-mouse IgG, and the mean fluorescence intensity (MFI) was measured with a flow cytometer. Samples are triplicates, and error bars show standard errors of the means. Statistical significance between groups at each time point was calculated based on two-way analysis of variance (ANOVA) to compare the growth kinetics (**, P < 0.01).

rJPV-ΔSH induces more apoptosis in L929 cells than rJPV.

To study the phenotype of rJPV and rJPV-ΔSH in L929 cells, confluent cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5 (Fig. 3A). At 2 days postinfection (dpi), more dead cells were seen with rJPV-ΔSH infection. To determine whether there was a difference in the apoptosis of rJPV- and rJPV-ΔSH-infected cells, we examined cellular DNA fragmentation, a hallmark of apoptosis, in rJPV- and rJPV-ΔSH-infected L929 cells. DNA was extracted from L929 cells infected with rJPV or rJPV-ΔSH or mock infected. Extracted DNA was resolved by gel electrophoresis. Fragmented DNA was not visible in mock-infected cells, but both rJPV- and rJPV-ΔSH-infected cells had DNA fragmentation. Increased DNA fragmentation was seen in rJPV-ΔSH-infected cells (Fig. 3B). To quantify the apoptosis induced by rJPV and rJPV-ΔSH, we performed a Pacific Blue annexin V apoptosis detection assay with 7-amino-actinomycin D (7-AAD) staining. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH. At 2 dpi, there were significantly more apoptotic cells in cells infected with rJPV-ΔSH (Fig. 3C).

FIG 3.

rJPV-ΔSH induces more apoptosis in L929 cells. (A) CPE induced by rJPV and rJPV-ΔSH infection in L929 cells. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. At 2 dpi, the cells were photographed. (B) DNA fragmentation assay in rJPV- or rJPV-ΔSH-infected cells. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. The cells were collected at 2 dpi. (C) Induction of apoptosis by rJPV and rJPV-ΔSH. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. The cells were collected for a Pacific Blue annexin V apoptosis detection assay with 7-AAD at 2 dpi. Statistical significance was calculated based on the Student t test. (D) Concentrations of TNF-α produced from rJPV- and rJPV-ΔSH-infected cells. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH. The medium was collected at different time points after infection. The amounts of TNF-α were measured by an ELISA. Samples are triplicates, and error bars show standard errors of the means. Statistical significance between groups at each time point was calculated based on two-way ANOVA (***, P < 0.001; **, P < 0.01; ns, not significant).

To investigate the role of the SH gene of JPV in the production of TNF-α, L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. Supernatants were collected at various time points to measure TNF-α production by an enzyme-linked immunosorbent assay (ELISA). The amount of TNF-α was larger in rJPV-ΔSH-infected cells than in rJPV-infected cells (Fig. 3D).

Inhibition of rJPV-ΔSH-induced apoptosis by neutralizing anti-TNF-α antibody.

To determine the role of TNF-α in apoptosis induced by rJPV- and rJPV-ΔSH-infected cells, L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. Cells were incubated with Dulbecco's modified Eagle's medium (DMEM)–2% fetal bovine serum (FBS) containing no antibody, neutralizing antibody against TNF-α, or the isotype control at 30 μg/ml. At 2 dpi, cells were examined for cytopathic effect (CPE). More CPE was seen in cells infected with rJPV-ΔSH than in those infected with rJPV. CPE was inhibited in rJPV-ΔSH-infected cells that were incubated with neutralizing antibody against TNF-α, while untreated cells and cells incubated with the isotype control induced more CPE (Fig. 4A). The ability of the TNF-α-neutralizing antibody to inhibit apoptosis was quantified by flow cytometry (Fig. 4B). At 2 dpi, rJPV-ΔSH induced less apoptosis in cells treated with TNF-α-neutralizing antibody than in untreated cells or cells treated with the isotype control. Only 11.6% (95% confidence interval [CI], 10.9 to 12.3%) of neutralizing-antibody-treated cells infected with rJPV-ΔSH were apoptotic, compared to 31.5% (95% CI, 28.9 to 34.1%) of apoptotic cells in untreated cells and 23.8% (95% CI, 21.4 to 26.2%) apoptosis in cells treated with the isotype control. However, the neutralizing antibody did not cause a significant inhibition in apoptosis induced by rJPV. The level of apoptosis induced by rJPV-ΔSH was higher than that induced by rJPV. These results indicate that SH plays a role in blocking TNF-α-mediated apoptosis.

FIG 4.

Inhibition of rJPV-ΔSH-induced apoptosis by neutralizing antibody against TNF-α. (A) Inhibition of CPE by the neutralizing antibody against TNF-α. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5 and incubated with no antibody, the isotype control (30 μg/ml), or TNF-α-neutralizing antibody (30 μg/ml) for 2 days. (B) Inhibition of apoptosis by neutralizing antibody against TNF-α. L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5 and incubated with no antibody, the control antibody (30 μg/ml), or neutralizing antibody (30 μg/ml) for 2 days. A Pacific Blue annexin V apoptosis detection assay with 7-AAD was carried out at 2 dpi. Error bars show standard errors of the means. All infections were performed in triplicates. Statistical significance between groups at each time point was calculated based on one-way ANOVA (***, P < 0.001; **, P < 0.01; *, P < 0.05).

rJPV-ΔSH is attenuated in vivo.

BALB/c mice were infected with rJPV or rJPV-ΔSH in three different doses, 10⁶, 6 × 10⁵, and 2 × 10⁵ PFU, or 100 μl of phosphate-buffered saline (PBS) intranasally. The mice were monitored for 14 days. rJPV infection caused more weight loss (Fig. 5A) and mortality (Fig. 5B) than rJPV-ΔSH. The 50% lethal dose (LD₅₀) values for rJPV and rJPV-ΔSH were calculated based on the method described by Miller and Tainter (26), using the back-titration titer. LD₅₀ values of rJPV and rJPV-ΔSH were determined to be 1.76 × 10⁵ PFU and 6.7 × 10⁵ PFU, respectively, confirming that JPV-ΔSH was attenuated.

FIG 5.

rJPV-ΔSH is attenuated in BALB/c mice. BALB/c mice in 7 groups, each with 8 animals, were infected with rJPV or rJPV-ΔSH in three different doses, 10⁶, 6 × 10⁵, and 2 × 10⁵ PFU, or 100 μl of PBS intranasally. (A) Body weight loss. Mice were monitored daily, and weight loss was graphed as the average percentage of their original weight (on the day of infection). (B) Survival rate (n = 8). Statistical analysis of the survival curve was performed based on a log rank (Mantel-Cox) test.

Recovery and growth characteristics of the SH chimera viruses rJPV-MuVSH and rJPV-RSVSH.

For a comparative study of the function of the SH genes of JPV, MuV, and RSV in pathogenicity, recombinant JPVs were designed in such a way that the SH ORF of the full-length JPV plasmid was replaced with SH of MuV or RSV (Fig. 6A). rJPV-MuVSH and rJPV-RSVSH were rescued and sequenced as described in Materials and Methods. After obtaining the rescued viruses, PCR amplification of cDNA with primers MA12F and MA09R was used to identify rJPV-MuVSH and rJPV-RSVSH (Fig. 6B). SH proteins in Vero cells infected with rJPV, rJPV-MuVSH, or rJPV-RSVSH were detected by Western blotting. JPV N was detected in all virus-infected cells (Fig. 6B). Both rJPV-MuVSH and rJPV-RSVSH have growth patterns similar to that of rJPV, as shown by low-MOI (Fig. 6D) and high-MOI (Fig. 6E) growth curves.

FIG 6.

Recovery and growth characteristics of SH chimera viruses rJPV-MuVSH and rJPV-RSVSH. (A) Schematics of rJPV, rJPV-MuVSH, and rJPV-RSVSH indicating the location where the ORF of JPV SH was replaced with SH of MuV or RSV. (B) To confirm the presence of JPV SH, EGFP, MuVSH, and RSVSH in rJPV, rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH, respectively, RT-PCR was performed using primers MA12F and MA09R to amplify the SH region. The expected sizes of PCR products are indicated above the gel. (C) Viral protein expression levels of rJPV, rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH. Six-well plates of Vero cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 1. Cell lysates were subjected to immunoblotting with JPV anti-SH, MuV anti-SH, RSV anti-SH, or JPV anti-N. (D) Low-MOI growth curves of rJPV, rJPV-MuVSH, and rJPV-RSVSH. Vero cells were infected with rJPV, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 0.1, and the medium was harvested at 24-h intervals. A plaque assay was performed on Vero cells to determine the virus titer. (E) High-MOI growth curves of rJPV, rJPV-MuVSH, and rJPV-RSVSH. Vero cells were infected with rJPV, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5, and the medium was harvested at 24-h intervals. A plaque assay was performed on Vero cells to determine the virus titer. Statistical significance between groups at each time point was calculated based on two-way ANOVA to compare the growth kinetics (***, P < 0.001; **, P < 0.01; *, P < 0.05).

MuV SH and RSV SH reduce the levels of apoptosis and TNF-α production.

It was previously shown that the SH proteins of MuV and RSV block apoptosis, and the deletion of SH from either virus resulted in increased TNF-α production in vitro (12, 25). To determine the in vitro effect of rJPV-MuVSH and rJPV-RSVSH infection, L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. rJPV-ΔSH infection produced significantly higher levels of apoptosis (Fig. 7A) and TNF-α production (Fig. 7B) than any of the other recombinant JPV-infected groups. Interestingly, rJPV-MuVSH and rJPV-RSVSH behaved like wild-type rJPV in terms of apoptosis and TNF-α production. TNF-α activates NF-κB and the nuclear translocation of the p65 subunit. It was described previously that rJPV-LW-ΔSH infection increases the nuclear translocation of p65. To study the effects of JPV SH, MuV SH, and RSV SH on p65 nuclear translocation, L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. There were increases in p65 expression and nuclear translocation in rJPV-ΔSH-infected cells compared to cells infected with the other viruses (Fig. 7C and D). These results suggest the functional similarity of SH proteins of JPV, MuV, and RSV.

FIG 7.

MuV SH and RSV SH reduce the levels of apoptosis and TNF-α production. (A) Induction of apoptosis by rJPV-MuVSH and rJPV-RSVSH. L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. The cells were collected for a Pacific Blue annexin V apoptosis detection assay with 7-AAD at 24 hpi and 48 hpi. Statistical significance between groups at each time point was calculated based on one-way ANOVA. (B) Concentrations of TNF-α produced from rJPV-MuVSH- and rJPV-RSVSH-infected cells. L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. The medium was collected at 24 hpi, 48 hpi, and 72 hpi. The amounts of TNF-α were measured by an ELISA. Samples are triplicates, and error bars show standard errors of the means. Statistical significance between groups at each time point was calculated based on two-way ANOVA (***, P < 0.001; **, P < 0.01; *, P < 0.05) (C) Detection of NF-κB p65 subunit expression in L929 cells by an IFA. L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. At 1 dpi, cells were fixed and stained with Zenon APC-conjugated JPV N and anti-p65 antibodies followed by staining with Cy3-conjugated secondary antibody. DAPI staining was performed with ProLong gold antifade mountant. Pictures were taken at a magnification of ×40. (D) Nuclear translocation of the NF-κB p65 subunit into the nucleus, expressed in terms of Mander's overlap coefficient.

rJPV-MuVSH and rJPV-RSVSH are pathogenic in mice.

MuV SH and RSV SH have not been studied in a suitable animal model to determine their role in pathogenesis. rJPV-MuVSH and rJPV-RSVSH were used to infect mice to determine their role in pathogenesis. BALB/c mice were intranasally infected with 100 μl of PBS or 8 × 10⁵ PFU of rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH. Mice were monitored for 14 days. rJPV, rJPV-MuVSH, and rJPV-RSVSH infection resulted in greater weight loss (Fig. 8A) and mortality (Fig. 8B) than rJPV-ΔSH infection. However, weight loss and mortality induced by infection with rJPV-MuVSH were intermediate between those induced by rJPV-ΔSH and those induced by rJPV or rJPV-RSVSH. High virus loads were observed in mice infected with rJPV, rJPV-MuVSH, and rJPV-RSVSH at 3 dpi, but interestingly, no difference was observed in the virus loads between the groups at 7 dpi (Fig. 9A). Increased serum levels of TNF-α were detected in rJPV-ΔSH-infected animals (Fig. 9B), suggesting that SH plays a role in reducing the production of TNF-α in JPV-infected animals. Reduced weight loss and higher survival rates in rJPV-ΔSH-infected mice were supported by histopathology, which showed less interstitial pneumonia in the lungs of mice infected with rJPV-ΔSH (Fig. 9C).

FIG 8.

rJPV-MuVSH and rJPV-RSVSH are pathogenic in mice. BALB/c mice were intranasally infected with 100 μl of PBS or 8 × 10⁵ PFU of rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH. (A) Body weight loss. Mice were monitored daily, and weight loss was graphed as the average percentage of their original weight (on the day of infection). (B) Survival rate (PBS, n = 10; rJPV, n = 8; rJPV-ΔSH, n = 9; rJPV-MuVSH, n = 8; rJPV-RSVSH, n = 9). Statistical analysis of the survival curves was performed based on a log rank (Mantel-Cox) test.

FIG 9.

Infection of BALB/c mice with rJPV, rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH. (A) Lung virus titers in mice. BALB/c mice were intranasally infected with 100 μl of rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at a dose of 8 × 10⁵ PFU. Mouse lungs were collected at 3 and 7 dpi. Virus titers were determined by a plaque assay on Vero cells. (B) Concentrations of serum TNF-α in infected BALB/c mice. Sera were collected from infected animals at different time points after infection. Levels of TNF-α were measured using an ELISA (n = 4). Statistical significance between groups at each time point was calculated based on two-way ANOVA (***, P < 0.001; **, P < 0.01; *, P < 0.05). (C) Histopathology of lungs. Lungs were collected from infected and mock-infected animals at 3 dpi and stained with H&E. Photomicrographs were taken at a magnification of ×20.

DISCUSSION

The families Paramyxoviridae and Pneumoviridae are in the order Mononegavirales, the members of which have a nonsegmented, negative-stranded RNA genome. Viruses in these two families have many similarities in gene order, gene expression strategies, and replication. Jeilongvirus is a newly proposed genus of the Paramyxoviridae family, which includes rodent viruses like JPV and BeiPV. JPV and BeiPV have similar genome organizations, and it was previously shown that their genome replication machineries could be interchanged (2, 6). Unlike other paramyxoviruses, the inhibition of STAT1 translocation by JPV and BeiPV is V protein independent, supporting the inclusion of these viruses in a new genus (27). The presence of two additional membrane proteins, SH and TM, in the genomes of JPV, TlmPV, and BeiPV makes these viruses unique in comparison to other paramyxoviruses (1, 6, 7). At present, very little is known about Jeilongviruses. JPV antibodies have been detected in wild mice, wild rats, pigs, and humans (3), suggesting a wide host range of JPV. JPV is able to cause severe disease in laboratory mice (10), which makes JPV an excellent model to study the pathogenesis of Jeilongviruses in vivo.

The function of the SH gene in blocking TNF-α-mediated apoptosis was described previously for MuV, PIV5, RSV, and JPV-LW. Deletion of the SH protein of these viruses did not affect replication and gene expression in vitro (12, 14, 16, 21). Recently, it was reported that MuV SH inhibits TNF-α, interleukin-1β (IL-1β), and NF-κB activation by interacting with TNF receptor 1 (TNF-R1), IL-1 receptor 1 (IL-1R1), and Toll-like receptor 3 (TLR3) complexes (28). These findings suggest the role of SH in evading the host immune response. However, the role of the SH protein has not been studied in a suitable pathogenic animal model, and the in vivo roles had not been examined. The laboratory mouse is not an ideal model in which to study RSV and MuV pathogenesis. The role of RSV SH in pathogenicity was previously reported (22–24). RSV lacking SH expression was attenuated in chimpanzees and mice. As RSV is a virus that affects humans, using chimpanzees for studying RSV pathogenicity is more relevant but difficult due to ethical and financial concerns. Similarly, MuV lacking SH expression was attenuated with reduced neurovirulence in a neonatal rat intracerebral infection model (25). Poor systemic replication of MuV was the major limitation in using this model for pathogenesis studies. As a rodent paramyxovirus containing SH, JPV is suitable to study the functions of SH of paramyxoviruses in mice. rJPV-ΔSH had no growth defect in tissue culture cells. Plaque size and expression levels of viral proteins were similar in rJPV- and rJPV-ΔSH-infected cells. Cells infected with rJPV-ΔSH underwent greater apoptosis and TNF-α production. Neutralizing antibodies against TNF-α inhibited apoptosis induced by rJPV-ΔSH in L929 cells, confirming the role of SH protein in blocking TNF-α-mediated apoptosis. No significant inhibition in apoptosis was seen in rJPV-infected cells incubated with a TNF-α-neutralizing antibody. Interestingly, L929 cells incubated with TNF-α-neutralizing antibodies were reported to inhibit apoptosis induced by rJPV (LW strain) in a previous study (21). This might be due to the increased production of TNF-α and good sensitivity to TNF-α-neutralizing antibodies in infection with the tissue culture-adapted JPV-LW strain. In contrast, JPV-BH is not a tissue culture-adapted strain, causing less CPE than JPV-LW (10).

We generated JPV-BH-based SH chimera viruses, rJPV-MuVSH and rJPV-RSVSH, to study the functional relationship between the SH proteins of JPV, MuV, and RSV. rJPV-MuVSH and rJPV-RSVSH infection in L929 cells resulted in reduced levels of apoptosis, inhibition of NF-κB p65 translocation, and decreased TNF-α production compared to rJPV-ΔSH infection. rJPV-ΔSH was attenuated in mice, with infection resulting in reduced weight loss. The LD₅₀ value of rJPV-ΔSH was higher than that of rJPV in infection in BALB/c mice. We also observed increased serum levels of TNF-α in rJPV-ΔSH-infected animals at 3 dpi and 5 dpi. The increased serum levels of TNF-α correlate with decreased lung viral loads and decreased interstitial pneumonia in rJPV-ΔSH-infected animals at 3 dpi, suggesting a role of increased TNF-α levels in the attenuation of rJPV-ΔSH. It is possible that increased TNF-α-mediated apoptosis of cells infected with rJPV-ΔSH limited the spread of virus infection and reduced lung lesions. rJPV-MuVSH and rJPV-RSVSH were pathogenic in animals, and for the first time, we have demonstrated functions of MuV SH and RSV SH in an animal model. rJPV-RSVSH behaved like rJPV in mouse infection; replacing JPV SH with RSV SH completely rescued the phenotype of JPV SH. Interestingly, rJPV-MuVSH was less pathogenic than rJPV-RSVSH and rJPV, indicating that MuV SH can only partially complement JPV SH functions. We speculate that the function of SH is to block signaling and amplification of TNF-α (since TNF-α is an autocrine factor). It is not yet clear what initially activates the expression of TNF-α in infected cells. Whereas SH blocks TNF-α signaling, it does not affect the initial activation of TNF-α, as indicated by the detection of TNF-α in JPV wild-type-infected cells (Fig. 3D). It is possible that this difference in pathogenicity might be due to the difference in the expressions of various proinflammatory cytokines. Further examination of the viruses in vivo may allow dissection of the different roles of JPV SH and MuV SH in the regulation of various cytokines. Even though there is no sequence homology between JPV SH and SH of MuV or RSV, the functions of these proteins are similar in a mouse model, suggesting that the SH proteins of various paramyxoviruses may have structural similarity and/or that pathways affected by SH proteins are conserved in both mice and humans.

MATERIALS AND METHODS

Cells.

Human embryonic kidney HEK293T cells (ATCC CRL-1573), mouse fibroblast L929 cells (ATCC CCL-1), and Vero cells (ATCC) were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 100 IU/ml penicillin, and 100 μg/ml streptomycin. All cells were incubated at 37°C in 5% CO₂. Cells infected with viruses were grown in DMEM containing 2% FBS. Vero cells were used to perform plaque assays.

Mice.

Six-week-old female BALB/c mice (Envigo) were used for the studies. Mice were infected with JPV in enhanced biosafety level 2 facilities in HEPA-filtered isolators. All animal experiments were performed in accordance with national guidelines provided by the Guide for Care and Use of Laboratory Animals (29) and the University of Georgia Institutional Animal Care and Use Committee (IACUC). The IACUC of the University of Georgia approved all animal experiments.

Construction of SH recombinant JPV plasmids.

In this work, we are exclusively using the JPV-BH backbone. Hence, JPV-BH is referred to as JPV. The construction of a recombinant JPV plasmid with a PvuI restriction site at the N gene was previously described (10). By using standard molecular biology techniques, the ORF of the SH gene was replaced by an enhanced green fluorescent protein (EGFP) gene. The construct lacking the SH gene and containing the EGFP gene was designated the pJPV-ΔSH plasmid. Similarly, the SH ORF of JPV was replaced with the ORFs of MuV SH and RSV (A2 strain) SH to generate pJPV-MuVSH and pJPV-RSVSH, respectively.

Virus rescue and sequencing.

To generate viable recombinant JPV without a SH gene (rJPV-ΔSH), a full-length pJPV-ΔSH plasmid, a plasmid expressing T7 polymerase (pT7P), and three plasmids encoding the N, P, and L proteins of JPV (pJPV-N, pJPV-P, and pJPV-L) were cotransfected into HEK293T cells at 95% confluence in a 6-cm plate with Jetprime (Polypus-Transfection, Inc., New York, NY). The amounts of plasmids used were as follows: 5 μg of the full-length pJPV-ΔSH plasmid, 1 μg of pT7P, 1 μg of pJPV-N, 0.3 μg of pJPV-P, and 1.5 μg of pJPV-L. At 2 days posttransfection, 1/10 of the HEK293T cells were cocultured with 1 × 10⁶ Vero cells in a 10-cm plate. Seven days after coculture, media were centrifuged to remove the cell debris, and the supernatant was used for plaque assays in Vero cells to obtain single clones of recombinant JPV-ΔSH. Vero cells were used to grow the plaque-purified virus. Also, full-length pJPV-MuVSH and pJPV-RSVSH plasmids were used to rescue the SH chimera viruses rJPV-MuVSH and rJPV-RSVSH.

The full-length genomes of plaque-purified rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH were sequenced. Total RNAs of rJPV-ΔSH-, rJPV-MuVSH-, and rJPV-RSVSH-infected Vero cells were purified using the RNeasy minikit (Qiagen, Valencia, CA). cDNAs were prepared by using random hexamers. PCR amplification of cDNA with primers MA12F and MA09R was used to identify rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH. In addition, sequences of all primers used for sequencing the whole genome of rJPV-ΔSH, rJPV-MuVSH, and rJPV-RSVSH are available upon request. DNA sequences were determined by using an Applied Biosystems sequencer (ABI, Foster City, CA).

Fluorescence microscopy.

To confirm the rescue of JPV-ΔSH (EGFP in the location of SH), Vero cells were mock infected or infected with rJPV or JPV-ΔSH. At 2 days postinfection (dpi), the cells were photographed using a Nikon FXA fluorescence microscope to look for EGFP expression. The lack of SH in rJPV-ΔSH was confirmed by an immunofluorescence assay. Vero cells were mock infected or infected with rJPV or JPV-ΔSH. At 2 dpi, cells were washed with phosphate-buffered saline (PBS) and fixed with 0.5% formaldehyde. The cells were permeabilized with a 0.1% PBS–saponin solution and incubated for 30 min with polyclonal anti-F or -SH rabbit serum at a 1:100 dilution (GenScript USA, Inc., Piscataway, NJ), and phycoerythrin (PE)-labeled goat anti-rabbit antibody was then added to the cells. The cells were incubated for 30 min and were examined and photographed using a Nikon FXA fluorescence microscope.

To detect the NF-κB p65 subunit, L929 cells on coverslips were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. At 1 dpi, cells were washed with PBS and then fixed with 1% formaldehyde for 15 min at room temperature. Cells were permeabilized using 0.1% Triton X-100 for 10 min. Cells were washed 3 times with PBS. Anti-mouse JPV N monoclonal antibody conjugated with Zenon allophycocyanin (APC)-mouse IgG2a (Thermofisher Scientific) was used to stain cells for 1 h. The cells were washed with PBS and then incubated for 1 h in a 1:100 dilution of rabbit monoclonal antibody specific for the p65 subunit of NF-κB. Cells were washed with PBS and incubated with goat anti-rabbit Cy3 for 1 h. The cells were washed 3 times with PBS and stained with DAPI (4′,6-diamidino-2-phenylindole) for 10 min. ProLong gold antifade mountant (Life Technologies) was applied directly to the fluorescently labeled cells. Fluorescence was examined and photographed using a Nikon FXA fluorescence microscope and a Nikon Eclipse Ti confocal microscope. Nuclear translocation of p65 was assessed based on the colocalization of anti-rabbit Cy3 into the DAPI region. Colocalization was expressed in terms of Mander's overlap coefficient using Nikon NIS-Elements software.

Electron microscopy.

rJPV and rJPV-ΔSH were grown in Vero cells and concentrated in a 20% sucrose gradient. Concentrated JPVs dissolved in PBS were adsorbed onto parlodion-coated grids for 30 s. Grids were washed with Tris-buffered saline (TBS) and then stained with 2% phosphotungstic acid (pH 6.6). These grids were then examined using a JEOL 1230 transmission electron microscope (JEOL, Tokyo, Japan).

Growth kinetics.

Vero cells in 6-well plates were infected with rJPV, JPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 0.1 or 5. The cells were then washed with PBS and maintained in DMEM–2% FBS. The medium was collected at 0, 24, 48, 72, and 96 h postinfection (hpi). The titers were determined by a plaque assay on Vero cells.

Detection of viral protein expression.

Expression levels of N and F in virus-infected cells were compared. L929 cells in six-well plates were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. The cells were collected at 2 dpi and fixed with 0.5% formaldehyde for 1 h. The fixed cells were resuspended in FBS-DMEM (50:50) and permeabilized with 70% ethanol overnight. The cells were washed once with PBS and then incubated with mouse anti-N monoclonal antibody or mouse anti-F monoclonal antibody in PBS–1% bovine serum albumin (BSA) (1:200) for 1 h at 4°C. The cells were stained with APC-goat anti-mouse IgG from BioLegend (1:500) for 1 h at 4°C in the dark and then washed once with PBS–1% BSA. The fluorescence intensity was measured with a flow cytometer (LSR II; Becton Dickinson).

For immunoblotting, Vero cells in a 6-well plate were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 1. At 2 dpi, cells were lysed with whole-cell extraction buffer (WCEB) (50 mM Tris-HCl [pH 8], 280 mM NaCl, 0.5% NP-40, 0.2 mM EDTA, 2 mM EGTA, and 10% glycerol). The lysates were run on SDS-PAGE gels and immunoblotted with primary antibody (anti-JPV SH, anti-MuV SH, anti-RSV SH, or anti-JPV N) and the corresponding secondary antibodies conjugated to Cy3.

Enzyme-linked immunosorbent assay for TNF-α.

L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. The medium was collected at 24 h, 48 h, and 72 hpi. The amounts of TNF-α were measured by using a mouse TNF-α detection kit (R&D Systems, Inc., Minneapolis, MN, USA) according to the manufacturer's instructions. Fifty microliters of medium from infected cells, standards, or controls was mixed with 50 μl of assay diluent and added to strips prelabeled with antibody against TNF-α. The strips were incubated at room temperature for 2 h. After the strips were washed five times with wash buffer provided by the manufacturer, 100 μl of polyclonal antibody specific for mouse TNF-α conjugated to horseradish peroxidase was added, and the strips were incubated at room temperature for 2 h. The strips were then washed five times, and 100 μl of tetramethylbenzidine substrate solution was added to each well. The strips were incubated in the dark at room temperature for 30 min, and 100 μl of stop solution was added to each well. The optical density at 450 nm was measured within 30 min. The amounts of TNF-α were calculated by using standard curves generated from known concentrations of TNF-α provided by the manufacturer.

To measure the amount of TNF-α in serum, 50 μl of sera, standards, or controls was mixed with 50 μl of assay diluent and added to strips prelabeled with antibody against TNF-α. An ELISA was then performed as described above.

Apoptosis assay.

Confluent L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5. At 2 dpi, cells were washed twice with PBS without Mg²⁺ or Ca²⁺ and incubated in 0.5 ml of TTE buffer (0.2% Triton X-100, 10 mM Tris, 15 mM EDTA [pH 8.0]) at room temperature for 15 min. Cell lysates were harvested and centrifuged at 14,000 rpm for 20 min. Supernatants were digested with 100 μg of RNase A/ml at 37°C for 1 h. Samples were purified by phenol-chloroform extraction, precipitated, and washed with 70% ethanol. Pellets were air dried and resuspended in 10 μl of Tris-EDTA. Electrophoresis was performed on 2% agarose gels with size markers.

For apoptosis assays, a Pacific Blue annexin V apoptosis detection assay with 7-amino-actinomycin D (7-AAD) from BioLegend (San Diego, CA, USA) was used. L929 cells were mock infected or infected with rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH at an MOI of 5. At 2 dpi, cells were trypsinized and combined with floating cells in the medium. Cells were washed twice with cold BioLegend cell staining buffer and then resuspended in annexin V binding buffer at a concentration of 0.25 × 10⁷ to 1.0 × 10⁷ cells/ml. One hundred microliters of the cell suspension was then transferred to a 5-ml test tube and mixed with 5 μl each of Pacific Blue annexin V and 7-AAD viability staining solution. Cells were then vortexed and incubated for 15 min at room temperature (25°C) in the dark. The cells were analyzed by flow cytometry (LSR II; Becton Dickinson).

Antibody treatment of infected cells.

Confluent L929 cells were mock infected or infected with rJPV or rJPV-ΔSH at an MOI of 5 and incubated in 0.5 ml of DMEM–2% FBS with neutralizing antibody against TNF-α (BD Pharmingen, San Jose, CA) or the isotype control at 30 μg/ml. At 2 dpi, the cells were photographed using a light microscope, and an apoptosis assay was performed as described above.

Infection of mice with JPV.

All animal experiments were carried out strictly according to the protocol approved by the University of Georgia IACUC. To study the pathogenesis of JPV in animals, 6-week-old female BALB/c mice (Envigo) were infected with 100 μl of PBS or 10⁶, 6 × 10⁵, or 2 × 10⁵ PFU each of rJPV or rJPV-ΔSH, intranasally. The weights of the mice were monitored for up to 14 dpi.

To study the pathogenesis of rJPV-MuVSH and rJPV-RSVSH, 6-week-old BALB/c mice (Envigo) were infected with 100 μl of PBS or 8 × 10⁵ PFU of rJPV, rJPV-ΔSH, rJPV-MuVSH, or rJPV-RSVSH, intranasally. The weights of the mice were monitored for up to 14 dpi. Blood was collected at 1, 3, 5, and 7 dpi to determine the serum level of TNF-α. Mice were euthanized at 3 and 7 dpi to collect lungs to determine the virus titer.

Histology studies.

BALB/c mice from the infection study were euthanized at 3 dpi. The lungs were inflated with 4% paraformaldehyde and collected. Samples were processed, embedded, and sectioned for hematoxylin and eosin (H&E) staining. Interstitial pneumonia was scored from 1 (minimal) to 4 (severe) by a board-certified veterinary pathologist blind to the study groups. Photomicrographs were taken using an Olympus BX41 microscope with an Olympus DP70 microscope digital camera and DP Controller imaging software.