Transcriptome Profiling of the Virus-Induced Innate Immune Response in Pteropus vampyrus and Its Attenuation by Nipah Virus Interferon Antagonist Functions

Abstract

Bats are important reservoirs for several viruses, many of which cause lethal infections in humans but have reduced pathogenicity in bats. As the innate immune response is critical for controlling viruses, the nature of this response in bats and how it may differ from that in other mammals are of great interest. Using next-generation transcriptome sequencing (mRNA-seq), we profiled the transcriptional response of Pteropus vampyrus bat kidney (PVK) cells to Newcastle disease virus (NDV), an avian paramyxovirus known to elicit a strong innate immune response in mammalian cells. The Pteropus genus is a known reservoir of Nipah virus (NiV) and Hendra virus (HeV). Analysis of the 200 to 300 regulated genes showed that genes for interferon (IFN) and antiviral pathways are highly upregulated in NDV-infected PVK cells, including genes for beta IFN, RIG-I, MDA5, ISG15, and IRF1. NDV-infected cells also upregulated several genes not previously characterized to be antiviral, such as RND1, SERTAD1, CHAC1, and MORC3. In fact, we show that MORC3 is induced by both IFN and NDV infection in PVK cells but is not induced by either stimulus in human A549 cells. In contrast to NDV infection, HeV and NiV infection of PVK cells failed to induce these innate immune response genes. Likewise, an attenuated response was observed in PVK cells infected with recombinant NDVs expressing the NiV IFN antagonist proteins V and W. This study provides the first global profile of a robust virus-induced innate immune response in bats and indicates that henipavirus IFN antagonist mechanisms are likely active in bat cells.

Importance: Bats are the reservoir host for many highly pathogenic human viruses, including henipaviruses, lyssaviruses, severe acute respiratory syndrome coronavirus, and filoviruses, and many other viruses have also been isolated from bats. Viral infections are reportedly asymptomatic or heavily attenuated in bat populations. Despite their ecological importance to viral maintenance, research into their immune system and mechanisms for viral control has only recently begun. Nipah virus and Hendra virus are two paramyxoviruses associated with high mortality rates in humans and whose reservoir is the Pteropus genus of bats. Greater knowledge of the innate immune response of P. vampyrus bats to viral infection may elucidate how bats serve as a reservoir for so many viruses.

FIG 1

Formation and characterization of Pteropus vampyrus cells. (a) PPVK cells were cultured from a kidney and then immortalized by ectopically expressing human telomerase using a Moloney murine leukemia virus system, creating the cell line PVK4. (b) qRT-PCR expression of NDV NP, IFN-β, and RIG-I in PPVK cells infected with NDV at an MOI of 0.2 or mock infected. The data shown are normalized to the level of actin expression. Error bars represent standard deviations. No data point signifies that no signal was detected for that sample. (c) qRT-PCR expression of NDV NP, IFN-β, and RIG-I in PVK4 cells infected with NDV at an MOI of 2 or mock infected. The data shown are normalized to the level of actin expression. Error bars represent standard deviations. No data point signifies that no signal was detected for that sample.

FIG 2

Functional characterization of the PVK4 cell innate immune response. (a) NDV-GFP cytokine bioassay. (b) NDV-GFP infection (MOI = 1) in PVK4 and Vero cells treated with universal IFN or UV-inactivated supernatant from NDV-infected (MOI = 2) or mock-infected PVK4 cells. The data shown are the ratio of the GFP signal to the ethidium bromide signal for staining of nuclei. The level of infection for supernatant containing DMEM only was normalized to 100%, and the results for the other samples are shown as a percentage of that value. Error bars represent standard deviations. (c) Immunofluorescence staining of STAT1 in PVK4 cells following universal IFN treatment over 6 h.

FIG 3

Canonical pathways enriched during NDV infection. Genes upregulated during NDV infection were clustered using the Ingenuity pathway analysis software. Shown here are the top 10 canonical pathways and the genes upregulated during NDV infection in each pathway. The area of the circle is inversely proportional to the P value of the pathway. Genes with darker colors are expressed at higher levels. See Table S2 in the supplemental material for a complete list of canonical pathways. IL-17A, interleukin-17A; PKR, RNA-dependent protein kinase.

FIG 4

Confirmation of genes upregulated during NDV infection of PVK4 cells. (a and b) qRT-PCR for gene expression in PVK4 cells infected with NDV at an MOI of 2. (a) Level of expression of NDV NP normalized to the level of expression of the housekeeping gene RPL11. (b) Heat map of qRT-PCR expression for the indicated genes, which were selected on the basis of mRNA-seq analysis (see Table S1 in the supplemental material). Data are normalized C_T values (C_Ttarget − C_TRPL11). Darker boxes indicate higher levels of expression.

FIG 5

Regulation of gene expression in PVK4 cells (a and b) and A549 cells (c) treated with IFN or infected with NDV at an MOI of 2 for 24 h. (a) Heat map of qRT-PCR data for the indicated genes, which were selected from those in Fig. 4b. Data are calculated as normalized C_T values (C_Ttarget − C_TRPL11). Darker boxes indicate higher levels of expression. (b and c) Bar graphs of the fold induction of select genes from PVK4 cells (b) and A549 cells (c). Data are presented as the fold induction relative to that for mock-infected cells, and error bars represent standard deviations. Statistical significance was determined by Student's t test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; NS, no statistically significant difference.

FIG 6

Regulation of RND1 and MORC3 in cell lines from different bat species. qRT-PCR analysis for NDV-NP (a), ISG56 (b), RND1 (c), and MORC3 (d) in PVK4, EidNi/41.2, EpoNi/22.1, RoNi/7.3, and MVI cells treated with IFN (2,000 U/ml) or infected with NDV (MOI = 2) for 24 h. Primers for the selected genes were designed from the P. vampyrus genome sequence. The data shown in panel a and for EidNi in panel b are normalized to the expression levels of RPL11. The data shown in the remainder of panel b and in panels c and d are shown as the fold change relative to mock infection. Error bars represent standard deviations. Statistical significance was determined by Student's t test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; NS, no statistically significant difference. No data point signifies that no signal was detected for that sample.

FIG 7

Response to henipavirus infection in PVK4 cells. (a) Multicycle growth curve of HeV and NiV in PVK4 cells. Data are shown as the log number of TCID₅₀s per milliliter at the indicated time points. Error bars represent standard deviations. (b) Immunofluorescence staining of NiV- and HeV-infected PVK4 cells at the indicated MOIs and time points. Virus-infected cells were detected with an antibody against the nucleocapsid protein. (c and d) qRT-PCR of PVK4 cells infected with HeV and NiV at an MOI of 10. (c) Levels of expression of NiV L and HeV L normalized to the level of expression of RPL11. Error bars represent standard deviations. No data point signifies that no signal was detected for that sample. (d) Heat map of qRT-PCR data of gene expression normalized to the level of expression of the housekeeping gene RPL11. The genes selected are a subset of those shown in Fig. 4b. Data are calculated as normalized C_T values ((C_Ttarget − C_TRPL11). Darker boxes indicate higher levels of expression. (e) NDV-GFP cytokine bioassay. NDV-GFP infection (MOI = 1) in PVK4 cells treated with universal IFN or UV-inactivated supernatant from henipavirus-infected (MOI = 10) or mock-infected PVK4 cells. Data are shown as the ratio of the GFP signal to the ethidium bromide signal for staining of nuclei. The level of infection for supernatant containing DMEM only was normalized to 100%, and the results for the other samples are shown as a percentage of that value. Error bars represent standard deviations.

FIG 8

Recombinant NDVs expressing NiV IFN antagonists block the innate immune response in PVK4 cells. (a) qRT-PCR analysis of genes in PVK4 cells infected with recombinant NDVs expressing either NDV V, NiV V, or NiV W at an MOI of 2. Genes were selected from those in Fig. 4b. The data shown are levels of expression normalized to the level of actin expression. Error bars represent standard deviations. No data point signifies that no signal was detected for that sample. (b) Western blot (immunoblot [IB]) showing the expression of NDV NP, NiV V, and NiV W from PVK4 cells infected with rNDVs expressing NDV V, NiV V, and NiV W. Actin levels are shown as a loading control.