key: cord-0748241-ma114p7u
authors: Papies, Jan; Sieberg, Andrea; Ritz, Daniel; Niemeyer, Daniela; Drosten, Christian; Müller, Marcel A.
title: Reduced IFN-ß inhibitory activity of Lagos bat virus phosphoproteins in human compared to Eidolon helvum bat cells
date: 2022-03-08
journal: PLoS One
DOI: 10.1371/journal.pone.0264450
sha: 4f78efb37b9055c6eb4668dc17193d8fd0f5fddf
doc_id: 748241
cord_uid: ma114p7u

Eidolon helvum bats are reservoir hosts for highly pathogenic lyssaviruses often showing limited disease upon natural infection. An enhanced antiviral interferon (IFN) response combined with reduced inflammation might be linked to the apparent virus tolerance in bats. Lyssavirus phosphoproteins inhibit the IFN response with virus strain-specific efficiency. To date, little is known regarding the lyssavirus P-dependent anti-IFN countermeasures in bats, mainly due to a lack of in vitro tools. By using E. helvum bat cell cultures in a newly established bat-specific IFN-promoter activation assay, we analyzed the IFN-ß inhibitory activity of multiple lyssavirus P in E. helvum compared to human cells. Initial virus infection studies with a recently isolated E. helvum-borne Lagos bat virus street strain from Ghana showed enhanced LBV propagation in an E. helvum lung cell line compared to human A549 lung cells at later time points suggesting effective viral countermeasures against cellular defense mechanisms. A direct comparison of the IFN-ß inhibitory activity of the LBV-GH P protein with other lyssavirus P proteins showed that LBV-GH P and RVP both strongly inhibited the bat IFN-β promotor activation (range 75–90%) in EidLu/20.2 and an E. helvum kidney cell line. Conversely, LBV-GH P blocked the activation of the human IFN-β promoter less efficiently compared to a prototypic Rabies virus P protein (range LBV P 52–68% vs RVP 71–95%) in two different human cell lines (HEK-293T, A549). The same pattern was seen for two prototypic LBV P variants suggesting an overall reduced LBV P IFN-ß inhibitory activity in human cells as compared to E. helvum bat cells. Increased IFN-ß inhibition by lyssavirus P in reservoir host cells might be a result of host-specific adaptation processes towards an enhanced IFN response in bat cells.

Lyssaviruses infect a wide range of mammals causing up to 50,000 fatal rabies cases in humans per year [1] . Most lyssaviruses are found in bats and transmissions to humans are sporadically a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

Here we developed bat-specific assays to compare distinct features of lyssavirus P proteins in cells of a natural lyssavirus host, E. helvum. We found conserved and strong IFN-ß inhibitory activity of all tested lyssavirus P proteins in bat cells. In human cells, the activity of several different LBV P was markedly reduced compared to RVP. Our data encourage further studies of the complex interplay of bat-associated viruses and host cells in the natural reservoir context.

Permission to capture and sample bats was obtained from the Wildlife Division, Forestry Commission, Accra, Ghana. Bats were captured at a sampling site in Kumasi, Ghana with following geographic coordinates: N06˚42 0 02.0@ W001˚37 0 29.9@. Animal handling and all involved procedures were performed by trained veterinary staff and all efforts were made to minimize the suffering of animals. Bats were caught using mist nets and animals were anesthetized with a Ketamine/Xylazine mixture and euthanized by cervical dislocation as covered by the obtained permits (permit no. CHRPE49/09; A04957), as described previously [41, 42] . Samples were exported under a state contract between the Republic of Ghana and the Federal Republic of Germany under export permission (permit no. CHRPE49/09; A04957) from the Veterinary Services of the Ghana Ministry of Food and Agriculture.

HEK-293T (Homo sapiens, kidney; ATCC CRL-3216), A549 cells (Homo sapiens, lung; ATCC CCL-185), VeroFM (Chlorocebus sp., kidney; ATCC CCL-81), and EidNi/41.3 (E. helvum, kidney; [42] ) were grown in Dulbecco's Modified Eagles Medium (DMEM, Gibco) supplemented with 1% Penicillin/Streptomycin (Gibco), 1% non-essential amino acids (Gibco), 1% sodium pyruvate (Gibco), 1% L-glutamine (Gibco) and 10% FBS (Gibco) at 37˚C and 5% CO 2 .

For the generation of the E. helvum lung cell culture EidLu/20.2, bats were caught with mist nets under the supervision of Ghana authorities. Animals were anesthetized with a Ketamine/ Xylazine mixture and euthanized to perform organ preparations (permit no. CHRPE49/09; A04957) as described previously [42] . Cells were immortalized by transduction of the large T antigen of Simian virus 40 (SV40). Subsequently, cells were expanded, subcloned, and frozen at -80˚C for cryopreservation. EidLu/20.2 and EidNi/41.3 cells were genotyped by cytochrome b PCR as previously described [43, 44] . All cells were tested negative for contaminations with mycoplasma [45] , simian virus 5 (in-house assay), lyssaviruses [46] , and filoviruses [47] by RT-PCR.

VeroFM cells (3x10 7 ) were infected with vesicular stomatitis virus (VSV) diluted in DMEM at a multiplicity of infection (MOI) of 0.01 plaque-forming units/ml. After 18 hours (h) cells were washed with phosphate-buffered saline (PBS) and lysed in 2.5 ml of β-mercaptoethanolcontaining buffer RA1 from the NucleoSpin RNA kit (Macherey-Nagel). The lysate was incubated at -80˚C overnight and RNA extraction was performed according to the manufacturer´s instructions.

Transient transfection of eukaryotic cells was performed using FuGENE 1 HD (Promega) and X-tremeGENE™ siRNA (Roche) according to the manufacturers´instructions. Briefly, cells (1x10 5 to 4x10 5 cells/ml) were cultured in 24-well plates and plasmid DNA was added after 15 minutes (min) incubation with transfection reagent in OptiPRO SFM™ (Gibco) at a 1:3 ratio. Transfection of total RNA from VSV-infected cells was performed with X-tremeGENE™ siRNA analogous to FuGENE 1 HD transfection.

A549 and EidLu/20.2 cells were infected with low-passage LBV Ghana isolate (LBV-GH; passage 4; INSDC LN849915) for comparative virus replication kinetics. Cells were seeded and incubated for 24 h and infected with LBV-GH with an MOI of 0.001 in OptiPRO SFM (Gibco). After a 1 h incubation at 37˚C, the inoculum was removed, cells were washed once with PBS and supplemented with DMEM. Supernatants were collected at the indicated time points (18, 24, 48, 72, 96 h) and 75 μl were subjected to viral RNA isolation with the NucleoSpin 1 RNA virus kit (Macherey-Nagel) according to the manufacturer´s instructions. Subsequently, viral replication was monitored by genotype-specific LBV RT-qPCR using the Superscript III One-step RT-PCR kit (Thermo Fisher) according to the following protocol: reactions were performed in 12.5 μl volume with 6.25 μl of 2x reaction buffer, 10 μM forward and reverse primers (see S1 Table for details), 5 μM probe, 0.5 μl of bovine serum albumin (BSA), 1.5 μl of RNase-free water, 0.5 μl of enzyme mix, and 100 ng of RNA template. RNA was reverse-transcribed at 55˚C for 20 min, followed by an initial denaturation at 95˚C for 120 seconds (s). Cycling and fluorescence signal acquisition was done for 45 cycles of 95˚C for 15 s and 58˚C for 40 s using the LightCycler 1 480 Real-Time PCR System (Roche). Additionally, previously generated LBV in vitro transcribed RNA standards (in-house, unpublished) were used for absolute quantification of viral RNA.

To investigate the functionality of IFN signaling pathways in the analyzed cells IFN bioassays were performed as described before [48] . Briefly, cells were cultured in 24-well plates for 24 h and the IFN response was induced by transfection of 0.5 μg, 0.75 μg, or 1.5 μg VSV-RNA with X-tremeGENE™ siRNA transfection reagent according to the manufacturer's protocol. After 20 h, supernatants were collected and serially diluted in used DMEM from the respective cultured cell line. Next, naïve cells were incubated with 100 μl of the diluted supernatants for 7 h, followed by infection with a Rift Valley fever virus Renilla luciferase-based reporter virus (RVFV-Ren) [48] at an MOI of 0.002 for 16 h. Cell lysis and read-out were performed with the Renilla Luciferase Assay System (Promega) according to an adapted protocol. Briefly, 16 h post-infection (hpi) supernatants were removed and cells were lysed in 50 μl Renilla luciferase assay lysis buffer (Promega). Luminescence was measured using the Synergy™ 2 microplate reader (BioTek).

Additionally, cells were treated with serial dilutions of recombinant universal type I IFN (rIFN; R&D Systems) ranging from 1 Unit (U)/ml to 1,000 U/ml IFN and infected with RVFV-Ren. Cell lysis and analysis were performed as described previously.

Isolation of total RNA was performed using the NucleoSpin 1 RNA kit according to the manu-facturer´s instructions. Nucleic acid concentration was determined by photometric analysis with the NanoDrop 2000c (Peqlab).

To quantify IFNB1 RNA levels a hydrolysis probe-based reverse transcription quantitative PCR (RT-qPCR) assay was used. Primers and probes for IFNB1 and ACTB for H. sapiens and E. helvum were designed as described previously [42] and purchased from Integrated DNA Technologies. RT-qPCRs were performed using the Superscript III One-step RT-PCR kit (Thermo Fisher) with primers and probes according to S1 Table. Cycling and fluorescence signal acquisition was done for 45 cycles of 95˚C for 15 s and 58˚C for 40 s. RT-qPCR and data processing was performed using the LightCycler 1 480 Real-Time PCR System (Roche). Relative quantification of IFNB1 against the reference gene ACTB was evaluated using the 2 -ΔΔCt method as described elsewhere [49] . Quantification of CCL5, IFIT1, and MX1 expression was performed in an analogous procedure, using TBP as reference gene. The primers and probes that were used in this study are listed in S1 Table. Directional cloning Generation of recombinant pCAGGS expression vectors and pGL4 reporter plasmids was performed using directional cloning with restriction enzymes KpnI and NotI. Restriction recognition sites were inserted into DNA sequences by PCR. Subsequently, vector and insert DNA were digested and ligated using the Rapid DNA Ligation kit (Roche) with 10 ng of digested vector and 30 ng of insert DNA. Next, One Shot 1 TOP10 chemically competent E. coli cells (Thermo Fisher) were transformed with recombinant vector DNA according to the manufac-turer´s instructions. Clones were checked for recombinant DNA by colony PCR using the indicated primers. Plasmid DNA was isolated using the NucleoSpin 1 Plasmid kit and the NucleoBond 1 Xtra Midi EF kit (Macherey-Nagel) according to the manufacturer's instructions. Sequence verification of recombinant DNA was accomplished by Sanger sequencing (Microsynth AG) and analysis was performed with Geneious version 9.1.8 software (Biomatters Limited).

Plasmids encoding lyssavirus P proteins were generated by directional cloning of the respective lyssavirus P coding sequence (CDS) into the pCAGGS expression vector (sequence IDs of applied lyssavirus species/strains and phosphoproteins: RVP: EU886634.1; RVP 1088: AB645847.1; DUVV P: AF049115.1; LBV GH P: INSDC LN849915; LBV Sen P: NC_020807.1; LBV Nig P: EF547407.1). Constructs were designed containing restriction enzyme recognition sites, N-terminal FLAG-tag, and a Kozak sequence. DNA was synthesized as gBlocks 1 Gene Fragments by Integrated DNA Technologies. Cloning into pCAGGS vector and DNA sequencing was accomplished as described above.

Cells were grown in 24-well plates and transfected with 0.5 μg of the above-mentioned expression plasmids. For Western blot analysis cell lysates of the P protein-expressing cells were prepared. Cells were washed with cold PBS and lysed with Pierce RIPA lysis buffer (Thermo Fisher), supplemented with Benzonase 1 (Sigma-Aldrich), dithiothreitol (DTT), and Proteinase Inhibitor Cocktail III (Sigma-Aldrich). After incubation on ice for 20 min, cell lysates were centrifuged and stored at -80˚C until analysis.

Analysis of protein lysates was performed as described elsewhere [50] . Briefly, protein lysates along with the Spectra™ Multicolor Broad Range Protein Ladder (Thermo Fisher) were separated using 12% polyacrylamide gels (120 V, 45 min). Western blot analysis was performed with monoclonal mouse-anti-FLAG (1:5,000; Sigma-Aldrich; F1804) and monoclonal pan-species mouse-anti-β-actin (1:5,000; Sigma-Aldrich; A5316) primary antibodies that detect conserved epitopes of bat β-actin. For signal detection, horseradish peroxidase (HRP) labeled polyclonal goat-anti-mouse secondary antibody (1:10,000; Dianova; 115-035-146) was used. Chemiluminescence signal was detected using the SuperSignal 1 West Femto Chemiluminescence Substrate (Thermo Fisher) and visualized with the Fusion FX7™ imaging system (Vilber Lourmat).

For indirect immunofluorescence assay (IFA) analysis cells were seeded onto glass coverslips in 24-well plates and transfected with pCAGGS and pCAGGS-P carrying expression plasmids. Cells were fixed with Histofix 1 (Carl Roth), permeabilized with Triton™ X-100 (Sigma Aldrich) in PBS and incubated with a mouse-anti-FLAG (1:100; Sigma-Aldrich) primary antibody. Secondary detection was performed with a Cy3-conjugated goat-anti-mouse (1:200; Dianova) antibody. Glass slides were mounted using ProLong 1 Gold Antifade Reagent with DAPI (Thermo Fisher) and analyzed by fluorescence microscopy.

IFN-β promoter sequences of E. helvum [42] , Hypsignathus monstrosus [51] , Rhinolophus alcyone [52] , Tadarida brasiliensis (ATCC CCL-88), and Myotis daubentonii [51] cells were amplified by PCR and sequenced using primers Eidolon IFNb-P Screen fw, Eidolon IFNb-P Screen rev MyoLucF378, MyoLucR1082, bIFNp F1, bIFNpF2, bIFNp R3.Afr, according to S1 Table. Sequence verification was accomplished by Sanger sequencing (Microsynth AG) and analysis was performed with Geneious version 9. 

Recombinant IFN-β reporter vectors were produced by directional cloning of human and Rousettus aegyptiacus (R. aegyptiacus) IFN-β promoter sequences into the pGL4.10[luc2] vector (Promega), encoding a genetically engineered firefly luciferase. Restriction recognition sites were introduced to the promoter sequence via PCR using 20 ng of previously generated plasmids containing the human IFN-β promoter sequence (p125Luc; Prof. Takashi Fujita, University of Kyoto) and the in-house amplified R. aegyptiacus IFN-β promoter sequence, respectively, with primers p189_KpnI_fw2, p189_complete_XhoI_rev2, and p125_KpnI_fw2 and p125_complete_XhoI_rev2 according to S1 Table. The cloning procedure for vector and IFN-β promoter sequences was performed as described above. Clones were checked for recombinant DNA by colony PCR using primers pGL4.10-luc2_MCS_Screen_fw and pGL4.10-luc2_MCS_Screen_rev (see S1 Table for details). DNA sequence verification was performed as described above.

Activation of the human IFN-β promoter was analyzed by IFN-β promoter activation reporter assay. HEK-293T and A549 cells were seeded in 24-well plates (1x10 5 Reporter Assay System (Promega) using the Mithras LB 940 multimode microplate reader (Berthold Technologies). IFN-β promoter activity was calculated as percent induction relative to the empty vector control. Intra-assay variations in transcriptional activity were compensated by normalization with pGL4.74[hRluc/TK]-based luminescence data. Statistical analysis was performed using independent t-tests and SOFA Statistics version 1.4.6. Optimization experiments were performed accordingly and as indicated in the text and figure legends.

Essentially, the ISG54/interferon-stimulated response element (ISRE) promoter activation reporter assay was performed in an analogous procedure. Briefly, the following variations were applied to the protocol: HEK-293T cells were transfected with 25 ng pRL-SV40 and 125 ng pISG54-luc [53] reporter plasmids, and promoter activation was achieved by incubating cells with 200 U/ml rIFN for 18 h before cell lysis.

To conduct in vitro studies in reservoir-derived cells, we established an immortalized monoclonal E. helvum bat lung cell line, designated EidLu/20.2 ( Fig 1A) from a previously published heterogeneous EidLu cell culture [54] . To confirm that immortalization and subcloning did not affect the integrity of the IFN pathway, we compared the IFN response of EidLu/20.2 with the prototypic human lung cell line A549. First, we addressed IFN-β mRNA expression levels. We triggered the IFN response according to our previous experience with an RVFV-Ren reporter virus that lacks its IFN antagonist NSs and is predominantly sensed by RIG-I [42, 48, 55] . In addition, we transfected total RNA from VSV-infected cells previously shown to trigger both RNA helicases RIG-I and MDA5 [42] . The IFN-β mRNA induction ranged from 1,500-fold in VSV-RNA-transfected A549 cells to a maximal 400,000-fold in RVFV-Reninfected EidLu/20.2 cells. Overall, we found a 10-fold higher IFN-β mRNA expression in EidLu/20.2 cells compared to A549 for both stimuli (Fig 1B) . Secondly, we analyzed the secretion of bioactive IFN and characterized the cell-specific IFN sensitivity. We applied a previously established species-independent IFN bioassay [42, 49, 56, 57] . Cells were pretreated with supernatants from VSV-RNA-transfected or untreated control cells for 8 h and subsequently Relative quantification was performed using the 2 -ΔΔCt method and actin-β (ACTB) as a reference gene; induction is presented as Log10 fold induction. (C) A549 and EidLu/20.2 cells were infected with LBV-Ghana isolate (MOI 0.001). Viral RNA was extracted from cell culture supernatants and RT-qPCR was performed. Absolute quantification was done using LBV RNA standard curves. Data are presented as mean values and SD, derived from three biological replicates. All experiments were performed at least in triplicate. ���� = p<0.0001; �� = p<0.01; � = p<0.05; ns = not significant; determined by student's t-test (see S2 Table for infected with the IFN-sensitive RVFV-Ren reporter virus. High levels of bioactive IFN cell supernatants correspond to delayed recovery of RVFV-Ren growth in diluted supernatants (shown as relative light units (RLU)) [48] . Whereas preincubation of cells with a 1:100 dilution of A549-derived supernatants resulted in only 40% RVFV growth, EidLu/20.2 already showed 80% RVFV growth recovery (S1A and S1B Fig; experimental procedure is illustrated in S1C Fig) suggesting reduced IFN secretion by EidLu/20.2 cells or altered IFN signaling dynamics. Based on these findings, we sought to further investigate the interplay between bat-associated viruses and the host IFN response. To initially compare LBV replication between human and bat cells we infected A549 and EidLu/20.2 with a low-passage bat-borne LBV isolate from E. helvum brain named LBV-GH [8] using an MOI of 0.001. Virus growth was assessed in cell supernatants by genotype-specific RT-qPCR over a time course of 96 h (Fig 1C) . LBV-GH replication was comparable between A549 and EidLu/20.2 cells at early time points (18 to 48 h) reaching up to 3.3x10 5 RNA copies/ml in EidLu/20.2 cells (A549: 3x10 5 RNA copies/ml; Fig  1C) . Final virus titers were significantly higher in EidLu/20.2 cells at 96 hpi reaching 6.2x10 5 RNA copies/ml as compared to 1.2x10 5 copies/ml in A549 cells. The higher final titers of LBV-GH in EidLu/20.2 might be linked to the observed reduction of IFN bioactivity in this cell line. Another factor possibly affecting LBV propagation and RNA levels in human and bat cells might be virus-dependent control of the cellular antiviral response.

Cellular countermeasures against lyssavirus infections might modulate virus replication and often involve the IFN response [14, 16, 19] . The highly conserved lyssavirus phosphoprotein (P protein) counteracts antiviral cellular responses of innate immunity [16, 19, 57] . To first evaluate the impact of LBV P on IFN induction and signaling in human cells we cloned the P protein of the LBV-GH isolate and performed established promoter activation reporter assays for IFN-β and ISG54/ISRE signaling [53] in human HEK-293T cells. As control and for comparison, we included RVP, known to efficiently block IFN induction and signaling in human cells [16, 19, 57] . As expected, RVP reduced the IFN-β (Fig 2A) and ISG54/ISRE (Fig 2B) promoter activation by more than 90%. Importantly, we detected a significantly reduced inhibitory effect on IFN-β promoter activation by LBV P (63%) as compared to RVP (95%) (Fig 2A) . At the same time, LBV P and RVP exhibited a comparable reduction of IFN-based ISG54/ ISRE promoter activation (LBV P and RVP both >90%), with LBV P showing slightly increased efficacy (Fig 2B) . To validate whether the increased activity of LBV P over RVP in blocking ISRE activation translates into meaningful differences in the expression of distinct downstream interferon-stimulated genes (ISG), we quantified ISG transcription in RVP or LBV P expressing HEK-293T cells ( Fig 2C) . As expected, we observed a robust and highly comparable inhibition of CCL5, IFIT1, and MX1 induction by RVP and LBV P in IFN-treated cells. In contrast, there was no significant difference detectable between RVP and LBV P efficacy to block ISRE activation. In addition, we assessed the activation of the conserved ISG54/ ISRE reporter in EidLu/20.2 bat cells (S1D Fig). We detected robust ISRE activation upon recombinant IFN treatment and also observed a strong inhibition of ISG54/ISRE activation by RVP and even stronger for LBV P. As the pattern of IFN signaling inhibition by RVP and LBV P was comparable between bat and human cells we focused our study on the inhibitory activity of lyssavirus P proteins on IFN-β induction.

The differential inhibition of IFN-β promoter activation by RVP and LBV P in human cells and the fact that functional aspects of lyssavirus P proteins have mainly been investigated in human and primate cells [14, [16] [17] [18] [19] [57] [58] [59] encouraged us to further investigate specific lyssavirus P protein functions in E. helvum bat cells. For this, we established a bat-specific IFN-β promoter-reporter assay. Five so far unknown sequences of the positive regulatory domains (PRDs) of different bat IFN-β promoters, which represent the central hub for the regulation of antiviral IFN-β transcription [60] , were amplified, sequenced, and aligned with all available bat sequences and representative sequences from selected mammalian taxa and Homo sapiens (Fig 3A) . GenBank entries were uploaded for the respective sequences (see Material and Methods). All applied Yinpterochiroptera PRDs were highly conserved except for the Pteropus alecto (P. alecto) PRD that exhibited two nucleotide polymorphisms. Sequences from other mammalian representatives showed a reduced nucleotide identity of <89% to the E. helvum PRD. This finding is in line with our own previous experience that human IFN-β promoter plasmids show limited activation in bat cell lines. For the generation of a bat-specific IFN-β promoter activation reporter plasmid, applicable to multiple bat species, a representative IFN promoter region (-278 to -89 upstream of ATG) from previously established Rousettus aegyptiacus kidney cells (RoNi/7) [42] was cloned into the commercially available reporter vector pGL4.10[luc2], encoding a genetically engineered firefly (FF) luciferase from Photinus pyralis. Subsequently, the bat-specific pGL4-IFN-β reporter construct, designated IFN-β-FF, was tested and titrated in E. helvum bat cells. To identify the optimal reference plasmid for the detection of the baseline transcription factor activity in bat cells we tested different Renilla reniformis (RL) luciferase-expressing plasmids controlled by an SV40 (pRL-SV40) or a thymidine kinase promoter (pGL4.74 [hRLuc/TK]) (Fig 3B) . Taking into account the luminometer settings, the linear range of detection is achieved above 1,000 RLU with background counts <100 RLU. Optimal RL luciferase read-out was therefore only achieved upon transfection of 25 ng pRL-SV40 plasmid with RLU values >1,000 (Fig 3B) . IFN-β promoter activation was inducible by VSV-RNA transfection and dose-dependent ranging from 0.1 to 0.4 relative induction (FF/RL). Comparison of non-transfected with VSV-RNA transfected cells showed that the fold induction was highest (94-fold) using 250 ng IFNb-FF, which was subsequently

Human HEK-293T cells were transfected with RL and FF luciferase reporter plasmids (A: pGL4.10[luc2]; B: pISG54-luc) and 50 ng of pCAGGS empty vector (EV) or pCAGGS vector encoding Rabies virus P protein (RVP) or LBV-GH P protein (LBV GH P). Cells were stimulated by VSV-RNA transfection (A) or recombinant IFN treatment (B) and lysed after 18 h. Luciferase signal was measured and IFN-β or ISG54/ISRE induction relative to EV was calculated. Inhibition of interferon-stimulated gene (ISG) expression was analyzed in HEK-293T (C) cells using RT-qPCR analysis. Cells were transfected with 250 ng of pCAGGS EV or pCAGGS vector encoding RVP or LBV-GH P. After 24 h, cells were stimulated using 100 U/ml recombinant IFN, RNA was extracted 18 h after treatment, and the expression of antiviral ISGs CCL5, IFIT1, and MX1 was quantified. Relative quantification was performed using the 2 -ΔΔCt method and TBP as a reference gene. All experiments were performed in triplicate and mean values and SD were calculated. ���� = p<0.0001; ��� = p<0.001; �� = p<0.01; � = p<0.05; ns = not significant; determined by student's ttest (see S2 Table for applied in follow-up experiments. To assess overall assay performance, we tested RVP-dependent inhibition of IFN-β promoter activation, which is a broadly active IFN antagonist. As shown in Fig 3D, IFN -β promoter activation was inhibited in a dose-dependent manner ranging from~50% (6.25 ng pCAGGS-RVP) to~90% inhibition (50 ng pCAGGS-RVP).

Next, we reassessed our findings on LBV P IFN-ß inhibitory activity in bat cells and performed IFN-β promoter activation reporter assays in E. helvum cells. To better reflect lyssavirus diversity, we included the P proteins of two additional LBV isolates (LBV-Nig56, lineage B, Nigeria, accession no. EU293110; LBV-Sen, lineage A, Senegal, EU293108) [61] , representing lyssavirus phylogroup II, as well as Duvenhage virus (DUUV; accession no. EU293120) and street rabies To exclude cell line-specific effects, we compared the IFN-ß inhibitory activity of all mentioned lyssavirus P in two human cell lines (HEK-293T, A549) and two E. helvum cell lines (EidLu/20.2, EidNi/41.3). Representative immunofluorescence assays and control Western blots were performed to illustrate transfection efficiency and P protein expression in the analyzed cell lines (S2 Fig). Protein levels were normalized using actin as a reference protein, showing generally comparable P protein expression. All LBV P showed significantly lower inhibition of IFN-β promoter activation than phylogroup I representatives RVP, RVP 1088, and DUVV P in human cell lines (Fig 4A and 4B) . Whereas RVP and DUVV P were able to reduce IFN-β promoter activation by~95% (HEK-293T), and~70% (A549), all applied LBV P variants exhibited significantly reduced efficiency in HEK-293T (73% to 86% inhibition), and A549 cells (42% to 46% inhibition). Noteworthy, RVP of the RABV-1088 street strain showed enhanced IFN-β inhibition in A549 cells, corroborating previous findings of Masatani et al. using a comparable luciferase-based approach [18] . In both bat cell lines, no significant decrease of IFN-β inhibition was detected between RVP and all tested LBV P constructs exhibiting a highly comparable inhibition of >87% (Fig 4C and 4D) . In contrast to the other tested cell lines, all LBV P variants showed slightly enhanced IFN-ß inhibitory activity in comparison to RVP in EidNi/41.3 cells (Fig 4D) . In sum, we show that, as compared to prototypic RVP, LBV P anti-IFN-β activity is substantially reduced in human cells but robust and comparable to other lyssavirus P in cells from the putative natural bat reservoir.

Bats host a plethora of viruses and seem to tolerate virus infections without showing severe clinical symptoms [29, 31] . This might be connected to an enhanced IFN response with reduced general inflammation upon virus infection, possibly facilitating virus persistence instead of acute infections [35, 62] . By using a newly established bat-specific IFN promoter activation assay we showed that LBV P has robust IFN-ß inhibitory activity in reservoir hostderived bat cells. In contrast, its capability to block IFN-ß induction was clearly reduced in human cell lines compared to the impact of prototypic RVP. The conserved and strong IFN-ß inhibitory activity of all tested LBV P proteins might be connected to long-term host adaptation processes to the bat cell environment [35] . In addition, a strong and continuously upregulated IFN response [34] would force bat-associated viruses to evolve highly active P proteins to prevent virus elimination. Although we did not observe the previously described [34] elevated basal transcriptional upregulation of IFNB1 mRNAs in E. helvum cells, we detected an enhanced bat cell-specific IFN response upon transfection of dsRNA molecules and infection with an RVFV reporter virus lacking its IFN antagonist NSs.

A previous study showed differential IFN-ß inhibitory activity of lyssavirus P proteins in human cells [18] . Using human HEK-293T cell-based assays, Masatani and colleagues found that all lyssavirus P proteins block IFN induction via TBK-1. In addition, street strain RVP variants showed enhanced IKKepsilon (IKKε)-dependent IFN-ß inhibitory activity, as compared to cell culture-adapted RVP or other lyssavirus P proteins. The authors suggested that the additional function might inhibit the NF-κB-dependent proinflammatory response. We observed that LBV P proteins (phylogroup II) derived from LBV isolates from wild bats show reduced IFN-ß inhibition in HEK-293T and A549 cells as compared to prototypic RVP. The strongest IFN-ß inhibitory activity was indeed seen for the street strain variant 1088 P protein. SAG2 strain-based RVP and DUVV P protein also showed strong IFN-ß inhibitory activity in HEK-293T cells that was, however, not seen in A549 cells. An explanation might be that we induced the IFN response with total RNA from VSV-infected cells, known to trigger both RIG-I and MDA5 helicases. Masatani et al. overexpressed specific signaling molecules, e.g. IRF-3, and infected cells with New Castle Disease virus known to be sensed by RIG-I. In addition, the authors applied different promoter activation plasmids focusing on IRF-3-dependent promoter binding, whereas we used the complete IFN-beta promoter region that is targeted by multiple transcription factors. An MDA5-triggered IFN induction might surpass the IKKεmediated IFN response and, for example, stimulate NF-κB-dependent IFN promoter activation. Of note, overexpression of signaling molecules as well as transfection of exogenous RNA

Human kidney (HEK-293T; A) and lung (A549; B), and E. helvum kidney (EidNi/41.3; D) and lung (EidLu/20.2; C) cells were transfected with Renilla (RL) and firefly (FF) luciferase reporter plasmids, and 50 ng of pCAGGS empty vector (EV) or following FLAG-tagged lyssavirus P proteins: Rabies virus P protein (RVP), RVP street rabies strain 1088 (RVP 1088), Duvenhage virus P protein (DUVV P), LBV Ghana isolate (LBV-GH), LBV Senegal isolate (LBV-Sen) and LBV Nigeria isolate (LBV-Nig), respectively. IFN induction was achieved by transfection of 1 μg VSV-RNA (total RNA from vesicular stomatitis virus-infected cells). Cells were lysed after 18 h and luciferase signal was measured. Data were normalized using RL luciferase reference plasmid. For comparability between cell lines, IFNβ induction relative to the EV negative control was determined. Experiments were performed in triplicate and mean values from three independent experiments and SD are shown. ���� = p<0.0001; �� = p<0.01; � = p<0.05; ns = not significant; determined by student's t-test (see S2 Table for are both well-established methods to analyze immune pathways and highly sensitive and finetuned IFN-ß promoter reporter assays. While the use of such stimuli enables specific targeting of certain components of immune pathways, virus infections are less artificial and elicit more complex cellular responses [63] . During replication, VSV produces different RNA species that induce both RIG-I and MDA5, although RIG-I seems to be predominantly involved in generating an efficient antiviral immune response [64, 65] . Comparably, RIG-I was identified as the primary sensor molecule for rabies virus as prototypic representative of the virus family Rhabdoviridae, which also encompasses VSV [66] . In addition. rabies virus was also found to induce MDA5-dependent IFN signaling [11] , whereas in vivo experiments suggested a superior role for RIG-I activation in survival and IFN production upon both rabies virus and VSV infection. While using exogenous RNA stimuli simplifies complex virus-host interactions and focuses on the relevant signaling pathways, detrimental effects of virus infection like transcriptional or translational shutoff are avoided [67, 68] .

In bat cells, all lyssavirus P proteins of both phylogroups showed strong activity, possibly indicating that all P proteins interfere with bat TBK-1 and bat IKKε, blocking IFN-ß induction and NF-κB activation at the same time. This would support previous suggestions that all street strain lyssavirus P proteins originally have a double function that might be lost through cell culture passaging in fixed, culture-adapted lyssavirus strains. Detailed protein interaction and promoter activation studies with bat TBK-1 and bat IKKε should further investigate this hypothesis. In human cells, the induction of other IFN subtypes, such as IFN-α largely relies on the same signaling molecules as IFN-ß, including TBK-1 and IKKε, but does not require NF-κB signals for gene expression [69] . IFN-α is not as well characterized as IFN-ß in the context of rabies virus infection and seems to play a subordinate role in antiviral signaling [70] and was therefore not assessed in this study. Due to signaling through TBK-1 and IKKε, the major targets of RVP interference, a comparable RVP-based inhibition of IFN-α promoter activation is highly likely. Interestingly, IFN-λ has very recently been shown to attenuate rabies virus infection and induce antiviral genes, although P protein-based interference with the induction of distinct IFN classes is less well described [71] .

By using an ISG54/ISRE promoter activation assay, we observed a slightly increased capacity of LBV P as compared to RVP to block IFN signaling in both human and bat cells using a human ISRE reporter construct. While this increase in efficiency might translate to an increased replicative capacity of LBV over rabies virus, we were not able to detect differences between RVP and LBV P in inhibiting ISG expression in human cells. In contrast, the decreased IFN-ß inhibitory action of LBV P was only detectable in human cells, highlighting the need for species-specific tools to evaluate antiviral immune responses. Nevertheless, future studies should not only further evaluate the consequences of P protein activity in bat cells in the context of lyssavirus infection but also evaluate the capacity of P proteins to block JAK--STAT signaling in bat cell systems with appropriate bat-specific assays.

To strengthen our findings on IFN-ß inhibition, we not only included two established human cell lines but also distinct E. helvum bat cells lines in the described IFN-ß promoter activation studies. Although these cell cultures are each derived from different organs (lung, kidney), observing cell line specific effects cannot be completely ruled out. In addition, the order Chiroptera is highly diverse and general conclusions on bat immune responses are difficult to draw. As we focused on LBV in this study, we chose cells from the putative reservoir host E. helvum for our analyses. However, it has to be noted that the described findings might not hold true for cell lines from other species of this large mammalian taxon. A study by Hölzer et al. has recently analyzed the IFN-induced transcriptome of Yangochiroptera bat cells in great detail, revealing substantial differences to published findings on Yinpterochiroptera bats [55] . The authors not only describe the lack of multiple ISGs that were described in Yinpterochiroptera bats but also failed to confirm earlier findings of IFN-α gene expansion or high baseline IFN expression in bat cell lines. These discrepancies further highlight the need for bat-specific tools and an overall caution regarding broad and general statements on bat immune responses. Future studies should not only consider multiple bat species but also implement an increased repertoire of cell culture models to further elucidate the interaction of lyssaviruses and the immune system of distinct bat representatives.

Noteworthy, the impact of IFN antagonists on virus replication and IFN signaling in different bat cells is still not fully understood and few studies have addressed this topic [72] [73] [74] . In the case of bats, it is largely unknown to what extent the immune system influences the manifestation of viral diseases or their progression. However, recent research suggests that bats have evolved unique features to fight RNA viruses [75] . One of the most important findings is the discovery of an attenuated inflammatory response in bats [33, 76] that is tightly connected to the IFN response. While Ahn et al. reported a dampened inflammasome activation in bat cells, others suggested enhanced virus restriction based on elevated metabolism or a particularly effective IFN response [33, 77, 78] . To analyze the extent of differential LBV P activity on viral replication and proinflammatory responses would rely on reverse genetics approaches and more bat-specific tools to investigate proinflammation in the context of virus infection.

Other studies also reported high ISG expression in unstimulated bat cells, possibly enhancing intrinsic virus resistance [79] . A recent study by Irving et al. elegantly describes the influence of interferon regulatory factors 1/3/7 on ISG expression in bat cells [80] . Besides showing high expression levels of these central antiviral regulators in bat cells and tissue, which translates to elevated ISG expression, the authors also describe a prolonged IFN-like antiviral signature and novel bat-specific antiviral mechanisms. In light of these findings, developing batspecific tools to integrate IFN induction pathways, on the one hand, and expression of ISGs as antiviral effectors, on the other hand, might be an important step towards decoding the unique antiviral immune response of bats.

In sum, our established bat-specific IFN-β promoter-reporter assay might serve as an in vitro risk assessment tool to analyze bat virus-derived IFN antagonists in bat cells. This approach enables comparisons with other mammalian cell systems to explore distinct aspects of virus virulence and understand specific features of the bat IFN system [18, [81] [82] [83] .

The functionality of IFN signaling pathways was analyzed in A549 (A) and EidLu/20.2 (B) cells. After seeding cells, increasing amounts of total RNA from Vesicular stomatitis virus (VSV)-infected cells were transfected as indicated. After 20 h the supernatants were collected, diluted as indicated, and added to naïve cells of the respective cell line. After 7 h cells were infected with RVFV-Ren (MOI 0.002), incubated for 16 h, lysed and luciferase signal was measured. Virus growth was calculated relative to the mock-treated cells (0 μg) for the respective dilution. High values indicate increased virus replication and low amounts of bioactive IFN in the supernatants. Experiments were performed in triplicate and mean values and SD are shown. The experimental procedure is depicted in (C). Upon detection of VSV-RNA type I IFNs are produced and secreted. Incubation of cells with IFN-containing supernatants induces an antiviral state characterized by the transcription of interferon-stimulated genes (ISGs). RVFV-Ren reporter virus growth is inhibited by various ISG proteins in IFN stimulated cells. EidLu/20.2 cells were transfected with RL and FF luciferase reporter plasmids (pISG54-luc) and 50 ng of pCAGGS empty vector (EV) or pCAGGS vector encoding Rabies virus P protein (RVP) or LBV-GH P protein (LBV GH P). Cells were stimulated by recombinant IFN treatment (100 U/ml) and lysed after 18 h. Luciferase signal was measured and ISG54/ISRE induction relative to EV was calculated (D). All experiments were performed in triplicate and mean values and SD were calculated. � = p<0.05; determined by student's t-test (see S2 Table  for Human kidney (HEK-293T) and E. helvum lung (EidLu/20.2) cells were transfected with pCAGGS vector encoding P proteins from human-and bat-associated lyssaviruses (Rabies virus P protein (RVP); RVP street rabies strain 1088 (RVP 1088); Duvenhage virus P protein (DUVV P); LBV Ghana isolate P protein (LBV-GH P); LBV Senegal isolate P protein (LBV-Sen P); LBV Nigeria isolate P protein (LBV-Nig P)). (A) For immunofluorescence assay cells were seeded on glass coverslips in 24-well plates and transfected with 0.5 μg plasmid DNA per well. Cell fixation and lysis were performed 24 h after transfection. For immunofluorescence microscopy, cells were stained with mouse-anti-FLAG antibody (1:100) and goat anti-mouse-Cy3 antibody (1:200). Cellular DNA was stained with DAPI. Bars represent 50 μm. (B) For Western blot analysis, cell lysates from transfected (2 μg/6-well, respectively) HEK-293T, and A549, EidLu/ 20.2, and EidNi/41.3 were prepared 48 h post-transfection with Pierce RIPA lysis buffer (Thermo Fisher). Protein lysates and Spectra™ Multicolor Broad Range Protein Ladder (Thermo Fisher) were separated by SDS-PAGE using 10% polyacrylamide gels (120 V, 45 min). Western blot analysis was performed with mouse-anti-FLAG (1:3,000; Sigma-Aldrich) and mouse-anti-β-actin (1:5,000; Sigma-Aldrich) primary antibodies and fluorescently-labeled goat-anti-mouse (1:15,000; 800 nm; Licor), or goat-anti-rabbit (1:15,000; 680 nm; Licor) secondary antibody. Fluorescent signal was visualized with the Licor Odyssey CLx imaging system (Licor), fluorescent signal intensities were measured and FLAG-based signal intensity was normalized to actin reference protein using the Image Studio™ software package. (TIF) S1 

Estimating the global burden of endemic canine rabies

Australian Bat Lyssavirus in a Child: The First Reported Case

Lyssaviruses and Bats: Emergence and Zoonotic Threat

Fatal human rabies caused by European bat lyssavirus type 2a infection in Scotland

Fatal Human Rabies due to Duvenhage Virus from a Bat in Kenya: Failure of Treatment with Coma-Induction, Ketamine, and Antiviral Drugs

Exposure to Lyssaviruses in Bats of the Democratic Republic of the Congo

Evidence of Lagos bat virus circulation among Nigerian fruit bats

Lagos bat virus transmission in an Eidolon helvum bat colony

Endemic Lagos bat virus infection in Eidolon helvum

Pathogenesis of bat rabies in a natural reservoir: Comparative susceptibility of the straw-colored fruit bat (Eidolon helvum) to three strains of Lagos bat virus

Rabies virus infection induces type I interferon production in an IPS-1 dependent manner while dendritic cell activation relies on IFNAR signaling

Experimental study of European bat lyssavirus type-2 infection in Daubenton's bats (Myotis daubentonii)

Phosphoprotein Gene of Wild-Type Rabies Virus Plays a Role in Limiting Viral Pathogenicity and Lowering the Enhancement of BBB Permeability

Role of interferon antagonist activity of rabies virus phosphoprotein in viral pathogenicity

The importance of immune evasion in the pathogenesis of rabies virus

Genetic dissection of interferonantagonistic functions of rabies virus phosphoprotein: inhibition of interferon regulatory factor 3 activation is important for pathogenicity

Inhibition of interferon signaling by rabies virus phosphoprotein P: activation-dependent binding of STAT1 and STAT2

Contribution of the interaction between the rabies virus P protein and I-kappa B kinase epsilon to the inhibition of type I IFN induction signalling

Conservation of a Unique Mechanism of Immune Evasion across the Lyssavirus Genus

Replication and shedding of MERS-CoV in Jamaican fruit bats (Artibeus jamaicensis)

Isolation of genetically diverse Marburg viruses from Egyptian fruit bats

Pteropid Bats are Confirmed as the Reservoir Hosts of Henipaviruses: A Comprehensive Experimental Study of Virus Transmission

Amplification of Emerging Viruses in a Bat Colony

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding

A pneumonia outbreak associated with a new coronavirus of probable bat origin

WHO. Coronavirus disease 2019 (COVID-19)-Situation Report

Investigating the zoonotic origin of the West African Ebola epidemic

Bats are natural reservoirs of SARS-like coronaviruses

Bat tolerance to viral infections

Host immunity to repeated rabies virus infection in big brown bats

Bat coronaviruses and experimental infection of bats, the Philippines

Experimental infection of serotine bats (Eptesicus serotinus) with European bat lyssavirus type 1a

Dampened NLRP3-mediated inflammation in bats and implications for a special viral reservoir host

Contraction of the type I IFN locus and unusual constitutive expression of IFN-alpha in bats

Accelerated viral dynamics in bat cell lines, with implications for zoonotic emergence

Tools to study pathogen-host interactions in bats

Uptake of rabies virus into epithelial cells by clathrin-mediated endocytosis depends upon actin

Experimental rabies infection of non-nervous tissues in skunks (Mephitis mephitis) and foxes (Vulpes vulpes)

Experimental Lagos bat virus infection in straw-colored fruit bats: A suitable model for bat rabies in a natural reservoir species

Rabies virus distribution in tissues and molecular characterization of strains from naturally infected non-hematophagous bats

Bat airway epithelial cells: a novel tool for the study of zoonotic viruses

Type I Interferon Reaction to Viral Infection in Interferon-Competent, Immortalized Cell Lines from the African Fruit Bat Eidolon helvum

Evolution of the cytochrome b gene of mammals

Variation of mitochondrial DNA in the Hipposideros caffer complex (Chiroptera: Hipposideridae) and its taxonomic implications

Broad-range real-time PCR assay for the rapid identification of cell-line contaminants and clinically important mollicute species

Heminested PCR assay for detection of six genotypes of rabies and rabies-related viruses

Diagnostic reverse-transcription polymerase chain reaction kit for filoviruses based on the strain collections of all European biosafety level 4 laboratories

Species-independent bioassay for sensitive quantification of antiviral type I interferons

Analysis of relative gene expression data using real-time quantitative PCR and the 2(T)(-Delta Delta C) method

Human coronavirus NL63 open reading frame 3 encodes a virion-incorporated N-glycosylated membrane protein

Comparative analysis of Ebola virus glycoprotein interactions with human and bat cells

Differential sensitivity of bat cells to infection by enveloped RNA viruses: coronaviruses, paramyxoviruses, filoviruses, and influenza viruses

IFN-Stimulated transcription through a TBP-free acetyltransferase complex escapes viral shutoff

A novel rhabdovirus isolated from the straw-colored fruit bat Eidolon helvum, with signs of antibodies in swine and humans

PubMed Central

Virus-and Interferon Alpha-Induced Transcriptomes of Cells from the Microbat Myotis daubentonii. iScience

Differential transcriptional responses to Ebola and Marburg virus infection in bat and human cells

Identification of the rabies virus alpha/beta interferon antagonist: phosphoprotein P interferes with phosphorylation of interferon regulatory factor 3

Interaction of Rabies Virus P-Protein With STAT Proteins is Critical to Lethal Rabies Disease

The nucleocytoplasmic rabies virus P protein counteracts interferon signaling by inhibiting both nuclear accumulation and DNA binding of STAT1

Enhanceosome formation over the beta interferon promoter underlies a remote-control mechanism mediated by YY1 and YY2

Phylogeny of Lagos bat virus: Challenges for lyssavirus taxonomy

Selection of viral variants during persistent infection of insectivorous bat cells with Middle East respiratory syndrome coronavirus

Interferons and viruses: an interplay between induction, signalling, antiviral responses and virus countermeasures

Characterization of retinoic acid-inducible gene-I expression in primary murine glia following exposure to vesicular stomatitis virus

Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses

5'-Triphosphate RNA is the ligand for RIG-I

Translational Control in Virus-Infected Cells

Viral subversion of the host protein synthesis machinery

Interferon-alpha and -beta in kidney inflammation

The type I interferon response bridles rabies virus infection and reduces pathogenicity

Interferon-lambda Attenuates Rabies Virus Infection by Inducing Interferon-Stimulated Genes and Alleviating Neurological Inflammation

The V Protein of Tioman Virus Is Incapable of Blocking Type I Interferon Signaling in Human Cells

Marburg Virus Viral Protein 35 Inhibits Protein Kinase R Activation in a Cell Type-Specific Manner

Lloviu virus VP24 and VP35 proteins function as innate immune antagonists in human and bat cells

Bats and Coronaviruses

Lack of inflammatory gene expression in bats: a unique role for a transcription repressor

Bat flight and zoonotic viruses

Comparative analysis of bat genomes provides insight into the evolution of flight and immunity

The IFN Response in Bats Displays Distinctive IFN-Stimulated Gene Expression Kinetics with Atypical RNASEL Induction

Interferon Regulatory Factors IRF1 and IRF7 Directly Regulate Gene Expression in Bats in Response to Viral Infection

Assessing the contribution of interferon antagonism to the virulence of West African Ebola viruses

Novel Bat Influenza Virus NS1 Proteins Bind Double-Stranded RNA and Antagonize Host Innate Immunity

Possible role of the Nipah virus V protein in the regulation of the interferon beta induction by interacting with UBX domain-containing protein1

We thank Nicolas Heinemann, Max Tschischka, Christina Metzger for excellent technical assistance and Dr. Sebastian Hauka for technical and scientific advice. We also thank the Friedrich-Löffler-Institute for providing the LBV-GH isolate, Prof. Klaus Conzelmann for the RVP plasmid, Prof. Friedemann Weber for sharing the RVFV-Ren reporter virus, and Prof. Takashi Fujita and Prof. David E. Levy for kindly sharing the p125-luc and ISG54-luc plasmids.