Activation of protein kinase receptor (PKR) plays a pro-viral role in mammarenavirus-infected cells

ABSTRACT

Many viruses, including mammarenaviruses, have evolved mechanisms to counteract different components of the host cell innate immunity, which is required to facilitate robust virus multiplication. The double-stranded RNA (dsRNA) sensor protein kinase receptor (PKR) pathway plays a critical role in the cell anti-viral response. Whether PKR can restrict the multiplication of the Old World mammarenavirus lymphocytic choriomeningitis virus (LCMV) and the mechanisms by which LCMV may counteract the anti-viral functions of PKR have not yet been investigated. Here we present evidence that LCMV infection results in very limited levels of PKR activation, but LCMV multiplication is enhanced in the absence of PKR. In contrast, infection with a recombinant LCMV with a mutation affecting the 3′−5′ exonuclease (ExoN) activity of the viral nucleoprotein resulted in robust PKR activation in the absence of detectable levels of dsRNA, which was associated with severely restricted virus multiplication that was alleviated in the absence of PKR. However, pharmacological inhibition of PKR activation resulted in reduced levels of LCMV multiplication. These findings uncovered a complex role of the PKR pathway in LCMV-infected cells involving both pro- and anti-viral activities.

IMPORTANCE

As with many other viruses, the prototypic Old World mammarenavirus LCMV can interfere with the host cell innate immune response to infection, which includes the dsRNA sensor PKR pathway. A detailed understanding of LCMV-PKR interactions can provide novel insights about mammarenavirus-host cell interactions and facilitate the development of effective anti-viral strategies against human pathogenic mammarenaviruses. In the present work, we present evidence that LCMV multiplication is enhanced in PKR-deficient cells, but pharmacological inhibition of PKR activation unexpectedly resulted in severely restricted propagation of LCMV. Likewise, we document a robust PKR activation in LCMV-infected cells in the absence of detectable levels of dsRNA. Our findings have revealed a complex role of the PKR pathway during LCMV infection and uncovered the activation of PKR as a druggable target for the development of anti-viral drugs against human pathogenic mammarenaviruses.

INTRODUCTION

Mammarenaviruses are enveloped viruses with a bi-segmented negative-stranded RNA genome (1) which cause chronic infections of their natural rodent host across the world, and human infections occur through mucosal exposure to aerosols or by direct contact of abraded skin with infectious materials (2). Upon a zoonotic event, mammarenaviruses can subvert the innate immune responses in the infected individual and interfere with the development of an effective adaptive immune response, which results in unrestricted virus multiplication and associated pathology and disease. Thus, several mammarenaviruses, chiefly Lassa virus (LASV), cause hemorrhagic fever disease in humans and pose important public health problems in their endemic regions (3). Moreover, evidence indicates that the worldwide distributed lymphocytic choriomeningitis virus (LCMV) is a neglected human pathogen of clinical significance (4–8) and a serious risk to immunocompromised individuals (9, 10). There are no Food and Drug Administration-licensed arenavirus vaccines, and current anti-arenaviral therapy is limited to an off-label use of ribavirin for which efficacy remains controversial (11). Therefore, there is a pressing need to develop effective strategies to combat human pathogenic mammarenaviruses, a task that will be facilitated by a better understanding of mammarenavirus-host innate defense interactions.

During multiplication in their host cells, RNA viruses generate a variety of pathogen-associated molecular patterns (PAMPs), including RNA-based PAMPs, that are recognized by host cellular pathogen recognition receptor (PRR) molecules, including TLR 3/TLR 7 and the retinoic acid-inducible gene I-like receptors (RLRs) retinoic acid-inducible gene I (RIG-I) and MDA5 (12, 13). Activated RLRs associate with the adapter mitochondrial anti-viral-signaling (MAVS) protein (14) to promote activation of the non-classical IkB kinase (IKK)-related kinases (15) TANK-binding kinase 1 and IKK epsilon (16, 17) that activate interferon regulatory factor 3 and nuclear factor kappa-light-chain-enhancer of activated B cells, which, together with ATF2/c-JUN, initiate transcription of interferon beta (IFN-β) promoter (18). Interaction of secreted IFN-β with its receptor (IFNAR) activates the JAK/STAT signaling pathway (19), resulting in induction of hundreds of type 1 IFN (T1IFN) stimulated genes to produce a cellular anti-viral state and control viral infection. Many viruses, including mammarenaviruses (20), have developed ways to subvert the T1IFN response (21, 22). Induction of IFN-β production in mammarenavirus-infected cells is greatly diminished by the virus nucleoprotein (NP) ability to inhibit activation of IRF3 (23, 24). The anti-IFN-β activity of NP has been linked to its C-terminally located functional 3′−5′ exonuclease (ExoN) domain characteristic of the DEDDh ExoN superfamily (25, 26). It is thought that NP’s ExoN activity promotes the degradation of viral RNA species that can activate the RIG-I/MAVS pathway and subsequent induction of IFN-β production to trigger the T1IFN pathway (27). The double-stranded RNA (dsRNA) sensor protein kinase receptor (PKR) is at the center of cellular responses to a variety of stress signals, including viral infection (28). Short viral dsRNA species generated during virus replication in infected cells can trigger PKR activation reflected in increased levels of phosphorylated protein kinase receptor (pPKR) (29–32). Increased levels of pPKR lead to phosphorylation of the eukaryotic translation initiation factor 2 alpha (eIF2α), resulting in the inhibition of cap-dependent protein translation initiation, which contributes to restricting virus propagation (33). Accordingly, many viruses have evolved mechanisms to counteract the activation of the PKR pathway (34–36). However, conflicting findings have been reported on mammarenavirus-PKR pathway interactions. The New World mammarenavirus Junin virus (JUNV) was reported to induce high levels of pPKR that did not result in increased levels of phosphorylated eukaryotic translation initiation factor 2 alpha (peIF2α), whereas infection with the Old World mammarenavirus LCMV resulted in significantly increased levels of peIF2α despite minimal increased levels of pPKR (37). In addition, infection with the New World mammarenavirus Tacaribe virus (TCRV) resulted in increased levels of both pPKR and peIF2α, which contributed to restricted TCRV multiplication (38). However, other reports documented minimal or negligible activation of both PKR and/or eIF2α in cells infected with mammarenaviruses (39–42). Hence, the need for a detailed characterization of PKR-LCMV interactions is warranted, which is the focus of the present work.

Here, we investigated the interaction of LCMV with the PKR pathway and showed that LCMV infection resulted in minimal increased levels of pPKR, but LCMV multiplication was enhanced in PKR knockout (PKR-KO) cells. However, unexpectedly, pharmacological inhibition of PKR activation resulted in severely restricted multiplication of LCMV, as well as LASV and JUNV. In contrast, cells infected with rLCMV/NP(D382A), impaired in its NP ExoN and anti-T1IFN activities (25), exhibited high levels of pPKR and peIF2α. However, levels of dsRNA remained below detectable levels in rLCMV/NP(D382A)-infected cells. Our findings have uncovered a previously unnoticed pro-viral activity of the PKR pathway during mammarenavirus infection, raising the intriguing possibility that PKR activation might be considered as a potential druggable target to combat infections by human pathogenic mammarenaviruses.

RESULTS

Effect of genetic ablation of PKR on multiplication of LCMV

To determine whether genetic ablation of PKR affected LCMV multiplication, we compared the multi-step growth kinetics of LCMV/wild type (WT) and the ExoN mutant rLCMV/NP(D382A) in WT and PKR-KO A549 cells. We reasoned that LCMV/WT has already a high multiplication fitness in cultured cells, and therefore, ablation of an anti-viral host cell factor would be expected to result in only a modest increase in virus peak titers. In contrast, the fitness of an attenuated LCMV mutant, like rLCMV/NP(D382A), may be more dramatically affected by the removal of an anti-viral host cell factor. We observed that LCMV/WT replicated to high titers in WT and KO A549 lines (Fig. 1A), but virus titers were consistently higher in PKR-KO compared to WT A549 cells. Multiplication of rLCMV/NP(D382A) was severely restricted in WT A549 cells, and production of infectious progeny in tissue culture supernatant (TCS) was only detected at 24 hpi and at very low levels (<10² focus-forming units [FFU]/mL), whereas at 48, 72, and 96 hpi, titers in TCS were below detection levels (Fig. 1A). In contrast, rLCMV/NP(D382A) was able to multiply in PKR KO A549 cells, resulting in titers of infectious progeny in TCS of >10³ FFU/mL (Fig. 1A). We further confirmed the effect of PKR on LCMV multiplication in cultured cells by determining levels of LCMV NP RNA using reverse transcription quantitative PCR (RT-qPCR) in WT- and PKR-KO A549-infected cells (Fig. 1B). Consistent with the RT-qPCR results, levels of rLCMV/NP(D382A), but not LCMV/WT, small (S) segment RNA (replication), and NP mRNA (transcription) were drastically reduced in WT compared to PKR-KO A549 cells as determined by northern blotting (Fig. 1C). We next asked if the restricted multiplication of rLCMV/NP(D382A) in A549 WT cells could be overcome by increasing the initial virus input. We infected WT and PKR-KO A549 cells with either LCMV/WT or rLCMV/NP(D382A) using different multiplicities of infection (MOIs) and determined the numbers of NP⁺ cells at 48 hpi (Fig. 1D). Propagation of LCMV/WT was similarly efficient in WT and PKR-KO A549 cells, even at the lowest (0.05) MOI. In contrast, propagation of rLCMV/NP(D328A) in WT, but not PKR-KO, A549 cells was greatly influenced by the MOI used to initiate infection (Fig. 1Di). Infection of WT A549 cells with rLCMV/NP(D382A) at an MOI of 0.05 or 1.35 resulted in low (6%) and high (90%), respectively, percentage of NP⁺ cells (Fig. 1Dii). These findings suggest that PKR exerts an anti-viral role during LCMV infection that is counteracted by the ExoN and anti-T1IFN activities of the virus NP.

Fig 1.

Multiplication of LCMV/WT and rLCMV/NP(D382A) in A549 cells. (A) WT or PKR-KO A549 cells (biological triplicate) were infected with LCMV/WT (MOI 0.01) or rLCMV/NP(D382A) (MOI 0.05), and at the indicated hpi virus titers in TCS were determined. Values correspond to the mean ± standard deviation (SD). (B) Determination of NP gene expression levels by RT-qPCR. Total cellular RNA from samples in panel A was isolated using TRI Reagent, and NP gene expression was determined by RT-qPCR. GAPDH was used for normalization utilizing the fold change (2^−ΔΔCt) calculation and standardized using the 24-hpi value. Normalized data were plotted as mean ± SD (error bars) with a statistically significant set at P < 0.05. (C) Northern blot analysis. RNA samples from panel B were analyzed by Northern blot. Methylene blue (MB) staining was used to confirm similar transfer efficiency for all RNA samples. Statistically significant values: **P < 0.01, ***P < 0.001, ****P < 0.0001. (D) Effect of MOI on multiplication of rLCMV/NP(D382A). WT or PKR-KO A549 cells were seeded at 2 × 10⁴ cells /well in 96-well plates (two biological replicates, three technical replicates) and infected with LCMV WT or rLCMV/NP(D382A) at the indicated MOIs. At 48 hpi, cells were fixed and analyzed by immunofluorescence (IF). (Di) LCMV-infected cells were identified by IF using the rat monoclonal antibody VL4 to NP, followed by an anti-rat antibody conjugated to AlexaFluor 568. Individual images were obtained using Keyence BZ-X710 imaging system using 1-second exposure time with PlanApo_λ 10 × 0.45/4.00-mm objective lens. Files containing the labeled images were transferred to a laptop for processing the data using ImageJ. PowerPoint (v.2019) was used to compile and arrange the individual images. Adobe Illustrator was used to align the panels within the composite. (Dii) Quantification of A549 WT cells infected with rLCMV/NP(D382A). Cell nuclei were stained with 4',6-diamidino-2-phenylindole (DAPI), and the percentage of infected cells was determined based on the ratio (DAPI⁺+ NP⁺):DAPI⁺. ImageJ software was used to quantify the signal, and GraphPad Prism software was used to blot the data. (E and F) Effect of PKR complementation on rLCMV/NP(D382A) multiplication in PKR-KO A549 cells. (E) PKR-KO A549 cells were transfected with pCAGGS empty (pC-E) or pCAGGS expressing PKR-P2A-mCherry (pC-PKR) in suspension, then seeded as indicated per plate format, and on the next day infected with rLCMV/NP(D382A) at an MOI of 0.05. At 48 or 72 hpi, cells were fixed for imaging analysis. Numbers (%) of infected cells were normalized to vehicle-treated infected cells. To ensure accurate comparisons, the signals of the virus (NP) were normalized to the pC-E conditions, which exhibited the highest signal intensity. Conversely, PKR was normalized to PKR transfected cells, as determined by the mCherry signal. This normalization approach was adopted due to the absence of the mCherry signal in both the WT and pC-E conditions. Quantification of colocalization of NP and PKR signals as a function of Pearson’s coefficient, Spearman’s rank coefficient, and area overlap. These measures were obtained from running BIOP JaCoP script (43) using ImageJ software. A control test was done using DAPI images, which were duplicated; Look-Up Table (LUT) values of Region of Interest (ROI) were changed to red and green, then merged, and analyzed using BIOP JaCoP to detect colocalization. The correlation showed 100% agreement. Then the images were split. One image was translated by transformation to variable degrees, re-merged, and re-analyzed using the software to assess the correlation. Correlation levels decreased, depending on the degree of transformation. Duplicate samples were overlayed. (F) PKR-KO A549 cells were transfected with pC-E or pC-PKR in suspension, then seeded at 1 × 10⁶ cells/well into an M6 well plate and on the next day were infected with rLCMV/NP(D382A) at an MOI of 0.05. At 72 hpi, single-cell suspensions were prepared using Accutase, and after two washes with fluorescence activated cell sorting (FACS) buffer, cells were fixed with 4% paraformaldehyde (PFA) for 20 min, permeabilized, then probed with primary and secondary antibodies, followed by two washes after each antibody, re-suspended in FACS buffer, and analyzed using ZE5 analyzer (Bio-Rad). Quantification was done using FlowJo v.10.9, then followed by statistical analysis using GraphPad Prism. Two-way analysis of variance with Šidák correction for multiple comparisons was used. ns, not significant.

To assess whether the enhanced multiplication of rLCMV/NP(D382A) in PKR-KO A549 cells was mainly due to the absence of PKR, we transfected PKR-KO A549 cells with pCAGGS PKR-P2A-mCherrry (pC-PKR), a plasmid expressing PKR tagged with mCherry and containing the self-cleaving peptide (P2A) sequence (44) between the PKR and mCherry open reading frames. This allowed us to use mCherry expression as a surrogate of PKR expression in transfected cells. As a control, we used empty pCAGGS plasmid (pC-E). At 24 h post-transfection, cells were infected with rLCMV/NP(D382A), and at 48 and 72 hpi, infected cells were identified by staining with an NP-specific antibody. We observed a reduced level of rLCMV/NP(D382A) infection at 48 and 72 hpi in A549 PKR-KO cells expressing pC-PKR (mCherry+) compared to cells transfected with the control pC-E plasmid (Fig. 1E), which correlated with greatly reduced colocalization of NP (LCMV infection) and mCherry (PKR expression) as determined by Pearson’s coefficient, Spearman’s rank coefficient, and area overlap measures obtained from running BIOP JaCoP script (43) using ImageJ software. We further validated these findings by flow cytometry (Fig. 1F).

We also examined the role of other interferon (IFN)-induced genes including RNase L and MAVS during LCMV infection and found that MAVS, but not RNAse L, also contributed to the restricted multiplication of rLCMV/NP(D3282A) (data not shown).

Effect of LCMV infection on PKR activation

Activation of PKR requires its autophosphorylation upon binding to viral dsRNA and pPKR negatively regulates translation via phosphorylation of eIF2a. We found that rLCMV/NP(D382A), but not LCMV/WT, activated PKR as determined by the detection of increased levels of pPKR (Fig. 2A) and peIF2α (Fig. 2B). We used infection with Sindbis virus (SINV) as a positive control, as SINV infection activates the PKR pathway (31, 45). Quantification of the immunoblots (IBs) showed that levels of pPKR were fourfold higher in rLCMV/NP(D382A) than LCMV/WT-infected cells, both normalized to levels of pPKR detected in mock-infected control cells (Fig. 2Aii). Using total protein staining to normalize the signal in the IBs, we observed a higher level of pPKR and significantly increased levels of peIF2α in cells infected with rLCMV/NP(D382A), but not in cells infected with LCMV/WT (Fig. 2Bi and Bii).

Fig 2.

(A) PKR activation in LCMV-infected cells. (Ai) WT or PKR-KO A549 cells were infected with LCMV WT (MOI 0.01), rLCMV/NP(D382A) (MOI 0.05), or SINV (MOI 3); at 48 hpi, cell lysates were prepared for Western blot using the indicated antibodies. Samples from SINV-infected cells containing eightfold lower amount of protein, compared to the other lysates, were used as a control (one-eighth). Levels of GAPDH were used for the normalization of samples. Membranes were probed with anti-PKR, anti-pPKR, or anti-GAPDH (primary), anti-GP and horseshoe peroxidase (secondary) antibodies. (Aii) Bands of IB membranes were quantified, and the value of the ratio of pPKR/PKR was plotted. (Bi) WT and PKR-KO A549 cells were infected with LCMV WT (MOI 0.01) or rLCMV/NP(D382) (MOI 0.05). At 48 hpi, cell lysates were prepared for Western blot analysis. IB membrane was probed with anti-PKR, anti-pPKR, or anti-phospho-eIF2α (primary) and the corresponding HRP-conjugated secondary antibodies. The arrow refers to the eIF2α band. (Bii) pPKR:total protein signal ratio. IB signals were quantified using Image Lab software (BioRad). Total protein values were used to normalize the values, and normalized results were plotted. (C) Interferon-dependent PKR-induced expression. (Ci) WT and MAVS-KO A549 cells were seeded in a six-well plate at a density of 1 × 10⁶ cells/well and infected with LCMV WT (MOI 0.01) or rLCMV/NP(D382A) (MOI 0.05). At 48 hpi, cell lysates were prepared for Western blot analysis using anti-PKR and LCMV-GP antibodies. (Cii) IB signals were quantified using Image Lab and normalized to the total protein signal.

PKR is an interferon-stimulated gene (ISG), and LCMV with mutations affecting NP’s ExoN activity (e.g., D382A) has been shown to induce much higher levels of T1IFN than LCMV/WT in infected cells, a process mediated by the cytosolic PRR RIG-I and subsequent activation of the MAVS adaptor and downstream activation of IRF3, a transcriptional factor that promotes induction of IFN-β expression (46). To assess whether the observed higher levels of total PKR protein in rLCMV/NP(D382A)-infected A549 cells was driven by the T1IFN pathway, we compared levels of PKR protein in WT and MAVS-KO A549 cells infected with LCMV/WT or rLCMV/NP(D382A) (Fig. 2C). Levels of PKR were higher in rLCMV/NP(D382A) than LCMV/WT-infected WT A549 cells, whereas both LCMV/WT and rLCMV/NP(D382A)-infected MAVS-KO A549 cells exhibited similar levels of total PKR protein. Consistent with previous results (Fig. 1), levels of LCMV GP2, as determined by Western blot, were slightly higher in MAVS-KO compared to WT-infected A549 cells, whereas infection with rLCMV/NP(D382A) resulted in detectable levels of GP2 expression in MAVS-KO but not in WT A549 cells. These findings supported that T1IFN played a role in driving the expression of total PKR protein in rLCMV/NP(D328A)-ected cells.

We next investigated whether the extremely low increased levels of pPKR in LCMV/WT-infected cells reflected LCMV’s ability to either escape sensing by PKR or actively interfere with the activation of PKR. For this, we infected WT A549 cells with LCMV/WT (MOI 0.5) for 24 h and then superinfected them with SINV (MOI 3). At 24 hpi with SINV, we prepared whole-cell lysates and analyzed them by Western blot (Fig. 3). SINV-mediated activation of PKR, as determined by increased levels of pPKR, was not significantly affected by LCMV infection, indicating that LCMV does not actively inhibit the activation of PKR. SINV infection did not affect LCMV multiplication in A549 cells based on levels of GP2 expression determined by Western blot (Fig. 3A and B) and numbers of LCMV NP⁺ cells determined by immunofluorescence (IF) (Fig. 3C).

Fig 3.

Effect of LCMV infection on PKR activation. (A) WT A549 cells were infected with LCMV/WT (MOI 0.5) for 24 h and then superinfected with SINV (MOI 3). At 24 hpi with SINV, whole-cell lysates were prepared for Western blot analysis using antibodies to PKR, pPKR, or LCMV-GP. (B) The IB signals were quantified using Image Lab and normalized to total protein signal. (C) WT A549 cells were infected with LCMV/WT (MOI 0.5) for 24 h and then superinfected with SINV (MOI 3). At 24 hpi with SINV, cells were fixed for IF analysis using antibodies against LCMV-NP. Images were taken at ×20 magnification with Keyence BZ-X710.

Role of dsRNA on PKR activation in rLCMV/NP(D382A)-infected cells

We next investigated whether increased levels of dsRNA correlated with increased levels of pPKR observed in rLCMV/NP(D382A) compared to LCMV/WT-infected cells. For this, we infected A549 WT cells with LCMV/WT or rLCMV/NP(D382A) for 24 h, or SINV as a positive control, and used the anti-dsRNA 9D5 antibody to detect dsRNA by IF. Both LCMV/WT- and rLCMV/NP(D382A)-infected cells, as well as mock-infected control cells, had undetectable levels of dsRNA, whereas dsRNA was readily detected in SINV-infected cells (Fig. 4A). We obtained similar results in IFN-deficient Vero E6 cells (Fig. 4B). Levels of dsRNA remained undetectable in Vero E6 cells that had been persistently infected with LCMV for 15 days, but SINV infection of LCMV persistently infected cells resulted in high levels of dsRNA detected by IF (Fig. 4B). Unexpectedly, in contrast to published findings (40), we did not detect dsRNA in cells infected with the Candid#1 live-attenuated strain of the New World Junin virus (JUNV). We obtained similar results using the anti-dsRNA J2 antibody (data not shown). These findings indicated that under our experimental conditions, infection with either WT or NP(D382A) mutant LCMV does not result in levels of dsRNA that can be detected with the validated 9D5 antibody to dsRNA (47, 48).

Fig 4.

Detection of dsRNA. (A) A549 cells were seeded at 1.5 × 10⁴ cells/well in a 96-well plate and were infected with LCMV/WT or rLCMV/NP(D382A) at an MOI of 0.5. At 24 hpi, cells were fixed and analyzed by IF using 9D5 anti-dsRNA antibody. Infection with SINV (MOI 3) was used as a positive control. (B) IFN-deficient Vero E6 cells were infected with LCMV/WT (MOI 0.01), rLCMV/NP(D382A) (MOI 0.05), Candid#1 live-attenuated vaccine strain of JUNV (MOI 0.5), and SINV (MOI 3). At 24 hpi cells were fixed and analyzed by IF as in panel A. Images were taken at ×20 magnification and zoomed in digitally twice. (C) Effect of LCMV persistence on the production of dsRNA. Vero E6 (left) or BHK21 (right) cells were infected with LCMV WT at an MOI of 0.01 in a six-well plate, then expanded to T25 and passaged every 3–4 days along parenteral mock-infected cells. On day 15, cells were seeded into a 96-well plate at a density of 2 × 10⁴ cells/well and incubated for 24 h, then superinfected with SINV (MOI 3). At 24 hpi with SINV, cells were fixed and processed for IF as in panel A. Images were taken using Keyence BZ-X710.

Assessing the contribution of T1IFN to PKR activation in rLCMV/NP(D382A)-infected cells

We have shown that rLCMV/NP(D382A), but not LCMV/WT, can trigger a robust T1IFN response (25). The highly enhanced multiplication of rLCMV/NP(D382A) in PKR-KO compared to WT A549 cells could reflect differences in induction of T1IFN expression between WT and PKR-KO A549 cells following infection with rLCMV/NP(D382A). To examine this possibility, we infected WT or PKR-KO A549 cells with LCMV/WT (MOI 0.01) or rLCMV/NP(D382A) (MOI 0.05) and at 72 h pi collected TCS samples that we used to treat IFN-deficient Vero E6 cells for 24 h, followed by their infection (MOI 0.1) with vesicular stomatitis virus (VSV). At 24 h pi with VSV we fixed the cells and stained them with crystal violet to assess VSV induced cytopathic effect (CPE) (Fig. 5A). Vero E6 cells treated with TCS from either WT or PKR-KO A549 cells infected with rLCMV/NP(D382A) were protected against VSV-induced CPE. As expected, Vero E6 cells treated with TCS from WT or PKR-KO A549 cells infected with LCMV/WT were fully susceptible to VSV induced CPE. Results of quantitative analysis of expression of the ISGs MX1 and ISG15 in Vero E6 cells treated with TCS from A549 cells infected with LCMV/WT or rLCMV/NP(D382A) (Fig. 5B) were consistent with the results of the T1IFN bioassay (Fig. 7A). ISG15 and MX1 genes showed a significant upregulation, ~4- and ~500-folds respectively, in Vero E6 cells treated with TCS from A549 cells infected with NP(D382A), but not LCMV/WT, compared to treatment with TCS from mock-infected A549 cells.

Fig 5.

T1IFN response in PKR-KO A549 cells infected with LCMV. (A) WT and PKR-KO A549 cells were seeded at 1 × 10⁵ cells/well in 24 well plate, and infected on the next day with LCMV/WT (MOI 0.01) or rLCMV/NP(D382A) (MOI 0.05). At 24 h pi, TCS were collected and used to treat Vero E6 cells seeded at 1 × 10⁵ cells/well in a 24-well plate. After 24-h exposure to TCS, Vero E6 cells were infected with VSV at an MOI of 0.1; at 24 hpi with VSV, cells were fixed and stained with crystal violet for 10 min and imaged with BioSpot machine. (B) TCS collected from 48 hpi of A549 WT-infected with LCMV/WT or rLCMV/NP(D382A) was used to treat Vero E6 cells seeded at 1 × 10⁵ cells/well in a 24-well plate. After 24-h exposure to TCS, cells were lysed with TRI Reagent and RNA was collected and pretreated with ezDNAse for 5 min at 37°C, then 10 ng/sample (three technical replicates) for RT-qPCR (two steps: reverse transcription [RT] with random hexamers, followed by quantitative PCR with specific primers). Sybr green was used for RT-qPCR.

Effect of pharmacological inhibition of PKR activation on LCMV multiplication

To assess the effect of PKR inhibition on LCMV multiplication, we examined the effect of the PKR inhibitor C16, and its inactive control C22 (49), on LCMV multiplication. Dose-response assays indicated that C16, but not C22, exerted a dose-dependent inhibitory effect on LCMV multiplication (Fig. 6A). C16 exhibited an EC₅₀ = 0.177 µM and CC₅₀ >20 µM, while C22 exhibited an EC₅₀ >20 µM and CC₅₀ >20 µM (Fig. 6B). Treatment (3 µM) with C16, but not with C22, strongly inhibited propagation of LCMV in A540 cells following infection at low (0.01) MOI (Fig. 6C). Treatment with C16 (3 µM) caused >5 logs reduction in peak titers of rLCMV/green fluorescent protein (GFP) in WT A549 cells (Fig. 6D). A much lower (~ 2 logs) reduction in peak titers of rLCMV/GFP was observed in PKR-KO A549 cells treated with C16, which may reflect C16 off-target effects that affected PKR-independent pathways and factors (50).

Fig 6.

Effect of PKR inhibition on LCMV multiplication. (A) C22 and C16 Dose-response. A549 cells were seeded at 2 × 10⁴ cells/well into a 96-well plate, infected with LCMV/WT (MOI of 0.05) and treated with C16 and C22 at the indicated concentrations. At 96 hpi, cells were fixed, and the numbers of infected cells were determined by IF using a rat mAb to NP. Numbers (%) of infected cells were normalized to vehicle control (VC) infected cells. Results show the percentage of infected WT vs PKR-KO A549 cells (four replicates). (B) C16 and C22 EC₅₀, CC₅₀, and selectivity index (SI). (C) Effect of C16 and C22 on LCMV propagation. A549 cells were seeded at 4 × 10⁴ cells/well into a 96-well plate, infected with LCMV (MOI of 0.05), and treated (3 µM) with either CC16, C22, or VC. At 96 hpi, cells were fixed and stained with DAPI. IF analysis was determined using Keyence BZ-X710. Images were taken at ×10 magnification. (D) Effect of C16 and C22 on the production of virus infectious progeny. A549 cells were seeded at 2 × 10⁵ cells/well in an M12-well plate, infected with LCMV/WT (MOI of 0.01), and treated (3 µM) with C16 or C22 compounds, or with VC. At 72 hpi, TCS was collected, and virus titers were determined by focus-forming assay using Vero E6 cells in a 96-well plate format. The repeated measures analysis of variance with mixed effect analysis and the Dunnet correction for multiple comparisons were used. (E) Time of addition assay. A549 cells were seeded into a 96-well plate at a density of 2 × 10⁴/well. Next day, cells were infected with the single-cycle infectious rLCMV∆GPC/ZsG (MOI = 0.5) and treated (3 µM) with C16 or C22 compounds, or with VC, starting 2 h before (−2 h) or after (+2 h) infection. The LCMV cell entry inhibitor F3406 (5 µM) was used as a control. At 48 hpi, ZsG⁺ cells were assessed using Cytation 5 reader. Values were normalized to vehicle control-treated and vehicle control-infected cells. (F) Budding assay. HEK293T cells were seeded at a density of on poly-l-lysine-coated wells in a M12-well format. Next day, cells were transfected with either pC.LASV-Z-GLuc or pC.LASV-Z-G2A-GLuc (mutant control) or pCAGGS-Empty(pC-E). At 5 h post-transfection, cells were washed three times and fed with fresh medium containing the indicated drugs and concentrations. At 48 h post-transfection, both TCSs were collected and whole-cell lysis (WCL) was prepared. GLuc activity was determined in TCS and WCL using SteadyGlo Luciferase Pierce: Gaussia Luciferase Glow assay kit utilizing Cytation5 reader. Raw data signal was normalized and then plotted using GraphPad Prism software (v.10). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

To gain insights about the mechanism by which the PKR kinase inhibitor C16 disrupted the LCMV life cycle, we used C16, and its negative control C22, in cell-based assays for different steps of the LCMV life cycle. To distinguish between an effect of C16 on a cell entry or post-entry step of LCMV, we conducted a time of addition assay using a single cycle infectious recombinant LCMV expressing GFP (rLCMV∆GPC/GFP) (51, 52). C16 inhibited multiplication of rLCMV∆GPC/GFP, whether applied 2 h before (−2 hpi) or 2 h after (+2 hpi) infection (Fig. 6E). In contrast, regardless of the time of its addition, C22 did not inhibit rLCMV∆GPC/GFP. As predicted addition at −2 hpi, but not at +2 hpi, of the LCMV entry inhibitor F3406 (51, 52), resulted in inhibition of rLCMV∆GPC/GFP. This observation indicated that C16 inhibited a post cell entry step of the LCMV life cycle. To assess the effect of C16 on mammarenavirus budding, a process driven by the viral matrix protein Z (53), we examined the effect of C16 on the budding activity of the LCMV matrix Z protein using a published cell-based assay where levels of Gaussia luciferase (GLuc) activity serve as an accurate surrogate of Z expression levels (54). C16, but not C22, inhibited Z budding activity (Fig. 6F).

Effect of pharmacological inhibition of PKR on multiplication of other mammarenaviruses

To determine whether our findings could be also extended to other mammarenaviruses, we examined whether multiplication of the New World JUNV and Old World LASV mammarenaviruses was also affected by pharmacological inhibition of PKR activation. For this, we infected A549 WT cells with r3JUNV-GFP (Candid#1 strain) (55) and treated infected cells with C16 (3 µM) or C22 (3 µM) compounds. At the indicated hpi, TCSs were collected, and viral titers were determined. Treatment with compound C16, but not with compound C22, caused three log reductions in virus titers, at 48 and 72 hpi, compared to vehicle control-treated cells (Fig. 7A). Similarly, treatment with C16, but not with C22, inhibited propagation of LASV (Fig. 7B). As with LCMV (Fig. 1), genetic ablation of PKR resulted in enhanced propagation of LASV (Fig. 7C).

Fig 7.

Contribution of PKR to JUNV and LASV multiplication in cultured cells. (A) Treatment with compound C16 inhibits multiplication of JUNV. A549 cells were seeded into an M24 well plate (1.5 × 105 cells/well) and next day (18 h), they were infected with r3JUNV-GFP (Candid#1 strain) (55) at an MOI of 0.05. After 90 min of adsorption (0.2 mL/well), the inoculum was removed; cells were washed once; and fresh medium (0.5 mL/well) containing C16 (3 µM) or C22 (3 µM) compounds, or VC, was added to the cells. At the indicated time points, TCS samples were collected and virus titers were determined by focus-forming unit assay (FFUA) on Vero E6 cells. (B) The PKR inhibitor compound C16 exhibits a dose-dependent inhibitory effect on multiplication of LASV. A549 cells were infected with r3LASV (MOI 0.001). After 1 h of viral adsorption, compounds C16 and C22 or RBV were added at a range of concentrations. At 72 h post-infection, cell culture supernatants were collected and Gluc activity was quantified. Data were normalized by assigning the value of 100 to vehicle control (VC)-treated and r3LASV-infected cells. Cells were then fixed in 10% formalin and imaged for GFP expression using an EVOS M5000 imaging system. (C) Genetic ablation of PKR results in enhanced r3LASV propagation. Parental and PKR KO A549 cells were infected with r3LASV-GFP/Gluc (MOI 0.001). At indicated time points post-infection, TCS samples were collected, and cells were fixed. GFP+ cells, corresponding to LASV-infected cells, were visualized by epifluorescence using an EVOS M5000 imaging (Ci). TCS samples were assayed for their Gluc activity (Cii). Two-way analysis of variance with Šidák correction for multiple comparisons was implemented for statistical analysis. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

DISCUSSION

The PKR pathway is one of the major arms of cell innate immunity response to viral infections. Accordingly, viruses have evolved different strategies to counteract the PKR pathway. In this work, we have documented that LCMV/WT infection does not result in robust PKR activation, as determined by its autophosphorylation. This finding is consistent with those reported by others showing that at early times of infection and up to 48 hpi, LCMV/WT-infected cells showed negligible activation of PKR (37), as well as results documented with other mammarenaviruses (41). Interestingly, LCMV multiplication was increased in PKR-KO cells, reflected in consistently 5- to 10-fold higher virus titers in PKR-KO than in WT A549 cells. It seems unlikely that LCMV actively interferes with PKR activation as A549-LCMV-infected cells superinfected with SINV exhibited robust PKR activation. In contrast to LCMV/WT, infection of A549 cells with rLCMV/NP(D382A) resulted in robust PKR activation, which correlated with a severely restricted multiplication of rLCMV/NP(D382A) that was alleviated in PKR-KO A549 cells. PKR reconstitution studies using transfection of PKR-KO A549 cells with a PKR-expressing plasmid showed that multiplication of rLCMV/NP(D382) was restricted in PKR-KO A549 cells upon reconstitution of PKR expression, further confirming that PKR can exert an anti-LCMV activity. PKR was found to be involved in apoptosis, the formation of stress granules (56, 57), and IFN-induced cellular necrosis (58). However, neither total protein levels nor numbers of apoptotic cells were significantly affected in cells infected with rLCMV/NP(D382A) despite activation of eIF2α, a finding that warrants further studies.

PKR is activated by conformational change and autophosphorylation upon binding of dsRNA generated during viral genome replication. The 9D5 antibody has been reported to detect dsRNA species ≥40 bp (47, 48, 59) with a motif A₂N₉A₃N₉A₂ (59), which is present in both GPC (802–826) and L (763–787, 1,007–1,031, 2,970–2,994, 3,711–3,735, and 5,974–5,998) LCMV genes. However, our repeated attempts to detect dsRNA in A549 cells infected with LCMV/WT or rLCMV/NP(D382A) using the 9D5 antibody were unsuccessful, whereas we readily detected dsRNA in cells infected with SINV. Unexpectedly and in contrast with published findings (40), the 9D5 antibody was also unable to detect dsRNA in cells infected with JUNV. Nevertheless, our findings are consistent with earlier observations documenting the detection of dsRNA by the 9D5 antibody in cells infected with dsRNA and positive, but not negative, stranded RNA viruses (60). It is possible that virus-specific dsRNA species generated in LCMV-infected cells are <40 bp in length either because of a fast binding of NP to the newly formed RNA (61) or the lack of immediate supercoiling of RNA (62), which might prevent the formation of the structural binding site for the 9D5 antibody. Alternatively, it might be possible that the 9D5 antibody is targeting a specific domain that is common in dsRNA and +ssRNA supercoiled structures with sizes ≥40 bp (62). The 9D5 antibody has been shown to have an affinity for the peptide motif SIGNAYSMFYDG (63), not present in LCMV proteins, and therefore it cannot be ruled out that the 9D5 antibody binds to a protein target expressed only under specific situations that were not recreated under our experimental conditions.

Our results with PKR-deficient cells supported an expected anti-viral activity of PKR, but intriguingly pharmacological inhibition of PKR activation using compound C16 uncovered a pro-viral activity of PKR. C16, aka imoxin, is a specific PKR inhibitor that has been validated in biochemical (49) and cell-based (64) assays, as well as different animal models of disease including type two diabetes (65), cancer (66), neurodegeneration (67), hypertension (68), rheumatoid arthritis (69), and neuroinflammation (70). C16 has been shown to affect also PKR-independent biochemical intracellular transduction mechanisms (71), and the activity of cyclin-dependent kinases CDK2 and CDK5 (72), which could have contributed to the C16 anti-viral activity against LCMV, LASV, and JUNV. However, we have documented that siRNA-mediated knock-down of the CDK2 activator cyclin A2 protein resulted only in 50% reduction in levels of LCMV multiplication (73), which cannot account for the over five log reductions in production of LCMV infectious progeny caused by C16 treatment we have document in the present work. C16 has been extensively characterized for its inhibitory effect on autophosphorylation of PKR (49). In addition, comprehensive kinomics studies showed that C16 has high specificity for PKR with a Gini score of 0.44 PKR (74). Moreover, other proposed kinase targets of C16, including JNK, MKKs, and GSK3β, are downstream PKR pathway, which may reflect an upstream targeting pf PKR (75–77). Importantly, the C16 closely related compound C22 does not inhibit PKR autophosphorylation (49), which was associated with its lack of anti-mammarenavirus activity, providing additional evidence supporting inhibition of autophosphorylation of PKR as the main factor contributing to the C16 anti-mammarenaviral activity.

PKR may exert functions in LCMV-infected cells that do not require its activation. Thus, PKR has been shown to activate IKK by protein-protein interactions, leading to NF-kB activation and the induction of the T1IFN pathway (78). PKR has also been implicated in the inhibition of gelsolin-mediated regulation of actin dynamics and cytoskeletal cellular functions contributing to innate immunity including restricted virus cell entry, activities that do not appear to require PKR activation via autophosphorylation (79). The actin filament severing activity of gelsolin is inhibited by the oligomeric molecular chaperone CCT (80), and inhibition of the CCT activity resulted in restricted multiplication of LCMV (81), suggesting a connection between PKR, the host cell cytoskeletal activity, and mammarenavirus multiplication.

Similar to findings reported for infectious pancreatic necrosis virus (82) and SARS-CoV-2 (83) treatment with the PKR inhibitor C16 resulted in strong inhibition of LCMV multiplication in WT A549, supporting the role of PKR in LCMV multiplication and uncovering PKR activation as a druggable target for the development of anti-viral drugs against human pathogenic mammarenaviruses, including LCMV, for which evidence indicates it is a neglected human pathogen of clinical significance (4–8) and a serious risk to immunocompromised individuals (9, 10). C16 has shown tolerability and efficacy in different animal models of disease (65–70), supporting its future evaluation, beyond the scope of the present work, in mouse models of LCMV infection.

MATERIALS AND METHODS

Antibodies and compounds

We used the following antibodies at the indicated dilutions: rat mAb VL4 to LCMV NP (BE0106, Bio X cells) 1 in 1,000; mouse mAb 9D5 anti-dsRNA (3361, EMD Millipore) 1 in 2; mouse mAb rJ2 (MABE1134, Millipore/Sigma) 1 in 125; rabbit mAb anti-PKR (D7F7) (12297S, Cell Signaling Technology) 1 in 1,000; antibody (E120) anti-pPKR (phosphor T446) (ab32036, Abcam) 1 in 1,000; rabbit mAb anti-peIF2α (Ser51) (D9G8) XP (3398T, Cell Signaling); Phospho-eIF2α-S51 mouse mAb (AP0692, Abclonal) 1:1,000; mouse Anti-LCMV-GP hybridoma G204 1 in 500, goat anti-rabbit IgG (H + L) cross-adsorbed secondary antibody, Alexa Fluor 488 (A-11008, Thermo Fisher Scientific); goat anti-mouse secondary antibody, Alexa Fluor 568 (Thermo Fisher Scientific); PKR inhibitor C16 (MGF# 15323–1, Thermo Fisher Scientific), and PKR inhibitor negative control (C22) (MGF# 527455–10MG, Fisher Scientific).

Plasmids

pSB819-PKR-hum was a gift from Harmit Malik (Addgene plasmid # 20030; https://www.addgene.org/20030/; RRID: Addgene_20030) (84), and it was used to subclone the human PKR isoform 2 into pCAGGS plasmid preceded by mCherry and separated by P2A self-cleaving peptide sequence (44). The PKR was HA tagged at its C-terminus. HD In-Fusion kit (Takara 638909 In-Fusion HD Cloning Plus) was used for cloning.

Cell lines

A549 (Homo sapiens) [American Type Culture Collection (ATCC), CCL-185], A549 PKR-KO, A549 RNase L-knockout, A549 MAVS-knockout (85), BHK-21 (Mesocricetus auratus) (ATCC, CCL-10), and Vero E6 (Chlorocebus aethiops) (ATCC CRL-1586) cell lines were maintained in Dulbecco’s modified Eagle’s medium (Thermo Fisher Scientific) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 2-mM L-glutamine, 100 µg/mL of streptomycin, and 100 U/mL of penicillin. BHK-21 medium also contained 5% tryptose phosphate broth.

Viruses

The recombinant LCMV, clone 13 (Cl-13), has been described (86); rLCMV/NP(D382A) was rescued via reverse genetics as described (86). Mutation D382A in NP was generated with the In-Fusion cloning method (Takara 638909 In-Fusion HD Cloning Plus). rLCMV/NP(D382A) was grown in a BHK-21 cells, and its identity was confirmed by sequencing. rSINV-GFP was obtained from Matthew Daugherty (UCSD). The recombinant trisegmented (r3) LASV, strain Josiah, expressing GFP and Gluc was rescued as described for other r3MaAv, including LCMV (87), JUNV (88), and TCRV (89), using described LASV reverse genetic approaches (90, 91). Briefly, the mPolI-S LASV plasmid was modified to contain two restrictions sites (BsmBI and BbsI) in opposite orientation (mPolI-S LASV backbone). To generate the mPolI-S1 LASV, the open reading frame (ORF) of EGFP was amplified by PCR and cloned into the BsmBI restriction site of the mPolI-S LASV backbone. Then, the ORF of LASV GPC was amplified by PCR and cloned using BbsI. To generate the mPolI-S2 LASV, the ORF of LASV NP was amplified and cloned into the BsmBI restriction site of the mPolI-S LASV backbone. Then, the ORF of Gluc was amplified and cloned using BbsI. All the plasmids were entirely sequenced by Plasmidsaurus to ensure the absence of unwanted mutations. To generate r3LASV, BHK21 cells in six-well plates were transfected with mPolI-L LASV, mPol-S1 LASV, and mPol-S2 LASV, using LPF2000. At 48 h post-transfection, BHK21 cells were scaled up into a T75 flask, and 48 h after the presence of the r3LASV was verified by GFP expression. Cell culture supernatants were collected to confirm Gluc expression and to generate viral stocks. Stocks of r3LASV were generated and titrated in Vero E6 cells (90, 91). All the experiments with r3LASV were conducted in the biosafety level 4 laboratory at Texas Biomedical Research Institute.

Virus titration

Virus titers were determined by focus-forming assay using Vero E6 cells. Briefly, cells were seeded in 96-well plates at a density of 2 × 10⁴ cells/well and fixed at 20 hpi using 4% paraformaldehyde in phosphate-buffered saline (PBS) for 25 min, washed twice with PBS, permeabilized with 0.3% Triton X-100 containing 3% bovine serum albumin (BSA) in PBS, incubated with the rat mAb VL4 against LCMV NP (Bio X Cell) for 1 h at room temperature, and then incubated with an anti-rat IgG conjugated to Alexa Fluor 568 for 1 h; cells were washed 3× with PBS after each antibody probing step prior to imaging.

RT-qPCR

Cells were infected with LCMV (MOI 0.01) or rLCMV/NP(D382A) (MOI 0.05). Total cellular RNA was isolated using TRI reagent (TR 118) (MRC), and 1 µg was reverse-transcribed to cDNA using the SuperScript IV first-strand synthesis system (Thermo Fisher Scientific). Powerup SYBR (A25742, Life Technologies) was used to amplify LCMV NP and the housekeeping gene GAPDH using the following primers: NP forward (F): 5′ CAGAAATGTTGATGCTGGACTGC-3′ and NP reverse (R): 5′-CAGACCTTGGCTTGCTTTACACAG-3′ (92); GAPDH F: 5′-CATGAGAAGTATGACAACAGCC-3′ and GAPDH R: 5′-TGAGTCCTTCCACGATACC-3′; ISG15 F: 5′-CAGGACGACCTGTTCTGGC-3′ and ISG15 R: 5′-GATTCATGAACACGGTGCTCAGG-3′; and MX1 F: 5′- GCAGCTTCAGAAGGCCATGC-3′ and MX1 R: 5′-CCTTCAGGAACTTCCGCTTGTC-3′.

Western blotting

Cell monolayers were washed with ice-cold PBS, and whole-cell lysates were prepared in cytoplasmic lysis buffer (50-mM Tris HCl, 150-mM NaCl, 1% NP-40, 10% glycerol, and 2-mM EDTA) supplemented with Halt Protease and Phosphatase Inhibitor Cocktails (PI78442, Thermo Fisher Scientific). Lysates were clarified by centrifugation at 10,000 relative centrifugal force (RCF) for 20 min. Samples (24 µg) were denatured by heating for 5 min at 95°C and separated by Stain-Free SDS-PAGE gel (4568096, Bio-Rad). The gel was activated using 1-min setting (Bio-Rad Imager), imaged, then transferred to the low fluorescence polyvinylidene difluoride membrane (1704274, Bio-Rad) that was probed with mAbs to PKR (Cell Signaling Technology) or phospho-PKR-446 (AP1134, Abclonal). Following the incubation with a secondary HRP antibody, the bands were visualized with the chemiluminescent substrate (Thermo Fisher Scientific). The bands’ signals were quantified and analyzed using Image Lab V6 (Bio-Rad).

Northern blotting

RNA samples (5 µg) were fractionated by 2.2-M formaldehyde-agarose (1.2%) gel electrophoresis. The gel was washed once with warm each H₂O and 10-mM NaPO₄, and RNA transferred in 20× saline sodium citrate (SSC) (3-M sodium chloride, 0.3-M sodium citrate) to a Magnagraph membrane (NJTHYA0010, Osmonics MagnaGraph nylon) using the rapid downward transfer system (TurboBlotter). Membrane-bound RNA was cross-linked by exposure to UV light; the membrane was washed with MilliQ water and stained with methylene blue (MB) to reveal the 18S and 28S RNA plus RNA ladder. After photo documentation, 1% SDS solution was used to remove MB staining, and the membrane was hybridized using QuickHyb (#201220–12 Agilent) to a ³²P-labeled dsDNA NP. Hybridization was performed at 65°C overnight. The DNA probe was prepared according to the supplier’s protocol using a DecaPrime kit (Ambion). After overnight hybridization, the membrane was washed twice with 2× SSC–0.2% SDS at 65°C, followed by two washes with 0.2 × SSC–0.2% SDS at 65°C, and then exposed to an X-ray film.

Immunofluorescence

Individual images were obtained using the Keyence BZ-X710. Files containing the labeled images were transferred to a laptop for processing the data using ImageJ. PowerPoint (v.2019) was used to compile and arrange the individual images. Each image was imported individually and organized within the corresponding composite, adjusting the canvas size to ensure a cohesive layout. Adobe Illustrator was used to align the panels within the composite.

Dose-response and cell viability assay

CellTiter 96 AQueous One Solution (G3580, Promega) was used, according to manufacturer protocol, to quantify viable cells. DAPI and GFP signals were quantified after fixation of cells with 4% PFA using a Biotek H4 plate reader, and the Celigo Image Cytometer (Nexcelom). Data were normalized using vehicle-treated infected and mock-infected cells as the highest and lowest signals, respectively. Ribavirin (100 µM) was used as a control of a validated inhibitor of LCMV multiplication. Four replicates were used to quantify each sample.

Flow cytometry

Cells were washed and treated with Accutase (490007–741, VWR), transferred into 15-mL tubes, washed twice with FACS buffer (1× PBS, 2% FBS, 1-mM EDTA), then fixed with 4% PFA for 20 min, and transferred to a 96-well plate (round bottom). After permeabilization for 30 min (Permeabilization Buffer, 00–8333-56, eBioscience), cells were washed and reacted with the indicated antibodies. Cells were washed twice with permeabilization buffer followed by centrifugation for 3 min after each antibody probe. The primary antibody was probed for 1 h and the secondary antibody for 30 min. Cells were re-suspended in FACS buffer for analysis.

Dose-response inhibition of r3LASV

A549 cells (96-welll plate format, quadruplicate) were infected with r3LASV (MOI 0.001). After 1 h of viral adsorption, indicated concentrations of inhibitors were added. At 72 h post-infection, cell culture supernatants were collected, and Gluc activity was quantified according to the manufacturer protocol using Pierce Gaussia Luciferase Glow Assay Kit (16160, Thermo Fisher Scientific) and GloMax plate reader (Promega). Data were normalized to mock-treated, r3LASV-infected cells. Cells were then fixed in 10% formalin and imaged for GFP expression using an EVOS M5000 imaging system (Thermo Fisher Scientific).

Infection of parental and PKR KO A549 cells with r3LASV-GFP

Parental and PKR KO A549 cells (96-welll plate format, triplicates) were infected with r3LASV-GFP/Gluc (MOI 0.001). At indicated time points post-infection, cell culture supernatants were collected and Gluc activity was quantified according to the manufacturer protocol using Pierce Gaussia Luciferase Glow Assay Kit (16160, Thermo Fisher Scientific) and GloMax plate reader (Promega). Data were normalized by subtracting Gluc signal from mock-infected cells. Cells were then fixed in 10% formalin and imaged for GFP using an EVOS M5000 imaging system.

Statistical analysis

Differences that are statistically significant in virus growth and NP gene expression were determined using an unpaired t-test with Welch correction. Bands’ signals of IB were quantified using Image Lab (Bio-Rad). All statistical analyses were conducted using GraphPad Prism software v.9.5.1 (GraphPad). Flow cytometry analyses were performed using FlowJo v.10.9 software.

ETHICS APPROVAL

Protocols were approved by Texas Biomedical Research Institute Biosafety and Recombinant DNA committees (BSC21-013 and RDC21-013, respectively).