Human glioblastoma (GBM) cells (DBTRG-05MG, GBM cells) (Bioresource Collection and Research Center, BCRC-60380, Hsinchu, Taiwan) were maintained in RPMI-1640 medium (Gibco, #31800022, Waltham, MA, USA). Vero cells (American Type Culture Collection (ATCC), CCL-81, Manassas, VA, USA) and human neuroblastoma cells (IMR-32 cells, ATCC, CCL-127, Manassas, VA, USA) were cultured in MEM/EBSS medium (HyClone, #SH30008.02, Logan, UT, USA). Cell culture mediums were supplemented with 10% fetal bovine serum (FBS) (Gibco, A4766801, Waltham, MA, USA) and penicillin-streptomycin (P/S) (Gibco, #15140122, Waltham, MA, USA). PeV-A3 was from a virology group, Department of Microbiology, Kaohsiung Veterans General Hospital (27). PeV-A3 was propagated in Vero cells. The full genome sequence of PeV-A3 (strain VGHKKS-2007 accession #KM986843) has been characterized (28).

For cell staining, cells were plated at 1 × 10⁴ cells/well in 96-well plates and incubated overnight. The cells were then infected with PeV-A3 at multiplicity of infection (MOI) = 5. After 8- and 24-h infection, the cells were fixed with 1:1 Methanol/Ethanol solution (Sigma-Aldrich, St. Louis, MO, USA) for 15 min. Cells were incubated with blocking buffer [10% skim milk in phosphate-buffered saline (PBS)] and incubated with anti-PeV VP0 polyclonal antibody at 4°C overnight (LTK BioLaboratories, Taoyuan, Taiwan) (10, 29). Then, the IRDye^® 800CW goat anti-rabbit IgG secondary antibody was added into the cells (Li-COR, #926-32211, Lincoln, NE, USA), followed by 2-h incubation. Images of immunofluorescence assay were captured, and the fluorescence intensity of PeV-A3-infected cells were quantified using Odyssey image system (Li-COR, Lincoln, NE, USA).

Total RNA in cells or tissues were extracted first with 1 ml of TRIzol™ reagent (Invitrogen, #15596018, Waltham, MA, USA) and then with 1-bromo-3-chloropropane (BCP) (Sigma-Aldrich, B9673, St. Louis, MO, USA). The emulsion was centrifuged, and the aqueous phase was transferred to a fresh tube. The RNA was precipitated with 500 µl of isopropanol (Sigma-Aldrich, #278475, St. Louis, MO, USA) and pelleted by centrifugation at 12,000 × rpm at 4°C for 15 min. RNA pellet was washed with 1 ml of 75% ethanol and then centrifuged at 12,000 × rpm at 4°C for 10 min. After removal of supernatant, RNA pellet was air-dried for 10 min in a laminar flow hood and resuspended in nuclease-free water.

The RNA concentration was measured using a spectrophotometer (Eppendorf, Hamburg, Germany). For reverse transcriptase, cDNA was synthesized using 50 mM of random primer with 2 μg of total RNA in a total reaction volume of 20 μl using Superscript III reverse transcriptase method (Invitrogen, 18080085, Waltham, MA, USA). qPCR amplification involved 4 ng of cDNA in Fast SYBR™ Green Master Mix (Thermo Fisher, #4385612, Waltham, MA, USA) with 3 μM of primers (Genomics, New Taipei City, Taiwan) ( Table S1 ). qPCR was performed at 95°C for 3 min for 1 cycle and then 95°C for 20 s and 60°C for 30 s for 40 cycles. A final melting curve stage was performed from 60°C to 95°C using ABI StepOnePlus Real-Time PCR System (Life Technologies, Waltham, MA, USA).

The cDNA was synthesized from 2 μg of total RNA with 50 mM of random primer in a total reaction volume of 14 μl using RT² First Strand Kit (Qiagen, #330401, Hilden, Germany). The cDNA quality control of genomic DNA contamination (GDC), first-strand synthesis (RTC), and real-time PCR efficiency (PPC) were confirmed according to the manufacturer’s instruction. Finally, 25 μl of PCR component mixture in each well was used for the analysis of human cell death pathway using RT² Profiler PCR plate array (Qiagen, #330231, Hilden, Germany) in ABI StepOnePlus Real-Time PCR System.

GBM cells (1 × 10⁴ cells/well) were grown in 96-well plates overnight and then infected with PeV-A3 at MOI = 0.1, 1, and 10. After infection for 6 and 24 h, cells were incubated with 10 μl of WST-1 reagent (Roche-Sigma-Aldrich, #116448807001, St. Louis, MO, USA) for 2 h. The absorbance at 450 nm was monitored, and the reference wavelength was set at 620 nm using a microplate spectrophotometer (Epoch, BioTek, Winooski, VT, USA).

GBM and IMR-32 cells (5 × 10⁵ cells/well) were grown in six-well plates overnight and then infected with PeV-A3 at MOI = 1 for 24 h. After infection, cells were harvested for JC-10 analysis (Abcam, #ab112133, Cambridge, England). Cells were suspended in 500 µl of JC-10 solution and incubated at 37°C for 45 min, protected from the light. Then, cells with fluorescence were analyzed using flow cytometry (BD FACSCalibur Franklin Lakes, NJ, USA). Fluorescence-activated cell sorting (FACS) analysis displayed cell subsets, which were estimated with respect to populations selected on the basis of membrane potential change. Apoptotic cells lost the mitochondria inner membrane potential, and the monomeric form of JC-10 was released into the cytoplasm. Cells displayed red fluorescence loss and turned green. The FL1 channel of FACS was used for detecting green fluorescent monomeric signal in apoptotic cells, while the FL2 channel was used for the detecting orange fluorescent aggregated signal in healthy cells.

Cells and mouse brain tissue lysates were lysed in protein lysis buffer (2% sodium dodecyl sulfate (SDS) and 50 mM of Tris-HCl, pH = 7.5) containing protease and phosphatase inhibitors (Thermo Fisher, #78442, Waltham, MA, USA). The lysates were homogenously sonicated, separated in 10% or 12% SDS–polyacrylamide gel electrophoresis (PAGE), and finally transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, IPVH00010, Burlington, MA, USA). PVDF membranes were treated with 10% milk blocking buffer and incubated with primary antibody in 1:2,000 dilution in 5% milk/TBST at 4°C overnight, Antibodies used included anti-PeV VP0 (10), anti-caspase 3 (Cell Signaling, #9662, Danvers, MA, USA), anti-caspase 7 (GeneTex, GTX123679, Hsinchu City, Taiwan), anti-PARP (Cell Signaling, #9532, Danvers, MA, USA), anti-caspase 8 (Cell Signaling, #9746, Danvers, MA, USA), anti-caspase 9 (Cell Signaling, #9508, Danvers, MA, USA), anti-FASLG (GeneTex, GTX31191, Irvine, CA, USA), anti-CD40 (GeneTex, GTX101447, Irvine, CA, USA), anti-LC3 A/B (Cell Signaling, #4108, Danvers, MA, USA), anti-p62 (Cell Signaling, #16177, Danvers, MA, USA), anti-beclin 1 (Cell Signaling, #3495, Danvers, MA, USA), anti-phospho-NF-κB p65 (Ser536) (Cell Signaling, #3033, Danvers, MA, USA), anti-NLRP3 (Cell Signaling, #13158, Danvers, MA, USA), anti-IL1β (Cell Signaling, #12242, Danvers, MA, USA), anti-RIPK1 (R&D, MAB3585, Minneapolis, MN, USA), anti-RIPK3 (R&D, MAB7604, Minneapolis, MN, USA), anti-GSDMD (GeneTex, GTX135366, Hsinchu City, Taiwan), anti-β-actin (ProteinTech, #66009-1-Ig, Rosemont, IL, USA), and anti-GAPDH (ProteinTech, #60004-1-Ig, Rosemont, IL, USA). Secondary antibodies conjugated with horseradish peroxidase (HRP)-anti mouse (LEADGENE, 20112, Tainan, Taiwan) and HRP-anti rabbit (LEADGENE, #20202, Tainan, Taiwan) were incubated at room temperature 1 h. The images were acquired from biospectrum imaging system (UVP Analytik Jena, Jena, Germany) by using WesternBright ECL kit (Advansta, K-12045-D20, Menlo Park, CA, USA).

C57BL/6 mice were purchased from the National Laboratory Animal Center (NARLabs, Taipei, Taiwan). This study was approved by the Institutional Animal Care and Use Committee or Panel (IACUC) in Kaohsiung Veterans General Hospital, Taiwan (IACUC Approval No. VGHKS-2018-2021-A016). According to the reported method of intracranial injection of virus (30), 3-day-old C57BL/6 neonatal mice were intracranially inoculated with PeV-A3 (5 × 10⁵ pfu/mouse) into the cerebral lateral ventricles using Hamilton 1700 Series Syringes (Hamilton, 80000, Reno, NV, USA). PBS inoculation on mice was the control. Survival rate, clinical scores, and body weight were then monitored daily after inoculation. Clinical scores 0–5 indicated the severity level of disease symptoms: 0, healthy; 1, reduced mobility; 2, reduced body weight; 3, forelimb or hindlimb weakness; 4, forelimb or hindlimb paralysis; and 5, death.

Paraffin-embedded brain tissues from PeV-A3 mice and PBS mice were cut and mounted on coated slides. The sections were de-paraffinized in xylene and rehydrated through graded alcohol solutions, and antigen retrieval in sodium citrate buffer at 100°C for 20 min was performed. The immunohistochemistry (IHC) procedures of Novolink Polymer Detection Systems (Leia #RE7150-K, Wetzlar, Germany) were modified. Brain sections were incubated with Protein Block buffer for blocking 15 min and then incubated overnight at 4°C with rabbit polyclonal antibody anti-PeV-VP0 and rabbit IgG isotype control (Thermo Fisher, #02-6102, Waltham, MA, USA). After being rinsed with PBS (Biological Industries #11-223-1M Cromwell, CT, USA), the slides were incubated with secondary antibody goat anti-rabbit IgG Alexa Flour 488 (1:500 dilution, Thermo Fisher, #A32731, Waltham, MA, USA) for 1 h at room temperature. The slides were mounted after DAPI staining (1:5,000 dilution, Thermo Fisher #62248, Waltham, MA, USA) for 30 min. The hippocampus and cortex regions were observed under an inverted fluorescence microscope (Olympus, Tokyo, Japan).

Statistical analysis was performed using Prism 9 (GraphPad, San Diego, CA, USA). Student’s t-test or Mann–Whitney U test were used for quantitative parameters. Survival rates were analyzed using Kaplan–Meier curve. Results were expressed as mean ± SD. p-Value of <0.05 was considered to be significant.

GBM cells and neuroblastoma cells were used in research on neurotropic virus infections such as ZIKA virus and enterovirus 71 (31–34). Therefore, this study used human GBM cells (DBTRG-05MG) and human neuroblastoma cells (IMR-32) to decipher the neuropathic effects of PeV-A3. These two cell lines were infected with PeV-A3 at MOI = 5 for 8 and 24 h, followed by immunofluorescence analysis conducted with anti-PeV VP0 antibody. Green fluorescence indicated VP0 expression in PeV-A3-infected cells ( Figures 1A, B , top panels). The fluorescence level was further quantified. Comparison with mock controls and results at 8 h post infection (h.p.i) revealed an increased VP0 expression level detected at 24 h.p.i. ( Figures 1A, B , bottom panels). The viral genome in GBM cells infected with PeV-A3 at MOI = 0.1, 1, 5, and 10 was measured. A gradual viral load-dependent increase in PeV-A3 VP1 mRNA level was observed ( Figure 1C ). Taken together, these results showed susceptibility of GBM and neuroblastoma cells to PeV-A3 infection.

During viral infection, cellular type I and type III IFN antiviral responses are evoked to restrict virus replication (35). In the CNS, type I IFN protects neurons from prion infection (36). To clarify the PeV-A3-induced antiviral response in GBM, we measured the IFN and associated antiviral protein expression after infection for 24 and 48 h. RT-qPCR results showed significant induction of the determined genes, which were IFN-β, type III IFN (IFNλ-1, IFNλ-2/3), TANK-binding kinase 1 (TBK1), interferon regulatory factor 3 (IRF3), melanoma differentiation-associated protein 5 (MDA5), C-X-C motif chemokine ligand 10 (CXCL10), retinoic acid-inducible gene I (RIG-I), and mitochondrial antiviral-signaling protein (MAVS) transcripts ( Figure 2A ). These findings suggested that PeV-A3 activated cellular antiviral responses in GBM cells.

We further investigated type I IFN activity against PeV-A3 in GBM. Cells were infected with PeV-A3 at MOI = 1 for 48 h after being pretreated or posttreated with type I IFN 100 IU/ml for 24 h. RT-qPCR results showed reduction of viral replication along with induction of type I and type III IFN and ISGs (MDA-5 and 2′-5′-Oligoadenylate Synthetase 2, OAS2) ( Figure 2B ). PeV-A3-activated antiviral responses and exogenous type I IFN treatment-suppressed PeV-A3 suggest a typical virus–host interaction in GBM cells.

Although type I IFN response was elicited by PeV-A3 in GBM cells, the host antiviral activity failed to efficiently restrict this virus, with both GBM and IMR-32 cells showing cytopathic effect (CPE) morphology upon PeV-A3 infection at MOI = 1 for 24 h ( Figure 3A ). The WST-1 assay of GBM cells proliferation was impaired by PeV-A3, revealing a decreased cell growth in a viral load-dependent manner ( Figure 3B ). This study further investigated whether PeV-A3 infection caused mitochondrial membrane potential change using cationic, lipophilic JC-10 dye staining in GBM and IMR-32 cells. Compared with 12.6% and 6.38% of mock-infected GBM and IMR-32 cells, respectively showing JC-10 background signaling, 36% and 24.3% of PeV-A3-infected GBM and IMR-32 cells respectively exhibited mitochondrial membrane potential loss, becoming apoptotic cells ( Figure 3C ). Collectively, these findings revealed a pathogenic effect of PeV-A3 in neuronal cell types.

To understand the expression pattern of cell death-related genes in PeV-A3-infected GBM cells, the human RT² profiler PCR array was conducted ( Table S2 ). Results showed that PeV-A3-infected GBM cells expressed death protein genes at least twofold upregulated over mock controls including myelin-associated glycoprotein (MAG), Fas ligand (FASLG), BCL2-related protein A1 (BCL2A1), ATPase H+ Transporting V1 Subunit G2 (ATP6V1G2), insulin-like growth factor 1 (IGF1), Forkhead box I1 (FOXl1), and tumor necrosis factor receptor superfamily, member 11b (TNFSR11B) ( Figures S1A, B ). The expression level of these genes was further confirmed using RT-qPCR analysis, which showed a significant induction, compared with mock controls ( Figure 3D and Table S3 ). Hence, results obtained evidenced that PeV-A3 infection triggered cell death-associated gene transcript in GBM cells.

Programmed cell death is an important machinery in host’s defense to limit the replication of invading intracellular pathogens. Types of viral-mediated cell death included apoptosis, programmed necrosis, autophagy, and pyroptosis; the events promote the formation of effective long-term host immune inflammation and innate responses (23, 37). After detection of PeV-A3-mediated cell death, death signaling was analyzed to understand the cellular activity in response to PeV-A3 infection. Immunoblotting of signaling proteins was conducted in GBM cells infected with PeV-A3 for 6, 24, and 48 h. The protein level was also quantified and statistically analyzed. The time dependently increased PeV-A3 VP0 protein level revealed the spread of viral infection status in GBM cells. We detected significant increase in signaling protein level underlying apoptosis and pro-apoptosis pathways, such as activated caspase-3, caspase-7, PARP, caspases 8, caspase 9, FASLG, and CD40, after PeV-A3 infection in a time-dependent manner ( Figure 4A left panel and Figure 4B ).

Autophagy is a process that consigns cytoplasmic components to lysosomes for degradation. Autophagy is usually associated with cell death. Autophagy-associated cell death depends largely on the physical conditions (37, 38). Autophagy-induced NFκB activity was also reported, which could be targeted by viral infection (39, 40). This study found significant accumulation of autophagic factors LC3 and p62 and a slight increase in the beclin-1 level of GBM cells during PeV-A3 infection periods. The induced phosphorylated NF-κB p65 Ser536 activity was also detected. The above results suggested that PeV-A3 infection-associated autophagy influx is possibly involved in PeV-A3-triggered cell death ( Figure 4A , right panel, and Figure 4B ). Since PeV-A3-caused neuronal cell death was revealed ( Figure 3C ), we also analyzed the apoptosis and pro-apoptosis and autophagy signaling responses in IMR-32 cells with PeV-A3 infection. Compared with that of mock controls, the increased protein level of activated caspase 3, CD40, FASLG, LC3, and p62 was detected ( Figure S2 ).

Pyroptosis is a type of cell death accompanied with inflammation and is essential for limiting pathogen infections. Infections stimulate pyroptosis activation and formation of NLR family pyrin domain containing 3 (NLRP-3) inflammasome, resulting in processing and activation of caspase 1 for expression of inflammatory cytokines interleukin 1β (IL-1β) and IL-18 (41, 42). In the signal pathways of NLRP3-mediated pyroptosis, IL-1β and receptor-interacting protein (RIP) family of serine-threonine kinases, RIPK1 and RIPK3, Gasdermin-D (GSDMD) were detected in PeV-A3-infected GBM cells ( Figure 4A , right panel, and Figure 4B ). This study detected activation of death signaling of apoptosis, autophagy, and pyroptosis as early as 6 h.p.i. in GBM cells, suggesting a fast pathogenic response promoted upon PeV-A3 replication. Together, these results suggested that PeV-A3 induced multiple cell death signaling pathways.

In view of PeV-A3 infection-triggered signaling activation of autophagy and pyroptosis, this study then accessed autophagy- and pyroptosis-mediated inflammatory cytokine expression responses in GBM and IMR-32 cells. These cells were infected with PeV-A3 for 6, 24, and 48 h, followed by the induction of TNF-α, IL-6, IL-1β, and IL-18 measured using RT-qPCR. Results showed similar transcript patterns in GBM ( Figure 5A ) and IMR-32 cells ( Figure 5B ), where inflammatory cytokines, TNF-α, IL-6, IL-1β, and IL-18 transcripts, were significantly induced at 6 h.p.i. and then declined at 24 and 48 h.p.i. These data suggested that, consistent with early activation of death signaling by PeV-A3 shown in Figure 6 , an early response of inflammatory transcript in GBM cells was also detected. The reduction of these inflammatory genes might be due to impaired cells’ viability and low activity of host gene transcription. Collectively, the above findings revealed the neuropathogenic mechanism of PeV-A3 in vitro.

The above findings evidenced that PeV-A3 infection inhibited cell growth and activated cell death pathways including apoptosis, autophagy, and pyroptosis in vitro. To investigate the PeV-A3-induced neuropathic disease in vivo, we established a PeV-A3 infection animal model by using 3-day-old C57BL/6 mice. Neonate mice were intracranially inoculated with PeV-A3 at 5 × 10⁵ pfu/mouse, while littermates that received PBS injection served as mock controls. After inoculation, the survival rate was monitored from day 0 to day 11. Results showed approximately 20% mortality in PeV-A3-infected mice ( Figure 6A ).

After infection for 11 days, the mice were sacrificed, and total RNA was extracted from the brain tissues to analyze the PeV-A3 VP1 mRNA expression. RT-qPCR results showed significant expression of PeV-A3 RNA in mouse brain as compared with PBS control ( Figure 6B ). The PeV-VP0 protein expression in each mouse brain was also detected by immunoblotting, and the protein level was quantified ( Figure 6C ). The PeV-A3 viral RNA and protein expression indicated replication of PeV-A3 in the neonatal mouse brain.

According to the severity of physical condition of the mice (health, reduced mobility, body weight loss, limb weakness, paralysis, and death; Figure 6D , top panel), clinical symptoms observed from days 0 to 11 post injection were scored. Days 2, 6, and 11 showed progression of disease with increasing clinical scores in PeV-A3-infected mice ( Figure 6D , bottom panel). The images showed mice with limb paralysis or moribund ( Figure 6E ). These results suggested that PeV-A3 induced strong neurovirulence in C57BL/6 mice. Changes in daily body weight in surviving PeV-A3-infected mice were measured. The findings showed lower average body weight in PeV-A3-infected C57BL/6 neonatal mice at days 2, 6, and 11 after inoculation as compared with PBS control mice ( Figure 6F ). This infection animal model demonstrated susceptibility of neonatal C57BL/6 mice to PeV-A3 infection, which caused neuropathic symptoms and decreased survival rate.

PeV-A3 RNA and protein were detected in the brain extracts of infected mice. We further conducted IHC analysis to understand whether the hippocampus and cortex of brain regions were infected by PeV-A3. The result showed that the PeV-VP0 expression was detected in the hippocampus and cortex in the PeV-A3-infected mice, but not in the PBS control mice ( Figures 7A, B ). Quantification data showed a significant high level of PeV-VP0 expression in the hippocampus and cortex; compared with PBS control, nearly 40-fold and 70-fold protein level increases were detected in the hippocampus and cortex sections, respectively ( Figures 7C, D ). Taken together, these results suggest that PeV-A3 directly infected hippocampus and cortex regions of the neonatal mouse brain.

PeV-A3 caused neurological symptoms in neonatal mice. Mouse brain tissues were examined to understand whether the present in vitro findings of PeV-A3-triggered inflammation and whether cell death signaling activation also occurred in the infected mice. RT-qPCR results showed a higher transcript level of inflammatory cytokines (Tnfa, Il-6, Il-8, and Il-1b) and interferon signature [Ifna, Ifnb, Ifng, interferon regulatory factor 3 (Irf3), Irf7, Viprin, and Toll-like receptor 3 (Tlr3)] in PeV-A3-infected brain tissues compared with PBS controls ( Figure 8 ). Furthermore, the protein extract from each mouse brain tissue was employed to analyze death signaling; the quantified immunoblots demonstrated that brain tissues of PeV-A3-infected mice expressed a higher protein level of activated cleaved-caspase 3, cleaved-PARP, LC3, phospho-NF-κB p65, IL-1β cytokine, and RIPK1 than did PBS controls ( Figures 9A, B ). These results suggested that PeV-A3 infection induced inflammatory cytokine expression and activated apoptosis, autophagy, and pyroptosis signaling pathways in neonatal mice, which might be associated with the neurovirulence of PeV-A3.