NAD ⁺ Homeostasis Attenuates Japanese Encephalitis Virus Infection Progression

ABSTRACT

Nicotinamide adenine dinucleotide (NAD⁺) is a crucial molecule involved in numerous interconnected metabolic processes. Due to its implication in multiple viral infection responses, maintaining NAD⁺ homeostasis has become a promising target for host‐directed therapies. Japanese encephalitis virus (JEV) causes severe, fatal encephalitis with irreversible brain damage and long‐lasting neurological deficits in survivors. However, the potential interaction between JEV infection and NAD⁺ metabolism remains largely unclear. In this study, we found that JEV infection dysregulates NAD⁺ metabolism and the expression of its pathway enzyme genes in Type I interferon (IFN‐α/β) receptor‐deficient (A129) mice and human glioblastoma (T98G) cells. Specifically, JEV infection altered the expression of de novo/kynurenine pathway (IDO, KATII, KMO) and salvage pathway (NAMPT, NMNATs) NAD⁺ biosynthetic enzymes, as well as NAD⁺‐consuming enzymes (PARPs, SIRTs), culminating in a substantial decrease in NAD⁺ levels. Furthermore, NAD⁺ depletion and JEV production increased when salvage biosynthesis was restrained through NAMPT knockdown, but these effects were reversed by supplementing nicotinamide riboside (NR) in NAMPT knockdown T98G cells. Importantly, restoring NAD⁺ levels with NR supplementation as an anti‐JE strategy in A129 mice reduced JEV production and improved infection outcomes. In conclusion, this study demonstrates that JEV infection disrupts NAD⁺ metabolism, and restoring NAD⁺ levels inhibits JE progression. Therefore, maintaining NAD⁺ homeostasis and regulating its metabolic pathway could be a promising therapeutic approach for JE.

JEV infection in brain tissue triggers inflammation and stress in neuronal cells, leading to NAD⁺ depletion via the overactivation of PARPs and SIRTs. While compensatory biosynthetic pathways (de novo and salvage) were upregulated, the NAD⁺ level fell critically. In vitro NAMPT inhibition exacerbates this deficit and viral load. Conversely, rescuing NAD⁺ via NR administration restores the salvage pathway and reduces viral production, mitigates cellular stress and damage, and ultimately improves infection outcomes both in vivo and in vitro.

1. Introduction

Japanese encephalitis (JE) is one of the most serious and oftentimes deadly forms of viral encephalitis in Asia, caused by Japanese encephalitis virus (JEV) belonging to the genus Flavivirus [1]. The virus is commonly transmitted through the bite of infected Culex mosquitoes, particularly Culex tritaeniorhynchus. JE is endemic to the rural and agricultural areas of Asia and the Western Pacific. There are cyclical outbreaks every 2–15 years, with the disease most prevalent in densely populated rural areas and slums [2]. While children between the ages of 0–14 years are the most and the earliest affected, the elderly population has also been increasingly affected and has been associated with high morbidity and mortality rates [3]. The global annual reported clinical JE cases are approximately 68 000, with 20%–30% of patients facing the threat of death [4]. JE patients present with severe neurological manifestations such as fever, headache, vomiting, and encephalitis, and 30%–50% of survivors are left with cognitive and movement disorders, as well as other neuropsychiatric complications in the form of long‐term disability [5].

Even though vaccines have been available for several decades, there remains low coverage, and there are still seasonal surges in JE cases and hospitalizations during the rainy season [6]. The emergence of novel JEV strains with distinct genotypes is another significant concern regarding the protective efficacy of commercially available live‐attenuated vaccines against circulating variants [7, 8, 9]. This ongoing breach underscores the potential value of antiviral drugs in compensating for the current limitations of JE immunization strategies. The lack of specific antiviral treatments for JE creates an unmet need for these vulnerable populations in endemic regions [10, 11], highlighting the critical need to investigate the pathogenesis of JE and develop targeted antiviral treatments.

Viruses may exploit NAD⁺ pathways to either directly enhance replication and elude immune detection or indirectly enact NAD⁺ metabolism changes, which involve immune responses, DNA repair, energy metabolism, and cell stress, many of which rely on NAD⁺ as a co‐substrate [11]. Disrupted NAD⁺ metabolism has been associated with microcephaly caused by Zika virus [12] and pneumonia caused by SARS‐CoV‐2 in mice [13]. Brain infections leading to NAD⁺ depletion also cause the compensatory overexpression of KP genes for de novo NAD⁺ biosynthesis, resulting in the overproduction of neurotoxic excitatory metabolites, including 3‐hydroxykynurenine (3‐HK) and QUIN. The increased accumulation of these neurotoxic metabolites, along with NAD⁺ depletion, can worsen neuronal damage during encephalitis [14]. NAD⁺ depletion raises the nicotinamide mononucleotide (NMN)/NAD⁺ ratio, activating SARM1 (sterile α and Toll/IL‐1 receptor motif–containing 1), which triggers a cascade of events leading to axon degeneration [15, 16, 17, 18, 19, 20]. However, the involvement of NAD⁺ metabolism in the pathogenesis of JE remains entirely uncharacterized.

While most individuals infected with JEV remain asymptomatic or only develop mild febrile illness, a small but significant proportion progress to severe encephalitis, often resulting in permanent neurological sequelae [4, 5]. In the mouse model, adult wild‐type mice show resistance to JEV, experiencing milder symptoms than those usually seen in humans. Using immunocompromised mice, which are highly susceptible to JEV and reliably develop lethal encephalitis [21, 22], offers a more accurate model for severe disease in humans, especially in vulnerable groups like young children and the elderly, who often have weakened immune systems [3, 23].

In this study, we found that JEV can cause acute infection with typical encephalitis symptoms in A129 mice and disrupt NAD⁺ metabolic pathways in the brain. Specifically, infected mice showed dysregulations in the gene expression of key enzymes involved in the NAD metabolic pathways, along with NAD⁺ depletion. Importantly, maintaining NAD⁺ levels reduced JEV yield in T98G cells and A129 mice, leading to improved infection outcomes in A129 mice, which deepens our understanding of NAD⁺ metabolic disruptions and highlights the potential of NAD⁺ restoration as a therapeutic strategy for JE.

2. Materials and Methods

2.1. Virus and Cells

For this study, we used the genotype I (GI) JEV GS strain, which was isolated in 2016 from mosquito vectors around Pingliang City, Gansu Province, and is stored at Lanzhou Veterinary Research Institute (LVRI). The viral strain was grown in Aedes albopictus clone (C6/36) cells (BCCC100985, RRID: CVCL_Z230). C6/36 cells were cultured in Eagle's Minimum Essential Medium (EMEM) with 10% fetal bovine serum (FBS) (Cellmax), 100 units/mL of penicillin, and 100 μg/mL of streptomycin, and incubated at 28°C with 5% CO₂. Human glioblastoma (T98G) cells (BNCC388721, RRID: CVCL 0556) were grown in Dulbecco's Modified Eagle Medium (DMEM) with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin, all kept in a controlled incubator at 37°C and 5% CO₂. For viral titer quantification, we used Baby Hamster Kidney (BHK‐21) (R700‐01, RRID: CVCL_1914) cells in plaque assays. These BHK‐21 cells were cultured in DMEM with 5% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin, maintained in a 37°C, 5% CO₂ incubator.

2.2. In Vitro and In Vivo Infection Experiments

2.2.1. T98G Cell Infection Experiment

T98G cells were cultured overnight in 12‐well plates until they reached 70%–80% confluence. The cells were then infected with JEV at 1.0 multiplicity of infection (MOI). The experiment was repeated three times for both JEV‐infected and mock‐infected (control) groups. Total RNA was extracted at 48 h post‐infection (hpi) from all samples to analyze gene expression related to NAD⁺ metabolism and associated pathways.

2.2.2. Mouse Infection Study

Wild‐type, female BALB/c (211, RRID: MGI:2683685) and C57BL/6 (RRID: IMSR_JAX:001913), and both sexes of type I IFN receptor‐deficient (A129) mice aged 7–8 weeks were obtained from the laboratory animal center of LVRI. The BALB/c mice were randomly divided into five groups (n = 5/group) and infected with JEV. Two groups of mice received 10⁵ and 10⁶ plaque‐forming units (PFU) intraperitoneally (IP), while one group was given 10⁷ PFU via intravenous injection. To validate an additional infection route, one additional group (n = 5/group) of mice was injected through the subcutaneous route. A negative control group received an equal volume of MEM. Following ethical guidelines for the care and use of laboratory animals, the challenge dose was started at the lowest possible dose of 10⁵ PFU. If mortality occurred with the initial lower dose infection, higher doses were not administered to subsequent groups. C57BL/6 mice were grouped into four (n = 5/group) and injected with 10⁷ PFU of JEV in three different routes (IV, IP, and SC), and the remaining one group (negative control) was injected with MEM.

To determine the infectious dose of GI JEV in A129 mice, four groups (n = 4/group with random allocation) were inoculated with GI JEV at doses of 10⁶, 10⁴, 10², and 1 PFU intravenously (IV). A fifth group injected with MEM served as the control. Following infection, all mice were monitored for symptoms of JEV infection over 15 days. Mice were checked daily in the morning and evening, and body weight was recorded each day. Mice showing terminal illness (TE) symptoms—defined as more than 20% weight loss, cachexia, and loss of food and water‐seeking behavior in a labored breathing state with non‐responsiveness—were gently anesthetized with isoflurane and euthanized. Temporal profiling of JEV in various tissues and the expression of NAD⁺ metabolic pathway‐related genes in the brain were studied in A129 mice (n = 12) infected with a 100 PFU dose of JEV, with tissues collected from three mice daily from 2 to 5 days post‐infection (dpi).

2.3. Real‐Time Quantitative PCR (RT‐qPCR)

Brain, liver, kidney, heart, lung, spleen, and thymus tissue samples were collected from the mice and stored at −80°C. The samples were weighed, and total RNA was extracted using TRIzol reagent (TransGen Biotech; RNA Kit, Beijing, China), followed by cell lysis. First‐strand cDNA was synthesized from 1 μg of total RNA using the PrimeScript RT Reagent Kit with gDNA Eraser (TAKARA BIO INC). For RT‐qPCR, 1 μL of cDNA, equivalent to 50 ng of total RNA, was used. Viral RNA was amplified with specific primers for the NS3 gene of the GI JEV isolate (Table S1). Quantification was performed using a standard curve generated from serial dilutions of pcDNA3.1 myc‐his (V80020) plasmid containing the NS3 gene, based on the qPCR cycle threshold values from the dilutions.

The mRNA expression levels of selected genes from the NAD⁺ metabolic pathway were analyzed by RT‐qPCR, using specific primers (Table S1). Changes in expression were presented as fold changes, calculated as ($2^{-∆∆C_{\mathrm{T}}}$) using β‐actin and GAPDH as reference genes for mouse tissues and T98G cells, respectively.

2.4. Plaque and Median Tissue Culture Infectious Dose (TCID₅₀ ) Assay

The infectious viral titers were measured using plaque assays from brain tissue samples. After weighing, the tissue was homogenized in 1 mL of DMEM. The homogenate was then separated via slow‐rate centrifugation, and all samples underwent a series of dilutions with DMEM. BHK‐21 cells were cultured until reaching 90%–100% confluence in a 24‐well plate and inoculated with the diluted samples. The viral titers in each sample were calculated per gram of tissue to estimate the viral load.

To determine the infectious viral titer from the infected cell maintained medium, a TCID₅₀ assay was performed using BHK‐21 cells. Ten‐fold serial dilutions of stock viruses were prepared in DMEM, from which 100 μL each was transferred onto monolayers of BHK‐21 cells grown on 96‐well plates in DMEM supplemented with 2% FBS and 1% penicillin–streptomycin and incubated at 37°C in a 5% CO₂ incubator. Virus titers were calculated at 3 dpi and expressed as TCID₅₀/mL value.

2.5. Histopathological and Immunohistochemistry (IHC) Analysis

Brain, liver, and spleen tissues were collected from a mouse infected with 100 PFU of JEV and euthanized at 5 dpi for histopathology and IHC analysis. The tissues were fixed in 10% neutral buffered formalin, sectioned, and stained with hematoxylin and eosin (H&E) for histopathology. To assess viral distribution in the brain, liver, and spleen, IHC staining was performed using an antibody against JEV NS1 (82707‐13‐RR, RRID: SCR_008986). Brain slices from NR‐treated or untreated mice were immunostained with antibodies against NS1 (JEV) (Gene Tex: GTX644073), cleaved caspase‐3 (Cas‐3) (82707‐13‐RR, RRID: SCR_008986), and 4,6‐diamidine‐2‐phenylindole (DAPI).

2.6. Analysis of NAD Contents

The total NAD (NAD⁺ /NADH) and NAD⁺ levels in mouse brain tissue and T98G cells were measured using the NAD⁺/NADH detection kit (WST‐8 Method, Product No. S0175). NAD was measured from the supernatant of tissue and cell lysates following the protocols provided by the kit's manufacturer. Consistent with the manual's instructions, NADH was assessed from samples of tissues and cells after heating to 60°C for 30 min to inactivate NAD⁺, and the NADH value was then calculated.

2.7. NAMPT siRNA Interference

Overnight‐cultured T98G cells with 70%–80% confluence were transfected with NAMPT‐specific siRNA or a negative control (NC) siRNA using the jetPRIME transfection reagent. The siRNA sequences were as follows: NAMPT (330‐sense: 5′‐GCAUCUUCCAAUAGAAAUATT‐3′, antisense: 5′‐UAUUUCUAUUGGAAGAUGCTT‐3′; 1281‐sense: 5′‐GGGCCGAUUAUCUUUACAUTT‐3′, antisense: 5′‐AUGUAAAGAUAAUCGGCCCTT‐3′) and negative control siRNA (sense: 5′‐UUCUCCGAACGUGUCACGUTT‐3′, antisense: 5′‐ACGUGACACGUUCGGAGAATT‐3′). Knockdown efficiency was validated at 48 h post‐transfection by RT‐qPCR and western blotting, and transfection‐related cytotoxicity was assessed using the CCK‐8 assay. To evaluate the effect of NAMPT knockdown on JEV infection, cells were incubated for 24 h after transfection; three wells per group (NAMPT siRNA or NC) were then infected with JEV at an MOI of 1. At 48 hpi, RNA was extracted to measure viral RNA levels, and viral titers were determined from the infected cell culture‐maintained media.

2.8. Western Blotting

Cells were lysed with RIPA buffer and heated at 100°C for 10 min in loading buffer. The samples were then separated by SDS–PAGE using a 12% gel and transferred to nitrocellulose membranes via a wet transfer system. After blocking with 5% skim milk, the membrane was incubated overnight at 4°C with primary antibodies. The following primary antibodies were used to detect target host and viral proteins: mouse anti‐IDO1 (1:5000), mouse anti‐KMO (1:5000), mouse anti‐SIRT4 (1:20 000), rabbit anti‐PARP1 (1:2000), rabbit anti‐NAMPT (1:2000), (11776‐1‐AP, RRID: SCR_008986), rabbit anti‐JEV prM (1:2000) (GeneTex: GNT131833). Mouse anti‐β‐actin (1:1000; 66009‐1‐Ig, RRID: SCR_008986) was used as a loading control. HRP‐conjugated goat anti‐rabbit IgG (1:6000) (SA00001‐2, RRID: SCR_008986) and HRP‐conjugated goat anti‐mouse IgG (1:6000) (B900620, RRID: SCR_008986) served as secondary antibodies. Membranes were visualized using a Bio‐Rad imager with ECL substrate.

2.9. Nicotinamide Riboside (NR) Supplementation

2.9.1. For T98G Cells

Using the jetPRIME reagent, cells at 70%–80% confluence were transfected with NAMPT‐specific siRNA or NC. Sterile PBS dissolved NR, then further diluted to 0.1 mM in DMEM as described by Ruszkiewicz [24], was used to maintain the cells infected with JEV at an MOI of 1.0 for 48 h. Afterward, the cells were analyzed for viral load (RT‐qPCR) and NAD⁺ levels following the protocol outlined above.

2.9.2. In Mice

Eight‐week‐old A129 mice (n = 12) were randomly divided into two groups, with six mice in each. NR was dissolved in PBS, filtered through a 0.22 μm filter, and injected IP at a dose of 200 mg/kg in 0.2 mL for the treatment group, as described by Pang [12]. NR supplementation commenced 24 h before JEV infection and continued daily for three dpi, totaling four doses. The second group received the same volume of sterile PBS via the same route. All mice were infected intravenously with 100 TCID50 of JEV at day 0. Mice were monitored daily for symptoms and weight changes. Blood samples were collected on 2 and 4 dpi to measure viremia. On 4 dpi, all mice were sacrificed, and organs were collected for viral load analysis. The brain was specifically targeted for viral load measurement, histopathology, and immunostaining analysis.

2.10. Statistical Analysis

All analyses, including both descriptive and statistical components, were conducted using GraphPad Prism 8 software (RRID: SCR_002798). Unpaired Student's t‐test was used for two‐group analysis (e.g., infected vs. control, treated vs. untreated). For assessing viral RNA load in tissues over time, values from 3, 4, and 5 dpi were compared to those from 2 dpi. Gene expression profiles were measured on 2, 3, 4, and 5 dpi and compared to naive controls. Statistical significance was defined as p < 0.05.

3. Results

3.1. Susceptibility of Wild‐Type and IFN‐α/β Receptor‐Deficient Mice to Field Isolate GI JEV

Our study evaluated the susceptibility of wild‐type (BALB/c and C57BL/6) and IFN‐α/β receptor‐deficient (A129) mice to the GI JEV field isolate. Prior research showed that Type I and II IFN‐deficient (AG129) mice develop severe infections after being injected IP with the SA14‐14‐2 vaccine strain [21] and following IP or intradermal administration of the Indian clinical isolate P20778 [22]. To explore different routes and validate the model for JEV research, we administered various doses (10⁶, 10⁴, 10², and 1 PFU) of JEV intravenously (IV) to A129 mice (Figure 1A). Results indicated that wild‐type mice showed resistance to the field‐isolate GI JEV infection. Neither IP, IV, nor SC inoculation with doses up to 10⁷ PFU caused noticeable symptoms in wild‐type mice. Although no overt symptoms appeared, in BALB/c mice, a brief weight loss of 2%–4% occurred between 5 and 7 dpi in the group injected with 10⁷ PFU via IV (Figure 1B). RT‐qPCR analysis of viremia (5–7 dpi) and tissue viral loads (10 dpi) showed cycle threshold (CT) values above 30, indicating no significant detection. Additionally, agarose gel electrophoresis of qPCR products revealed no amplified bands (data not shown), confirming that JEV replication and spread were suppressed in wild‐type BALB/c mice. Similarly, the C57BL/6 mice also didn't show any symptoms, and their post‐infection continuous body weight gain was maintained (Figure 1C). However, in the A129 mouse model, severe, fatal infection developed even at the lowest dose (1 PFU). The viral dose impacted disease progression (Figure 1D): mice that received 10⁶ and 10⁴ PFU survived for 3 and 4 dpi, respectively (data not shown); those given 10² PFU survived up to 5 dpi; and the 1 PFU group had a 25% mortality rate (1 out of 4 mice) (Figure 1E). Infected mice displayed typical signs, including lethargy, decreased activity, behavioral changes, piloerection, hunched posture, discharge from both eyes, and paralysis of the hind limbs (Figure 1F).

FIGURE 1.

Field isolate GI JEV causes acute and deadly infections in A129 mice. 7–8 week old female BALB/c and A129 mice of both sexes were inoculated with GI JEV. (A) Schematic diagram of the A129 mice JEV challenge experiment, including inoculum doses, post‐infection monitoring, and key time points for euthanasia (humane endpoints). (B–D) Changes in body weights of BALB/c, C57BL/6, and A129 mice were recorded over 15 days. (E) Survival curves of A129 mice infected with 100 and 1 PFU, along with a control group, and of BALB/c mice infected with 107 PFU, all monitored over 15 dpi. (F) Main symptoms observed in JEV‐infected A129 mice.

3.2. Viral Distribution and Histopathology in the Infected A129 Mice

We assessed viral load in different tissues of infected A129 mice, which were collected at various times. To measure viral RNA levels (copy numbers of genes per 1 μL of cDNA), we used a regression equation based on a standard curve (Figure 2A). The results showed that the spread of the virus in the brain, lung, heart, liver, and kidneys continued to increase between 3 and 5 dpi compared to 2 dpi (Figure 2B–F). The spleen and thymus maintained persistently high levels of viral RNA (Figure 2G,H). The qPCR findings were validated by the band intensity of agarose gel electrophoresis (Figure 2I). The plaque assay also indicated a similar trend in infectious viral titers in brain homogenates, with a progressive increase from 2 to 5 dpi (Figure 2J).

FIGURE 2.

Dissemination of JEV in the brain and peripheral organs of A129 mice: GI JEV (100 PFU) was inoculated into A129 mice (n = 12) of both sexes aged 7–8 weeks. Three mice were sampled at each time point (2, 3, 4, and 5 dpi). (A) Standard curve generated using a 10‐fold serial dilution of the NS3 gene cloned into the pcDNA3.1 myc‐his plasmid. (B–H) Viral RNA load quantified in tissues (brain, thymus, lung, heart, liver, spleen, and kidney); data represent the mean ± standard deviation of log10 copies of viral RNA/μL of cDNA (equivalent to 50 ng of total tissue RNA), derived from three samples with duplicate measurements per sample. (I) Representative agarose gel electrophoresis image of the qPCR‐amplified NS3 gene of JEV. (J) Infectious viral load was measured in brain tissue sampled at 2, 3, 4, and 5 dpi using a plaque assay, with data showing the mean ± standard deviation of log10 plaque‐forming units (PFU)/g of tissue based on triplicate measurements. Statistical comparisons were made between results at 3, 4, and 5 dpi and those at 2 dpi using Student's t‐test. Significance thresholds were set at **p < 0.01, and ***p < 0.001. ns: not statistically significant (p > 0.05).

Histopathological examination revealed hyperchromic neurons, necrosis, and mild gliosis, along with a glial compensatory response in the brain. In the liver, there was steatosis, hepatocyte swelling, focal necrosis, and lymphocyte infiltration, indicating inflammation. The spleen showed necrosis, granulocyte infiltration mainly by neutrophils, and fibrin deposition, suggesting an active immune response, as shown in the Figure S1A–C. IHC analysis identified abundant viral antigens; the brain contained the highest levels, indicating significant viral replication in this vital organ (Figure S1D–F).

3.3. NAD ⁺ Metabolism Pathway Dysregulation

3.3.1. JEV Disrupts the Expression of Rate‐Limiting Genes in De Novo NAD⁺ Biosynthesis Within the Kynurenine Pathway (KP)

The KP is a crucial metabolic pathway that converts the essential amino acid tryptophan (Trp) into several bioactive compounds, including NAD⁺. These compounds play a key role in regulating the immune system and in neuron protection or degeneration (Figure 3A). The rate‐limiting enzymes controlling the KP are indoleamine 2,3‐dioxygenase (IDO), kynurenine aminotransferase II (KATII), and kynurenine 3‐monooxygenase (KMO), which facilitate the production of neuroactive metabolites like KYNA, 3‐HK, and QUIN [25, 26]. Neurotropic infections boost the production of the neurotoxic QUIN, leading to neuronal damage [27]. Although neuronal damage/encephalitis is the most common characteristic of JEV infection, its effects on KP neuroactive metabolites are still being studied. In this study, we explored how JEV infection affects the expression of key regulatory genes in the KP using the A129 mouse model and T98G cells.

FIGURE 3.

JEV infection upregulated KP rate‐limiting enzymes gene expression: (A) Schematic of tryptophan metabolism via the KP. Enzymes in red (IDO, KATII, and KMO) were tested for gene expression and found to be upregulated. (B–D) Gene expression levels of IDO, KATII, and KMO in brain tissues from 100 PFU JEV‐infected and mock‐infected A129 mice are shown as fold change. (E–G) Kynurenine pathway enzymes (IDO, KATII, KMO) gene expression in JEV‐infected and mock‐infected T98G cells at 48 hpi. (H) Immunoblot showing PARP1 and SIRT4 protein expression at different time points post‐infection. Data are expressed as the mean ± standard deviation of gene expression ($2^{-∆∆C_{\mathrm{T}}}$), with β‐actin and GAPDH serving as the reference genes for brain and T98G cell samples, respectively, from three samples in each group, with duplicate measures per sample. Statistical analysis employed student's t‐test. Significance levels: *p < 0.05, **p < 0.01, ***p < 0.001. 3‐HAO, 3‐hydroxyanthranilate 3,4‐dioxygenase; ACMS, 2‐amino‐3‐carboxymuconate semialdehyde; DAMPs, damage‐associated molecular patterns; FMD, formamidase; IDO, indoleamine 2,3‐dioxygenase; KATs, kynurenine aminotransferases; KMO, kynurenine 3‐monooxygenase; KYN, kynurenine; KYNA, kynurenic acid; N‐FKYU, N‐formylkynurenine; PAMDs, pathogen‐associated molecular patterns; PRRs, pattern recognition receptors; QUIN, quinolinic acid; TDO, tryptophan 2,3‐dioxygenase.

Using the A129 mouse model, we observed an increase in IDO expression (2.89‐ and 3.13‐fold on 3 and 4 dpi, respectively; Figure 3B). KATII showed the most notable rise, with 14.6, 33.7, 16.5, and 9.3‐fold increases on 2, 3, 4, and 5 dpi, respectively (Figure 3C). KMO was also moderately elevated to 2.8, 6.4, 2.9, and 1.4 from 2 to 5 dpi, respectively (Figure 3D). To verify these results, we examined infected T98G cells. We similarly observed upregulation changes in these genes compared to mock‐infected controls. IDO transcripts increased 4.97‐fold, matching peak levels observed in A129 mice (Figure 3E). KATII transcripts showed notable upregulation (Figure 3F), whereas KMO levels remained unchanged, with no statistically significant difference (Figure 3G). To further validate whether the transcriptional mRNA dysregulation translated into changes at the protein level, we conducted western blotting analysis for selected genes. Consistent with our mRNA findings, both IDO and KMO proteins exhibited overexpression (Figure 4H). This confirmed that the observed shifts in gene expression are indeed reflected in the abundance of these critical enzymes.

FIGURE 4.

NAD⁺ dependent poly ADP‐ribose polymerases (PARPs) and Sirtuin (SIRT) gene expression changes induced by JEV infection: (A) A schematic of NAD⁺ breakdown in the infected brain, where NAD⁺ functions as a co‐substrate for enzymes activated during neuroinflammation (CD38/CD157 and SARM1), DNA damage (PARPs), and cellular stress (SIRTs). (B) PARPs gene expression levels in the brain tissues of JEV‐infected A129 mice compared to mock‐infected controls. (C) Relative PARPs gene expression in JEV‐infected T98G cells versus mock‐infected controls at 48 hpi. (D) Relative expression of SIRTs genes in brain tissues of JEV‐infected A129 mice compared to controls. (E) Relative SIRTs gene expression in JEV‐infected T98G cells versus controls at 48 hpi. (F) Immunoblot showing PARP1 and SIRT4 protein expression at different time points post‐infection. Data are expressed as the mean ± standard deviation of gene expression ($2^{-∆∆C_{\mathrm{T}}}$), with β‐actin and GAPDH serving as the reference genes for brain and T98G cell samples, respectively, from three samples in each group, with duplicate measures per sample. Statistical comparisons used student's t‐test, with significance thresholds of *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. ADPR, adenosine diphosphate ribose; cADPR, cyclic adenosine diphosphate ribose; CD157, cluster of differentiation 157 (also known as BST‐1, bone marrow stromal cell antigen 1); CD38, cluster of differentiation 38; NAM, nicotinamide; NMA, nicotinic acid mononucleotide; Quin, quinolinic acid; SARM1, sterile alpha and TIR motif‐containing 1.

3.3.2. Dysregulation of Poly (ADP‐Ribose) Polymerases (PARPs) Gene Expression

Poly (ADP‐ribose) polymerase (PARP) enzymes use NAD⁺ as a substrate to ADP‐ribosylate proteins, attaching ADP‐ribose covalently. This biochemical process is essential for DNA damage repair and immune response regulation. As shown in Figure 4A, inflammation, DNA damage, metabolic stress, and disruption of cellular homeostasis can lead to NAD⁺ depletion due to hyperactivation of NAD⁺‐dependent enzymes like PARPs, SIRTs, CD38, and others [28, 29]. Our study indicates that JEV infection causes upregulation of most PARP family genes both in vivo and in vitro. In the A129 mouse brain, PARP1 and PARP10 were slightly upregulated (significant at 3 and 4 dpi), while PARP3, PARP9, PARP12, and PARP14 showed sustained elevation from 2 to 5 dpi. PARP14 exhibited the most significant increase (4.7 to 19.4‐fold), and PARP6 was notably downregulated, with the most significant decrease observed at 4 dpi (Figure 4B).

In T98G cells, all the tested PARPs were upregulated. PARP14 showed the greatest increase, confirming the in vivo result. PARP6 expression exhibited a 2.5‐fold upregulation in vitro, in striking contrast to its downregulated expression pattern observed in vivo. This indicates a response influenced by cellular microenvironments and stress signaling pathways, which may differ depending on the cell type (Figure 4C).

3.3.3. Disruption of Sirtuins (SIRTs) Gene Expression

Sirtuins (SIRTs), a family of NAD⁺‐dependent deacylases, play a crucial role in regulating cellular metabolic processes. However, activating them, especially during stress responses, can worsen NAD⁺ depletion, which has significant implications for cellular metabolism and energy balance [30, 31] (Figure 4A). To assess the impact of JEV infection on genes regulating NAD⁺‐dependent deacylases, we examined the expression levels of selected SIRTs in A129 mice and T98G cells. In infected mice, SIRT1 was upregulated (3.1‐ to 5.3‐fold), while SIRT3 remained elevated (> 2‐fold at 3–5 dpi), and SIRT4 consistently increased by more than 1.5‐fold. In contrast, SIRT5 was substantially suppressed (0.6–0.33‐fold) between 2 and 5 dpi, which might be due to a distinct regulatory mechanism of this cell type during infection (Figure 4D). In vitro, SIRT1 showed a slight increase, whereas SIRT3 exhibited a more moderate decline compared to in vivo results. SIRT4 showed a non‐significant positive trend, and SIRT5 remained unchanged and stable (Figure 4E).

The expression of two selected NAD⁺‐dependent enzymes, PARP1 and SIRT4, was further analyzed by western blot to validate whether their upregulation in mRNA expression was reflected at the protein level. Our result showed a noticeable overexpression of both PARP1 and SIRT4 proteins (Figure 4F), aligning with their mRNA expression in infected T98G cells, which suggests a heightened activation of NAD⁺ dependent pathways due to JEV infection.

3.3.4. Dysregulation of NAD⁺ Salvage Pathway Biosynthetic Enzyme Genes

Two primary regulators of NAD⁺ biosynthesis through the salvage pathway are nicotinamide mononucleotide adenylyltransferases (NMNATs) and nicotinamide phosphoribosyl transferase (NAMPT) [32, 33]. NAD‐dependent PARPs and SIRTs, along with KP rate‐limiting enzymes, showed upregulation during JEV infection, suggesting that JEV infection might cause further disruption in the salvage pathway enzymes and decrease NAD⁺ levels. Therefore, we analyzed the expression of NAMPT, NMNAT1, and NMNAT2 in JEV‐infected A129 mice and T98G cells.

The mRNA levels of NAMPT, a key enzyme in the salvage pathway, increased in infected brains (Figure 5A). In contrast, NMNAT1 and NMNAT2 expression levels were reduced (Figure 5B,C). Consistent with in vivo findings, T98G cells exhibited a 3.5‐fold increase in NAMPT expression (p < 0.01). However, the results differ from in vivo; NMNAT1 showed a slight rise, and NMNAT2 increased by 1.5‐fold (p < 0.05) (Figure 5D). The upregulation of NAMPT and NAD⁺‐dependent enzymes underscores the impact of JEV infection on NAD⁺ metabolism. The consistent increase of NAMPT in both in vivo and in vitro models could be the activation of salvage biosynthesis of NAD⁺ as a compensatory mechanism to maintain cellular NAD⁺ homeostasis.

FIGURE 5.

NAD⁺ salvage pathway biosynthetic enzyme gene expression changes induced by JEV infection: (A–C) Gene expression levels of NAMPT, NMNAT1, and NMNAT2 in brain tissues from 100 PFU JEV‐infected A129 mice compared to mock‐infected controls. (D) Gene expression levels of NAMPT, NMNAT1, and NMNAT2 in JEV‐infected T98G cells at 48 hpi compared to mock‐infected controls. For panels (A–D), data are expressed as the mean ± standard deviation of gene expression ($2^{-∆∆C_{\mathrm{T}}}$), with β‐actin and GAPDH serving as the reference genes for brain and T98G cell samples, respectively, based on three samples in each group, with duplicate measures per sample. (E) Total NAD (NAD⁺ and NADH) and NAD⁺ content in brain tissues of JEV‐infected and mock‐infected A129 mice. (F) Total NAD (NAD⁺ and NADH) and NAD⁺ content in JEV‐infected and mock‐infected T98G cells at 48 hpi. For panels (E) and (F), data represent the mean ± standard deviation of NAD_total in pmol/mg tissue and pmol/10⁶ cells from triplicate samples, with duplicate technical measurements per sample. Statistical comparisons were performed using student's t‐test. Significance thresholds were set at *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

3.3.5. JEV Infection Led to NAD⁺ Depletion in Both A129 Mice and T98G Cells

We measured NAD⁺ and NAD_total (NAD⁺ + NADH) levels in brain tissue and T98G cells to determine whether gene expression dysregulation in the pathway affects NAD⁺ levels. As shown in Figure 5E, infected brain tissue displays lower NAD⁺ levels (p < 0.05), and a similar trend is observed for NAD_total, although it is not statistically significant. Similarly, the infected T98G cell line shows relatively higher depletion in both NAD⁺ and NAD_total levels (p < 0.01 and < 0.001) (Figure 5F). This NAD⁺ metabolism disruption in JEV infection supports the conclusion that transcriptional dysregulation of NAD⁺ metabolism pathway‐associated genes impairs NAD⁺ homeostasis.

3.4. Restoration of NAD ⁺ Attenuates JEV Yield in NAMPT Knockdown T98G Cells

It is increasingly recognized that dysregulation of NAD⁺ metabolism contributes to disease progression, including various viral infections [11]. Infection with viruses like Zika, coronaviruses including SARS‐CoV‐2, and the mouse hepatitis virus (MHV) disrupts the NAD⁺ metabolome by activating PARP and other NAD⁺‐dependent enzymes [11, 12, 13, 34]. NAMPT is crucial for NAD⁺ biosynthesis and inhibits virus production [31, 32, 35, 36]. To investigate the role of NAMPT in JEV infection, we first assessed NAMPT expression using Western blotting in JEV‐infected T98G cells. Results showed that JEV infection increased NAMPT protein levels (Figure 6A), suggesting a potential role of NAMPT in regulating JEV infection. Subsequently, we used siRNA to knockdown NAMPT in T98G cells. This siRNA‐mediated knockdown reduced both mRNA and protein levels of NAMPT (Figure 6B,C) without a significant effect on T98G cell viability (Figure 6D). Interestingly, NAMPT knockdown led to increased JEV RNA levels (Figure 6E) and viral titers (Figure 6F) compared to the siRNA‐NC control, indicating that NAMPT acts as a restriction factor for JEV production. To further determine whether this effect is directly related to NAD⁺ homeostasis, we supplemented NAD⁺ precursor NR, which is converted into nicotinamide mononucleotide (NMN) via the NAD⁺ salvage pathway, ultimately generating NAD⁺ [37] in NAMPT knockdown T98G cells. This supplementation lowered JEV RNA levels (Figure 6G) and viral titers (Figure 6H), while also restoring NAD⁺ levels (Figure 6I) compared to infected groups without NR treatment. These findings demonstrate that NAD⁺ homeostasis is essential for suppressing JEV infection in T98G cells.

FIGURE 6.

Restoration of NAD⁺ restricts JEV yield in knockdown T98G cells. (A) Immunoblot showing NAMPT protein expression at different time points post‐infection. (B) NAMPT mRNA levels in mock‐ and JEV‐infected T98G cells treated with targeted siRNA or siRNA‐NC. (C) Effect of NAMPT knockdown on protein expression shown by western blot. (D) Cell viability after transfection with NAMPT siRNA and a negative control siRNA‐NC, measured by CCK‐8 assay. (E) Relative viral RNA abundance in NAMPT knockdown and siRNA‐NC‐treated cells. (F) Viral titers of the cell supernatant from NAMPT knockdown and siRNA‐NC‐treated T98G cells collected at 48 hpi. (G) Intracellular viral RNA abundance in NAMPT knockdown and siRNA‐NC‐treated cells with NR supplementation. (H) Viral titers of the cell supernatant in NAMPT knockdown and siRNA‐NC‐treated cells with NR supplementation. (I) Intracellular NAD⁺ content. RT‐qPCR data are presented as mean ± standard deviation of NAMPT mRNA gene expression or viral RNA levels based on the NS3 gene in $2^{-∆∆C_{\mathrm{T}}}$, using GAPDH as the reference gene; samples from three biological replicates collected at 48 hpi, with duplicate technical measurements per sample. Statistical comparisons were performed using Student's t‐test. Significance thresholds *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ns: not statistically significant (p > 0.05).

3.5. Restoration of NAD ⁺ Alleviates JE Progression and Reduces Viral Load in A129 Mice

Building on the clinical studies that suggest NAD⁺ restoration can improve outcomes in viral infections [13], we further evaluated NAD⁺ restoration as an anti‐JE strategy in A129 mice. We selected NR for NAD⁺ supplementation due to its ability to effectively cross the blood–brain barrier and increase cerebral NAD⁺ levels [12, 38, 39, 40]. Mice were administered NR (200 mg/kg/day) via IP injection from 1 day before infection until three dpi (Figure 7A). The results showed that NR supplementation reduced weight loss and delayed the onset of disease symptoms (Figure 7B,C). By 4 dpi, all control mice (6/6) displayed symptoms, with five becoming severely ill. In contrast, only two of six NR‐treated mice showed symptoms, and only one became severely sick. Furthermore, NR treatment significantly lowered viremia (Figure 7D) and viral load in tissues (Figure 7E). H&E staining of brain tissue revealed that the control group had more extensive hemorrhages across various brain regions, whereas the NR‐treated group showed a reduced extent of hemorrhage (Figure 7F).

FIGURE 7.

Nicotinamide riboside (NR) supplementation improved infection outcomes in JEV‐infected A129 mice: (A) Schematic illustration of the experiment: eight‐week‐old A129 mice (n = 12) were divided into two groups (n = 6 per group); the treatment group received 200 mg/kg NR, while the control group received an equivalent volume of sterile PBS via IP. (B) Post‐infection body weight changes in mice tracked over 4 days. (C) Percentage of mice showing disease symptoms and terminal illness. (D) Relative viremia analyzed by RT‐qPCR targeting the NS3 gene of JEV, with β‐actin as a reference gene (50 μL of blood was collected from each mouse, and total RNA was extracted by pooling blood from two mice in the same group as one sample). (E) Relative viral RNA load quantified in tissues (brain, lung, heart, spleen, and kidney). (F) Representative H&E‐stained brain sections from NR‐treated and untreated groups. Shrunken and deeply stained neurons (orange arrow), neuronal edema (yellow arrows), and hemorrhage (green curve polygons). Images are from three individual mice per group. (G, H) Immunostaining images showing DAPI (blue), caspase‐3 (green), and JEV NS1 (red) in NR‐treated and untreated mice. Three mouse brains from each group were examined; Scale bars, 500 μm (G) and 100 μm (H). (I) mRNA expression levels of inflammatory markers: tumor necrosis factor‐alpha (TNF‐α), interleukin‐6 (IL‐6), C‐C motif chemokine ligand 3 (Ccl3), and nitric oxide synthase 2 (NOS2), measured by RT‐qPCR. Data for viral RNA and inflammatory marker mRNA are presented as mean ± standard deviation of fold change ($2^{-∆∆C_{\mathrm{T}}}$), with β‐actin as a reference gene, derived from three samples collected at 4 dpi, with duplicate measurements per sample. Statistical comparisons were performed using student's t‐test. Significance thresholds were set at *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. ns: not statistically significant (p > 0.05).

Using the JEV NS1 antibody for immunostaining enabled us to closely examine how JEV spreads within the brain. The group treated with NR showed a decreased fluorescence intensity of JEV NS1, with levels barely detectable compared to the untreated group (Figure 7G,H). Notably, the NR‐untreated group had multiple infection areas with higher viral density in the cortex and hippocampus (Figure 7G,H). This suggests that supplementing with NR reduces JEV spread or replication within the brain. Neurotropic virus infection can trigger neuronal cell death through multiple mechanisms, including direct cytotoxicity, indirect immune‐mediated damage, apoptosis, and neuroinflammation [41]. To assess apoptosis induced by JEV‐infection, we immunostained brain slices from mice treated with or without NR for activated caspase‐3 (Cas‐3), a key marker of apoptosis. Overall, the fluorescence intensity of activated caspase 3 in the NR untreated group was higher than that in the NR treated group (Figure 7G,H). The mRNA expression of inflammatory marker genes in the brain was measured to determine whether NR supplementation could ameliorate the neuroinflammatory response induced by JEV infection. The results revealed that NR treatment reduced the expression of interleukin‐6 (IL‐6), C‐C motif chemokine ligand 3 (CCL3), and nitric oxide synthase 2 (NOS2) as compared to NR‐untreated controls (Figure 7I). These findings demonstrate that NR treatment not only helps inhibit viral replication but also reduces neuroinflammation during JEV infections in A129 mice.

4. Discussion

Japanese encephalitis virus (JEV) poses a significant public health challenge in Asia due to its ability to cause severe encephalitis with long‐term disabilities, including seizures, amnestic episodes, and cognitive impairments [1, 3, 4, 5]. From a therapeutic perspective, understanding the molecular mechanisms of neuroinflammation and necrosis caused by JEV infection is essential for developing treatments targeting the neurological sequelae. This study examined and identified changes related to the regulation of NAD⁺ metabolism in the JEV‐infected T98G cells and A129 mice. Notably, restoring NAD⁺ homeostasis reduced JEV yield, offering valuable insights into NAD⁺ metabolism as a potential therapeutic target for JE.

Establishing an animal model susceptible to JEV infection is essential for studying the viral pathogenesis. We tested the virulence of the laboratory‐preserved GI strain of JEV in both immunocompetent (BALB/c and C57BL/6) and immunodeficient A129 mice. Results showed that this strain does not cause noticeable disease in immunocompetent BALB/c and C57BL/6 mice, but it is highly virulent in A129 mice. This aligns with previous research by Tripathi et al. [42], who observed similar resistance to P20778 JEV in teenage C57BL/6 mice, along with earlier studies reporting low infection and death rates in adult BALB/c mice [43]. The failure of the field‐isolated JEV strain to produce consistent disease in wild‐type models necessitated the selection of the A129 mouse model for this study. Although the A129 model lacks a functional Type I interferon (IFN‐α/β) receptor, it allowed us to isolate and rigorously evaluate the role of a functional cellular pathway, specifically NAD⁺ metabolism, in combating viral infection independently of this key antiviral axis. The rapid onset of symptoms and viral spread to the brain and other organs in immunocompromised mice highlights how JEV causes severe disease in humans, especially among vulnerable groups like young children and the elderly, who are often unimmunized and have weakened immune responses. This underscores the importance of immune status in determining JEV infection outcomes [44].

To understand dysregulations in NAD⁺ metabolism, we examined it mechanistically by analyzing mRNA and protein expression of selected genes related to NAD⁺ synthesis and consumption, as well as measuring NAD⁺ levels after JEV infection. NAD⁺ metabolism disruption can stem from dysregulation in biosynthetic pathways (de novo and salvage) or NAD⁺‐dependent enzyme activity [31]. The activation of the kynurenine pathway during neuroinflammation supports NAD⁺ production through de novo synthesis from tryptophan [25, 27, 45]. In this study, we investigated how key enzyme genes in the KP are transcriptionally and at the protein level altered to show the impact of JEV infection on de novo NAD⁺ synthesis. We observed that IDO, KATII, and KMO were upregulated in JEV‐infected brain tissue and T98G cells, indicating activation of de novo biosynthesis to meet NAD⁺ demands driven by the antiviral immune response and possibly leading to the production of neuroactive metabolites, as seen in other viral infections [46].

NAD⁺ is an essential coenzyme and co‐substrate involved in many biological reactions, non‐redox processes, and sirtuin‐mediated protein deacetylation. PARPs are the main NAD⁺ consumers through ADP‐ribosylation of NAD⁺ during DNA repair and immune response activation [12, 33]; this makes them a primary focus of this study to suggest possible causes of NAD⁺ depletion. NAD⁺‐dependent PARPs and SIRTs are crucial enzymes that regulate host metabolism, inflammation, and antiviral defense [34, 47]. For instance, PARP14 activates IFN‐I and IFN‐III responses, restricting the replication of ARH‐deficient murine hepatitis virus (MHV), SARS‐CoV‐2, and herpes simplex virus 1 (HSV‐1) [48]. Conversely, PARP1 promotes infections by RNA viruses such as Influenza A [49], Hepatitis C [50], and PRRSV [51]. Desingu et al. [52] demonstrated that during JEV infections, PARP1 supports viral replication in neuronal cells through autophagy, and its inhibition reduces infection severity. Despite these specific pro‐viral and antiviral roles, overexpression of PARPs during infection disrupts cellular NAD⁺ levels, leading to disturbed bioenergetic homeostasis and worsening the severity of the infection [31, 32, 33, 53]. Our data showed that JEV infection causes overexpression of PARPs, which could be a potential cause of NAD⁺ depletion. Other viral infections, such as Zika [12, 54] and SARS‐CoV‐2 [34, 55, 56], also upregulate PARPs, resulting in NAD⁺ depletion and supporting the idea that PARP overexpression is a key factor in disrupting cellular NAD⁺ metabolism. Similar to PARPs, excessive SIRTs can also deplete NAD⁺, leading to energy metabolism crises, impairing stress responses, and reducing cell survival [47, 57]. Further research is needed to clarify the distinct roles of PARPs and SIRTs, as well as each enzyme's individual contribution to JEV replication and antiviral responses.

The salvage pathway is another essential metabolic route that contributes to NAD⁺ production, as demonstrated by its role in maintaining NAD⁺ levels in the brains of ZIKV‐infected mice [12]. Salvage NAD⁺ biosynthesis enzymes (NAMPT, NMNAT1/2) were found to be dysregulated due to JEV infection. Specifically, the pathway's rate‐limiting enzyme, NAMPT [31, 32, 33, 36], was largely overexpressed in both models, indicating an increased need for NAD⁺ [58]. However, the rise in NAMPT levels did not fully offset the NAD⁺ depletion caused by JEV infection. These findings demonstrate that the compensatory increase in NAMPT during JEV infection is eventually overwhelmed, resulting in severe NAD⁺ depletion despite high NAMPT levels.

As noted by Yang et al. [59] and Pang et al. [12], administering NAD⁺ or its precursors like NR can reverse NAD⁺ depletion and correct deficiencies across various models. We examined how NR supplementation impacts NAD⁺ homeostasis and its subsequent effect on JEV infection. The link between JEV infection and disrupted NAD⁺ metabolism indicates that NAD⁺ levels can influence disease progression. The antiviral effect of maintaining NAD⁺ homeostasis against JEV was confirmed by NR supplementation in NAMPT knockdown T98G cells, which reduces viral loads and alleviates NAD⁺ depletion. Complementing this, reports showed that restoring NAD⁺ with NR appears to lessen the clinical symptoms caused by viral infections such as SARS‐CoV‐2 and Zika [12, 13]. Similarly, this study demonstrated that restoring NAD⁺ through NR supplementation not only relieved JE symptoms but also lowered viremia and tissue viral loads in A129, highlighting its potential as an anti‐JE treatment in vivo. As previously discussed, NR supplements have increased NAD⁺ levels during SARS‐CoV‐2 infection, potentially bolstering antiviral defenses and helping to control widespread inflammation [13, 60, 61]. In addition, NR supplementation improved cortical thickness, body and brain weight, and modestly increased survival rates in aged mice infected with Zika virus [12]. NAD⁺ depletion also activates SARM1, leading to axon degeneration in neuronal cells [15, 16, 17, 18, 19, 20]. Therefore, maintaining NAD⁺ homeostasis could help protect against neurotropic virus infection‐induced neuronal damage. Our data demonstrate that bolstering cellular resilience through NAD⁺ precursor supplementation can significantly mitigate disease severity even in the context of a compromised innate immune response. This underscores a promising therapeutic paradigm focused on supporting core cellular health, which may provide a beneficial adjunct to conventional antiviral strategies. Whether transient pharmacological inhibition of IFN signaling in wild‐type hosts recapitulates the therapeutic effect of NAD⁺ augmentation observed here, an approach not feasible in the current study due to the lack of a robust wild‐type infection model, is the future research question.

Overall, causing deadly infections and serious long‐term cognitive, motor, and neuropsychiatric problems in JE survivors highlights the urgent need for therapies that go beyond just antiviral effects. In this study, we demonstrate that NAD⁺ replenishment is an intervention that reduces viral replication and improves disease outcomes in JEV‐infected A129 mice. NAD⁺ restoration supports cellular energy production, which can enhance neuronal health and function in JE patients. Therefore, maintaining NAD⁺ levels in JE patients is a dual‐action intervention that could be key to improving acute disease outcomes and reducing long‐lasting neurological disabilities.

Author Contributions

T.A. conducted most experiments, organized and analyzed the data, and wrote the initial manuscript, including revisions. Z.T. contributed to the conceptualization, experimental design, material procurement, and review and editing of the manuscript. G.G. and J.D. managed the project and assisted with editing and review. Q.N., J.Y., and Z.Z. helped with manuscript review and editing. Z.H., Y.S., and F.C. provided technical support during the experiments. H.Y. secured funding and supervised the research. All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by NBCITS, CARS‐37, the Natural Science Foundation of Gansu Gansu Province, 22JR5RA029, the Innovat Innovation Program of the Chinese Academy of Agricultural Sciences, CAAS‐ASTIP‐2021‐LVRI, the Science Fund for Creative Research Groups (22JR5RA024) of Gansu Province, 22JR5RA024, the National Parasitic Resources Center, NPRC‐2019‐194‐30, the National Key Research and Development Program of China, 2024YFD1800100.

Ethics Statement

This study was carried out following the applicable regulations from the Ministry of Science and Technology of the People's Republic of China regarding the humane treatment of experimental animals. The procedures were reviewed and approved by the IACUC at the Lanzhou Veterinary Research Institute (LVRI) of the Chinese Academy of Agricultural Sciences (GSCAAS), with permit number LVRIAEC‐2024‐034.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Histopathological and immunohistochemical analysis of tissues from JEV‐ and mock‐infected A129 mice. (A–C) Histopathological examination images of brain, liver, and spleen tissues collected at 5 dpi, captured at 20× and 200× magnifications. Tissues were fixed in 10% formalin and stained with hematoxylin and eosin (H&E). In the infected brain, black, yellow, and blue arrows highlight hyperchromic neurons, degenerated neurons with vacuolated cytoplasm, and gliosis, respectively. In the infected liver, yellow, orange, black, red, and purple arrows indicate hepatocyte steatosis, swelling, focal necrosis with pyknotic and lytic nuclei, and eosinophilic cytoplasm with lymphocytic infiltration, respectively. In the infected spleen, black, blue, and yellow arrows denote punctate necrotic cells, granulocyte infiltrates, and insoluble fibrin, respectively. (D–F) IHC analysis of brain, liver, and spleen tissues from JEV‐ and mock‐infected groups harvested at 8 dpi. Tissues were stained for the NS1 viral antigen, with antigen–antibody reactions visualized as brown staining.

Table S1: Primers designed and utilized in this study, along with their corresponding target gene bank accession numbers.

Kassegn T. A., Tian Z., Du J., et al., “ NAD ⁺ Homeostasis Attenuates Japanese Encephalitis Virus Infection Progression,” The FASEB Journal 40, no. 4 (2026): e71526, 10.1096/fj.202503827R.

Data Availability Statement

All relevant data supporting the findings of this study are included in the article and its Supporting Information. Any additional information or materials are available from the corresponding author upon reasonable request.