Integrated in vivo and transcriptomic analyses of lethal Oropouche virus infection reveal suppression of pathogenic host responses by antiviral therapy

Abstract

Oropouche virus (OROV) is an emerging arbovirus responsible for large outbreaks of febrile illness in Central and South America, with increasing reports of severe neurological disease and fatal outcomes. Despite its growing public health impact, no approved antiviral therapies or vaccines are currently available. Here, we show that favipiravir, a broad-spectrum nucleoside analogue, robustly suppresses OROV replication and disease in vivo. In a lethal Syrian hamster model, favipiravir treatment provided complete protection against OROV infection, preventing viral dissemination to peripheral organs and the central nervous system, and remained highly effective when administration was initiated after infection. In contrast, insufficient antiviral control resulted in viral neuroinvasion and fatality. To define host responses associated with OROV pathogenesis and their modulation by antiviral therapy, we performed transcriptomic profiling of liver and brain tissues. OROV infection induced interferon-driven inflammatory programs accompanied by marked disruption of metabolic and tissue homeostatic pathways, whereas these transcriptional signatures were largely abrogated by favipiravir treatment. Together, our findings identify favipiravir as a potent antiviral candidate against OROV and provide the first in vivo, tissue-resolved transcriptomic framework of OROV infection, linking effective viral suppression with the prevention of neuroinvasion and pathogenic host responses. These results highlight antiviral intervention as a viable strategy to mitigate OROV-associated disease and mortality.

Author summary

Oropouche virus (OROV) is an emerging zoonotic arbovirus that causes large outbreaks of Oropouche fever across Central and South America. Although most infections are self-limiting, some progress to severe neurological manifestations, and recent reports of deaths and adverse pregnancy outcomes have heightened public health concern. Despite thousands of reported cases every year, no approved antiviral treatments or vaccines are available. In this study, we evaluated favipiravir, an antiviral drug licensed for use against influenza, as a potential treatment for OROV infection. Favipiravir strongly inhibited viral replication in vitro and completely protected hamsters from lethal infections, preventing viral invasion of the central nervous system and the development of neurological symptoms. Notably, treatment remained highly protective even when antiviral administration was initiated after infection had begun. Transcriptomic analyses further revealed that OROV infection triggers strong antiviral and inflammatory responses, together with disruption of tissue homeostasis in the liver and brain, which are largely prevented by favipiravir treatment. Consequently, our findings demonstrate that favipiravir effectively inhibits viral spread and mitigates tissue damage and severe disease outcomes by suppressing virus-driven host responses, highlighting its potential as a promising therapeutic option for OROV infection and providing new insight into host responses associated with disease progression.

1. Introduction

Oropouche virus (OROV) is an emerging arbovirus endemic to Central and South America which causes recurrent outbreaks of an acute febrile illness, collectively referred to as Oropouche fever [1]. Although most infections are self-limiting, an increasing number of cases have been associated with severe neurological manifestations, fatal outcomes, and adverse pregnancy events [2–8]. Since 2023, large-scale outbreaks involving tens of thousands of confirmed cases have been reported, including regions which had no prior evidence of OROV circulation, underscoring the escalating public health threat posed by this virus [1]. Despite the growing recognition of OROV as a clinically relevant pathogen, there are currently no approved antiviral therapies or vaccines, and clinical management remains entirely supportive.

OROV belongs to the genus Orthobunyavirus and can cause systemic infection with central nervous system (CNS) involvement [9]. Clinical reports and experimental studies have documented viral dissemination to peripheral organs and the brain, accompanied by neurological manifestations ranging from meningitis to encephalitis [7,8]. However, the mechanisms by which OROV infection leads to tissue damage, neuroinvasion, and severe disease remain poorly understood. In particular, it is unclear how viral replication interfaces with host antiviral, inflammatory, and metabolic pathways in different tissues to drive disease progression. Such interactions are critical determinants of disease outcome, as excessive or dysregulated host responses have been implicated in the pathogenesis of numerous neurotropic viral infections [10].

A major limitation in this field is the lack of comprehensive in vivo analyses of host transcriptional responses during OROV infection. Although neuroinvasion and systemic disease have been demonstrated in both humans [8] and animal models [11], tissue-resolved transcriptomic studies which capture host responses in relevant target organs have not yet been reported. This gap has hindered a mechanistic understanding of OROV pathogenesis and limited the rational development of therapeutic interventions.

Favipiravir is a broad-spectrum nucleoside analogue that inhibits viral RNA-dependent RNA polymerases and is approved for the treatment of influenza in several countries [12,13]. This has demonstrated antiviral activity against a wide range of RNA viruses, including emerging and neurotropic pathogens [12,14–17]. Antiviral studies against OROV have so far been largely limited to in vitro discovery efforts. Broad-spectrum nucleoside analogues such as favipiravir, ribavirin and remdesivir inhibit viral replication in cultured cells, and recent screening approaches identified additional small-molecule and endonuclease-targeting inhibitors including acridones, lysergol, and 4′-fluorouridine [18–21]. Although in vivo antiviral activity has been reported [21], the impact of favipiravir on systemic viral dissemination and host transcriptional responses during Oropouche virus infection has not been comprehensively investigated.

Here, we investigated the antiviral efficacy of favipiravir against OROV infection using in vitro assays and a lethal Syrian hamster model. We further performed tissue-resolved transcriptomic analyses of the liver and brain to define host responses associated with OROV pathogenesis and their modulation by antiviral therapy. We show that effective viral suppression by favipiravir prevents systemic dissemination and neuroinvasion, suppresses pathogenic inflammatory and metabolic transcriptional programs, and confers robust protection against lethal disease. Together, these findings establish favipiravir as a promising antiviral candidate against OROV and provide an in vivo framework that integrates viral replication with host responses and disease-associated pathways during infection.

2. Results

2.1. Favipiravir potently inhibits OROV replication in vitro

To evaluate the antiviral efficacy of favipiravir against OROV (TRVL 9760 strain), we performed in vitro dose–response assay with Vero E6 cells and compared its activity to ribavirin. Favipiravir showed a higher inhibitory effect, with mean half-maximal effective concentration (EC₅₀) of 28.4 µM, whereas ribavirin was less effective with EC₅₀ of 115.7 µM (Fig 1A). Consistently, ribavirin treatment produced only a modest decrease in virus titer across the tested concentrations. On the other hand, favipiravir exerted a robust, dose-dependent antiviral effect, reducing infectious OROV production (Fig 1B). Untreated OROV-infected cells exhibited marked cytopathic effect (CPE), and ribavirin-treated cultures demonstrated only limited protection. In contrast, favipiravir treatment preserved cellular morphology, with a progressive reduction in CPE as the drug concentration increased. (Fig 1C).

Fig 1. In vitro antiviral effects of favipiravir and ribavirin against OROV infection.

(A-C) Vero E6 cells were infected with OROV TRVL 9760 at an MOI of 1. Favipiravir, or ribavirin was added 1 hour prior to infection. (A) Dose–response curves of antiviral activities of favipiravir and ribavirin were assessed by MTT assays at 72 hpi. Red dashed lines indicate 50% and 90% effective concentrations (EC₅₀ and EC₉₀). (B) Virus titers in the supernatants were measured by plaque assay at 72 hpi. A dashed line indicates the limit of detection. Bars indicate the geometric means ± geometric standard deviation (s.d.) of three independent biological replicates of a representative experiment. Statistical significance was assessed by two-way ANOVA with Tukey’s post-hoc test (*p < 0.05, **p < 0.01). (C) Representative microscopy images of cells at 72 hpi. Scale bars: 100 µm. (D, E) Time-of-addition assays. Vero E6 cells were infected with OROV at an MOI of 1, and favipiravir (200 μM) was administered at the indicated time points. (D) Schematic image of the assay. (E) Virus titers in the supernatants at 24 hpi. A dashed line indicates the limit of detection. Bars indicate the geometric means ± geometric s.d. of three independent biological replicates of a representative experiment. Statistical significance was assessed by one-way ANOVA with Dunnett’s post-hoc test (*p < 0.05).

To clarify which stage of the viral cycle is affected by favipiravir, we performed a time-of-addition assay (Fig 1D). Cells exposed to the drug during the post-entry stage (0 hours post infection [hpi]), or the early replication stage (3 hpi) showed significantly reduced viral titer at 24 hpi compared with non-treated cells, whereas drug treatment during the maturation stage (6 hpi) did not exhibit a significant decrease in virus titers (Fig 1E). These results demonstrated a strong inhibition after the onset of replication, consistent with favipiravir’s role as a nucleoside analogue that interferes with viral RNA synthesis, exerting its main antiviral effect during the post-entry replication phase.

Overall, these findings demonstrate that favipiravir efficiently inhibits OROV replication in vitro and offers superior protection from virus-induced CPE compared to ribavirin.

2.2. Antiviral suppression by favipiravir prevents systemic dissemination and lethal OROV disease in Syrian hamsters

Given the broad antiviral activity of favipiravir as demonstrated in multiple experimental animal models, including viral hemorrhagic fever and encephalitic viral diseases [17], we assessed its therapeutic efficacy against OROV infection employing a Syrian hamster infection model.

Animals were infected with a lethal dose of the OROV TRVL 9760 strain (5.4 × 10⁵ plaque-forming units [PFUs]) via the intraperitoneal route. Favipiravir (100 or 600 mg/kg/day) or ribavirin (40 or 100 mg/kg/day) was administered twice daily at 8-hour intervals via the perioral route beginning at 1 hour before infection and continuing for 6 days (Fig 2A).

Fig 2. Prophylactic oral administration of favipiravir protects Syrian hamsters from lethal OROV infection.

(A) Schematic illustration of the experimental design. Syrian hamsters were infected intraperitoneally with OROV (5.4 x 10⁵ PFU) and treated with favipiravir (100 or 600 mg/kg/day) or ribavirin (40 or 100 mg/kg/day) starting 1 hour before infection, twice daily at 8-hour intervals until 6 days post-infection (dpi). The dashed grey box indicates the treatment window. Animals were monitored for (B) survival and (C) body weight changes until 26 dpi (n = 6 per group). (D-G) Quantification of viral RNA by RT-qPCR in the (D) serum, (E) brain, (F) liver and (G) spleen collected at the indicated time points. Copy numbers were calculated based on standard curves. (H-K) Infectious viral titers determined by plaque assay in the (H) serum, (I) brain, (J) liver and (K) spleen at the indicated time points. (D-K) Vehicle-treated animals succumbed to infection after 2 dpi; therefore, samples were only available at 2 dpi for this group. N/A: not applicable. Dashed lines indicate the limit of detection. Bars represent geometric means ± geometric s.d. and dots represent individual animals (n = 4 per group) from a representative experiment. Statistical significance was assessed by Kruskal-Wallis test followed by Dunn’s multiple comparison post-hoc test: *p < 0.05, **p < 0.01 and ***p < 0.001. Panel (A) was created in BioRender. Tabata, K. (2026) https://BioRender.com/jl0r5d6.

All vehicle-treated animals succumbed to infection by 3 days post-infection (dpi) without preceding overt clinical symptoms or body weight loss. In contrast, animals treated with high-dose favipiravir (600 mg/kg/day) survived the 26-day study period without exhibiting any clinical signs and steadily gained weight throughout the trial. Treatment with a low dose of favipiravir (100 mg/kg/day) or ribavirin (40 or 100 mg/kg/day) conferred partial protection (Fig 2B and 2C). Notably, following treatment cessation at 6 dpi, some animals in each of the ribavirin groups exhibited weight loss and developed neurological symptoms, including ataxia, hindlimb weakness, and paralysis (S2 Fig.).

Viral replication in tissues and sera at 2, 6, 8, and 27 dpi was assessed by plaque assay and RT-qPCR. While high viral loads were detected in all tissues and sera of vehicle-treated animals at 2 dpi, no viral RNA or infectious virus was detected in animals treated with high-dose favipiravir at any time point across all tissues (Fig 2D–K). Treatment with low-dose favipiravir or ribavirin effectively reduced infectious virus production in serum, liver, and spleen (Fig 2H, 2J, and 2K). Importantly, infectious virus was detected in the brains of these groups at 6 and 8 dpi (Fig 2I), indicating incomplete clearance of OROV from the CNS at suboptimal dosing, and which was associated with the neurological symptoms observed after treatment cessation.

Collectively, these results demonstrate that favipiravir provides complete protection against lethal OROV infection in hamsters, suppressing systemic viral replication and limiting viral neuroinvasion.

2.3. Therapeutic administration of favipiravir after infection blocks viral neuroinvasion and improves survival

To assess the therapeutic potential of favipiravir, we examined its efficacy when treatment was delayed following OROV infection. Hamsters infected with OROV were administered favipiravir (600 mg/kg/day) starting at –1, 0, 6, 24, or 48 hpi (Fig 3A). All animals receiving favipiravir from –1, 0, 6, and 24 hpi were fully protected, with 100% survival (Fig 3B). Remarkably, even when administration was initiated as late as 48 hpi, favipiravir still conferred partial protection, with 50% survival and consistent weight gain during the observation period (Fig 3B–C and S3 Fig.).

Fig 3. Delayed favipiravir treatment protects Syrian hamsters from lethal OROV infection.

(A) Experimental design of the delayed-treatment model. Syrian hamsters were infected intraperitoneally with 5.4 × 10⁵ PFU of OROV and received favipiravir (600 mg/kg/day, orally, twice daily at 8-hour intervals for 6 days) starting at the indicated time points. (B) Survival and (C) body weight was monitored for 26 days (n = 4 per group). Panel (A) was created in BioRender. Tabata, K. (2026) https://BioRender.com/3dbf2zq.

These findings highlight favipiravir’s potential for clinical use when treatment cannot always be initiated immediately post-exposure.

2.4. Tissue-resolved transcriptomic analyses reveal pathogenic host programs induced by OROV infection and abrogated by favipiravir

The mechanisms underlying OROV-induced disease remain poorly characterized in both human cases and animal models. To define OROV infection–associated host responses and determine how favipiravir modulates these processes, we performed RNA sequencing (RNA-seq) of liver and brain tissues collected at 2 dpi, guided by the in vivo observation of hepatic and neurological involvement. Because the Syrian hamster genome remains less comprehensively annotated than the mouse genome, the identification and functional annotation of hamster genes were performed using their corresponding mouse orthologs. Principal component analysis (PCA) demonstrated a clear separation between uninfected and vehicle-treated infected animals in both tissues (Fig 4A–B). In the liver, one infected animal with low viral replication clustered closer to uninfected animals (Fig 4A–B and S4 Fig). Importantly, favipiravir-treated animals clustered near the uninfected control group, indicating that marked infection-driven transcriptional changes were largely suppressed by favipiravir treatment. Gene ontology (GO) enrichment analyses revealed that OROV infection induced robust systemic inflammatory and antiviral responses. In the liver, high viral replication induced upregulation of biological processes related to chemotaxis, cytokine signaling, myeloid cell differentiation, and interferon responses, along with downregulation of metabolic pathways related to energy production and fatty acid metabolism (Fig 4C, 4E, 4G, and S5A Fig.). In the brain, antiviral and inflammatory pathways were more strongly induced, and accompanied by the promotion of immune cell recruitment (Fig 4D, 4F, and S4B Fig.). These transcriptional modulations were markedly attenuated in favipiravir-treated animals, which maintained expression profiles similar to baseline in both tissues (Fig 4C–F, and S5 Fig.).

Fig 4. OROV infection alters host transcriptional responses in the liver and brain and favipiravir reversed these changes.

(A-F) The liver and brain (n = 3 per group) at 2 dpi were subjected to the RNAseq analysis. Principal component analysis (PCA) in the liver (A) and brain (B). Gene ontology (GO) enrichment analysis of differentially expressed genes (DEGs) in the liver (C) and brain (D). The n represents the number of genes related to each pathway that are differentially expressed. Heatmaps of standardized expression levels of genes in the selected metabolic processes in the liver (E) and brain (F). KEGG pathway–gene network analysis of the hepatic transcriptional response to OROV infection (G).

To validate the RNA-seq findings and assess time-course changes, we quantified the expression of selected inflammatory and IFN-related genes (Cxcl10, Rsad2, and Isg15) in the liver (Fig 5A) and brain (Fig 5B) at 2, 6, and 8 dpi by RT-qPCR. At 2 dpi, expression of these genes was strongly induced in vehicle-treated animals in both tissues, whereas both favipiravir (600 mg/kg/day) and ribavirin (100 mg/kg/day) treatment suppressed this early induction. Because no vehicle-treated animals survived to 6 or 8 dpi, the effects of favipiravir on gene expression at these later time points could not be directly compared with those in vehicle-treated infected animals. At 6 and 8 dpi, Cxcl10 expression in both antiviral-treated groups remained at levels comparable to those in mock-infected animals. In contrast, expression of Rsad2 and Isg15 in brain and liver tissues tended to remain elevated at 6 and 8 dpi in ribavirin-treated animals, correlating with increased viral loads (Fig 2E and 2I), whereas their expression remained relatively low in favipiravir-treated animals.

Fig 5. Validation of differential gene expression associated with immune and inflammatory responses.

Relative gene expression of cytokines and interferon-stimulated genes in the liver (A) and (B) brain of hamsters at 2, 6 and 8 dpi. mRNA expression was quantified using the ΔΔCt method. Data were normalized to the housekeeping gene Actb and presented as fold changes relative to mock-animals. All graphs show geometric means ± geometric (s.d.) from four biological replicates (n = 4 per group). Because vehicle-treated animals reached the humane endpoint before 6 dpi, samples from this group were only available at 2 dpi. Statistics were calculated by one-way ANOVA with Tukey’s post-hoc test: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

To further characterize host inflammatory responses during OROV infection, we evaluated the expression of Il6, Il10 and Tnfα, three canonical inflammatory cytokines. At 2 dpi, none of these cytokines showed significant changes in either liver or brain tissues across the experimental groups. Although their expression decreased at later stages in both favipiravir- and ribavirin-treated animals, these changes did not correlate with viral replication, suggesting that these cytokines are unlikely to be major drivers of OROV-associated disease in this model.

To further validate the transcriptional changes at the protein level, we evaluated the expression of Viperin, encoded by RSAD2. At 2 dpi, Viperin expression in the liver was significantly higher in the vehicle-treated group than in the favipiravir-treated group. In contrast, in the brain, animals treated with favipiravir at 100 mg/kg/day or ribavirin at 40 mg/kg/day showed significantly higher Viperin protein expression compared with the mock control group (S6 Fig).

Collectively, these data indicate that favipiravir provides sustained suppression of OROV-induced inflammatory and IFN responses in both the liver and the brain, whereas incomplete viral control by ribavirin is associated with re-emergence of IFN-stimulated gene expression at later stages of infection.

3. Discussion

OROV has historically been regarded as a cause of a range of self-limited febrile illnesses; however, recent outbreaks, confirmed human fatalities, and increasing reports of neurological and obstetric complications have redefined OROV as an emerging pathogen of significant public health concern [22–25]. Despite this growing clinical relevance, the mechanisms underlying severe disease and associated host responses remain poorly characterized, and no approved antiviral treatments are currently available. In this study, we demonstrate that favipiravir exhibits potent antiviral activity against OROV in vitro and provides robust therapeutic protection in a lethal Syrian hamster model. Importantly, our integrated virological and transcriptomic analyses have defined host response pathways associated with OROV infection and show that these disease-linked responses are effectively attenuated by antiviral treatment.

Favipiravir showed greater antiviral efficacy than ribavirin in vitro, with a substantially lower EC₅₀ and strong suppression of OROV-induced CPE (Fig 1A–C). Time-of-addition experiments indicated that favipiravir acts predominantly during the post-entry replication phase (Fig 1D–E), consistent with its established mechanism as a nucleoside analogue targeting the viral RNA-dependent RNA polymerase [14,16].

In vivo, favipiravir (600 mg/kg/day) conferred complete protection against lethal OROV challenge in hamsters, even when administration was initiated at 24 hpi, fully suppressing viral replication in systemic organs and the CNS (Fig 2A–K and 3A–C). In contrast, suboptimal antiviral treatment or ribavirin administration resulted in incomplete viral clearance from the brain (Fig 2I) and was associated with delayed onset of neurological signs alongside relative increases in inflammatory responses (Fig 5B). Disease progression, including motor impairment and, in severe cases, paralysis, is consistent with the hematogenous spread of the virus to neural tissues, as suggested by previous animal studies [11,25]. These findings parallel human clinical observations in which neurological manifestations occur in approximately 4% of clinical cases [26] along with reports of OROV RNA detection in the cerebrospinal fluids [8,27]. Collectively, these data suggest that timely and potent antiviral intervention by favipiravir may prevent CNS invasion and subsequent neurological disease during OROV infection, highlighting a critical therapeutic window relevant to human disease.

Although liver involvement is not a prominent feature of typical Oropouche fever, emerging clinical evidence indicates that hepatic dysfunction may occur in severe cases. Recent fatal OROV infections have been associated with markedly elevated aminotransferase levels, liver injury biomarkers [7], suggesting that hepatic infection may contribute to disease pathogenesis in humans. In the present study, OROV exhibited robust viral replication in the liver, accompanied by strong inflammatory and IFN-driven responses, as well as suppression of metabolic and mitochondrial pathways (Fig 4C, 4E, 4G, and 5A). These alterations were effectively normalized by favipiravir treatment, consistent with complete suppression of viral replication (Fig 2F, 2J, and 4C). Although species-specific differences may partly explain the limited recognition of liver involvement in human OROV infection, recent studies demonstrating hepatic infection and inflammatory responses in human liver organoids and an experimental mouse model [28,29] further support the potential contribution of hepatic responses to OROV-associated disease.

To our knowledge, this study provides one of the first comprehensive in vivo characterizations of host transcriptional responses to OROV infection and identifies pathways likely linked to disease severity. Transcriptomic analyses revealed robust innate immune activation in both the liver and brain, including IFN-stimulated gene expression, inflammatory signaling, and immune cell recruitment. In animals with higher viral burden, the simultaneous enrichment of pathways associated with positive regulation of innate immune responses and negative regulation of defense responses suggests a complex and potentially dysregulated host response to OROV infection. Rather than representing mutually exclusive processes, the concurrent activation of these pathways may reflect sustained antiviral and inflammatory signaling together with regulatory mechanisms that constrain excessive immune activation. When viral replication remains high, however, such regulatory responses may be insufficient to effectively control infection [30,31]. Importantly, the immune-related transcriptional changes were accompanied by downregulation of energetic, lipid metabolic, and mitochondrial pathways in the liver. Because metabolic and mitochondrial functions are closely linked to innate immune signaling and cellular homeostasis [30,32], their disruption during sustained antiviral and inflammatory responses may reflect increased cellular stress and potentially contribute to tissue dysfunction and injury during high-burden OROV infection. Together, these findings suggest that OROV-associated pathogenesis may result not only from direct viral replication but also from the interplay between persistent viral burden, dysregulated host immune responses, and disruption of cellular metabolic homeostasis.

Previous studies have established several animal models of OROV infection and have provided important evidence for viral dissemination, tissue tropism, and disease manifestations. For example, Rodrigues et al. (2011) [11] demonstrated that OROV infection in hamsters can result in systemic viral dissemination and neurological disease, establishing the Syrian hamster as a relevant model for studying OROV pathogenesis. The present study extends these observations by integrating tissue-resolved transcriptomic profiling with therapeutic intervention in the same experimental framework. By characterizing transcriptional responses separately in the liver and brain, we identified tissue-specific host responses associated with high viral burden, including metabolic and mitochondrial alterations in the liver and inflammatory/neuroinflammatory responses in the brain. Furthermore, favipiravir treatment allowed us to examine whether these virus-associated host responses were altered following suppression of viral replication. Thus, our study links viral burden, tissue-specific host transcriptional responses, and antiviral efficacy, providing a complementary framework for understanding OROV pathogenesis and treatment.

An important technical consideration is the limited availability of comprehensive Syrian hamster genome annotation, requiring the use of mouse orthologs for gene annotation and downstream functional enrichment analyses. Although this enabled identification of conserved host pathways, hamster-specific transcriptional features and genes without a confidently annotated mouse ortholog may have been underrepresented. Furthermore, a limitation of the temporal RT-qPCR analysis is the absence of untreated infected animals at 6 and 8 dpi, as all animals in this group reached the humane endpoint before these time points. Although favipiravir-treated animals did not show the marked induction of gene expression relative to mock-infected animals observed in vehicle-treated animals at 2 dpi, direct comparisons between vehicle-treated and favipiravir-treated animals at 6 and 8 dpi were not possible. Therefore, it cannot be determined whether the gene expression profiles at these later time points reflect the effect of favipiravir, as the corresponding gene expression levels in untreated infected animals at these time points are unknown.

Recent studies have demonstrated the efficacy of another polymerase-targeting antiviral, 4′-fluorouridine against OROV in vivo [21,33]. Favipiravir offers distinct translational advantages, including oral bioavailability and prior clinical use against multiple viral diseases [14,34]. Importantly, the observed efficacy of delayed treatment initiation enhances its relevance for outbreak settings, where early diagnosis and immediate therapy may not always be feasible (Fig 3B). Nevertheless, the potential clinical use of favipiravir for OROV infection should be considered in the context of its known safety limitations. Favipiravir has demonstrated teratogenic and embryotoxic effects, and its use has also been associated with increased serum uric acid levels and hyperuricemia [35,36]. These considerations are particularly relevant to OROV infection given the increasing recognition of congenital and pregnancy-associated disease. Therefore, although our findings provide encouraging preclinical evidence for the antiviral efficacy of favipiravir, further studies addressing pharmacokinetics, optimal dosing, treatment duration, and safety in relevant populations will be required before its clinical application to OROV infection can be considered.

In conclusion, this study has established favipiravir as a potent antiviral candidate against OROV and provides one of the first in vivo frameworks linking viral replication, host transcriptional responses, and disease-associated pathways in OROV infection. Our findings provide strong preclinical evidence that favipiravir can effectively suppress systemic and neuroinvasive viral replication while preventing the activation of inflammatory and metabolic pathways associated with pathology. Our findings advance understanding of OROV pathogenesis and support the potential repurposing of favipiravir as a targeted therapeutic option for Oropouche fever.

4. Materials and methods

4.1. Study design

This study was designed to evaluate the antiviral drugs against OROV infections. All viral experiments were performed in a biosafety level 3 laboratory at the International Institute for Zoonosis Control, Hokkaido University. Animal experiments with viral infection were performed according to the National University Corporation, Hokkaido University Regulations on Animal Experimentation. The protocol was reviewed and approved by the Institutional Animal Care and Use Committee of Hokkaido University (approval no. 24–0034). Experiments were done at least twice. All experiments with biological replicates are indicated in the figure legends.

4.2. Cells and viruses

Vero E6 (African green monkey kidney, clone E6) and BHK-21 cells were cultured and maintained in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco) supplemented with 10% heat-inactivated fetal bovine serum (FBS) at 37˚C in a humidified incubator with 5% CO₂. OROV TRVL 9760 strain was kindly provided by World Reference Center for Emerging Viruses and Arboviruses (WRCEVA), University of Texas Medical Branch at Galveston, UTMB Health and propagated in Vero E6 cells, and working virus stocks were stored at −80˚C until use.

4.3. Compounds

For in vitro studies, favipiravir (FUJIFILM Wako Pure Chemical Corporation) and ribavirin (Tokyo Chemical Industry Co. Ltd.) were dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich). For in vivo administration, favipiravir was prepared in 0.5% (w/v) carboxymethylcellulose (CMC; Sigma-Aldrich) and delivered via oral gavage. Ribavirin was dissolved in a vehicle solution containing 5% DMSO in 0.5% CMC.

4.4. MTT assays and EC₅₀ determination

Cell viability was assessed using the 3-(4.5-dimethylthiazol-2yl)-2,5diphenyltetrazolium bromide (MTT) assay. Cells and virus were diluted in MEM supplemented with 2% FBS and added simultaneously with compounds in 96-well plates. Briefly, 50 µL of two-fold serially diluted favipiravir or ribavirin was dispensed into each well, followed by 100 µL of cell suspension (2.5 × 10⁴ cells) and 50 µL of virus suspension. The amount of virus used corresponded to approximately 4 TCID₅₀. Plates were incubated at 37 °C for 72 h. After incubation, 30 µl of MTT solution (5 mg/L) was added, and the plates were incubated at 37˚C for 4 hours. Then, the supernatant was removed, and 140 µl of MTT solubilizer buffer was added to dissolve the formazan crystal overnight at room temperature. MTT reactions were measured using a microplate reader at an optical density (OD) of 570 nm. The results are presented as the percentage of viable cells relative to untreated control cells. The 50% effective concentration (EC₅₀) was determined using logarithmic interpolation (% of inhibition was calculated as (OD_sample  −  OD_{virus control})/(OD_{cell control}  −  OD_{virus control})) using GraphPad Prism with a four-parameter curve-fitting analysis.

4.5. Time of addition assay

Vero E6 cells were infected with OROV at a MOI of 1. Following virus adsorption for 1 hour at 37°C, cells were washed with PBS and 2% FBS DMEM was added. Favipiravir (50 µM) was added at: 1 hour before infection (−1 hpi), at the time of infection (0 hpi), and at 1-, 3-, and 6- hpi. Supernatants were harvested at 24 hpi, and viral replication was quantified by plaque assay.

4.6. Plaque assays

Monolayers of Vero E6 were inoculated with serial ten-fold dilutions of the specimens and incubated for 1 hour at 37˚C. After incubation, cells were overlaid with 0.5% methyl cellulose in Eagle’s Minimum Essential Medium (EMEM, Nissui Pharmaceutical Co., Ltd.) supplemented with 5% FBS and GlutaMAX supplement (Gibco). After 3 days of incubation, cells were fixed with a neutral 10% formalin neutral buffer solution and stained with 1% crystal violet to visualize plaques. Virus titers were calculated as PFU per milliliter (PFU/mL).

4.7. In vivo antiviral efficacy assessment in hamsters

Five-week-old male Syrian hamsters were intraperitoneally infected with 5.4 × 10⁵ PFU of the OROV TRVL 9760 strain in 200 µL of PBS. Favipiravir (100 or 600 mg/kg/day) or ribavirin (40 or 100 mg/kg/day) was administered orally beginning 1 hour before infection (-1 hpi) and continued twice daily with 8-hour intervals until 6 dpi. Animals were weighed daily before drug administration, and individual doses were adjusted according to the corresponding body weight. The doses of favipiravir and ribavirin were selected based on previous pharmacokinetic and safety data in hamsters, in which no toxicity was observed [15,37–40].

For survival analysis, animals were monitored daily for clinical symptoms and body weight changes up to 28 dpi. Disease severity was semi-quantitatively evaluated using a daily clinical scoring system based on four mutually exclusive categories: healthy, defined as no apparent symptoms or clinical signs; mild symptoms, defined as ruffled fur, hunched posture, and decreased exploratory activity; neurological symptoms, defined as hindlimb weakness, paralysis, or ataxia; and terminal/euthanasia, defined as animals reaching the humane endpoint or found dead.

For tissue or serum sample collection, animals (n = 4, per group) were euthanized at 2, 6, 8 dpi, with survivors sacrificed at 27 dpi, and the brain, liver, spleen, and serum were collected. Tissues were homogenized either in PBS, for viral titration and RT-qPCR or RNA Shield (Zymo Research) for RNAseq. For delayed treatment with favipiravir, hamsters were orally administered 600 mg/kg/day of favipiravir twice daily at 8-hour intervals, starting at −1, 0, 6, 24, or 48 hpi. The humane endpoints were defined by daily observation of severe clinical symptoms or weight loss exceeding 30% of the baseline body weight.

4.8. Reverse transcription - quantitative PCR (RT-qPCR) assay

Total RNA was extracted from hamster tissue homogenates using TRIzol LS reagent (Invitrogen) and Direct-zol RNA Miniprep kit (Zymo Research) according to the manufacturer’s instructions. RT-qPCR targeting OROV was performed using Thunderbird Probe One-step RT-qPCR kit (Toyobo) following the manufacturer’s recommended cycling conditions: 50˚C for 10 minutes, 95˚C for 1 minute, followed by 45 cycles of 15 seconds at 95˚C and 45 seconds at 60˚C. Reactions were carried out on a LightCycler 96 Real Time System (Roche). Primers and probe targeting the S segment of OROV genome were designed as previously described (S1 Data). The viral copy number was estimated by the standard curve method.

Measurements of host gene expression were performed using One Step TB Green PrimeScript RT-PCR Kit II (Perfect Real Time) (Takara). Primers for hamster Actb, Cxcl10, Il6, Il10, Isg15, RSAD2 and Tnf𝛼, are listed in the S1 Data. Target mRNA levels were normalized to hamster Actb mRNA and calculated as fold change relative to mock infected samples with the delta-delta cycle threshold (ΔΔCT) method.

4.9. Transcriptomic analysis

For mRNA-seq analysis, livers and brains collected at 2 dpi from 3 animals per group were stored in RNA Shield for 24 hours, then total RNA was extracted and purified using TRIzol LS reagent (Invitrogen), chloroform, and Direct-Zol RNA Miniprep kit (Zymo Research) according to the manufacturer’s protocol. Sequencing libraries were prepared using NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (New England BioLabs) and sequenced on an Illumina NovaSeq with 150 base pair-end reads at Azenta Life Science (S2 Data). Reads were processed with Trim Galore v0.6.10 for quality control, trimming and filtering. Preprocessed paired reads were aligned to a reference genome (accession number: GCA_017639785.1, Mesocricetus auratus) with Bowtie v2. RNA-seq counts data were filtered out to remove genes with low counts (<10 reads). RNA-seq data were analyzed in R using DESeq2 version 1.44.0 for normalization and differential expression. Principal component analysis (PCA) was performed on normalized gene expression data using DESeq2, and the results were visualized using ggplot2 (v4.0.0). Functional enrichment analysis was conducted with clusterProfiler version 4.12.6 [41]. Due to incomplete annotation of Mesocricetus auratus, differentially expressed hamster gene symbols were mapped to their Mus musculus ortholog gene symbols prior enrichment analysis. The conversion was done via Biomart (Ensembl) mapping. Any gene lacking a valid mouse ortholog was removed from the downstream analysis. Over-representation analysis (ORA) was performed using GO. For ORA, significance was determined with P values <0.05 (adjusted by the Benjamini-Hochberg method) of a one-sided version of Fisher’s exact test. Heatmaps were generated with the R package pheatmap version 1.0.13. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was performed using the enrichKEGG function from the clusterProfiler R package, using the mapped gene set described above. Genes with adjusted p values < 0.05 and an absolute log₂ fold change > 1 were included in the enrichment analysis. KEGG pathways with adjusted p values < 0.05 were considered significantly enriched. Network-based visualization of enriched KEGG pathways and their associated genes was generated using the cnetplot function from the enrichplot package version 1.30.4.

4.10. Western blot

Hamster brain and liver tissue homogenates were lysed using RIPA lysis and extraction buffer. The extracted protein samples were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred onto polyvinylidene difluoride (PVDF) membranes at 300 mA for 1 h. The membranes were blocked with 5% skimmed milk for 30 min and incubated overnight at 4°C with the primary antibodies, anti-RSAD2 (1:1000; #28089–1-AP; CosmoBio) and anti-β-actin-pAb-HRP-Direct (1:1000; #PM053–7; MBL). Following incubation, the membranes were washed and incubated with the appropriate Anti-Rabbit-IgG (Goat) HRP-conjugated secondary antibody (1:3000; #SA00001–2; CosmoBio) at 37°C for 1 h. Protein bands were visualized using an Immobilon Western chemiluminescent HRP substrate kit (Sigma) and ATTO lumino graph II EM imaging system (ATTO). Band intensities were quantified using ImageJ software. Viperin protein expression was normalized to β-actin and expressed as fold change relative to the mock-infected group.

4.11. Statistical analysis

All statistical analyses were performed using Prism version 10.4.2 (GraphPad Software). Comparisons of RNA copies by RT-qPCR and quantification of viable virus were performed using Kruskal-Wallis test followed by Dunn’s multiple comparison post-hoc test. P values of ≤ 0.05 were considered significant. For the validation of host expression genes, statistical significance of the fold difference between the infected and the uninfected samples was calculated with the ordinary one-way ANOVA followed by Tukey’s post-hoc test.

Transcriptomic data were analyzed and plotted in R version 4.4.0, using the above-mentioned R packages.

4.12. Data availability

Raw sequencing data were deposited at the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1416565.

Supporting information

S1 Fig. Individual body weight changes of hamsters in Fig 2C.

Bold lines represent the average of the body weight in each group while grey lines represent individual body weight values (n = 6 animals per group) in (A) mock, (B) vehicle, (C-D) favipiravir (100 or 600 mg/kg/day), and (E-F) ribavirin (40 or 100 mg/kg/day) groups. Dashed grey box indicates the treatment window (B-F).

(TIF)

S2 Fig. Daily clinical scores and body weight of Oropouche virus-infected hamsters under different therapeutic regimens.

Clinical progression was monitored over 27 days post-infection (dpi) in hamsters infected with OROV (n = 4 animals per group). Animals were monitored for (A) survival and (B) body weight changes. Dashed grey box indicates the treatment window (A-B). Clinical data are presented across six groups: (C) Mock; (D) Vehicle; (E-F) Favipiravir (100 or 600 mg/kg/day); (G-H) Ribavirin (40 or 100 mg/kg/day). Stacked bar charts represent the daily percentage of the total animal population within each group classified into four distinct clinical outcomes: healthy (green), mild symptoms (yellow), neurological symptoms (red), and terminal/euthanasia (black).

(TIF)

S3 Fig. Individual body weight changes of hamsters in Fig 3C.

Bold lines represent the average of the body weight in each group while grey lines represent individual body weight values (n = 4 animals per group) from a single independent experiment in (A) mock, (B) vehicle, (C) -1 hour before infection (-1 hpi), (D) at the time of infection (0 hpi), (E) 6 hpi, (F) 24 hpi, and (G) 48 hpi groups. Dashed grey box indicates the treatment window (B-G).

(TIF)

S4 Fig. Viral RNA copies in the liver and brain used for RNAseq analysis.

Viral RNA copies were measured based on the standard curve by RT-qPCR in the brain (A) and liver (B) of hamsters at 2 dpi. Dashed lines represent the limit of detection. Bars represent geometric mean ± geometric s.d., with individual biological replicates shown as dots (n = 3 animals per group).

(TIF)

S5 Fig. Transcriptomic profiling of OROV-infected hamsters related to Fig 4. RNA-seq analysis was performed using the biological replicates described in Fig 4 (n = 3 per group).

Differentially expressed gene (DEG) analysis was performed with DESeq2 (v1.44.0) in R (v4.4.0). Genes with log₂ fold change absolute values ≥ 1 and adjusted p-value (padj) < 0.05 were considered differentially expressed with statistical significance. Volcano plots were generated using ggplot2 version 4.0.0 comparing vehicle (High viral load) vs. uninfected (left) and favipiravir vs. uninfected (right) in the liver (A) and brain (B).

(TIF)

S6 Fig. Protein expression of Viperin (RSAD2) in the liver and brain tissues of OROV-infected hamsters.

RSAD2 protein expression was measured and normalized to β-actin protein levels in the liver (A) and brain (B) tissues of hamsters at 2 days post-infection (dpi). Bars represent the geometric mean ± geometric standard deviation (s.d), with individual biological replicates shown as dots (n = 4 animals per group) from a single independent experiment.

(TIF)

S1 Data. List of primers used in this study.

MGB = minor-groove-binding; NFQ = non-fluorescent quencher.

(PDF)

S2 Data. Summary of quality of isolated RNA and raw data statistics of the next generation sequencing related to Fig 4.

(PDF)

Data Availability

The minimal data set is available within the manuscript and its Supporting Information files. Raw sequencing data were deposited at the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1416565.

Funding Statement

This study was supported in part by the Japan Society for the Promotion of Science (JSPS) KAKENHI under grant numbers 25K18810 (YI) and 25K09431(HS); and the Japan Agency for Medical Research and Development (AMED) under grant numbers JP223fa627005 (HS) and JP24wm0225044 (YI). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.