The role of toll-like receptor 7 signaling in mice infected by severe fever with thrombocytopenia syndrome virus

Abstract

Severe fever with thrombocytopenia syndrome virus (SFTSV) is a negative-sense RNA virus that can cause high fever, thrombocytopenia, and multiple-organ dysfunction. Toll-like receptor 7 (TLR7) can recognize single-stranded RNA viruses and induce type I interferons (IFNs). Although type I IFNs play a critical role in SFTSV infection, the role of TLR7 remains incompletely understood. To investigate the role of TLR7 signaling in SFTSV pathogenesis, TLR7-deficient (TLR7^−/−) mice and wild-type (WT) mice were inoculated with SFTSV. TLR7^−/− mice exhibited severe weight loss and elevated viral RNA levels in various organs compared to WT mice. These mice also showed marked white pulp atrophy in the spleen and increased immunoreactivity for SFTSV nucleoprotein antigen. To further assess the significance of TLR7 signaling, resiquimod (R848), a TLR7 agonist was administered to SFTSV-infected mice. R848-treated mice exhibited significantly reduced viral loads in multiple organs and showed decreased levels of tumor necrosis factor alpha and interleukin-1 beta in the spleen and serum compared to mice in the control group. Consistently, R848-treated mice presented with mild white pulp atrophy and reduced antigen reactivity to the SFTSV nucleoprotein in the spleen. Collectively, modulation of TLR7 signaling, demonstrated by R848 activation and TLR7 deficiency, provides a novel approach to regulating antiviral immune responses against SFTSV infection.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-37472-4.

Introduction

Severe fever with thrombocytopenia syndrome (SFTS) is a significant zoonotic disease caused by the SFTS virus (SFTSV), a negative-sense RNA virus of the genus Bandavirus in the family Phenuiviridae¹. SFTSV was first identified in China in 2009. It has been reported in Korea, Japan, and other East Asian countries since 2011^2,3. The virus is primarily transmitted through bites from infected ticks, with incidence peaking during the active tick season from spring to autumn⁴. Clinical manifestations of SFTS include fever, nausea, vomiting, malaise, and diarrhea, often accompanied by a rapid decrease of platelet and white blood cell counts⁵. In severe cases, the disease progresses from a febrile phase to multiple organ dysfunction characterized by hemorrhagic symptoms, neurological disturbances, disseminated intravascular coagulation, and ultimately death⁶. SFTSV is known to target immune-related organs such as the spleen and liver, where viral replication and immune dysregulation contribute to disease severity⁷. Host-related factors such as age have been shown to influence disease severity. Fatal cases of SFTS are predominantly reported in individuals over 50 years of age⁸. This trend highlights advanced age as a major risk factor, likely due to age-related decline in immune function and increased systemic inflammation^6,9. This age-related susceptibility has also been observed in animal models. A previous study has shown that young adult ferrets did not exhibit mortality after SFTSV infection¹⁰. In contrast, aged ferrets displayed delayed innate immune responses and persistent inflammatory activation, which were associated with severe disease progression and high mortality¹⁰. In addition to host-related factors, a recent study has highlighted the expanding geographic distribution of SFTS in Asia¹¹. The inter-regional spread of competent tick vectors further suggests an increasing potential for regional epidemics and even global outbreaks of SFTSV^11,12. Although SFTS is classified as a biosafety level 3 (BL3) infectious disease, no standardized treatment protocols have not been established yet. Commercial vaccine is currently unavailable. Consequently, SFTS poses a serious public health threat, necessitating urgent and concerted research efforts.

Pattern-recognition receptors (PRRs) are proteins that can detect molecules commonly associated with pathogens or released from damaged host cells¹³. Upon pathogen recognition by PRRs, the host triggers an innate immune response as a first line of defense, subsequently initiating adaptive immunity to eliminate the pathogen¹³. PRRs include Toll-like receptors (TLRs), retinoic acid inducible gene I (RIG-I)-like receptors (RLRs), and C-type lectin receptors^14,15. Among these, endosomal TLRs including TLR3, TLR7, TLR8, and TLR9 play crucial roles in detecting viral RNA during infections^16,17. Specifically, TLR3 can detect double-stranded RNA viruses, whereas TLR7 and TLR8 recognize single-stranded RNA (ssRNA) viruses¹⁶. Previous studies have demonstrated that TLR7 plays a critical role in mediating antiviral immune responses against vesicular stomatitis virus and hepatitis C virus^18,19. However, the functional role of TLR7 during SFTSV infection remains to be fully elucidated. Considering the role of TLR7 in sensing ssRNA viruses and inducing antiviral type I interferon (IFN) production, TLR7 might serves as an important target for understanding host antiviral responses to SFTSV. Therefore, this study aimed to investigate the role of TLR7 in host responses to SFTSV infection using both TLR7-deficient (TLR7^−/−) mice and resiquimod (R848), a TLR7-specific agonist that could stimulate the receptor.

Materials & methods

Biosafety and ethics statement

All experiments were conducted in a BL3 facility at the Zoonotic Infectious Disease Research Institute, Jeonbuk National University. All research processes were carried out following experimental protocols approved by the Animal Ethics Committee of Jeonbuk National University (approval number: NON2023-094). All designated researchers completed the required safety training for handling the SFTSV and adhered to the safety guidelines of the BL3 facility. The study design, conduct, and reporting fully complied with the ARRIVE guidelines to ensure animal welfare and transparency.

Virus preparation and experimental animals

The Korean SFTSV strain KH1 (GenBank accession no. MH491547, MH491548.1, MH491549.1) was used in this study. The virus was propagated through five serial passages in Vero E6 monolayers maintained in Dulbecco’s Modified Eagle’s Medium supplemented with 5% fetal bovine serum and antibiotics (penicillin: 100 U/mL; streptomycin: 100 μg/mL)^20,21. For virus titration, Vero cells were seeded into 75 cm² tissue culture flasks and infected with the KH1 strain. After incubation at 37 °C for five days, infected cells were subjected to three cycles of freezing and thawing to release viral particles. The resulting viral suspension was clarified and titrated on Vero cells in 96-well microplates using tenfold serial dilutions. Viral antigens were detected by staining with an SFTSV nucleoprotein (NP)-specific hyperimmune serum, followed by incubation with an Alexa Fluor 488-conjugated anti-rabbit IgG secondary antibody (Invitrogen, Waltham, MA, USA)²². Viral infectivity was quantified using a fluorescence-activated infectious dose 50% assay.

Animals used in the experiment were five-week-old female C57BL/6 mice (WT; wild-type) (Samtako, Osan, Republic of Korea), and five-week-old female TLR7^−/− mice. These mice were kindly provided by Dr. Akira. (Osaka University, Suita, Japan).

Experiment design

To investigate the role of TLR7 in SFTSV infection, WT and TLR7^−/− mice were intraperitoneally (IP) inoculated with 2.61 × 10⁶ FAID₅₀/ ml of SFTSV diluted in phosphate-buffered saline (PBS; 100 μL per mouse). To partially inhibit type I IFN signaling, mice were administered 100 μg of the MAR1-5A3 monoclonal antibody (anti-mouse IFN alpha receptor, IFNAR, IgG1; BioxCell, Lebanon, NH, USA) by IP injection one day before and two days after SFTSV infection, considering the antibody’s half-life²¹. Mice were monitored daily for 9 days. On the day of necropsy, blood samples were collected, and organs including the brain, lung, liver, spleen, kidney, and small intestine were harvested for analysis (Fig. 1a).

Fig. 1.

Experimental design of a severe fever with thrombocytopenia syndrome virus (SFTSV) infection model and resiquimod (R848) treatment in mice. (a) Wild-type and toll like receptor 7-deficient mice were intraperitoneally (IP) inoculated with SFTSV (2.61 × 10⁶ FAID₅₀/ml). (b) Control and R848-treated mice were IP injected with SFTSV, followed by daily IP administration of either dimethyl sulfoxide (0.8%) or R848 (1 mg/kg), respectively, from day 0 to 2 dpi. To partially inhibit type I interferon (IFN) signaling, all mice received IP injections of IFN alpha receptor blocking antibody one day before SFTSV infection and again at 2 dpi. Sera and organs were collected at designated time points (3, 6, and 9 dpi) for virological, histopathological, and immunological analyses.

To reaffirm the importance of the TLR7 signaling mechanism in SFTSV infection, WT mice were treated with R848, a TLR7 agonist, following viral inoculation. R848 was diluted in dimethyl sulfoxide (DMSO; Sigma-Aldrich, Saint Louis, MO, USA). The control group received DMSO, a concentration confirmed to be non-toxic. R848 was administered IP at a dose of 1 mg/kg at 0, 1, and 2 days post-infections (dpi). The experiment was conducted over 6 days. Blood samples were collected from the retro-orbital venous plexus and the lung, liver, spleen, kidney, and small intestine were harvested at necropsy (Fig. 1b).

Body weight was recorded daily at the same time throughout the experiment. Necropsy procedures were performed after euthanasia. Prior to necropsy, mice were anesthetized by inhalation of isoflurane (3–5% for induction) (Hana Pharm. Co., Ltd., Hwaseong, Republic of Korea) delivered in room air. After confirming deep anesthesia, euthanasia was performed by cervical dislocation. Organs were aseptically placed in 500 μL of Tri-RNA reagent (Favorgen, Pingtung, Taiwan) and stored at − 40 °C for RNA analysis. For histopathological examination, tissues were fixed in 10% neutral phosphate-buffered formalin.

R848 toxicity test

R848 (Sigma-Aldrich), a TLR7 agonist, was used in this study to determine its maximum non-toxic dose in mice^23,24. Mice were divided into five groups and administered R848 IP at doses of 1, 2, 5, or 10 mg/kg. DMSO was used as vehicle control^25,26. DMSO at concentrations below 10% has been reported to be non-toxic in in vivo studies²⁷. To assess body weight changes induced by R848, measurements were taken prior to administration. Necropsies were performed at 3 and 6 dpi, corresponding to the time points used in the SFTSV infection experiment. During necropsy, blood samples were collected, and the liver, kidney, spleen, and mesenteric lymph nodes were harvested for further analysis. Histopathological quantification was performed in consultation with a veterinary pathologist on hematoxylin and eosin (H&E)-stained sections of the spleen and liver. Representative microscopic fields were randomly selected across animals within each group, and non-overlapping fields were used for quantification. For the spleen, megakaryocytes were manually counted at 200 × magnification in a total of eight non-overlapping fields per group. For the liver, inflammation was quantified by counting inflammatory foci at 100 × magnification in a total of ten non-overlapping fields per group.

Biochemical measurements

Serum samples were collected at 3 and 6 dpi. Supernatants were obtained by centrifugation at 3000 × rpm for 15 min at 4 °C. To evaluate systemic toxicity, particularly hepatic injury, serum aspartate aminotransferase (AST) levels were measured using a spectrophotometric assay kit (AM101-K; Asan Pharmaceutical, Hwaseong, Republic of Korea). Optical density (OD) was read at 490 nm using an EMax spectrophotometer (Molecular Devices, San Jose, CA, USA).

Quantitative polymerase chain reaction (qPCR)

Total RNA was isolated using Hybrid-R™ (GeneAll, Seoul, Republic of Korea). Complementary DNA was synthesized by reverse transcriptase using the ReverTra Ace™ qPCR RT Master Mix with gDNA Remover (TOYOBO, Osaka, Japan). Viral copy number was quantified by RT-PCR using primers targeting the L segment of SFTSV²⁸. Viral RNA copy numbers were quantified based on a standard curve generated from serial tenfold dilutions. Relative mRNA expression levels of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, IFN-α, interferon regulatory factor 7 (IRF7) and IL-28B (IFN-λ3) were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and calculated using the 2^−ΔΔCt method (Table 1)²⁹. To confirm the effect of IFNAR blockade, interferon-stimulated gene 15 (ISG15) and myxovirus resistance protein 1 (Mx1) were additionally measured as interferon-stimulated gene markers. Primer sequences are provided in Supplementary Table S1. qPCR was performed using qPCRBIO SyGreen Blue Mix (PCRbiosystems, Seoul, Republic of Korea) with a CFX Opus 96 Real-time PCR system (Bio-Rad Laboratories, Hercules, CA, USA).

Table 1.

Sequences of primers used for qPCR analysis.

Gene	Forward (5′-3′)	Reverse (5′-3′)
TLR7	CCACATTCACTCTCTTCATTGG	GGTCAAGAACTTCCAGCCTG
SFTSV-L	AACATCCTGGACCTTGCATC	CAATGTGGCCATCTTCTCCA
IRF7	GAGACTGGCTATTGGGGGAG	GACCGAAATGCTTCCAGGG
TNF-α	CCACCACGCTCTTCTGTCTAC	AGGGTCTGGGCCATAGAACT
IL-1β	GGTCAAAGGTTTGGAAGCAG	TGTGAAATGCCACCTTTTGA
IL-6	TACCACTTCACAAGTCGGAGGC	CTGCAAGTGCATCATCGTTGTTC
IFN- α	GCAATGACCTCCATCAGCAGCT	GTGGAAGTATGTCCTCACAGCC
IL-28B	GTTCAAGTCTCTGTCCCCAAAA	GTGGGAACTGCACCTCATGT
GAPDH	ACGGCAAATTCAACGGCACAG	GAAGACTCCACGACATACTCAGCAC

Enzyme-linked immunosorbent assay (ELISA)

To isolate proteins, serum samples were processed using Tri-RNA reagent and chloroform (Sigma-Aldrich, St. Louis, MO, USA) to separate RNA, DNA, and protein. RNA was removed. DNA and protein were then separated using ethanol. Proteins were precipitated with 2-propanol (Merck, Darmstadt, Germany) and centrifuged at 12,000 × g for 10 min at 4 °C, after which supernatants were discarded. The resulting protein pellets were washed with 0.3 M guanidine hydrochloride and 100% ethanol and then solubilized in 1% sodium dodecyl sulfate (Affymetrix, Santa Clara, CA, USA). After centrifugation at 10,000 × g for 10 min at 4 °C, the supernatant containing solubilized proteins was collected. Protein concentrations were measured using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions and normalized prior to ELISA. Cytokine levels of TNF-α, IL-1β, and IL-6 in serum samples were quantified using commercial ELISA kits (Invitrogen, Waltham, MA, USA) following the manufacturer’s protocol. Serum IFN-α levels were quantified using a commercial ELISA kit (PBL Assay Science, Piscataway, NJ, USA). IL-28A/B (IFN-λ2/3) levels were measured using a DuoSet ELISA kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. OD was measured at 450 nm using an EMax spectrophotometer (Molecular Devices, San Jose, CA, USA).

Histopathology and Immunohistochemistry (IHC) analysis

The brain, lung, liver, spleen, kidney, and small intestine were collected from each group for histopathological examination. Tissues were fixed in 10% neutral phosphate-buffered formalin and embedded in paraffin using a Modular Tissue Embedding Center EC 500 (Myr, Tarragona, Spain). Paraffin blocks were sectioned using a Microtome CR-603 (Craftek®, Jinhua, China). Slides were stained with H&E using a Myr Eva-slide stainer SS-30 (Myr) according to standard protocols. H&E-stained spleen and liver sections were semi-quantitatively scored according to predefined criteria (Supplementary Table S2). Each lesion parameter in the spleen and liver was scored for each mouse, and the parameter scores were summed to calculate a total pathology score per mouse. For the spleen, the scores for white pulp atrophy and decreased red pulp cellularity were summed. For the liver, the scores for inflammation and necrosis were summed. IHC was performed for spleen sections to detect SFTSV and to assess IRF7 expression. For SFTSV detection, a monoclonal antibody against the SFTSV NP was used. This antibody was kindly provided by Dr. Jun-Gu Choi (Animal and Plant Quarantine Agency, Gimcheon, Republic of Korea). For IRF7 detection, a primary antibody against IRF7 (Santa Cruz Biotechnology, CA, USA) was applied. Phosphorylated IRF7 (p-IRF7) was additionally evaluated using an anti-p-IRF7 antibody (Bioss, MA, USA). In all cases, sections were incubated overnight at 4 °C to allow antigen–antibody binding. A horseradish peroxidase-conjugated anti-mouse/rabbit IgG secondary antibody (MP-7500, Vector Laboratories, Newark, CA, USA) was subsequently used for detection. Visualization was performed using 3,3′-diaminobenzidine (SK-4105, Vector Laboratories) at a dilution of 1:33 for 20 s per slide. For SFTSV NP and IRF7 staining, slides were counterstained with methyl green for 5–10 min, whereas p-IRF7-stained slides were counterstained with hematoxylin. To quantify p-IRF7, eight randomly selected high-power fields per group were imaged, and only cells with nuclear p-IRF7 staining were manually counted. All slides were examined under an Olympus microscope (DP80, Olympus, Tokyo, Japan). Images were captured using Olympus digital imaging software.

Statistical analysis

All data are presented as mean ± standard deviation. Statistical significance between groups was determined using an unpaired t-test, one-way ANOVA, or the Mann–Whitney U test, as appropriate, in GraphPad Prism version 8.00 (GraphPad Software, San Diego, CA, USA). Asterisks were used in figures to indicate statistically significant differences between groups: *, p < 0.05; **, p < 0.01; ***, p < 0.001; and ****, p < 0.0001.

Results

TLR7 deficiency renders mice lethal to SFTSV infection

As a preliminary experiment, we evaluated TLR7 mRNA expression in spleen tissues of SFTSV-infected mice and uninfected WT mice. TLR7 expression was significantly reduced in SFTSV-infected mice compared to that in uninfected controls (Fig. 2a), indicating that SFTSV infection might suppress TLR7 signaling. Based on this observation, we hypothesized a potential association between TLR7 and the host antiviral response to SFTSV and subsequently conducted a comparative infection study using WT and TLR7^−/− mice. WT and TLR7^−/− mice were administered an IFNAR blocking antibody one day before and again two days after SFTSV inoculation to maintain inhibition of type I IFN signaling. To confirm effective IFNAR blockade, we measured ISG15 and Mx1 expression by qPCR. These genes are representative IFN-stimulated genes induced downstream of IFNAR. Under IFNAR blockade, the expression levels of ISG15 and Mx1 were suppressed at both 3 and 6 dpi. Notably, at 6 dpi, these expression levels were significantly lower in TLR7^−/− mice than in WT mice (Supplementary Fig. S1). Mice were subsequently IP inoculated with SFTSV (2.61 × 10⁶ FAID₅₀/ml). Clinical signs were monitored for 9 days. WT mice exhibited 100% survival and showed mild weight loss during infection. In contrast, TLR7^−/− mice showed a 20% mortality rate and significant body weight reduction compared to WT mice (Fig. 2b).

Fig. 2.

Toll like receptor 7 (TLR7) deficiency alters virological and clinical responses to severe fever with thrombocytopenia syndrome virus (SFTSV) infection. (a) TLR7 mRNA expression level in spleens of SFTSV-infected wild-type (WT) mice and uninfected controls were assessed by a quantitative polymerase chain reaction (qPCR). (b) Survival rates and body weight changes in WT and TLR7^−/− mice following intraperitoneally infection with SFTSV (2.61 × 10⁶ FAID₅₀/ml). (c) SFTSV RNA copy numbers in serum, lung, liver, spleen, kidney, and small intestine at 3, 6, and 9 dpi were quantified by qPCR. WT and TLR7^−/− mice were compared at each time point using unpaired t-tests. For figure clarity, only statistically significant differences (p < 0.05) are annotated, and non-significant comparisons are not shown. All data are presented as mean ± standard deviation. *, p < 0.05; **, p < 0.01; ***, p < 0.001; and ****, p < 0.0001.

To quantify viral loads in sera and organs, samples were collected at 3, 6, and 9 dpi. In sera, viral RNA was detected in both groups at 3 dpi. Serum viral RNA levels in WT mice decreased from 6 dpi. They were undetectable in some samples by 9 dpi. Although viral RNA levels in the serum of TLR7^−/− mice were higher at 6 or 9 dpi, statistical difference was not shown between WT and TLR7^−/− mice (Fig. 2c). Notably, at 3 dpi, viral loads were reduced by approximately tenfold in the spleen and 100-fold in the kidney of WT mice compared to those in TLR7^−/− mice. In WT mice, viral loads gradually decreased, whereas TLR7^−/− mice maintained high viral loads in various organs throughout the infection period. These findings suggest that TLR7 deficiency can lead to exacerbated disease progression following SFTSV infection.

Absence of TLR7 exacerbates pathological lesions during SFTSV infection

To further elucidate the impact of TLR7 signaling on SFTSV-induced pathogenesis, we next performed histopathological examinations of major organs. The spleen is considered a primary target organ in SFTSV infection with white pulp atrophy as the main histopathological lesion³⁰. This characteristic lesion was observed in both WT and TLR7^−/− mice at 3 dpi (Fig. 3a). By 6 dpi, the white pulp atrophy appeared more extensive in TLR7^−/− mice than in WT mice. At 9 dpi, the atrophy was reduced in WT mice, whereas it remained evident in TLR7^−/− mice. Similarly, in the liver, both groups exhibited coagulative necrosis at 3 dpi (Fig. 3b). Inflammatory cell infiltration around necrotic areas was detected in TLR7^−/− mice at 6 dpi. By 9 dpi, no liver lesions were apparent in WT mice, while TLR7^−/− mice still showed mononuclear inflammatory foci. Semi-quantitative histopathological scoring further supported these qualitative observations, showing significantly higher pathology scores in TLR7^−/− mice than in WT mice, with elevated spleen scores at 6 and 9 dpi and elevated liver scores at 3 dpi (Fig. 3c). To complement the lower-magnification images in Fig. 3b, representative higher-magnification H&E images of the same spleen and liver lesions are provided in Supplementary Fig. S2 to facilitate evaluation of tissue architecture and lesion areas.

Fig. 3.

Toll like receptor 7 (TLR7) deficiency exacerbates histopathological changes and reduces interferon regulatory factor 7 (IRF7) expression and activation following severe fever with thrombocytopenia syndrome virus (SFTSV) infection. Spleens and livers of the wild-type (WT) mice and TLR7^−/− mice were collected 3, 6, and 9 dpi. Representative hematoxylin & eosin (H&E)-stained sections of the spleen (a) and liver (b) from WT and TLR7^−/− mice at 3, 6, and 9 dpi are shown. Spleen sections showed white pulp atrophy. In liver sections, crosshatch marks (#) indicate coagulative necrosis and arrowheads ( →) indicate inflammatory cell infiltration. (C) Semi-quantitative histopathological scores for spleen and liver lesions were assigned and compared between groups. (d) SFTSV nucleoprotein antigen-positive cells were detected in the marginal zone of the splenic white pulp. (e) Percentage of SFTSV positive cell areas was quantified. (f) Relative mRNA expression levels of IRF7 in spleens of WT and TLR7^−/− mice at 3, 6, and 9 dpi determined by quantitative polymerase chain reaction. (g) Immunohistochemistry (IHC) staining of spleen sections for IRF7 expression. (h) Quantification of IRF7-positive cell area in the spleen. (i) Phosphorylated IRF7 (p-IRF7) staining in spleen and (j) its quantification as p-IRF7-positive area (%) and (k) nuclear p-IRF7-positive cells per high-power field (HPF, 600 ×) are shown. WT and TLR7^−/− mice were compared at each time point using unpaired t-tests. For figure clarity, only statistically significant differences (p < 0.05) are annotated and non-significant comparisons are not shown. All data are presented as mean ± standard deviation. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. H&E staining scale bar: 500 μm and 200 μm. IHC staining scale bar: 200 μm and 50 μm. p-IRF7 IHC staining scale bar: 20 μm.

To confirm viral antigen expression in the spleen, IHC was performed using an SFTSV NP-specific monoclonal antibody. NP antigen was detected in the marginal zone of the white pulp in both WT and TLR7 mice at 3 dpi (Fig. 3d). As shown at 3 dpi, the area of NP-positive cells was comparable between the two groups (Fig. 3e). From 6 dpi, WT mice showed a reduction in NP-positive cell areas, whereas TLR7^−/− mice continued to exhibit persistent NP-positive cells. Quantitative analysis further demonstrated that NP-positive cell areas persisted in TLR7^−/− mice at 6 and 9 dpi but, progressively decreased in WT mice (Fig. 3e).

To investigate antiviral immune factors associated with aggravated histopathological lesions and elevated viral antigen levels observed in TLR7^−/− mice, we focused on IRF7, a transcription factor positioned downstream of TLR7 signaling and recognized as a key regulator of antiviral responses³¹. Based on this, we hypothesized that TLR7 deficiency might suppress IRF7 expression and compromise the host antiviral immunity. We therefore examined IRF7 transcript levels during SFTSV infection. IRF7 transcript levels measured in the spleen were not significantly different between WT and TLR7 mice at 3, 6, or 9 dpi (Fig. 3f). We next evaluated IRF7 distribution in the spleen using IHC staining. At 3 dpi, IRF7-positive cells were observed in both WT and TLR7^−/− mice, showing no notable difference in their abundance (Fig. 3g). From 6 dpi, WT mice exhibited increased IRF7-positive cell areas compared to TLR7^−/− mice (Fig. 3h). Although IRF7-positive cells gradually increased in TLR7 mice, it remained at a reduced level compared to those in WT mice at both 6 and 9 dpi.

To further evaluate IRF7 activation, we examined p-IRF7 in spleen sections by IHC as an indicator of IRF7 activation and nuclear localization (Fig. 3i). We quantified both the p-IRF7-positive area and the number of nuclear p-IRF7-positive cells. WT mice showed higher p-IRF7 positive staining than TLR7 mice during SFTSV infection, with significant differences at 6 and 9 dpi (Fig. 3j). Consistently, WT mice had significantly higher numbers of nuclear p-IRF7-positive cells at 6 and 9 dpi compared with TLR7^−/− mice (Fig. 3k). Together, these findings indicate that IRF7 activation, as reflected by p-IRF7 staining and nuclear localization, is reduced in the absence of TLR7 during SFTSV infection.

Determination of R848 concentration for administration as a TLR7 signaling stimulator

Although previous studies have utilized R848 in in vivo models of viral infection^25,26, experimental conditions were not fully identical to those of our study. Therefore, preliminary dose-validation experiments were conducted based on previously reported concentrations to assess the applicability of R848 for subsequent experiments in the SFTSV infection model. Mice were administered with R848 at concentrations of 1, 2, 5, or 10 mg/kg. As shown in Fig. 4a, mice treated with 2, 5, or 10 mg/kg exhibited a concentration-dependent reduction in body weight compared to mice in the control group at 3 dpi. In contrast, mice treated with 1 mg/kg of R848 did not exhibit significant weight loss compared to mice of the control group. Serum AST levels were measured at 3 and 6 dpi to evaluate potential hepatotoxicity (Fig. 4b). There were no statistically significant differences in AST levels between R848-treated groups and the control group. Given these findings, histopathological analysis was performed to assess tissue lesions. In the spleen, megakaryocytes were observed in all groups from 3 dpi, showing an increase of frequency in mice treated with 2, 5, or 10 mg/kg of R848 compared to mice in the group treated with 1 mg/kg of R848 (Fig. 4c). These lesions were present at both 3 and 6 dpi. A previous study has reported that activation of TLR7 can promote megakaryocyte differentiation³². Such an immune response might partly explain the increased presence of megakaryocytes observed in spleens of R848-treated mice in this study. Furthermore, R848 administration has been associated with mild liver inflammation³³. In contrast, no inflammatory changes were detected in the group treated with R848 at 1 mg/kg (Fig. 4e). These histopathological findings were further supported by quantitative analyses showing increased megakaryocyte counts in the spleen and mild inflammatory foci in the liver in the higher-dose groups, whereas minimal changes were observed at 1 mg/kg (Fig. 4d, 4f). Hence, these results support the selection of 1 mg/kg R848 as a non-toxic dose for subsequent experimental applications²⁵.

Fig. 4.

Resiquimod (R848) induces dose-dependent body weight changes and histological lesion, with minimal toxicity at 1 mg/kg. (a) Body weight was monitored daily for 6 days in mice treated intraperitoneally with R848 at doses of 1, 2, 5, or 10 mg/kg, or vehicle control (dimethyl sulfoxide). (b) Serum was collected from each group at 3 and 6 dpi to assess Aspartate aminotransferase. (c) Hematoxylin & Eosin (H&E)-stained spleen sections show an increased number of megakaryocytes (#) in mice treated with 2, 5, or 10 mg/kg R848. (d) Megakaryocytes were quantified as the number of cells per 200 × field using eight images per group. (e) H&E-stained liver sections reveal multifocal inflammatory cell infiltrates ( →) in mice treated with 2, 5, or 10 mg/kg R848. (f) Liver inflammation was quantified as the number of inflammatory foci per 100 × field using ten images per group. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple-comparisons test. For clarity, non-significant comparisons are omitted and only statistically significant differences (p < 0.05) are annotated. All data are presented as mean ± standard deviation. *, p < 0.05; **, p < 0.01; ***, p < 0.001. H&E staining scale bar: 100 μm.

R848 administration alters organ viral RNA levels during acute SFTSV infection

To examine whether R848-induced TLR7 activation is accompanied by changes in viral RNA levels during acute SFTSV infection, R848 (1 mg/kg) was administered IP immediately after viral inoculation and continued daily until 2 dpi. As shown in Fig. 5a, no mortality was observed in either group during a 6-day observation period. Both groups showed weight loss following infection, with the R848-treated group exhibiting significantly less reduction than the control group at 4 and 5 dpi (Fig. 5b). At the time of necropsy, serum and organ samples were collected for viral RNA quantification. No statistically significant differences in serum SFTSV RNA levels were detected between R848-treated and control groups at 3 and 6 dpi (Fig. 5c). In contrast to serum data, spleen viral RNA levels in R848-treated mice were significantly reduced at 3 dpi. In addition, viral RNA copy numbers in lung, liver, kidney, and small intestine were significantly reduced in the R848-treated group compared to those in controls at 6 dpi. Collectively, these data show lower viral RNA levels in multiple organs in the R848-treated group during SFTSV infection.

Fig. 5.

Resiquimod (R848) treatment reduces severe fever with thrombocytopenia syndrome (SFTS) viral loads and results in less body weight loss in infected mice. (a) Survival rate and (b) body weight were monitored in SFTS virus (SFTSV)-infected mice treated with R848 or vehicle. Mice were intraperitoneally inoculated with SFTSV (2.61 × 10⁶ FAID₅₀/ml), and body weight was measured daily. (c) At 3 and 6 dpi, serum and organs (lung, liver, spleen, kidney, and small intestine) were collected for viral RNA quantification using quantitative polymerase chain reaction with absolute quantification. Statistical analysis was performed using unpaired t-test. For clarity, significance is indicated only for comparisons with p < 0.05. All data are presented as mean ± standard deviation. *, p < 0.05; **, p < 0.01; ***, p < 0.001; and ****, p < 0.0001.

R848 treatment modulates pro-inflammatory cytokines and pathological lesions in SFTSV infection

To assess the immunomodulatory effect of R848 on SFTSV-induced inflammation, pro-inflammatory cytokine expression was evaluated in the spleen and serum of treated and control mice. At 3 dpi, splenic TNF-α, IL-1β, and IL-6 mRNA levels showed no significant differences between groups. At 6 dpi, splenic TNF-α and IL-1β mRNA levels were significantly decreased in R848-treated mice compared with controls, whereas IL-6 did not differ significantly between groups (Fig. 6a). In serum, TNF-α was significantly decreased at 6 dpi in the R848-treated group, whereas IL-1β and IL-6 showed no significant differences between groups (Fig. 6b). Type I and III IFN responses were also evaluated. Splenic IFN-α and IL-28B mRNA levels showed no statistical difference between groups at 3 or 6 dpi (Fig. 6a). Splenic IL-28A mRNA was not detected by qPCR and was not shown. Serum IFN-α and IL-28A/B levels likewise showed no significant differences between groups at 3 or 6 dpi (Fig. 6b). Histopathological evaluation of the spleen revealed that white pulp atrophy was more prominent in the control group than in the R848-treated group at 6 dpi (Fig. 6c). Coagulative necrosis was observed in the liver of the control group, whereas lesions appeared relatively less pronounced in R848-treated mice at 3 dpi (Fig. 6d). To complement these qualitative observations, spleen and liver lesions were semi-quantitatively scored, and the overall pathology scores did not differ significantly between groups at either 3 or 6 dpi (Fig. 6e). Higher-magnification H&E images corresponding to the representative spleen and liver fields in Fig. 6c and 6d are provided in Supplementary Fig. S3. IHC analysis showed that the area of SFTSV NP antigen-positive cells was more expanded in the control group than in the R848-treated group (Fig. 6f). Data in Fig. 6g revealed that R848 treatment markedly reduced NP antigen-positive cell areas at 6 dpi. Collectively, these results show that R848-treated mice exhibited lower organ viral RNA levels and reduced NP antigen distribution, together with attenuation of selected inflammatory markers, during acute SFTSV infection.

Fig. 6.

Resiquimod (R848) treatment decreased levels of pro-inflammatory cytokines and histopathological lesions in severe fever with thrombocytopenia syndrome virus (SFTSV) infection. Samples were collected from control and R848-treated mice at 3 and 6 dpi. (a) mRNA expression levels of tumor necrosis factor alpha (TNF-α), interleukin-1 beta (IL-1β), IL-6, interferon alpha (IFN-α), and IL-28B in spleen tissues were measured by quantitative polymerase chain reaction. (b) Serum levels of TNF-α, IL-1β, IL-6, IFN-α, and IL-28A/B were quantified by enzyme-linked immunosorbent assay. (c) Hematoxylin & Eosin (H&E)-stained spleen sections showed white pulp atrophy in the R848-treated group and control group. (d) Liver sections showed coagulative necrosis, marked with crosshatches (#) and arrowheads ( →) indicating inflammatory cell infiltration. (e) Semi-quantitative histopathology scores were assigned for each mouse and compared between groups. (f) Immunohistochemistry (IHC) staining detected SFTSV nucleoprotein (NP) antigen-positive cells in the marginal zone of the spleen. (g) The percentage of SFTSV NP-positive area was quantified. Data were analyzed using unpaired t-test. Statistically significant differences only (p < 0.05) are annotated. All data are presented as mean ± standard deviation. *, p < 0.05; **, p < 0.01; ***, p < 0.001. H&E staining scale bar: 200 μm and 500 μm. IHC staining scale bar: 200 μm and 50 μm.

Discussion

SFTSV is a highly pathogenic virus that can be transmitted between animals, from animals to humans, and between humans^34,35. However, there remains a lack of suitable animal models to comprehensively study the pathogenicity and lethality of SFTSV. In a previous study, WT mice infected with SFTSV typically exhibited mild clinical signs and subsequently recovered¹¹. In contrast, immunocompromised mice, such as IFNAR^−/− mice, succumbed to even low titers of SFTSV (5 × 10² FAID₅₀/ml)¹⁷. Therefore, partial blocking of type I IFN signaling in mice by treatment with an IFNAR blocking antibody has been suggested as an SFTSV infection mouse model²¹. Antibody-mediated IFNAR blockade primarily attenuates receptor signaling without abolishing upstream PRR sensing. Accordingly, this approach was intended to increase susceptibility while preserving upstream PRR signaling during the early phase of infection. However, potential discrepancy issue was noted since TLR7 signaling is closely related to type I IFN signaling. To better interpret this discrepancy, it is relevant to revisit the biological role of TLR7. As a member of the PRR family, TLR7 can recognize pathogen-associated molecular patterns and promote type I IFN production through distinct signaling mechanisms^8,14. However, it has been demonstrated that the secretion of type I IFNs is regulated not only by TLR7 but also by other signaling pathways¹⁴.

In addition to endosomal TLR7, SFTSV RNA may also be detected by cytosolic RNA-sensing pathways, which may influence viral control during infection. For example, cytosolic RLRs, including RIG-I and melanoma differentiation-associated protein 5, signal via the mitochondrial antiviral signaling adaptor protein and can activate IRF3/IRF7 and nuclear factor kappa B (NF-κB), thereby inducing antiviral and inflammatory responses^36,37. In addition to promoting type I IFN production, TLR7- myeloid differentiation factor 88 adaptor protein (MyD88) signaling can also regulate IFN-independent responses. These include NF-κB-associated cytokine responses and innate immune cell activation, which may influence viral control and tissue pathology even when type I IFN receptor signaling is attenuated. The relative contribution of these PRR pathways is context dependent and can vary across tissues and cell types, reflecting differences in PRR expression and viral RNA accessibility^36,37. In this regard, endosomal TLR7-mediated sensing is relatively prominent in certain immune cell compartments, whereas RLR-mediated sensing can occur in a broad range of infected parenchymal and myeloid cell types^38,39. Thus, while both WT and TLR7^−/− mice are expected to retain shared RLR-mediated sensing, only WT mice preserve the additional TLR7-dependent endosomal input. Collectively, these considerations support the interpretation that the impaired viral control and exacerbated pathology observed in TLR7 mice are consistent with loss of this TLR7-dependent signaling in our model.

To further understand downstream mediators involved in TLR7-driven antiviral immunity, we focused on IRF7, a transcription factor known to play a central role in type I IFN induction⁴⁰. Upon sensing viral RNA, TLR7 can signal through the MyD88, leading to IRF7 phosphorylation, dimerization, and nuclear translocation, which ultimately drives the expression of type I IFNs³⁷. Disruption of this cascade can attenuate IFN responses and contribute to impaired antiviral immunity¹⁶. The critical role of IRF7 has been well established in several viral infection models. For instance, in severe acute respiratory syndrome coronavirus 2 infection, IRF7 deficiency has been shown to impair both type I and II IFN responses, resulting in increased pulmonary viral loads and exacerbated disease outcomes^40,41. Similarly, IRF7 mice infected with Sendai virus exhibit elevated viral replication and reduced survival due to defective IFN induction⁴². These studies underscore IRF7’s essential role in mounting an effective early antiviral response. In line with these findings, IHC quantification in our study indicated reduced IRF7 immunoreactivity in TLR7^−/− mice, as shown by a reduced IRF7-positive area in the spleen. Consistent with altered upstream activation, nuclear p-IRF7-positive cells were also decreased in TLR7^−/− mice during infection. Notably, this reduction was observed under IFNAR blockade in both groups, supporting impaired TLR7-MyD88-dependent IRF7 activation in vivo. These observations support the notion that TLR7-dependent signaling contributes to IRF7 activation in vivo during SFTSV infection, and may be associated with the higher viral burden and more severe pathological lesions observed in TLR7^−/− mice.

Considering the established role of TLR7 in antiviral immunity, we investigated whether stimulation of this pathway could influence host responses during SFTSV infection. To this end, we used R848, a synthetic agonist that targets both TLR7 and TLR8 in humans but selectively activates TLR7 in mice^43,44. R848 has been increasingly utilized in antiviral studies and investigated as an adjuvant to enhance vaccine efficacy^23,45. R848 has demonstrated antiviral activities in vitro against several viruses, including herpes simplex virus type 1 and type 2, human immunodeficiency virus type 1, Zika virus, and hepatitis B virus^46–48. R848 has also been tested in previous in vivo studies for various immunological applications, although experimental conditions are different between studies^26,45. Accordingly, we conducted a dose-finding experiment based on dose ranges reported in previous studies to determine an appropriate concentration for our SFTSV mouse model. In the present study, we observed the presence of megakaryocytes in the spleens of mice treated with high doses of R848. Megakaryocytes are primarily produced in the bone marrow under homeostatic conditions. However, their presence in the spleen may raise the possibility of extramedullary hematopoiesis⁴⁹, a compensatory process that can occur under pathological or inflammatory conditions, such as impaired bone marrow function or increased peripheral hematopoietic demand. A recent study has shown that TLR7 activation can promote megakaryocyte differentiation from human CD34⁺ progenitor cells and regulate platelet production during viral infections³². Taken together, the splenic megakaryocyte increase observed in this study is more likely to reflect a physiological immune response to TLR7-driven activation rather than pathological extramedullary hematopoiesis. Also, mice administered 2, 5, and 10 mg/kg of R848 exhibited body weight loss (Fig. 4a), consistent with a previous report showed that R848 induced weight reduction⁵⁰. Importantly, the selected dose of 1 mg/kg did not result in weight loss, histopathological abnormalities, or signs of toxicity, supporting its suitability for subsequent antiviral evaluation.

To assess systemic viral burden in this model, we measured SFTSV RNA levels in serum and in major target organs. Because circulating viral RNA and tissue viral RNA can reflect different aspects of infection at a given time point, we interpreted these readouts together. Across our experiments, serum viral RNA did not show statistically significant between-group differences at the sampled time points, whereas significant differences were detected in selected organs. This can be understood by considering that serum and organ RT-qPCR capture viral RNA in different biological compartments. Serum qPCR reflects viremia at the time of sampling and is sensitive to short-term fluctuations and inter-individual variability, which can reduce the ability to detect between-group differences at selected dpi⁵¹. In contrast, organ qPCR more directly reflects viral RNA burden within tissues and is more sensitive to group differences at a given time point. Consistent with this interpretation, in a mouse SFTSV infection model with regulated IFNAR signaling, viral RNA was detected in multiple organs early after infection and observed at high levels around 3 dpi. In contrast, serum viral RNA remained detectable over time at lower levels than those in organs depending on the sampling time point²¹. Accordingly, a non-significant serum RT-qPCR result at a given dpi does not exclude meaningful differences in tissue viral RNA burden. For example, in the R848 experiment, organ viral RNA levels were significantly reduced, whereas serum viral RNA levels were not significantly different at the same time points. Taken together, these findings suggest that tissue viral RNA burden can vary between groups even when serum viral RNA levels do not show significant differences at the assessed time points.

In conclusion, this study provides in vivo evidence that TLR7 signaling is essential for effective antiviral defense against SFTSV. We showed that TLR7 deficiency led to reduced IRF7 activation, increased viral burden, and severe organ damage. Moreover, administration of R848, a TLR7 agonist, effectively alleviated SFTSV-induced pathology and inflammation in mice. These findings can enhance our understanding of host-virus interactions.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing—Review & Editing: G.H.; Investigation, Formal analysis: B.O.; Writing—Original Draft Preparation, Visualization: G.H.; Supervision, Validation, Visualization: S.-C.P., S.-I.O.; Resources: D.T., J.-G.C.; Conceptualization, Resources, Project administration, Supervision, Validation, Visualization, Funding acquisition, Writing—Review & Editing: B.K. All authors have reviewed the manuscript and approved it for submission and publication.

Funding

This research was funded by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (RS-2019-NR040075).

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.