All virus-related experiments in our study were conducted in the biological safety level 2+ laboratory of the Eastern Theater Disease Control and Prevention Center. We collected blood samples from 29 SFTS patients ranging from 4 to 18 days after onset who were admitted to the Department of Infectious Disease, First Affiliated Hospital of Nanjing Medical University between January 2020 and November 2021. Among the 29 SFTS patients, 3 patients were in the fever stage, and 26 patients were in the multiple organ failure (MOF) phase based on the clinical manifestations and examinations. The viral load of SFTS patients was 3.03–9.12 lg copies/ml. According to previous research (Deng et al., 2013; Liu et al., 2017), we divided the SFTS patients into 7 non-severe cases and 22 severe cases, and 6 severe cases died. Detailed information on patients is present in Supplementary Table 1. In addition, blood samples were collected from 29 healthy individuals aged 18–55 years who donated blood at the Jiangsu Provincial Blood Center from October 2020 to October 2021. This study was performed with the approval of the Ethics Committee of the Institutional Review Board at the First Affiliated Hospital of Nanjing Medical University, and each participant signed an informed consent form. Animal experiments were approved by the Huadong Medical Institute of Biotechniques (HMIB) ethics committee and conducted as per HMIB's ethical regulations.

Using the principle of homologous recombination, the TRIM3 gene was modified by targeting ES cells. The ES cell targeting vector was constructed using the Infusion method. After linearization, the vector was electrotransfected into C57BL/6J × 129S3 ES cells. Positive ES cell clones were amplified and injected into the blastocysts of C57BL/6J mice to obtain chimeric mice. Then these mice were maintained and bred in an environmentally controlled and specific pathogen-free (SPF) animal facility at the Eastern Theater Disease Control and Prevention Center. When these mice grew to 7–10 days, their tails were cut for gene identification of TRIM3^−/− homozygote.

IFNAR^−/− C57BL/6 mice were purchased and identified from Cyagen (Cyagen Biosciences Inc.). All mice were bred and maintained in an environmentally controlled and SPF animal facility at HMIB. Mice aged 6–8 weeks were randomly divided into three groups of five mice. On day 0, mice were inoculated subcutaneously with 100 μl 1.0 × 10⁵ TCID₅₀/ml, and 1.0 × 10⁶ TCID₅₀/ml virus solution, while the mock mice were inoculated with 100 μl of DMEM. Every mouse in each group was weighed daily, and the livers, spleens, lungs, and kidney tissues were obtained after death.

The Chinese DBV strain JS14 was provided by the Jiangsu Provincial Center of Disease Control and Prevention and was extensively proliferated using Vero cells. The viruses used for experiments were all confirmed to be alive through plaque testing, and they were properly packaged and stored in an ultra-low-temperature freezer.

THP-1 and Vero Cells were obtained from Huzhen Industrial Company (Shanghai, China) and Cell Signaling Technology (Beverly, MA, USA) separately. THP-1 cells were cultured in 1,640 medium with 10% FBS and 1% penicillin-streptomycin, while Vero cells were cultured in DMEM medium. Peripheral Blood Mononuclear Cells (PBMCs) were derived from the blood donor cell filters. They were separated by the FICOII separation solution. The separated PBMCs were resuspended in DMEM, laid flat in a 6-hole plate with 1.5 × 10⁷ cells per hole, and then put overnight in a 5% CO₂ incubator. The DMEM medium was slowly sucked out and discarded the next day. Instead, they were cultured in a 1,640 medium containing 10% FBS and 1% penicillin-streptomycin. Peritoneal macrophages were derived from corresponding C57BL/6 mice (TRIM3^−/−, IFNAR^−/−).

The total length of the CDS region of the human TRIM3 gene was found in GenBank. According to the CDS region of the TRIM3 gene and the selected vector, enzyme digestion was performed on upstream EcoR I and downstream BamH I, and primers were designed. The primers and TRIM3 overexpression (TRIM3^OE) Lentivirus were synthesized by Shanghai Genepharma Company. Lentivirus-mediated overexpression was constructed both in PBMCs and THP-1 cells successfully.

TRIzol regent (Invitrogen, USA) was used for isolating total RNA, PrimeScript RT Master Mix (Takara, Japan) was performed for cDNA synthesis, and TB Green Premium Ex Taq (Takara, Japan) was used for real-time PCR on the ABI QuantStudio 5 sequence detection system. Primers for qPCR are shown in Supplementary Table 2. GAPDH was used as an endogenous control for relative mRNA expression.

According to the manufacturer's instructions (BestBio, China), we extracted total proteins. All proteins were denatured at 100°C for 10 min, then separated by SDS-PAGE using 12% polyacrylamide gel, and finally transferred to a nitrocellulose membrane (Sangon Biotech, China). After blocking in Tris buffer saline (TBS) containing 4% fetal bovine serum albumin (BSA; Roche), the membrane was incubated with the first antibody and a corresponding horseradish peroxidase coupled secondary antibody, and then washed thoroughly using TBS-T (supplemented with 0.1% TBS of Tween 20, pH 7.4). Protein signals were detected using an enhanced chemiluminescence assay kit (ECL; Tanon) and visualized using a chemiluminescence imaging system (Tanon 5200, China). We put primary antibodies (TRIM3, ProteinTech; Toll-like receptor 3 (TLR3), K48, TRIF, TRAF6, TAB1, TAB2, phosphorylation p38, phosphorylation JNK1/2, phosphorylation IKKα/β, phosphorylation NF-κB, p38, JNK1/2, NF-κB, and GAPDH, Cell Signaling Technology) as target bands, and the anti-rabbit and/or anti-mouse IgG antibodies (Cell Signaling Technology, Beverly, MA, USA) as the secondary antibodies.

We carried out coimmunoprecipitation (CO-IP) and immunoblotting experiments as described below. For transient transfection and CO-IP experiments, THP-1 cells were lysed in 1 ml of NP-40 buffer (20 mM Tris HCl, pH 7.5150 MM NaCl, 1 mM EDTA, 1% NP-40, 10 μg/ml aprotinin, 10 μg/ml leucin peptide, and 1 mM phenylmethyl sulfonyl fluoride) and then centrifuged at 12,000 rpm for 10 min at 4°C. For each immunoprecipitation, the supernatant was kept at 4°C for 2 h with a 1:1 serum of 0.5 μg antibody or control IgG and 30 μl protein G Sepharose. Then, we washed the agarose beads three times with 1 ml of lysis buffer containing 500 mM NaCl and analyzed the sediment using standard immunoblotting with the indicated antibody.

Following the instructions of the corresponding ELISA kit (R&D Systems, USA), we tested pro-inflammatory cytokines (IL-6, IL-1β, and TNF-α) in the cell supernatant.

After dehydrating the livers, spleens, lungs, and kidneys, which were soaked in a 4% paraformaldehyde solution, we embedded them in paraffin and cut them with a thickness of 4 μm into parts. Then, they were incubated with primary anti TRIM3 (1:100) overnight at 4°C and incubated in the dark for 3–5 min at room temperature with fluorescence-labeled secondary antibodies. We stained them with DAPI without sunlight for 1 min and finally sealed them for microscopic examination to collect images after adding quenching agents.

The unpaired, two-tailed Student's t-test was used for statistical analysis, and the F-test was performed to confirm whether the two populations had homogeneity of variance. Mann-Whitney U test was performed for data not normally distributed. For mice survival studies, we generated Kaplan-Meier survival curves and analyzed them with logarithmic rank tests. All statistical analysis was performed by GraphPad Prism 8.0 software. A value of P < 0.05 was considered significant.

The relative expression level of TRIM3 mRNA in PBMC samples of SFTS patients was significantly lower than that of healthy participants, especially in severe cases (Figure 1A). After subcutaneous injection of DBV, the mice were observed daily, and their body weight was recorded. Infected mice manifested abnormal consciousness, a weakened response to external stimuli, hair loss, and a bowed back. All infected mice showed changes in body weight, and the weight loss in the 10⁶ TCID₅₀ group was more significant than that in the 10⁵ TCID₅₀ group (Figure 1B). The mice began to die on day 5 after infection, and all the mice in the 10⁶ TCID₅₀ group died on day 6 after infection, while all the mice in the 10⁵ TCID₅₀ group died on day 7 after infection (Figure 1C). An autopsy showed congestion and tissue swelling in the livers, spleens, lungs, and kidneys, with the 10⁶ TCID₅₀ group being more pronounced than the 10⁵ TCID₅₀ group. qRT-PCR indicated that DBV had a high level of replication in the livers, spleens, lungs, and kidneys, with the most significant replication in tissues of the livers and kidneys (Figure 1D). Immunofluorescence analysis and qRT-PCR of the 10⁶ TCID₅₀ group mice tissues showed a significant decrease in the TRIM3 expression in these organs (Figures 1E–I).

Innate immunity is the first line of defense against pathogen invasion (Chatterjee et al., 2016), and peripheral blood monocytes are the main target cells during DBV infection (Peng et al., 2016). Therefore, human PBMCs, THP-1 cells, and mouse peritoneal macrophages were selected for subsequent experiments. Human PBMCs were extracted and co-cultured with DBV at MOI = 0.5, MOI = 1, and MOI = 2.0 for 2 h. Cell viability was detected by the CCK-8 assay. The PBMC's viability was 98.9, 94.2, and 92.3% when MOI = 0.5, MOI = 1, and MOI = 2.0, respectively. Therefore, MOI = 0.5 was chosen as the working concentration for this experiment.

To verify the expression of TRIM3 in the cell model, qRT-PCR, and Western blot were used to detect the expression level of TRIM3 mRNA and protein after DBV infection with PBMCs. Compared with the mock group, the expression of TRIM3 mRNA, and protein was decreased in the DBV-infected group (data were not shown).

To confirm the role of TRIM3 in SFTS, we tested whether overexpression of TRIM3 could inhibit DBV-induced inflammation. According to the instructions, we transfected PBMCs and THP-1 cells with lentiviral vectors to overexpress TRIM3. qRT-PCR and Western blot analysis showed that TRIM3 mRNA and protein in PBMCs (Figures 2A, B) and THP-1 cells (Figures 2C, D) increased significantly in the TRIM3^OE group. Total mRNA was extracted from cells 24 h after DBV infection. The relative expression levels of IL-1β, IL-6, and TNF-α mRNA in PBMCs and THP-1 cells increased after DBV infection. After overexpression of TRIM3, the relative mRNA expression of the above inflammatory cytokines decreased significantly (Figures 2E, F). Meanwhile, at 24 h after DBV infection, the expressions of IL-1β, IL-6, and TNF-α in the supernatant were significantly increased, while the overexpression of TRIM3 could significantly reduce the secretion of these cytokines (Figures 2G, H).

To further explore the mechanism of increased inflammatory factors, total intracellular proteins were extracted 1 h after DBV infection, and the expression of TLR3 and the activation of the downstream Mitogen-activated protein kinase (MAPK) pathway in THP-1 cells of different groups were detected by Western blot. After DBV infection, TLR3, and MAPK signaling pathways are activated, with p38 and JNK1/2 phosphorylation being more pronounced. After overexpression of TRIM3, TLR3 activation decreased, and the MAPK signaling pathway was significantly inhibited. It is suggested that overexpression of TRIM3 can inhibit the activation of TLR3 and MAPK signaling pathways induced by DBV (Figure 2I).

We further investigate the effect of TRIM3 deletion on DBV-induced inflammatory signaling pathway activation and cytokines secretion. Peritoneal macrophages from TRIM3^−/− and wild-type (WT) mice were extracted. Total mRNA and cell supernatant were collected 24 h after DBV infection. The relative expression levels of IL-1β, IL-6, and TNF-α mRNA increased 24 h after infection, while the mRNA expression of the above cytokines showed a remarkable rise after TRIM3 knockout (Figure 3A). The protein expressions of IL-1β, IL-6, and TNF-α in the supernatant changed synchronously. It was significantly increased 24 h after DBV infection, and knockout of TRIM3 significantly promoted DBV-induced cytokine secretion (Figure 3B). After DBV infection in mouse peritoneal macrophages, TLR3 was activated, and proteins in the MAPK signaling pathway were phosphorylated. TLR3 activation was increased, and protein phosphorylation in the MAPK pathway was significantly promoted in the TRIM3 knockout group. These results suggest that the deletion of TRIM3 can promote the activation of TLR3 and MAPK pathway proteins induced by DBV infection (Figure 3C).

TLR3-mediated signaling is important for the innate immune response against RNA viruses (Zang et al., 2020). TLR3 expression was increased in DBV-infected THP-1 cells. To observe the expression of TLR3 in SFTS patients, flow cytometry was used to detect monocytes in 29 patients and 12 normal subjects. It revealed that compared with normal subjects, monocyte activation and TLR3 expression levels were significantly increased in SFTS patients (Figures 4A–C).

TRIM3 is an E3 ubiquitin ligase that can exert its immune regulatory effect by modifying proteins through ubiquitination. TRIM3 can inhibit DBV-induced elevated TLR3 expression, but the specific mechanism remains unclear. Western blot and IP analysis showed that in DBV-infected THP-1 cells with overexpression of TRIM3, DBV infection would lead to corresponding changes in TLR3 along with the change in TRIM3 expression, and there was an interaction between TRIM3 and TLR3 (Figures 4D, E). In DBV-infected TRIM3^OE THP-1 cell lines, TLR3 expression levels decreased compared to controls, which was associated with increased K48-linked ubiquitin (Figure 5A). In DBV-infected TRIM3^−/− mice's peritoneal macrophages, compared with the WT group, K48-linked ubiquitin was decreased and TLR3 expression was increased (Figure 5B). These results indicate that TRIM3 can bind to TLR3 and degrade TLR3 through K48-linked ubiquitination, which plays a crucial role in inhibiting cytokine storm.