ABSTRACT

   Severe Fever with Thrombocytopenia Syndrome Bunyavirus (SFTSV) is a
   highly lethal pathogen with expanding endemic regions in Asia. While
   primarily transmitted by ticks, recent evidence suggests potential
   airborne transmission, raising significant public health concerns. This
   study investigates the potential for respiratory transmission and
   pathogenesis using humanized NCG mice inoculated with SFTSV via
   subcutaneous injection challenge (SIC) or intranasal drop challenge
   (IDC). Both groups demonstrated rapid systemic dissemination, marked by
   viremia, weight loss, and multi-organ injury, with hemorrhagic
   manifestations observed in high-dose infection groups.
   Histopathological evaluations revealed lung pathology in the intranasal
   drop challenge mice, including extensive alveolar disruption and
   inflammatory cell infiltration. Transcriptomic analyses further
   confirmed that respiratory route inoculation resulted in heightened
   expression of inflammatory signalling pathways such as IL-17 and NF-κB,
   potentially contributing to severe local immunopathology. Subcutaneous
   infection provoked an earlier systemic immune response, with
   significant upregulation of antigen-processing genes in peripheral
   blood mononuclear cells. Nevertheless, both routes ultimately
   culminated in widespread injury to the liver, spleen, kidney,
   highlighting the systemic nature of SFTSV pathogenesis. These findings
   underscore the need for preventive strategies addressing respiratory
   spread.

   KEYWORDS: SFTSV, humanized mouse model, respiratory transmission
   potential, intranasal drop challenge, next-generation sequencing

Background

   Severe Fever with Thrombocytopenia Syndrome (SFTS) is a highly lethal
   infectious disease caused by Dabie bandavirus, initially discovered in
   the Dabie Mountains of China in 2009, also known as Severe Fever with
   Thrombocytopenia Syndrome Bunyavirus (SFTSV) [[50]1–3]. Subsequent
   studies have unveiled varying levels of endemicity across distinct
   regions of China, signalling an expansion of affected areas and a rise
   in morbidity rates over recent years [[51]2]. Reports have further
   emanated from other Asian territories, including Taiwan (China)
   [[52]4], Japan [[53]5]and South Korea [[54]6]. Data reveal that the
   cumulative mortality rate of SFTS infections in China approximates
   5.11% with 18,902 cases [[55]6]. Other regions exhibit mortality rates
   ranging between 6.3% and 30% [[56]2].

   While ticks are the primary vectors for SFTSV transmission
   [[57]1,[58]7], epidemiological evidence indicates that human-to-human
   spread can occur through contact with blood, secretions, or excreta of
   infected patients [[59]8–11]. Recent case reports highlight the
   possibility of airborne transmission or respiratory route
   transmissibility. Outbreak investigations in Xinyang (China) [[60]12],
   South Korea [[61]13], and Zhejiang (China) [[62]14] suggest that SFTSV
   may infect individuals through droplets or mucosal membranes, even when
   standard protective measures are in place. Additionally, study has
   confirmed the airborne transmission potential of SFTSV through a mouse
   aerosol infection model [[63]15]. These findings raise concerns that
   conventional tick-bite prevention alone may be insufficient to curb
   SFTSV spread.

   Although observations suggest the possibility of aerosol transmission,
   limited research has been conducted on the capacity and mechanisms of
   SFTSV respiratory route transmissibility. To explore this, a nose-drop
   inoculation model was established to study SFTSV infection through the
   upper respiratory route mucosa or lungs. Traditional mouse models often
   fail to replicate human immune responses [[64]15,[65]16]. To address
   this limitation, humanized NCG mice were used [[66]17]. These mice
   better simulate human immunity, providing a more relevant platform to
   study SFTSV infection [[67]18,[68]19]. The study compared infection
   outcomes between subcutaneous injection challenge (SIC) and intranasal
   drop challenge (IDC). Key parameters such as viral load dynamics, organ
   injuries, immune responses, and disease outcomes were analyzed to
   determine whether SFTSV can establish infection through the respiratory
   tract. The findings aim to improve understanding of SFTSV pathogenesis
   and guide public health strategies and therapeutic interventions.

Method

Viruses, cells, and animals

   In this study, the SFTSV-JS-14 virus strain was obtained from the
   Jiangsu Provincial Center for Disease Control and Prevention. Vero
   cells (provided by the Naval Medical University) were cultured in DMEM
   (Gibco, Beijing, China) supplemented with 5% fetal bovine serum, 100
   U/ml penicillin, and 100 μg/ml streptomycin. Employing gene editing
   techniques, NCG (NOD/ShiLtJGpt-Prkdc^em26Cd52Il2rg^em26Cd22/Gpt,
   GenPharmatech Co., Ltd, Nanjing, China) mice were engineered based on
   the NOD/ShiltJGpt strain, with knockouts of Prkdc (DNA-activated
   protein kinase) and Il2rg (common gamma-chain receptor) genes,
   resulting in low B cell and T cell levels [[69]17,[70]20].

Establishment of humanized mouse infection model

   Each mouse was administered an intraperitoneal injection of 500 μl of
   Peripheral Blood Mononuclear Cells (PBMCs) at a concentration of
   3 × 10^7/ml [[71]17,[72]21]. 100 μl blood sample was collected on day
   21 for humanization assessment via flow cytometry. Humanization was
   confirmed when CD3 ^+ CD45^+ cells of PBMCs exceeded 8% [[73]17].

   The humanized mice were assigned to the following groups: SIC (5 × 10^5
   TCID[50] SFTSV, n = 10), high-dose IDC (5 × 10^5 TCID[50], n = 10),
   moderate-dose IDC (5 × 10^4 TCID[50], n = 10), low-dose IDC (5 × 10^3
   TCID[50], n = 10), SIC sequencing (5 × 10^5 TCID[50], n = 8), IDC
   sequencing (5 × 10^5 TCID[50], n = 8), and control (n = 12). IDC mice
   received 50 μl of SFTSV via nasal drops post-anesthesia, while SIC mice
   received 50 μl subcutaneously. All mice were housed in specific
   pathogen-free (SPF) facility at 22–24°C, with 50–70% relative humidity
   and a 12-hour light cycle (8:00–20:00). Their weight, temperature,
   activity, and food intake were recorded daily.

   Non-sequencing group: On days 0, 1, 3, and 7 post infection (dpi), 150
   μl of blood was collected from each mouse. The plasma collected on dpi
   0 was previously used for flow cytometry. On dpi 14, surviving mice
   were euthanized, and blood was collected along with tissues (heart,
   lung, spleen, liver, kidney, testis, and brain). Any mouse that
   displayed severe adverse symptoms or died prematurely was immediately
   sampled by the same procedure. Sequencing group: No daily blood
   collection was performed. On dpi 3 and 7, three mice from each infected
   group were selected for blood collection and euthanasia, followed by
   tissue sampling. On dpi 3, three control mice were randomly chosen for
   the same procedure.

   Blood samples were mixed with EDTA and stored at 4°C. Tissue samples
   were either stored in liquid nitrogen or fixed in paraformaldehyde for
   subsequent histological analysis. The detailed flowchart was provided
   in Appendix Figure 1. The flow cytometry results of humanization were
   shown in Appendix Figure 2.

Enzyme-linked immunosorbent assay (ELISA) for plasma antibody and cytokine
detection

   Commercial ELISA kits were used to measure SFTSV-specific antibodies
   (Human SFTSV Ab Kit, JM-03042M2, Nanjing Jinmei, China), IL-6 (Human
   IL-6 ELISA Kit, LA137702H, Nanjing Lapuda, China) and Tumor necrosis
   factor α (TNF-α) (Human TNF-α ELISA Kit, LA137701H, Nanjing Lapuda,
   China) levels in plasma. Plasma was diluted due to limited volume. The
   minimum detection limits were 30 pg/ml for SFTSV antibody, 15.5 pg/ml
   for IL-6, and 117 pg/ml for TNF-α.

Real-time polymerase chain reaction (PCR) for viral load detection

   100μg Samples were finely pulverized and mixed with 900 μl of
   physiological saline, while plasma was diluted from 20 μl to 200 μl.
   SFTSV RNA from both tissue and plasma was extracted using the QIAamp
   MinElute Virus Spin Kit. The viral load in tissues and plasma was
   determined using the SFTSV RNA detection kit (Detection Kit for Severe
   Fever with Thrombocytopenia Syndrome Bunyavirus, DA0340, Daan Gene,
   China). The minimum detectable concentrations were 5 × 10^3 copies/ml
   for plasma and 5 copies/mg for tissue.

Hematoxylin and eosin (H&E) staining

   Tissues (lung, liver, spleen, kidney, and brain) were fixed in
   paraformaldehyde for 24 hours, paraffin-embedded, sectioned, and
   stained with hematoxylin and eosin (H&E). For lung tissues, injury was
   semi-quantitatively scored (1–5 scale) based on alveolar integrity and
   inflammatory infiltration (including macrophages, lymphocytes,
   neutrophils, eosinophils, pleuritis, bronchiolar and perivascular
   inflammation, edema, and epithelial necrosis). The average scores were
   classified as mild (1, +), moderate (2–3, ++), or severe (4–5, +++),
   and pathological differences among IDC, SIC, and control groups were
   compared based on these scores. Histopathological changes including
   structural integrity, inflammatory cell infiltration, cellular
   necrosis, and morphological abnormalities were qualitatively assessed
   in liver, spleen, kidney, and brain tissues, and the differences among
   IDC, SIC, and control groups were comparatively analyzed.

Next-generation sequencing

   After euthanasia, RNA was extracted from PBMCs and lung tissues of the
   sequencing and control groups (RNeasy Plus Mini Kit, QIAGEN), with
   concentration evaluated by Qubit, purity and integrity by Agilent 5400.
   Library preparation and sequencing were performed on the Illumina
   NovaSeq 6000 platform. The data were aligned to a reference genome
   using HISAT2 (v2.0.5), and read counts per gene were calculated with
   featureCounts (1.5.0-p3). Differentially expressed genes were
   identified via edgeR (3.22.5) with a cutoff of |log2(FC)| ≥ 1 and
   padj ≤ 0.05. GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes
   and Genomes) pathway enrichment analysis was performed using
   clusterProfiler (3.8.1).

Statistical analysis

   All experimental data were analyzed using R software. Unless otherwise
   specified, P ≤ 0.05 was considered statistically significant.

Laboratory safety and ethics

   All experiments involving infected animals and handling of infectious
   samples were conducted in the BSL-3 laboratory, while nucleic acid
   testing was performed in the BSL-2 laboratory. The animal experiments
   were approved and registered by the relevant CDC facility, adhering to
   the ARRIVE guidelines and Directive 2010/63/EU of the European
   Parliament on the protection of animals used for scientific purposes.

Results

Rapid onset of infection symptoms post SFTSV intranasal exposure

   After SFTSV exposure, infection symptoms quickly appeared in mice. Both
   SIC and IDC routes induced infections. In the IDC-high group, 9 mice
   were infected (9 tested positive for viral RNA, 3 positive for
   SFTSV-specific antibodies), and 7 mice died during the observation
   period. In the IDC-moderate group, 9 mice were infected (8 tested
   positive for viral RNA, 5 positive for antibodies; notably, 1 mouse was
   RNA-negative but antibody-positive), and 6 mice died during the
   observation period. In the IDC-low group, 6 mice were infected (5
   tested positive for viral RNA, 2 positive for antibodies; notably, 1
   mouse was RNA-negative but antibody-positive), and 4 mice died during
   the observation period. In the SIC group, all 10 mice were infected (10
   tested positive for viral RNA, 5 positive for antibodies), and 6 mice
   died during the observation period. ([74]Figure 1A and B). Most deaths
   occurred in the second week ([75]Figure 1B). Symptoms such as bloody
   stool, subcutaneous hemorrhage, hematuria, and cerebral hemorrhage were
   observed in 6 mice, including 3 in the high-dose IDC group, one in the
   moderate-dose IDC group, and 2 in the SIC group. (Appendix Figure 3)
   Additionally, weight loss was consistently observed throughout the
   infection period in the SIC and IDC groups, particularly in the
   high-dose IDC group and moderate-dose IDC group (P < 0.05) ([76]Figure
   1C). In the control group, no fatalities or symptoms were observed, and
   they showed a slight increase in weight ([77]Figure 1A, B and C).

Figure 1.

   [78]Figure 1.
   [79]Open in a new tab

   Progression of SFTSV Infection in Mice: Survival, Viral Load, and
   Immune Response Dynamics Across Infection Groups. A. Distribution of
   total, infected, and deceased mice across experimental groups. B.
   Survival probability following SFTSV infection across different
   infection routes and doses (log-rank test, p = 0.11). C. Longitudinal
   changes in body weight during infection. D. Infection status of
   individual mice over time by group. E. Plasma viral load (copies/mL)
   measured at multiple time points post-infection. F. SFTSV-specific
   antibody levels (pg/mL) in plasma at days 0, 7, and 14. G. Plasma IL-6
   concentrations (pg/mL) in individual mice over time. H. Number of mice
   with elevated TNF-α levels above 117 pg/mL detection threshold. Groups:
   Control (n = 12), SIC (subcutaneous injection challenge group, n = 10),
   IDC-High (high dose intranasal drop challenge group: 5×10^5 TCID₅₀, n =
   10), IDC-Moderate (moderate dose: 5×10⁴ TCID₅₀, n = 10), IDC-Low (low
   dose: 5×10³ TCID₅₀, n = 10)

   Assessments showed that on day 1 before exposure, no SFTSV RNA was
   detected in any of the mice. By dpi 1, infection was detected in each
   group, with positive rates ranging from 20% to 50%, and infection rates
   continued to rise over time. By dpi 14, detection rates in the infected
   groups ranged from 50% to 100% ([80]Figure 1D). In all infected groups,
   plasma viral loads increased over time across all infection groups
   (P < 0.001) ([81]Figure 1E), but with no significant differences
   between the groups (Appendix Figure 4A). SFTSV antibody detection
   showed no antibodies present on day 0. With disease progression,
   virus-specific antibody levels in infected mice notably increased. At
   dpi 7, antibody concentrations ranged from 30 to 100 pg/mL across
   infected groups, rising further to 100–370 pg/mL by dpi 14 ([82]Figure
   1F).

   Plasma Interleukin 6 (IL-6) concentrations increased significantly
   across IDC high, IDC moderate and SIC groups as the infection
   progressed ([83]Figure 1G and Appendix Figure 4B). Notably, in dpi14,
   IL-6 levels in the IDC high and IDC moderate groups remained
   significantly higher compared to the IDC low group (Appendix Figure
   4C). TNF-α detection was sporadic, with one instance in the high-dose
   IDC group and two in the moderate-dose IDC group initially. By dpi 14,
   TNF-α was detected 2 in both the high-dose and moderate-dose groups,
   and 1 in the SIC group ([84]Figure 1H).

SFTSV invaded multiple organs and caused substantial injury

   Histopathological examination revealed substantial organ damage induced
   by SFTSV infection, with lung tissues showing clear differences across
   groups ([85]Figure 2A). Control group lungs showed intact alveolar
   structures with mild inflammation. SIC group lungs exhibited increased
   infiltration of macrophages and lymphocytes, predominantly
   moderate-to-severe, along with notable perivascular inflammation and
   moderate interstitial inflammation. IDC group lungs presented the
   highest severity, characterized by extensive inflammatory infiltration
   predominantly scored as severe, widespread perivascular inflammation,
   and marked interstitial inflammatory responses ([86]Figure 2B).

Figure 2.

   [87]Figure 2.
   [88]Open in a new tab

   Histopathology and Viral Dynamics in Tissues of SFTSV-Infected Mice. A:
   Comparative Histopathology of Lung Tissues Control Group: Lung Tissue:
   Lung tissue exhibits intact pulmonary architecture, well-defined
   bronchioles, tightly arranged blood vessels, and clearly structured
   alveoli. Alveolar spaces show moderate and uniform expansion, with mild
   perivascular inflammatory cell infiltration (long arrows) observed.
   Lung IDC Group: a: Lung sections display moderate histopathological
   changes, characterized by focal mononuclear cell infiltration (long
   arrows) extending into alveolar and perivascular regions. b: Severe
   structural disruptions in alveolar integrity are observed, accompanied
   by fibrin-like exudates and necrotic cellular debris. c: Numerous
   infiltrating cells with eosinophilic cytoplasm (long arrows) and
   abundant foamy cells (arrowheads) are observed. The overall pulmonary
   structure shows severe damage indicative of exacerbated viral
   pathogenicity. Lung SIC Group: a and b: Lung tissues demonstrate severe
   pathological alterations, including extensive mononuclear cell
   infiltration (long arrows), pronounced alveolar structural disruption,
   and indistinct vascular boundaries. c: Multiple focal infarctions
   (pentagrams) occur within alveolar regions, characterized by extensive
   epithelial and endothelial cell necrosis, fibrin-like remnants (long
   arrows), and prominent inflammatory exudates. B: Inflammatory Scores of
   Lung Tissue Lung injury was evaluated using a semi-quantitative scoring
   system (1-5) based on key histopathological features. Shown are the
   average scores for four representative indicators: macrophage
   infiltration, lymphocyte infiltration, perivascular inflammation, and
   interstitial inflammation, across control, SIC, and IDC groups. The
   boxplots reveal distinct inflammatory patterns among the groups, with
   IDC lungs exhibiting the most pronounced changes. C: Comparative
   Histopathology of Liver, Kidney, Spleen, and Brain Control Group: Liver
   Tissue: Neat hepatocyte arrangements with no abnormalities. Mild focal
   infiltration of mononuclear cells is seen around the central vein, with
   clustered distribution (pentagrams). Spleen Tissue: Clearly defined red
   and white pulp regions, with a normal lymphocyte count. Occasional
   clusters of extramedullary hematopoietic cells are observed in the red
   pulp (pentagrams). Kidney Tissue: Orderly structures, with
   well-organized renal tubules and glomeruli. Brain Tissue: Healthy
   neuronal integrity, with normal neuron size, Nissl bodies, and moderate
   glial cells. Infection Group: Liver Tissue: Pericentral hepatocyte
   necrosis with inflammatory cell infiltration (pentagrams). Histological
   structure is disrupted, with hepatic cords replaced by inflammatory
   cells. Necrotic hepatocytes display strong eosinophilia with nuclear
   pyknosis, karyorrhexis, or karyolysis (long arrows). Spleen Tissue:
   Loosely organized structure; blurred red and white pulp boundaries;
   shrunken lymphocytes detached from tissues (long arrows). Kidney
   Tissue: Dilated renal tubules (long arrows); mononuclear cell clusters
   with eosinophilic cytoplasm around glomeruli (pentagrams). Brain
   Tissue: Disorganized structure; reduced neurons with eosinophilic
   cytoplasm, Nissl body loss, satellite cells (long arrows), activated
   microglia (arrowheads), and neutrophils (green long arrows). D:
   Tissue-Specific Detection Rates of SFTSV in Infected Mice. E: Tissue
   Viral Loads Across Experimental Groups.

   In the liver, there was noticeable hepatocyte necrosis around hepatic
   venules, along with massive macrophage infiltration replacing the
   hepatic cords. The spleen’s architecture was loosened, with a blurred
   demarcation between red and white pulp, and reduced lymphocyte volume
   with detachment from surrounding tissue. In the kidneys, mononuclear
   cell infiltration around the glomeruli disrupted the normal structure,
   while in the brain, neuronal details are obscured, and Nissl bodies are
   absent. Additionally, there was increased activation of microglial
   cells and neutrophil infiltration ([89]Figure 2C). No substantial
   histopathological differences between IDC and SIC were observed for
   these tissues.

   Tissue viral RNA detection results are shown in [90]Figure 2D, which
   presents the detection rates across different groups. Viral load
   analysis indicated that the spleen showed the highest levels of virus,
   while brain tissues showed relatively lower loads. The highest viral
   burdens were found in the high-dose IDC and SIC groups ([91]Figure 2E).

Transcriptomic profile of mice infected via different routes

   In this study, we performed transcriptomic sequencing of lung tissues
   and PBMCs from SIC and IDC groups at dpi3 and dpi7 aiming to compare
   the global and dynamic gene expression patterns induced by different
   infection routes. At dpi3, the IDC group showed 3684 differentially
   expressed genes (DEGs) in lung tissue (2017 upregulated, 1667
   downregulated), while the SIC group had 3102 DEGs (1872 upregulated,
   1230 downregulated). In PBMCs, 336 DEGs (115 upregulated, 221
   downregulated) were found in the IDC group, versus 373 DEGs (171
   upregulated, 202 downregulated) in SIC. By dpi7, DEG counts rose
   notably in the IDC lungs (4792 total; 2443 upregulated, 2349
   downregulated) compared to SIC lungs (2919 total; 1533 upregulated,
   1386 downregulated). PBMCs also showed increased DEGs in IDC (713
   total; 290 upregulated, 423 downregulated) relative to SIC (304 total;
   129 upregulated, 175 downregulated) at day 7. Volcano plots were
   provided in Appendix Figure 5 and 6.

   In lung tissue, several DEGs were consistently upregulated across both
   IDC and SIC groups, including genes such as IFITM3, CXCL13, SOCS3, and
   MT2. These genes are primarily involved in antiviral responses,
   chemokine signalling, and immune cell recruitment. Genes consistently
   downregulated (e.g. OAS1E, TRIM5, IFIT1, ISG20, etc.), associated with
   antiviral defense, immune regulation, and cellular structural
   integrity. In PBMCs, the consistently upregulated genes included FOS,
   ATF3, KLF4, SIK1 and NR4A1, primarily related to inflammation
   regulation and transcription factor activation. Several genes, such as
   VSTM4, CLEC12A, SIGLEC17P, and TRGV3, were consistently downregulate.
   The Venn diagram and heatmap of DEGs are shown in [92]Figure 3.

Figure 3.

   [93]Figure 3.
   [94]Open in a new tab

   Overlap and Expression Patterns of Differentially Expressed Genes
   (DEGs) in Lung Tissues and PBMCs During SFTSV Infection. Venn diagrams
   (left) show the number of differentially expressed genes (DEGs)
   identified in the lung tissues and PBMCs at dpi3 and dpi7, across
   subcutaneous injection challenge (SIC) and intranasal drop challenge
   (IDC) routes compared to uninfected controls. Heatmaps (right) display
   representative DEGs from the overlapping regions, reflecting
   transcriptional responses in both compartments.

   Gene Ontology (GO) enrichment analyses highlighted both common and
   distinct biological pathways activated by different infection routes
   and over the course of infection. In lung tissue, Several GO terms were
   enriched across both infection routes, including “antimicrobial humoral
   response”, “human immune response”, “leukocyte chemotaxis”, “cytokine
   activity” and “oxidative phosphorylation” indicating a general
   activation of innate immunity, inflammatory signalling, and metabolic
   adaptation during infection. However, distinct differences emerged
   between IDC and SIC infections. The IDC group exhibited marked
   enrichment of “antimicrobial humoral response”, “chemokine activity”
   and “oxidative phosphorylation”, suggesting that infection via the
   respiratory route induces a more pronounced stimulation in the lungs
   ([95]Figure 4). In PBMCs, The MHC protein complex, antigen processing
   and presentation and plasma membrane protein complex pathways were
   commonly enriched across all groups, reflecting the activation of
   immune recognition, and intercellular signalling processes., suggesting
   that infection via the subcutaneous route may elicit a more direct
   stimulation of the systemic immune response. Compared with the SIC
   group, pathways related to chemokine activity and cytokine activity
   were more significantly enriched in the IDC group. Although the antigen
   processing and presentation pathway was enriched under both infection
   routes, its activation was relatively less prominent in the IDC group.
   Similarly, pathways associated with leukocyte migration and
   inflammatory response were enriched during the dpi7 stage. The
   enrichment was more pronounced in the IDC group ([96]Figure 4).

Figure 4.

   [97]Figure 4.
   [98]Open in a new tab

   GO Enrichment Analysis of Differentially Expressed Genes in Lung
   Tissues and PBMCs at dpi3 and dpi7. Gene Ontology (GO) enrichment
   analysis was performed on differentially expressed genes (DEGs)
   identified in lung tissues (top panels) and PBMCs (bottom panels) at
   dpi3 and dpi7, in both subcutaneous injection challenge (SIC) and
   intranasal drop challenge (IDC). Dot plots display significantly
   enriched GO biological processes and molecular functions, with dot size
   representing the number of enriched genes and color indicating the
   adjusted p-value.

   Differential gene expression analysis between the IDC and SIC groups
   was conducted, followed by KEGG pathway enrichment analysis to explore
   their functional differences. In lung tissues ([99]Figure 5), multiple
   inflammation and immunity related signalling pathways were
   significantly upregulated at dpi 3, including cytokine-cytokine
   receptor interaction, IL-17 signalling pathway, and pathways associated
   with viral protein – cytokine receptor interactions. In addition,
   pathways related to cell proliferation and homeostasis maintenance,
   such as the p53 signalling pathway and cell cycle, also showed an
   upward trend. By dpi 7, enrichment of the TNF signalling pathway and
   chemokine signalling pathway was observed, with further enhanced
   enrichment of the IL-17 signalling pathway, cytokine-cytokine receptor
   interaction and viral protein – cytokine receptor interaction pathway.
   In contrast, several pathways associated with energy metabolism and
   cellular signal regulation were relatively downregulated, including
   glycolysis/gluconeogenesis, ECM-receptor interaction, protein digestion
   and absorption, calcium signalling pathway, cAMP signalling pathway,
   and the PPAR signalling pathway in dpi3. Notably, protein digestion and
   absorption and calcium signalling pathway remained downregulated on dpi
   7. In PBMCs ([100]Figure 5), several immune-related pathways were
   enriched at dpi 3, including natural killer (NK) cell mediated
   cytotoxicity, cell adhesion molecules, phagosome, and antigen
   processing and presentation. These findings indicated that, during the
   early stage of infection, circulating immune cells exhibited a robust
   activation of recognition and cytotoxic responses against viral or
   pathogenic stimuli, along with enhanced cellular interactions,
   phagocytic activity, and antigen presentation functions. In the dpi 7,
   continued upregulation of immune regulatory pathways was observed in
   the IDC group, including primary immunodeficiency, PI3K-Akt signalling
   pathway, NF-κB signalling pathway, Fc gamma R-mediated phagocytosis,
   and cytokine-cytokine receptor interaction. In contrast, pathways
   related to cellular senescence, endocytosis, graft-versus-host disease,
   cell adhesion, and antigen presentation were relatively downregulated.

Figure 5.

   [101]Figure 5.
   [102]Open in a new tab

   Comparative KEGG Pathway Enrichment Analysis Between IDC and SIC Groups
   in Lung Tissues and PBMCs. KEGG pathway enrichment analysis was
   performed to compare the transcriptomic differences between the
   subcutaneous injection challenge (SIC) and intranasal drop challenge
   (IDC) in both lung tissues (left) and PBMCs (right) at dpi3 and dpi7.
   Enriched pathways were separated into relatively upregulated (top) and
   downregulated (bottom) in the IDC group compared to the SIC group. Dot
   size represents the number of enriched genes per pathway, while color
   indicates the adjusted p-value.

Discussion

   In this study, we aimed to compare the pathogenic and immunological
   consequences of SFTSV infection via intranasal drop and subcutaneous
   injection in a humanized NCG mouse model. Our comprehensive analysis
   integrates clinical signs, histopathological damage, plasma cytokine
   profiles, viral loads. High-throughput RNA-sequencing was performed on
   lung tissues and PBMCs from infected mice at dpi 3 and dpi 7. Our
   findings demonstrate that SFTSV, regardless of whether it enters via
   nasal mucosa or subcutaneous tissue, can rapidly proliferate in the
   host, leading to viremia and uncontrollable inflammation and
   hemorrhage. By dpi 14, plasma viral loads steadily increased in all
   groups, correlating with severe clinical manifestations. Hemorrhagic
   symptoms were common in both IDC and SIC infection groups, especially
   at moderate and high doses, suggesting a synergistic role of pulmonary
   inflammation in exacerbating systemic disease. Meanwhile, in the IDC
   group, the administered doses showed a significant correlation with
   infection rates, suggesting a link between SFTSV transmissibility and
   viral load.

   Significant symptoms appeared swiftly after infection, that both IDC
   and SIC routes led to rapid onset of symptoms, with detectable viral
   RNA in peripheral blood by day dpi 1. Notably, the IDC route, despite
   its mucosal entry point, did not delay systemic dissemination,
   suggesting that SFTSV efficiently breaches epithelial barriers.
   Previous studies suggested that SFTSV primarily targets monocytes and
   dendritic cells (DCs) [[103]22–24], and dendritic cell abundant in
   mucosal and lymphoid tissues [[104]25,[105]26]. This highlights a
   potential mechanism whereby SFTSV exploits DCs for respiratory route
   transmissibility, as DCs may inadvertently assist the virus in crossing
   epithelial barriers and entering the lymphatic system [[106]10]. Once
   inside, SFTSV leverages the immune system’s cellular machinery for
   systemic dissemination, a phenomenon reminiscent of other viral
   infections such as Dengue and measles [[107]27–29]. Nevertheless,
   further experimental evidence is needed to fully validate this proposed
   mode of transmission.

   Histological examination revealed severe pulmonary injury in the IDC
   group, which exhibited more extensive mononuclear infiltration,
   alveolar wall thickening, and interstitial edema compared to the SIC
   group. This difference is plausible, as direct nasal inoculation
   presumably primes an immediate local immune response in lung, leading
   to localized immunopathology. These findings are further supported by
   transcriptomic data showing significant upregulation of proinflammatory
   pathways, including IL-17, TNF, and NF-κB signalling, in IDC lung
   tissues. IL-17, a key cytokine in mucosal immunity that promotes
   neutrophil recruitment and sustained inflammation [[108]30–32], was
   notably elevated in IDC lungs, suggesting that respiratory infection
   uniquely triggers local cytokine storms contributing to alveolar damage
   and vascular leakage [[109]33]. Notably, such localized inflammatory
   responses may drive systemic disease through increased vascular
   permeability and cytokine spillover into circulation [[110]34,[111]35].
   In contrast, SIC lungs showed moderate inflammatory responses, with
   less severe histological disruption and relatively lower expression of
   inflammatory genes. By dpi 7, the synergy among IL-17, TNF, and NF-κB
   can set off a “cytokine storm,” fuelling alveolar necrosis,
   perivascular inflammation [[112]36], and foam cell infiltration –
   pathological features consistent with those observed in the IDC group
   [[113]33,[114]37].

   Beyond the lungs, SFTSV induced widespread histopathological damage
   across multiple organs, including the liver, spleen, kidneys, and
   brain. No substantial histopathological differences between IDC and SIC
   were observed for these tissues, likely because sampling predominantly
   occurred at advanced disease stages. Nonetheless, these tissues showed
   common pathological features, including hepatocyte necrosis, glomerular
   infiltration, and neuronal damage. In the spleen, blurred boundaries
   between the white and red pulp alongside significant lymphocyte
   depletion indicated a severe compromise of this critical immune hub,
   closely associated with the targeting of macrophages by SFTSV
   [[115]38,[116]39]. These macrophages, once heavily infected, not only
   failed to perform normal antigen presentation and pathogen clearance
   but also became replication sites for SFTSV [[117]40,[118]41], leading
   to extensive viral proliferation and systemic immunosuppression
   [[119]34,[120]42,[121]43]. Moreover, immunopathological observations
   revealed that SFTSV infects neurons, inducing cerebral hemorrhagic
   lesions. Such findings align with neurological manifestations reported
   in SFTSV encephalitis cases [[122]31] and mirror patterns observed in
   other viral brain infections, including COVID-19 [[123]35].

   Significant increases in plasma IL-6 levels during infection reflect a
   robust innate immune reaction to viral intrusion, consistent with
   previous findings [[124]17,[125]30]. Although TNF-α detection was
   suboptimal, it remained detectable in moderate – and high-dose groups
   at later stages; together, these cytokines promote vascular leakage and
   tissue necrosis, resulting in widespread hemorrhage
   [[126]39,[127]44–46].

   Focusing on PBMCs, we observed that subcutaneous infection was strongly
   associated with the enrichment of “MHC protein complex” and “antigen
   processing and presentation” pathways [[128]3,[129]30], suggesting a
   more direct and earlier onset of systemic-specific immunity compared to
   IDC. Thus, while both routes effectively initiate immune activation,
   the SIC route appears to prime PBMCs earlier for systemic-specific
   responses, whereas IDC elicits a delayed but highly localized
   inflammatory storm in the respiratory tract.

   The p53 signalling pathway and cell cycle processes were also impacted.
   While some upregulation of p53 was initially noted, the net effect over
   time seemed to be insufficient to curb viral replication. Proteomic
   studies [[130]47] further highlight the upregulation of antiapoptotic
   proteins, such as SOD2 and BCL3, which suppress apoptosis and extend
   the survival of infected cells, creating a favourable environment for
   viral replication [[131]30]. Moreover, the viral NSs protein plays a
   critical role in manipulating the host cell cycle, inducing G2/M phase
   arrest to optimize conditions for viral replication [[132]38,[133]48].
   These combined strategies amplify tissue damage and systemic
   inflammation, contributing to the severe clinical manifestations of
   SFTSV infection, including multi-organ dysfunction and hemorrhagic
   symptoms.

   Hemorrhage and organ injury highlight the challenges of treating severe
   SFTSV infection, emphasizing the importance of individualized
   therapeutic approaches. Suppressing hyperactive inflammatory pathways,
   such as IL-17 and NF-κB, should be balanced to avoid compromising
   antiviral defense. In parallel, targeting virus-induced suppression of
   apoptosis and cell cycle checkpoints, such as restoring p53 signalling,
   could enhance immune cell viability and regenerative capacity
   [[134]49]. These strategies must be tailored to disease progression,
   ensuring early intervention focuses on viral load reduction, while
   later stages address immune exhaustion and tissue repair [[135]50].

   The study has several limitations. First, we used humanized NCG mice,
   which, while partially reflecting the human immune system, still have
   key immunological and physiological differences from humans, including
   a low B-cell level [[136]51]. Second, due to sample volume and
   technical constraints, our plasma cytokine analysis focused on IL-6 and
   TNF-α, leaving out other important markers such as IL-10 and TGF-β.
   Third, although transcriptomic analyses indicated changes in
   immune-related pathways, we did not perform functional validation, so
   the exact roles of these pathways remain unclear. Finally, our study
   employed intranasal drop challenge (IDC) rather than direct
   animal-to-animal transmission experiments, which does not fully
   replicate the natural process of respiratory transmission, including
   aerosol generation, droplet spread, and subsequent infection.

   In conclusion, this study employed a humanized NCG mouse model to
   investigate the pathogenic features and immunological responses of
   SFTSV via two routes of infection: subcutaneous injection challenge
   route or intranasal drop challenge route. The results indicate that
   regardless of whether the virus enters through intranasal or
   subcutaneous pathways, it rapidly disseminates systemically and induces
   severe immunopathological damage. The subcutaneous route triggered a
   more direct and earlier systemic immune response. The intranasal
   infection group displayed a more intense localized immune reaction in
   the lungs. Moreover, in both infection groups, multi-organ damage and
   hemorrhagic manifestations were observed at later stages.
   Transcriptomic data further revealed that intranasal infection
   prominently activates IL-17, TNF, and NF-κB, contributing to a more
   pronounced “cytokine storm” and tissue injury. By contrast,
   subcutaneous infection more strongly invoked antigen presentation and
   adaptive immune pathways. These findings confirm that SFTSV can
   efficiently utilize the respiratory route for transmission and cause
   severe pulmonary injury, highlighting a need for enhanced public health
   measures targeting both tick-bite prevention and respiratory
   protection.

Supplementary Material

   Appendix Figure.pdf
   [137]TEMI_A_2511134_SM1184.pdf^ (1.4MB, pdf)

Acknowledgments