In addition to playing a central role in hemostasis, platelets participate in regulating the immune response against infection with viruses such as DENV, Hantavirus, and SARS-CoV-2 [19, 38]. Here, we investigated the platelet response to SFTSV infection based on the transcriptome data of platelets purified from SFTS patients and revealed the functional platelet alterations. The results suggest that the functional changes in platelets in response to SFTSV infection may be associated with the outcomes of SFTS patients. First, in surviving SFTS patients, the platelet response was very active, as 99.6% of the identified DEGs (468/470) were upregulated, suggesting an enhanced platelet response to SFTSV infection. In platelets of fatal patients, although 71.22% of the identified DEGs (146/205) were upregulated, 28.78% of DEGs were downregulated, indicating suppressed platelet functions. Second, platelet functioning in surviving patients was more diverse than that in fatal SFTS patients, as shown by the various diverse functional modules in surviving patients compared to those in fatal patients. Third, the expression levels of the SFTS common DEGs, the fatal-related/specific DEGs, and the survival-related/specific DEGs significantly differed between the fatal and survival patients. Therefore, subsequent analyses focused on the functions in which the DEGs are involved and their differences between the fatal and surviving patients.
We found that platelets had enhanced functioning in mediating neutrophil activation, which is a common platelet response in SFTS, and this function was more enhanced in surviving patients than in fatal patients. This finding suggested a stronger platelet-neutrophil interaction in surviving patients. However, the fatal cases showed higher levels of NETs than the surviving cases, which was also described in a previous study that reported higher levels of cfDNA in severe SFTS patients than in mild/moderate patients [34]. Our results suggested that platelets cross-talk with neutrophils in SFTS to facilitate neutrophil activation and NET formation. However, the importance and molecular mechanism of platelets in mediating neutrophil activation in SFTS need to be addressed. Enhanced IFN signaling and regulation of the virus life cycle were the other common platelet responses in SFTS, which were also more enhanced in surviving patients than in fatal patients. A previous study reported that SFTSV infection and proliferation in human platelets potentiated platelet activation via in vitro mechanistic tests [35]. Recently, SFTSV infection in human platelets was studied both in vivo and in vitro [7, 27]. The findings of these studies indicated that SFTSV infection triggers platelet interactions with factors involved in virus infection and proliferation and the IFN signaling-dependent immune response. Besides the above common responses, human platelets exhibited unique functions, as shown by the different transcription levels of the platelet DEGs between the surviving and fatal SFTS patients. Based on our transcriptome data, impaired vesicle organization of platelets in fatal patients was consistent with less effective protein transport to the membrane. Upon platelet activation, cytokines and chemokines are translocated to the cell membrane or released to mediate platelet-platelet aggregation, platelet-leukocyte adhesion, intercellular signaling, and interaction with endothelial cells [39, 40]. Increased translocation of CD62P onto the membrane surface and release into plasma were the key biomarkers of platelet activation [39]. We recently found lower levels of CD62P released into the plasma of fatal patients than in surviving patients, demonstrating insufficient platelet activation in fatal SFTS patients [7]. Therefore, we speculated that impaired vesicle organization and less effective protein transportation may negatively affect platelet activation. Free histone H2A circulating in the blood system can bind to and activate platelets in Dengue fever patients [32]. In contrast, the transcriptome data in our study indicated enhanced regulation of histone H3 methylation, and the level was significantly higher in the platelets of fatal SFTS patients than in those of surviving patients. In chronic thromboembolic pulmonary hypertension (CTEPH), reduced histone trimethylation (H3K27me3) and increased histone acetylation (H3K27Ac) of the von Willebrand factor promoter (vWF), a critical factor mediating platelet adhesion in thrombosis, in the endothelium could promote vWF transcription by facilitating the binding of nuclear factor (NF)-κB2, which suggested that increased histone H3 methylation inhibited RNA transcription, resulting in reduced mRNA levels [41]. Since most of RNA transcripts in platelets are derived from progenitor megakaryocytes during thrombocytopoiesis [42], the altered transcriptome of platelets may also represent gene expression changes in megakaryocytes. We speculated that the enhanced histone H3 methylation in platelets may indicate suppressed RNA transcription in megakaryocytes, resulting in fewer RNA transcripts in the platelets of deceased patients than in those of surviving patients. This hypothesis was also supported by our transcriptome data, which showed that, compared to the fatal patients, the surviving patients had elevated transcript levels of mRNA in platelets. The dynamic regulation of RNA biogenesis by eukaryotic cells maintains the homeostasis of a stable mRNA pool through the reciprocal balancing of RNA synthesis rates and mRNA decay [43]. Changes in transcription impact the efficiency of RNA decay and are followed by a compensatory change in mRNA transcription [43]. Therefore, the significantly increased levels of RNA catabolic process-related transcripts in SFTS patients, especially surviving patients, may be a compensatory strategy to maintain the homeostasis of a stable mRNA pool in platelets.
Thrombocytopenia is the primary clinical symptom of SFTS. In fatal/severe patients, platelet counts could be even less than 50×109/L, and recovery from the severe drop is difficult [8, 44–46]. However, the pathogenesis of thrombocytopenia and the mechanisms underlying the severe drop in platelet numbers remain unclear. Our transcriptome data identified DEGs related to platelet apoptosis in surviving patients, and two recent studies reported platelet apoptosis is induced by SFTSV in vitro [35, 36]. Therefore, platelet loss may occur through cell death pathways, which could play a role in the pathogenesis of thrombocytopenia. The upregulated death-related DEGs were involved in pyroptosis, apoptosis, necroptosis, and autophagy based on the transcriptome data suggested accelerated platelet destruction via multiple pathways in SFTS, while the data also revealed that ferroptosis was suppressed. Pyroptosis in platelets contributes to thrombocytopenia and inflammation by activating inflammasomes and releasing cytokines such as IL-1β and IL-18, promoting platelet aggregation, endothelial permeability, and a cascade inflammatory response [47]. Platelet inflammasome activation has been confirmed in diseases such as dengue fever and sepsis [48, 49], suggesting the association of pyroptosis with thrombocytopenia. Here, increased platelet pyroptosis was demonstrated by the higher percentage of platelets expressing CASP1+ (4.16%, 95% CI: 2.76–5.56%) and GSDMD+ (2.11%, 95% CI: 1.34–2.88%) than the proportion of platelets expressing the other biomarkers and by its positive association with viremia, which suggested a leading role of pyroptosis in thrombocytopenia induced by SFTS. However, the mechanism and regulatory factors by which SFTSV infection induces platelet pyroptosis need to be investigated. Mild thrombocytopenia has been described in COVID-19 patients, which could be attributed to the antibody-induced apoptosis of platelets mediated by IgG via Fcγ-receptor IIA on the platelet surface [50]. A recent study found that in vitro incubation with SFTSV increased the surface level of phosphatidylserine on platelets, resulting in platelet apoptosis [35]. Another study showed that SFTSV infection induced platelet apoptosis in SFTSV-infected IFNar-/- mice and in human platelets after incubation with the virus [36]. Consistent with these findings, our study identified platelet apoptosis by transcriptome analyses and measured the percentage of platelets undergoing apoptosis in SFTS patients. Studies have suggested that SFTSV Gn and Gc could upregulate phosphatidylserine expression on the platelet surface [35] and that SFTSV Gn can trigger platelet apoptosis in an ROS-MAPK-dependent manner [36]. The molecular mechanism of apoptosis in platelets needs to be investigated in depth. In COVID-19 patients, platelets undergo necroptosis, as suggested by pathway enrichment analysis of the platelet transcriptome and increased levels of pMLKL [51]. Platelet necroptosis may also play a critical role in primary immune thrombocytopenia (ITP), as the MLKL mRNA levels in ITP patients were inversely correlated with platelet count [52]. In SFTS patients, although the increase in platelet MLKL levels was not as great as the levels of pyroptosis and apoptosis biomarkers, their increase still indicated a role of necroptosis in thrombocytopenia. Previous studies reported that SFTSV infection induces autophagic processes in HeLa cells [53] and inhibits autophagic degradation in Vero cells [54]. LC3 accumulation, which is required for autophagosome formation, was observed in both studies [53, 54]. In this study, increased levels of LC3B were observed, which suggested the occurrence of autophagy in the platelets of SFTS patients. Complete autophagic flux, from the formation of autophagosomes to autophagic degradation, is important for the number, size, and activation of platelets and could be a potential therapeutic target for hemorrhagic and thrombosis diseases [55]. Moreover, upregulated autophagy in platelets promotes platelet activation by decreasing hemostasis and thrombus formation [55], which reduce the number of platelets and shorten the lifespan of platelets [56]. Whether SFTSV infection induces complete autophagic flux in human platelets needs to be verified. In addition, ferroptosis is an iron-dependent cell death pathway that affects the functions of blood cells. Ferroptosis promotes platelet activation and thus causes thromboembolism [57] and contributes to megakaryopoiesis dysfunction [58]. Suppressed ferroptosis promotes megakaryocyte differentiation and platelet production [58]. This study revealed that ferroptosis was suppressed in the platelets of SFTS patients, as shown by the elevated levels of GPX4 and FTH1, which indicated enhanced megakaryocyte differentiation and increased platelet production to compensate for the loss of platelet numbers. This finding was also supported by our recent study, which found the elevated levels of thrombopoietin (a key factor to positively regulate platelet production) in SFTS patients and an increased capability of megakaryocytes to release platelet-like particles incubated with SFTSV [7]. Since SFTSV infects human platelets, which may promote virus spreading in the peripheral blood circulation [7, 35], we speculate that increased platelet death via pyroptosis, apoptosis, necroptosis, and autophagy are important antiviral strategies to prevent viral spreading in vivo. However, this process leads to excessive platelet loss, resulting in a drop in platelet counts. Ferroptosis may be a measure to compensate for platelet production.
Mice are valuable experimental models for studying the pathogenesis of SFTS and for developing antiviral drugs and vaccines [27]. SFTSV-infected C57BL/6 mice were first established to investigate virus infection mechanisms and pathogenesis in vivo [37]. Here, we evaluated the platelet response in this model and found that platelets from mice responded to SFTSV infection differently from those in SFTS patients. In contrast to the enhanced innate immune response in human platelets, the adaptive immune response of mouse platelets was activated, as shown by the significantly increased transcription levels of DEGs related to antigen processing and presentation. The association of host cytokine storms with disease severity in SFTS patients has been suggested in previous studies [59–61]. Moreover, the patients had deficient humoral immunity, especially fatal patients. SFTSV infects the B-cell lineage population and inhibits the secretion of plasma B cells, thus suppressing antibody production specific to SFTSV [62]. Platelets exert adaptive immune regulatory functions by sensing immune complexes through Fc γ RIIa to activate T cells and antigen presenting cells to dendritic cells, thus playing an irreplaceable role in fighting against viral infections [63, 64]. Therefore, the high-level acquired immune response by enhancing antigen presentation in mouse platelets may promote antibody production, which plays an important role in the nonlethal outcome of SFTSV infection in mice. A recent study characterized the platelet response in IFNAR-/- mice with SFTSV infection, which were found to have elevated platelet activation and apoptosis [36]. Although the study also characterized thrombocytopenia in wild-type C57BL/6 mice with SFTSV infection and found temporary platelet activation and apoptosis, a low dose of SFTSV (5×104 TCID50 per mouse) was used for the infection, and activation and apoptosis were observed only one day after infection [36]. In our study, C57BL/6 mice were infected using a high dose of virus (1×107 TCID50 per mouse), which was capable of causing platelet loss until 7 days post infection (Additional file 1: Fig. S2) and pathological lesions in mouse tissues as previously described [29]. Unlike the results in the recent study [36], platelet activation in this mouse model was insignificant [7], which might be the reason why the C57BL/6 mice did not exhibit a severe inflammatory cytokine storm [37, 65]. In contrast to model mice, SFTS patients have significantly high levels of platelet activation [7], as supported by our transcriptome data, which may contribute to the extensively active platelet response.
Our results, together with the findings in previous and recent studies [7, 35–37], concluded the possible mechanisms by which SFTS induces thrombocytopenia (Fig. 6). In humans with SFTSV infection, platelets are activated and then bound to SFTSV particles, which promotes phagocytosis by macrophages as previously described [1, 35]. Activated platelets cross-talk with neutrophils, which induces neutrophil activation and the formation of NETs. These interactions of platelets with macrophages and neutrophils enhance the antiviral response to promote viral clearance in patients but also reduce platelet numbers. SFTSV infection alters human platelet functioning, which may affect the regulation of viral replication and the inflammatory response mediated by platelets. This process also causes a reduction in platelets and should be further clarified. Moreover, SFTSV infection induces platelet death in humans via pyroptosis, apoptosis, necroptosis, and autophagy, accelerating platelet loss. In mice with SFTSV infection, platelets are consumed by macrophage phagocytosis [37]. Here, we found increased platelet-mediated antigen presentation, suggesting enhanced interaction of platelets with leucocytes to promote antibody production. Therefore, the humoral response and viral clearance in mice were promoted, thus resulting in platelet loss. However, we did not find increased NET formation in these mice (unpublished data), and the DEGs relating to neutrophil activation were not found in mouse platelets, suggesting that platelets were not consumed in this way. Moreover, platelet death in mice was not as severe as that in humans. All these results suggested a modest consumption or destruction of platelets in mice, which may be a reason that the mice had high antibody levels and quickly recovered from SFTSV infection. Furthermore, the different platelet responses in experimental mice due to SFTSV infection suggested that the platelet response and functional changes should be considered when using mice for studying the pathogenesis of SFTS and developing therapeutic strategies.
This study characterized the platelet response to SFTSV infection in humans and mice. However, there are still unsolved issues, such as the molecular mechanisms underlying platelet functioning and death, and the gaps in knowledge regarding the different responses between humans and mice remain to be investigated. There are also limitations in this study. RNA transcriptome analyses were performed using platelets purified from four fatal patients and four surviving patients, which were randomly selected from the identified SFTS patients. Therefore, the platelet response, as revealed by the data obtained from their platelets, may also be affected by the individual differences among them. Due to the severe loss of platelets in fatal patients, it was not easy to purify sufficient numbers of platelets from the blood samples of the fatal patients equal to those purified from the blood samples of the surviving patients. Therefore, we failed to analyze the association of viremia with necroptosis, autophagy, and ferroptosis in the platelets by including all fatal patients. Nevertheless, our results still provided insights for further research on platelet functioning and the development of disease therapy strategies that regulate platelet functioning.