Background
Severe fever with thrombocytopenia syndrome (SFTS) is a novel infectious disease caused by SFTS virus (SFTSV), a new type of bunyavirus, which is a phlebovirus in the Bunyaviridae family [
1‐
3]. SFTSV undergoes rapid changes owing to evolutions, gene mutations, and reassortments between different strains of SFTSV. From 2010 to 2017, SFTS cases were reported in 23 provinces of China and the numbers are increasing each year [
4]. SFTS is often characterized by fever, thrombocytopenia, leukocytopenia and gastrointestinal symptoms. The clinical symptoms of SFTS patients usually become exacerbated quickly to the multiple organ dysfunction syndromes (MODS). Due to this reason, the mortality rate of SFTS reached a high number of 30%, with higher risks to elder people [
1]. SFTS greatly threatened public health. Further explorations SFTS pathological mechanisms are urgently needed.
The number of neutrophils was abnormally decreased in SFTS patients [
5‐
7]. Although the number of reported SFTS has increased in the recent years, the neutrophils remains one of the less studied in SFTS. It is well known that neutrophils are the first immune cell population recruited to sites of infection [
8] and their functions in virus infection was controversial because neutrophils exhibit both protective and pathological functions [
8]. Accumulating evidence support the existence of distinct neutrophil subsets in circulating blood in density gradient preparations and this population of neutrophils is termed LDNs and NDNs [
9‐
12].Circulating LDNs are seldomly detected in healthy controls, however, LDNs are abundant in mycobacterial infections and are associated with the severity of tuberculosis [
11].More importantly, the number of LDNs were increased in the circulating blood of patients with human immunodeficiency virus [
13]. It has reported that LDNs display impaired neutrophil function and immunosuppressive properties, characteristics that are in stark contrast to those of NDNs. Recently, we have demonstrated the existence of abnormal percentage of LDNs in the circulation blood of SFTS patients [
14]. However, the source of LDNs in SFTS remains unclear. Therefore, this study was to explore the origin and functional roles of LDNs in SFTS, and may explain pathogenesis of SFTS disease.
Methods
Patients
Forty SFTS patients and 40 volunteers were recruited from the Department of Infectious Diseases of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology between May 2015 and July 2017. The procedures were in accordance with the ethical standards of the Helsinki Declaration. Subjects with the SFTS were diagnosed according to clinical symptom and laboratory examination.
The median interval of the SFTS patients between the onset of illness and extraction of peripheral blood at admission was 3 days (range, 1–5 days). There were 25 women and 15 men. The median age of the patients was 59.50 ± 8.85 years (range, 40–72 years). The healthy volunteers without any diseases, were recruited from April 2015 to July 2017. The volunteers’ health conditions were verified via health examination, and the volunteers were subjected to no treatments with drugs prior to and during the investigation phase. There were 20 women and 20 men. The median age of normal controls was 54.70 ± 7.82 years old (range, 40–66 years). There were no significant differences in the average age and sex ratio between SFTS patients and controls.
Blood samples
Collection of blood from SFTS patients and healthy volunteers was approved by Tongji Medical College, Huazhong University of Science and Technology institutional review board. Peripheral blood from SFTS patients or the healthy volunteer blood was drawn by venipuncture and collected using EDTA containers. Blood samples were transferred to the lab for analysis for different experiments.
Density separation
Peripheral blood was diluted with phosphate buffered saline (PBS) and was centrifuged on a cushion of Ficoll-Histopaque in a centrifuge tube for 30 min at 400×g at room temperature. LDNs were collected from the PBMCs interface, diluted with PBS and centrifuged for 5 min at 300×g. Red blood cells were eliminated by hypotonic lysis. The collected LDNs were washed twice with PBS. NDNs were collected from the granulocyte-erythrocyte pellet by brief hypotonic lysis and were washed twice with PBS. The cells were suspended respectively in RPMI-1640 for analyzing immediately by flow cytometry or other experiments. LDNs and NDNs were identified as CD15+, CD45 + .
Isolation of LDNs and NDNs
LDNs and NDNs were purified and performed on a FACS Vantage cell separator. Under sterile conditions, tube was incubated with APC-labeled CD15 +, PerCP-labeled CD45+ monoclonal antibody, then 488 mm laser activated fluorescence to separate cells, and 2% FBS in PBS as sheath fluid.
Adoptive transfers
The isolated NDNs from health volunteers were incubated for 4 h, 12 h, 24 h, 48 h in RPMI-1640 medium supplemented with 100 μl plasma of SFTS patient. The cells were harvested at different time points and isolated LDNs and NDNs using Histopaque 1077 and centrifuged for 30 min. The LDNs and NDNs were suspended in RPMI-1640 for analyzing by flow cytometry.
Measurements of SFTSV RNA in LDN, NDNs and serum from SFTS
Next, we tested the SFTSV load in LDNs and NDNs from SFTS patients and the SFTS serum. The Blood RNA Kit (Omega) were used to extract SFTSV RNA from serum, LDNs and NDNs of SFTS according to the manufacturer’s instruction. The SFTSV RNA in serum, LDNs and NDNs were detected by a commercially available quantitative real-time PCR diagnostic kit (SFDA, China), according to the manufacturer’s instructions.
Cytokine quantifications
We also detected inflammatory cytokines concentration from the serum of SFTS and normal controls, the cultured supernatant of LDNs and NDNs from SFTS. The LDNs, NDNs from SFTS were cultured in RPMI-1640/1% FBS for 48 h in the presence of SFTSV (3260 copies/ml).Supernatants were then harvested. The concentration of the cytokines IL-2, IL-4, IL-6, IL-8, IL-10, IL-17, TNF-α and IFN-γ were quantified using Human Cytometric Bead Array (CBA) Kit (BD BioScience) as described by the manufacturer’s instructions.
The endothelial cell cytotoxicity assay
The HUVECs were purchased from ATCCTM Cell Biology, USA and were cultured in RPMI-1640 medium supplemented with 10% FBS. The cells were grown in 5% CO2 at 37 °C. HUVECs were plated into a 6-well plate at a density of 2 × 105/well in RPMI-1640 supplemented with 10% FBS. Transwell (Corning, USA) inserts with an 0.4 μm pore size were inserted into the 6-well plates, while 900 μL (2 × 105cells) of activated and unactivated LDNs were, added respectively to the upper wells in RPMI-1640 supplemented with 10% FBS. These cells were then co-cultured in 5% CO2at 37 °C for 24 h or 48 h. After incubation, the HUVECs were trypsinized. Following the instructions of Annexin V-FITC apoptosis detection kit (BD Pharmingen TM, USA), apoptosis rate of HUVECs was detected by flow cytometry.
The ethics statement
All participants signed informed consents before participation. The research protocol was approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology.
Statistical analyses
Unless indicated otherwise, results were summarized as means and Standard Error of Mean (mean ± SEM). Results were analyzed using the statistical software package SPSS 18.0 (SPSS, an IBM Company, Armonk, NY, USA). Comparison of various groups was performed using Student’s t tests when the data were normally distributed; otherwise, the Mann-Whitney test was used. The correlation analysis of 2 variables among virus load, LDNs, NDNs, and cytokines were calculated using Pearson and Spearman test. P < 0.05 was considered statistically significant. The statistical graphs were performed by Graph Pad Prism 5.00 (GraphPad Software, San Diego, CA, USA).
Discussion
In our current work, we further demonstrated that circulating neutrophils in SFTS were presence of distinct subpopulations, LDNs and NDNs. SFTSV infections resulted in high proportion of LDNs in PBMCs. The proportion of LDNs not only elevated in mycobacterial infection [
11], but also increased in circulating blood of patients with viral infection [
13], cancer [
12]. Our results showed that the percentage of LDNs was (46.80 ± 14.1)% in circulating blood SFTS patients, increasing dramatically to the extent that LDNs often became the dominant circulating neutrophil population with The LDNs may play more important role in SFTS disease.
From prior studies, neutrophils are formed within the bone marrow during hematopoiesis in response to several cytokines, principally granulocyte colony–stimulating factor [
21]. Denny MF, et al. reported that the LDNs do not seem to represent a population of in vivo activated and degranulated neutrophils [
10]. We employed highly purified populations of the NDNs cells from the healthy donor and tested whether the transition between NDNs and LDNs can be mimicked in SFTS conditions, ex vivo. As expected, NDNs from the healthy donor could convert to LDNs under cultured with SFTS patients’ plasma, suggesting that the NDNs retained the developmental plasticity and could convert to LDNs which was one of the source increased percentage of LDNs in SFTS.
Neutrophils have the capacity to respond to a wide range of stimuli. There was not only presence of increased pro-inflammation cytokines, but also presence of SFTSV in SFTS patients’ plasma. At this moment, it is unclear which regulates the transition from NDNs to LDNs. Although we are making SFTS animal models, but the difficulty in handling SFTS animal model has hampered our study in vivo for confirming the experiments on transition NDNs to LDNs. Anyway, our results challenge the current concept of circulating neutrophils as a fully differentiated homogenous cell population that has limited plasticity.
We demonstrated that SFTSV load in LDNs and NDNs were diversified. From a global perspective, the load of SFTSV was lower in LDNs than that in NDNs of the SFTS patients. But careful analysis showed that the SFTSV loads of LDNs were much higher from severe SFTS patients than that from moderate SFTS patients.
Although the SFTSV were in LDNs, it was not clear whether this was due to uptake of the virus itself or by active infection and propagation of the virus within the LDNs. The neutrophils can actively take up IAV in vitro [
22]. One study showed that LDNs had the function of phagocytose and could phagocytose fewer FITC-labeled microbeads [
12]. Based on those studies, we inferred that LDNs in SFTS could engulf SFTSV and disseminated the viruses.
In this study we found that platelet number was lower and the APTT was longer, and the LDH level was higher when SFTSV load were high in LDNs. This phenomenon may be a contributing factor of bleeding by patients with SFTS, but the specific mechanism is not fully understood. In our study, 60% of SFTS patients were in moderate patient group, and 40% of SFTS patients were in severe patient group. That is why generally the load of SFTSV was lower in LDNs than that in NDNs of the SFTS patients. In clinical trials, almost all the SFTS patients were curable in moderate patient group, but almost all the death patients were in the severe patient group. Together, those findings suggested that high SFTSV load in LDNs was an indicator of SFTS disease severity and prognosis.
Controversy surrounds the neutrophil function in viral infection because neutrophils in viral infections were shown to have both beneficial and harmful effects [
23,
24]. On the one hand, the neutrophils exhibit a strong ability to mediate direct antiviral effects after viral infection. They can rapidly initiate an antiviral program and contributes to its clearance [
25]. On the other hand, neutrophils have also been proposed to act as “Trojan horses” for the dissemination of virus, acting as a vehicle for viral replication, proliferation [
26,
27]. LDNs from severe SFTS patients can engulf more SFTSV. It is not known whether LDNs from SFTS patients take up viral particles and kill them or act as a vehicle for viral replication, proliferation. Nonetheless, further investigations are needed to elucidate the functional role of LDNs in SFTSV infection.
The percentages of LDNs are not only elevated in bacterial infection patients PBMCs [
11], but also increased in viral infection patients PBMCs [
13]. LDNs may have special functional role in viral infection. Generally, SFTSV infection induces a cytokine storm in SFTS patients [
28]. Our results showed that LDNs secreted more IL-6, IL-8, IL-17 and TNF-α levels than that of NDNs. IL-6, IL-8, TNF-α are proinflammatory cytokines, indicating that the LDNs secreted higher proinflammatory cytokines than NDNs upon SFTSV infection, and these might be one of the sources of proinflammatory cytokines in SFTS.
Our results diaplayed that the active LDNs infected by SFTSV could induce more endothelial cytotoxic when compared to control LDNs. This observation that LDNs from SFTS were cytotoxic to the endothelium suggested that they might play an important role in the induction of vascular damage in SFTS. LDNs were active and could synthesize more cytokines. Based on those, we inferred that cytokines secreted by active-LDNs might be resulted in the violent immune reaction and resulting in vascular damage. The neutrophil function in virus infection exhibit both protective and pathological functions [
8]. Based on our findings as mentioned above, we reasoned that NDNs might play a protective role, whereas the LDNs may associate with proinflammation activity in SFTSV infection.