Introduction
Sepsis, defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, is a major public health concern [
1]. The in-hospital mortality rate of sepsis is recorded as over 10% worldwide, whereas that of septic shock is greater than 40% [
2]. Multiple organ dysfunction, including acute lung injury (ALI), is a common manifestation of sepsis. Sepsis is characterized by excessive inflammation in response to infection; however, the molecular mechanisms remain to be fully elucidated [
3].
Polymorphonuclear neutrophils (PMNs) are the most abundant leukocytes in mammals, playing crucial roles in innate immune responses during sepsis. In addition to degranulation and phagocytosis, neutrophils also release neutrophil extracellular traps (NETs) to ensnare and kill microbes [
4]. NETs are extracellular strands of decondensed DNA decorated with histones and neutrophil granule proteins. However, NETs may function as double-edged swords; overzealous NET formation during sepsis can lead to the development of multiple organ dysfunction [
5]. NETs have been found in the lungs and could induce lung endothelial injury mediated by extracellular histones, neutrophil granular proteins, and a tangled web of extracellular DNA [
6].
Platelets are not only elements of primary importance in hemostasis and thrombosis, but also essential elements of an integrated inflammatory response [
7]. Increasing evidence in vitro and in vivo indicates that platelets may contribute to the NET formation process [
8]. However, the mechanism underlying platelet-mediated NET formation remains unclear [
9]. Previous studies demonstrated that circulating plasma exosomes, which are small lipid packages that carry a variety of molecules, mostly originated from platelets, induces inflammation and myocardial dysfunction during sepsis [
10‐
12]. Exosomes serve as vehicles to transport nucleotides, lipids, and proteins between cells in physiological and pathological conditions [
13]. Study has shown that platelet-derived exosomes sustain autophagy-associated NET formation in systemic sclerosis [
14]. However, whether platelet-derived exosomes play a role in NET formation in sepsis has yet to be addressed.
MicroRNAs (miRNAs) are important modulators of gene expression as they induce messenger RNA degradation or prevent translation [
15,
16]. Exosome-contained miRNAs may serve as an important mediator of intercellular communication in sepsis [
17]. In addition, recent studies have also shown that human platelet high-mobility group box 1 (HMGB1) induces PMNs autophagy and NET formation [
14,
18]. Platelets are a source of HMGB1 protein, which is a ubiquitous nuclear and cytosolic protein [
19]. Released circulating HMGB1 mediates lethality in LPS-induced endotoxemia and cecal ligation and puncture (CLP)-induced polymicrobial sepsis mouse models [
20]. However, whether platelet-derived exosomal HMGB1 and/or miRNAs could modulate NET formation during septic shock needs to be further addressed.
Therefore, the current study aimed to test the hypothesis that exosomal HMGB1 and/or miRNAs from platelets might induce excessive NET formation in septic shock and subsequent acute lung injury. The results show that in sepsis, IκB kinase controls platelet-derived exosome secretion, which in turn induces NET formation. Exosomal HMGB1 and/or miR-15b-5p and miR-378a-3p regulate NET formation through the Akt/mTOR autophagy pathway in PMNs.
Materials and methods
Patients
We recruited 21 patients with early (less than 24 h) diagnosis of septic shock who were admitted to the ICU of Xinhua Hospital, Shanghai Jiaotong University, China, from March 2018 to July 2019 (Table
1). Septic shock was defined as persisting hypotension requiring vasopressors to maintain mean arterial pressure (MAP) ≥ 65 mmHg and a serum lactate level >2 mmol/L (18 mg/dL) despite adequate volume resuscitation [
1]. We excluded patients under 18 years old; those with pregnancy, severe anemia, active bleeding, platelet disorders, or chemotherapy; or those using full-dose heparin or any other medications that interfere with platelet function. The enrolled patients had 30 mL blood samples collected from a central venous catheter. Twenty-two healthy volunteers provided blood samples that served as controls. The study was approved by the institutional ethics and review board of Xinhua Hospital, and informed consent was obtained from the patients or their representatives.
Table 1
Demographic data of the included patients
Male sex, n (%) | 16 (76) |
Age, years | 71 ± 9 |
Mortality at 28 days, n (%) | 5 (24) |
Comorbidities, n (%) |
Arterial hypertension | 9 (43) |
Diabetes mellitus | 2 (9) |
Others | 10 (48) |
Source of sepsis, n (%) |
Urinary | 3 (14) |
Abdominal | 18 (86) |
Hemodynamic data |
Heart rate/min | 100 ± 24 |
Mean arterial pressure, mmHg | 81 ± 13 |
Norepinephrine dosage, μg/kg/min | 0.21 (0.18–0.64) |
Ventilatory data |
Respirate rate/min | 20 ± 7 |
PaCO2 (mm Hg) | 31 ± 8 |
PaO2/FIO2 (Kpa) | 366 ± 112 |
Use of mechanical ventilation, n (%) | 16 (76) |
Hematologic and inflammatory data |
Neutrophils, 109/L | 13.6 ± 8.9 |
Hemoglobin, g/dL | 102.3 ± 19.6 |
Platelets, 109/L | 150.5 ± 77 |
Lactate, mmol/L | 2.5 (2.1–6.1) |
CRP, mg/dL | 160 (102–160) |
Procalcitonin, ng/mL | 47 (18.9–100) |
Glasgow score | 9.8 ± 3.3 |
SOFA score | 9 (9–11) |
Exosome isolation and characterization
Exosomes were isolated from the plasma of healthy controls and septic shock patients or from the supernatant of platelets isolated from healthy volunteers stimulated ex vivo using Total Exosome Isolation Reagent (Thermo Fisher Scientific, Waltham, MA, USA). The detailed isolation procedure and the methods used to determine exosomal morphology, size distribution, and surface marker expression are described in Additional file
1: supplementary methods.
Platelet purification and activation
Human platelets were isolated from EDTA-anticoagulated venous blood as previously described [
21]. Platelet activation was induced upon incubation with PBS or 1 μg/mL LPS (from
Escherichia coli O111:B4, Sigma) or 0.1 U/mL thrombin for 3 h at 37 °C. More detailed information on all methods is provided in Additional file
1: supplementary methods.
PMNs isolation and NET induction
PMNs were isolated from the venous blood of healthy volunteers by discontinuous density gradient centrifugation with two commercially available solutions (Histopaque-1077 and Histopaque-1119) of differential density (#10771 and #11191, Sigma; St Louis, MO, USA), according to the manufacturer’s instructions. PMNs were treated with exosomes (100 or 200 μg/mL) up to 14 h or with 50 nM PMA for 3 h at 37 °C as the positive control. To determine the role of ROS and autophagy in NET formation, the following inhibitors and enhancers were used: 3-Methyladenine (3-MA, 5 mM), wortmannin (150 nM), bafilomycin A1 (1 μM), rapamycin (100 nM), or N-acetyl-l-cysteine (NAC, 50 μg/mL).
HL-60 cell culture and transfection
The human acute promyelocytic leukemia cell line HL-60 (ATCC-CCL-240) was maintained in phenol red-free RPMI-1640 medium supplemented with 2 mM L-glutamine, 10% FBS, and 1.25% DMSO for 3 days as described [
22]. Cells were then transfected with microRNA control (50 nM), mimics (miR-24-3p, miR-15b-5p, miR-25-3p, miR-126-3p, miR-378a-3p, and miR-155-5p), or inhibitors (miR-15b-5p, miR-378a-3p) (Shanghai GenePharma Co., Ltd., Shanghai, China) for 24–48 h prior to analysis or cocultured with exosomes as per the manufacturer’s instructions.
Animals
Wild type, male C57BL/6 mice aged 6–8 weeks were purchased from Shanghai Sippr-BK Laboratory Animal Co. Ltd. The animals were fed under a specific pathogen-free environment in the Shanghai Xinhua Hospital animal laboratory. All animal experiments were conducted under the rules approved by the Shanghai Xinhua Hospital Ethics Committee.
Platelet depletion
In order to investigate the role of platelets in NET formation and ALI development, platelets were depleted in WT C57BL/6 mice by injecting busulfan i.p. or anti-platelet antibody i.v. as previously described [
23,
24], resulting in a 40% or 60% reduction of platelets, respectively, without affecting leukocytes.
Mouse model of cecal ligation and puncture (CLP) and in vivo exosome administration
The mouse CLP model was prepared as previously described [
25]. In total, 40 mice were used in sham group or in CLP group. At 24 h after surgery, WT male C57BL/6 mice were euthanized with a phenobarbital overdose (100 mg/kg body weight), after which bronchoalveolar lavage fluid (BALF) and citrate-anticoagulated whole blood were collected at a 1:7 ratio. In order to visualize morphological changes and NET formation in sepsis-induced ALI, lung tissue was harvested and fixed in 4% paraformaldehyde 24 h for H&E and immunofluorescence staining.
To explore exosome function in vivo, mice were treated with exosomes isolated from the plasma of sham or CLP mice (300 μg/mouse) through i.p. injection using 31-gauge insulin syringes. After 24 h, BALF, venous blood, and lung tissue were harvested as described previously.
NET quantification assay
To quantify NETs in the cell culture supernatant, plasma, and mouse BALF, we used the PicoGreen dsDNA Quantification Kit (Invitrogen, Carlsbad, CA, USA) and a capture ELISA based on myeloperoxidase (MPO) associated with DNA [
6]. For ELISA analysis of NET concentration, 1 μg/mL anti-MPO mAb was used as a capture antibody with Cell Death Detection ELISA (Roche, Indianapolis, IN, USA) according to the instructions.
Sequencing of miRNA and data analysis
Exosomes were isolated from the supernatant of platelets stimulated with PBS (PBS-Exo) or LPS (LPS-Exo) ex vivo. Total RNA was extracted from exosomes using the miRNeasy Serum/Plasma Kit (Qiagen, Valencia, CA, USA). The final ligation PCR products were sequenced using the BGISEQ-500 platform (BGI Group, Shenzhen, China).
After acquiring the raw data, the differently expressed miRNAs were calculated using the t test. Those with ≥ 1.2-fold change and a P value < 0.05 were regarded as significantly different. A heat map was generated using the R 3.5.3 software. Pathway analysis was conducted using the reactome pathway database. The 20 most enriched pathways are listed and were used to reveal the most associated pathways.
Statistics
The normal distribution of the data was tested using the Shapiro-Wilk test. For normal distribution data, data are presented as means ± SEM. Comparisons between 2 groups under identical conditions were performed by the 2-tailed Student’s t test. Multiple group comparisons were performed by one-way ANOVA followed by calculating the least significant difference to compare means. For data are not normally distributed, data are presented as median (25th–75th percentile). Comparisons between 2 groups were performed by Mann-Whitney U test. A value of P < 0.05 was considered statistically significant.
Sample size was determined by PASS 11 software (NCSS, LLC, Kaysville, UT, USA). The MPO-DNA complexes in plasma of three septic shock patients or healthy volunteers were taken in the preliminary test. The MPO-DNA complexes in plasma of septic shock patients and healthy volunteers were 0.23 ± 0.11 and 0.09 ± 0.04, respectively. Based on the difference between groups and assuming a two-sided type I error rate of 0.05 and a power of 0.80, 6 patients in each group were required to reveal a statistically significant difference. Besides, the power analysis was also performed to determine the number of animals used in this study to reach statistical significance. The MPO-DNA complexes in plasma of three sham or CLP mice were taken in the preliminary test. Three mice in each group were required to reveal a statistically significant difference.
Discussion
In the current study, we demonstrate that NET components are significantly increased in the plasma of septic shock patients and correlate positively with disease severity and outcome. Platelet depletion in mice reduced the plasma exosome concentration, NET formation, and lung injury following sepsis. We also showed that exosomes isolated from the plasma of septic shock patients or from the supernatant of LPS-stimulated platelets induced NET formation. We further demonstrated that platelet-derived exosomes induce activation of the intra-neutrophil Akt/mTOR-related autophagy pathway and subsequent NET formation via exosomal HMGB1 and/or miR-15b-5p and miR-378a-3p. In addition, we revealed that IKK activity in platelets regulates exosome secretion during septic shock.
Septic shock remains one of the most challenging medical conditions, with increasing incidence in recent years. Accumulating experimental and clinical evidence indicates that overactive NET formation during sepsis can lead to the development of multiple organ dysfunction, highlighting the pathophysiological role of NETs in sepsis [
30,
31]. Platelets have been shown to be potent activators of NET formation by direct platelet-neutrophil interactions via cell adhesion molecules or soluble mediators, such as HMGB1, platelet-derived chemokines, or thromboxane A2 [
8]. Platelet depletion could inhibit NET formation and improve lung injury in multiple lung injury murine models [
6,
23,
32]. In accordance with previous studies, we demonstrated that NETs are a hallmark of septic shock, and platelets promoted NET formation and lung injury during sepsis. Although platelets may serve as a target to alleviate lung injury, they also function as key elements in hemostasis and thrombosis. Current anti-platelet therapies invariably cause bleeding as an undesired adverse effect [
33]. In addition, NETs have been proposed as an innate defense mechanism that is responsible for pathogen clearance, and impaired NET generation results in the dissemination of infections. As shown in our results, platelet depletion slightly increased the bacterial load in the lungs. Therefore, a critical balance of NETs is necessary to prevent lung injury and maintain microbial control [
34].
This study shows that platelet-derived exosomes are major mediators that induce NET formation in septic shock. The results revealed that platelet depletion prior to CLP significantly decreased the plasma exosome concentration and reduced NET generation, and a positive relationship was found between platelet-derived exosomes and NETs. Previous studies have demonstrated that most exosomes accumulated in the blood of patients with sepsis are derived from platelets and correlate with organ dysfunction [
11,
35,
36]. Our results suggested that sep-Exo or LPS-Exo from platelets promoted NET generation. Platelet TLR4 is involved in inducing NETs in mice and humans [
24]; moreover, it plays prominent roles in sensing high circulatory LPS levels during sepsis and in neutrophil-mediated pathogen clearance [
37]. However, the exact role of TLR4 in LPS-activated platelet exosome secretion needs to be further addressed.
The signaling pathways that mediate NET formation remain inadequately elucidated and have been found to vary in response to different stimuli [
26,
38]. ROS generation by NADPH oxidase is an integral, but not essential, cellular process involved in NET formation [
4]. The activation of NADPH and the MAPK-ERK pathways appears to be relevant, at least for NETs induced by PMA-activated neutrophils [
39]. In addition to ROS generation, NET formation also requires autophagy [
40]. Our previous study suggested that autophagy plays an important role in maintaining the function of NET formation in response to infection and in regulating neutrophil death [
25]. Our present study showed that platelet-derived exosomes increased both the ROS level and autophagic activity in PMNs during septic shock, while pharmacological inhibition of ROS and autophagy demonstrated that only the autophagy pathway participated in platelet-derived exosome-induced NET generation. Our results were consistent with previous research that showed that NADPH inhibition did not interfere with NET formation induced by LPS-, Pam3CSK4-, or AA-stimulated platelets, suggesting that ROS are not NET mediators under these conditions [
9].
One key regulator of autophagy is mTOR, a serine/threonine kinase that regulates cell growth, proliferation, and protein synthesis [
27]. The PI3K/Akt/mTOR signaling pathway is a negative regulator of both autophagy and apoptosis. The inhibition of Akt/mTOR activity is known to play an essential role in initiating autophagy [
41,
42]. In our study, platelet-derived exosomes promoted NET formation by inhibiting Akt/mTOR pathway activity during septic shock. Previous studies also demonstrated that HMGB1, either soluble or presented from activated platelets, induces autophagy in neutrophils, thus promoting NET generation [
14,
18]. Therefore, we decided to address whether platelet-derived exosomes induced NET formation via HMGB1 during septic shock. Our results showed that the HMGB1 expression level was elevated during septic shock in platelet-derived exosomes, which recapitulated the effect of HMGB1 on the inhibition of Akt/mTOR activity and the induction of NET formation. BoxA abrogated all events elicited by platelet-derived exosomes.
Exosomally transferred miRNAs are emerging as novel regulators of cellular function. Evidence has been found in both immune cells and other cell types that transferred miRNAs repress target mRNAs in recipient cells. In the current study, we determined that exosomal miR-15b-5p and miR-378a-3p were involved in negatively regulating the activity of the Akt/mTOR pathway by repressing PDK1 expression. Recent study also demonstrated that miR-378 promotes autophagy initiation through the mTOR/unc-51-like autophagy activating kinase 1 pathway and sustains autophagy by targeting PDK1 [
43]. Although the isoforms of miR-378 (miR-378a/b/c/d/e/f/g/h/i/j) are encoded by different genomic loci, they share identical seed sequences and are thus considered to have common regulatory targets [
44]. In addition, a recent study by Zhu et al. suggested that miR-15b-5p mediates the autophagy of endothelial progenitor cells and influences coronary atherosclerotic heart disease via the mTOR signaling pathway [
45]; however, its effect on PDK1 expression was not demonstrated previously.
Lastly, we demonstrated that IKK controlled platelet-derived exosome secretion during sepsis, which was in alignment with previous research showing that IΚΚ controls platelet secretion through regulating SNAP-23 phosphorylation [
29]. SNAP-23 phosphorylation enables the formation of the soluble
N-ethylmaleimide sensitive factor attachment protein receptors (SNAREs) complex to allow exosome release [
46]. NF-κB plays a pivotal role in sepsis, and its activation is initiated by signal-induced ubiquitylation and subsequent degradation of inhibitors of kappa B (IκBs) primarily via IKK activation. There is now compelling evidence that IKK inhibition reduces multiple organ dysfunction caused by sepsis in mice [
47]. However, we may be the first to show that the IKK inhibitor alleviated lung injury during sepsis through inhibiting platelet-derived exosome secretion.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.