Erythrocyte-derived extracellular vesicles aggravates inflammation via facilitating pro-inflammatory phenotype of macrophages

Transfusion of aged erythrocytes is associated with increased morbidity and mortality in the critically infections with incompletely understood mechanism. Previous studies suggested red blood cell (RBC)-derived extracellular vesicles (EVs) may be potential risk factors for the occurrence of transfusion-related immunomodulation (TRIM). The purpose of our study is to evaluate the effects of EVs under the inflammation condition and explore the underlying mechanisms. In vivo, the activity of EVs was evaluated in the caecal ligation and puncture ( CLP)-induced sepsis. Our results showed EVs significantly aggravated the inflammatory response of sepsis in serum and lung tissue by promoting the production of pro-inflammatory factors, TNF- α , IL-6, IL-1 β , and reduced the survival rate of septic mice in vivo. Importantly, adoptive transfer EVs pretreated bone marrow-derived macrophages (BMDM) obviously aggravated the systemic pro-inflammatory factors in mice after CLP surgery. In vitro, the pro-inflammatory properties of EVs were shown as elevating the levels of TNF- α , IL-6, IL-1 β in LPS-stimulated BMDM. Moreover, EVs promoted the LPS-induced macrophages polarization into pro-inflammatory phenotype. The underlying mechanism was possibly that EVs could up-regulate NF- κ B and MAPKs activity to favor macrophage cytokines production.


Introduction
In spite of advantages of packed red blood cell (RBC) transfusion, it also carries complications as well. Previous clinical and animal trials demonstrated that RBC transfusion could aggravate inflammations of the receptor, such as increasing the risk of bacterial infection. [1][2][3][4][5] Moreover, transfusion of aged erythrocytes is associated with increased morbidity and mortality in the critically infections with unknown cause. 6-

Discussion
The impetus for this work was the clinical observation that transfusion of aged erythrocytes is associated with increased morbidity and mortality in the critically infections with incompletely understood mechanism. In our study, the proinflammatory activity of RBC-derived EVs on infection disease was identified and the underlying mechanisms were further elucidated for the first time. RBC-derived EVs, mainly taking advantages of the NF-κB and MAPK signaling pathways, prompted the pro-inflammatory phenotype of macrophages thus aggravating the inflammatory conditions in CLP-induced sepsis.
Firstly, the pro-inflammatory activity of EVs was evaluated in vivo. It is vital to rebalance the inflammatory response and prevent the organ dysfunction and the sequential organ failure in infection illness. RBC-derived EVs obviously reduced survival rate of mice with CLP-induced sepsis. This was accompanied by aggravated systemic pro-inflammatory factors in serum of sepsis model mice ( Fig.1 A, B). EVs intensified the damage of lung tissue and increased the infiltration of proinflammatory macrophages (Fig.1). The process of inflammation was closely related to the phenotype of macrophages. 32, 33 Therefore, further investigation indicated that high expressed CD11c, a marker of pro-inflammatory macrophages, was significantly up-regulated by EVs treatment in lung tissue of CLP-induced sepsis mice (Fig.1F).
These results hinted us that the pro-inflammatory activity of EVs for sepsis may be related to the regulation of macrophage polarization into pro-inflammatory phenotype.
Furthermore, the targeting of macrophage by EVs in vivo was further confirmed in our experiments using adoptive transfer of EVs pretreated-BMDMs to macrophagedepleted mice (Figure 2). This pro-inflammatory properties of EVs, however, was barely observed in normal mice suggesting that EVs function may be mild and body could compensate for itin the healthy mice. However, EVs' action could be amplified by LPS infection.
Next, the activity of RBC-derived EVs on macrophage was evaluated in vitro. As shown in Fig. 3, the BMDM retained almost the same viability exposed to EVs at concentrations less than 2.5 mM in 24 hour so that the effect of EVs on cell viability is excluded in subsequent studies. RBC-derived EVs increased both the mRNA and protein levels of pro-inflammatory factors (TNF-α, IL-6 and IL-1β) on its own in a moderate degree, and this promotion effect was more prominent under LPS stimulation (Fig. 4). Furthermore, the result of RT-PCR suggested EVs could promote LPS and IFN-γ polarized into pro-inflammatory phenotype with increased mRNA expression of inos, Tnf-α and Il-6, but not IL-4 polarized into antiinflammatory phenotype with unchanged mRNA expression of Ym-1, Arg-1 and Fizz1 in macrophages (Fig. 5). Other functional characteristics of these macrophages, such as phagocytosis, were also inhibited, further confirming the pro-inflammatory activity of EVs (Fig.5). This selectivity for the pro-inflammatory, over the antiinflammatory phenotype might aggravate the infection as the secondary-hit for the sepsis patients which received blood transfusion therapy.
In the previous studies, Zecher and co-workers 19 found RBC-derived microvesicles could amplify both pulmonary leukocyte sequestration and induce higher serum levels of pro-inflammatory cytokines, such as IL-6, keratinocyte-derived chemoattractant (KC), and monocyte chemoattractant protein-1 (MCP-1) which was consistent with our in vivo data. However, in his study, administration of microvesicles in LPS-primed mice did not lead to invasive lung injury. In our experiments, RBC-derived EVs not only intensified the damage of lung tissue, but also obviously reduced survival rate of mice with CLP-induced sepsis. The possible reason is that 1) the differences in RBC and erythrocyte-derived EVs purification and storage conditions; 2) the differences in infection models. In our study, we chose the CLP-induced sepsis model which is much more similar to clinical endotoxin infection. In terms of mechanistic studies, they focused on the role of microvesicles in the complement activation. However, our point was the effect of EVs on macrophage, which played an important role in sepsis.
To confirm the underlying mechanism(s) and the molecular target(s) of EVs, we measured the phosphorylation levels of NF-κB and MAPKs pathways. Results in Figure 6 suggested EVs slightly enhanced the activation of NF-κB through promotion of IκBα degradation and phosphorylation of NF-κB p65 subunit. And LPS-mediated NF-κB activity increased significantly by EVs synergistically. Furthermore, the elevated TNF-α secretion by EVs treatment can be inhibited through NF-κB inhibitor