Flow cytometric analysis of Calcein-AM-labeled intact EVs in plasma showed that the sepsis group exhibited higher plasma EV concentrations than the SIRS group (Fig. 1A). However, there were no significant differences in the plasma EV concentrations between the SIRS group and the control group of 18 healthy volunteers (Fig. 1A).
EVs were then isolated from plasma samples using centrifugation and their size was measured by DLS analysis. The diameters of EVs, expressed as median (interquartile range), were 360 nm (124 nm–1444 nm) for the sepsis group, 265 nm (165 nm–776 nm) for the SIRS group, and 845 nm (514 nm–1317 nm) for the control group (Fig. 1B). Thus, the sizes of EVs in the SIRS group were significantly smaller and were more broadly distributed than those of the control group.
Sepsis EVs express higher levels of β2 integrin
We studied the expression of integrins that are potentially involved in the regulation of organ-distributed EVs. We focused on 3 major integrin subfamilies: β1, β2, and β3. EVs from the sepsis group expressed higher surface β2 integrin levels compared with the control group (Fig. 2B, Fig S1, http://links.lww.com/SHK/A792). By contrast, there were no significant differences among the 3 groups in the expression of β1 and β3 integrins (Fig. 2A and C, Fig S1, http://links.lww.com/SHK/A792).
SIRS and sepsis EVs express higher levels of PD-L2
We then investigated the expression of PD-1 ligands, PD-L1 and PD-L2, which are potentially involved in the negative regulation of lymphocyte activation and proliferation. EV expressions of PD-L1 in the sepsis and control groups were similar. Interestingly, EV expression of PD-L1 in the SIRS group was significantly decreased compared with those in the sepsis and control groups (Fig. 3A). By contrast, EV expressions of PD-L2 in the sepsis and SIRS groups were significantly higher than that of the control group (Fig. 3B, Fig S1, http://links.lww.com/SHK/A792).
sPD-L1 levels increased in SIRS and sepsis
Thus far, we have found significant increases in the EV expression of PD-L2, but not PD-L1, in the sepsis and SIRS groups. Therefore, we sought to identify the sPD-L1 and sPD-L2 forms in circulation. ELISA-based measurements of the plasma sPD-L1 concentrations showed that the sepsis group expressed significantly higher sPD-L1 levels than the control group (Fig. 4A). sPD-L1 is thought to be released via proteolytic cleavage of the membrane-bound intact PD-L1 on the cell surface; however, whether sPD-L1 could be released from the surface of EVs remains to be elucidated. Our results showed that the EV expression of PD-L1 did not correlate with sPD-L1 levels in the sepsis and SIRS groups (Fig S2, http://links.lww.com/SHK/A792). In contrast to sPD-L1, no significant differences were observed in sPD-L2 levels in sepsis, SIRS, and control groups (Fig. 4B).
Correlation of EV integrin and PD-1 ligand expression with clinical parameters
We next sought to investigate any possible correlations of the EV parameters with clinical parameters related to the sepsis pathogenesis (Fig. 5). Although we found many statistically significant correlations between the EV and clinical parameters in the SIRS and sepsis groups, we focused only on the EV parameters that were significantly increased in the sepsis and/or SIRS group. These included EV β2 integrin expression, EV PD-L2 expression, and plasma sPD-L1 levels (Fig. 5).
In the sepsis group, we found that EV β2 integrin expression correlated with hypotension (R = −0.56, P < 0.05) and impaired renal function, as indicated by reduced glomerular filtration rate (eGFR) (R = −0.38, P < 0.05). Although we did not find significant correlations of EV PD-L2 expression with clinical parameters, we did in terms of sPD-L1 levels and impaired renal function. This was evident based on reduced eGFR (R = −0.58, P < 0.05) and increased creatinine (R = 0.64, P < 0.05). We also noted the correlations of sPD-L1 with the following: dysregulated coagulation, as indicated by increased fibrin degradation products (R = 0.57, P < 0.05) and increased D-dimer (R = 0.46, P < 0.05), and worsened organ failures as shown by Glasgow Coma Scale (GCS) (R = −0.41, P < 0.05) and SOFA score (R = 0.40, P < 0.05) (Fig. 5B). In the SIRS group, EV PD-L2 expression levels correlated with reduced lymphocyte counts, reduced hematocrit, and reduced total protein levels (Fig. 5A).
There appears to be heterogeneity with respect to EV sizes and surface marker expression in the sepsis and SIRS samples. One might think that analyzing less heterogenous subgroups could produce additional valuable information. First, based on the mean diameters of EVs, samples in the SIRS and sepsis groups were subdivided into either a large EV subgroup (>1000 nm, n = 4, SIRS: n = 9, sepsis) or a small EV subgroup (<1000 nm, n = 23, SIRS; n = 18, sepsis). We then compared the EV expression of integrins and PD-L1 and PD-L2 between the large and small EV subgroups in the SIRS and sepsis parent groups (Fig S3, http://links.lww.com/SHK/A792), but did not observe any significant intersubgroup differences.
Second, based on the β2 integrin expression levels, samples in the sepsis parent group were subdivided into either a β2 integrin-positive (n = 19) or a β2 integrin-negative (n = 8) subgroups. Studying the correlations of the EV parameters (including sPD-L1) with the clinical parameters, we identified several significant intersubgroup differences (Fig S4, http://links.lww.com/SHK/A792). For example, a significant correlation of sPD-L1 levels with increased neutrophil counts was observed in the β2 integrin-positive subgroup. However, this did not hold true in either the β2 integrin-negative subgroup or the parent sepsis group.
Similarly, based on the EV PD-L1 or PD-L2 expression levels, samples in the SIRS and sepsis parent groups were subgrouped. We studied the correlations between the EV and clinical parameters and found several significant intersubgroup differences (Fig S5-S7, http://links.lww.com/SHK/A792). For example, a significant correlation of β2 integrin levels with liver injury, as indicated by increased liver enzymes (aspartate aminotransferase [AST], alanine aminotransferase [ALT]), was observed in the EV PD-L1-positive subgroup (Fig S5A, http://links.lww.com/SHK/A792). However, this was not the case in either the PD-L1-negative subgroup or the parent sepsis group. In addition, a significant correlation of sPD-L1 levels with reduced lymphocyte counts was observed in the PD-L1-positive sub-group, but not in the PD-L1-negative sub-group or the parent sepsis group.
Sepsis results from heterogenous etiologies, as is also the case with the present study. We examined subgroups of sepsis cases stemming from either pneumonia or nonpneumonia etiologies (Table S1, http://links.lww.com/SHK/A792). Comparisons between the pneumonia subgroup (n = 12) and the nonpneumonia subgroup (n = 15) revealed that, compared with the control group, sPD-L1 levels were significantly increased in the pneumonia subgroup, the nonpneumonia subgroup, and the parent sepsis group (Fig S8, http://links.lww.com/SHK/A792). However, EV expression levels of integrins, PD-L1, and PD-L2 did not differ significantly between the pneumonia and nonpneumonia subgroups (Fig S9-S12, http://links.lww.com/SHK/A792).
A pilot investigation on a mouse sepsis model to study the EV expression of integrins and PD-1 ligands
To further substantiate our findings that EV expression of β2 integrins and PD-L2 were upregulated in sepsis patients, we performed a pilot animal investigation, using a mouse model of sepsis, that aimed to recapitulate the similar changes in the characteristic of EVs. Specifically, we utilized a CLP model, studying the expressions of integrins and PD-1 ligands on EVs isolated from sepsis and control mouse plasma samples. Unexpectedly, we were unable to observe any significant changes in the EV expression levels of integrins and PD-1 ligands (Fig S13, http://links.lww.com/SHK/A792).
In this study, we have investigated the expression of integrins and PD-1 ligands on EVs as well as soluble forms of PD-1 ligands in SIRS and sepsis patients. We have shown that the sepsis patients exhibited increased EV expression of β2 integrin and PD-L2 as well as increased plasma levels of sPD-L1, compared with control group. EV expression of β2 integrin correlated with hypotension and reduced renal function, whereas sPD-L1levels correlated with severity of organ failures, reduced renal function, and impaired central nervous system function. In addition, the SIRS patients exhibited increased EV expression of PD-L2 as well as increased plasma levels of sPD-L1, compared with control group. EV expression of PD-L2 correlated with reduced lymphocyte counts, reduced hematocrit, and reduced total protein levels, whereas sPD-L1levels correlated with severity of organ failures and impaired central nervous system function. As discussed in the following paragraphs, the results in this study support the septic pathogenesis that the EV β2 integrin, as well as EV PD-L2 and sPD-L1increase in response to the systemic immune activation that leads to multiple organ dysfunctions.
EVs isolated in this study using a centrifugation at 16,500 g for 1 h predominantly included microvesicles, alternatively known as microparticles. Microvesicles are released from cells via the shedding of the plasma membrane, as opposed to exosomes which are much smaller in size and secreted from cells via endosomal pathways (2). Several previous studies have investigated microvesicles and microparticles in critically ill ICU patients. One previous study found that plasma microvesicle counts were higher in sepsis and burn patients (18), consistent with the current study showing that plasma microvesicle levels are higher in sepsis patients. Interestingly, some studies have shown that increased levels of plasma microparticles in critically ill patients are associated with a reduced risk of acute respiratory distress syndrome and sepsis (19, 20) and that microparticles expressing α-2-microgroblin may play a protective role in sepsis resulting from community-acquired pneumonia (8). By contrast, other studies have shown that increased levels of microvesicles correlated with poor outcomes in cases of septic shock (9–11).
One novel finding we have made is that EV β2 integrin expression levels were significantly increased in the sepsis group. Furthermore, we found significant correlations with clinical parameters, such as reduced blood pressure and reduced renal function, as shown by diminished eGFR. As β2 integrin is expressed only on leukocytes, β2 integrin-expressing EVs are thought to originate from leukocytes, including neutrophils, lymphocytes, and monocytes/macrophages. As expression of β2 integrin is upregulated upon leukocyte cellular activation (21), and activated leukocytes secrete β2 integrin-expressing EVs (22, 23), increased EV expression of β2 integrin suggests that enhanced leukocyte activation might be occurring systemically. This is consistent with the type of systemic inflammation that leads to aberrantly induced vascular dilatation and decreased blood pressure. Reduced eGFR could be caused either directly by an inflammatory injury to the kidney parenchymal and stromal cells and/or indirectly by renal hypoperfusion secondary to reduced blood pressure levels. Along with EVs from platelets and endothelial cells, EVs from leukocytes appear to represent a major fraction of the EVs found in the circulation of SIRS and sepsis patients (24). β2 integrin on EVs might interact with its ligand intercellular adhesion molecule-1, which is upregulated on inflamed endothelial cells, thereby, potentially regulating EV distribution in the body.
Another novel finding of our study is that EV PD-L2, but not PD-L1, expression levels were significantly increased in the sepsis and SIRS groups compared with the control group. PD-L2 was thought to be primarily expressed in macrophages and dendritic cells, but has recently been shown to be induced by microenvironmental stimulations in other various cell types (6). PD-L1 and PD-L2 are similar in that they both bind to and signal through PD-1. However, an important distinction is that PD-L1 is upregulated preferentially in Th1-type immune activation, whereas PD-L2 favors Th2-type stimulation (25), which might be relevant to immune suppression induced by sepsis and SIRS. PD-L1 and PD-L2 on EVs are assumed to be as functionally active to transmit the inhibitory signals through PD-1 as those on cells. However, the activities of PD-L1 and PD-L2 on EVs appear to change depending on the types of immune reactions involved. In delayed-type hypersensitivity responses, PD-L1 and PD-L2 on EVs were shown to be dispensable for damping T-cell activation (26). By contrast, in the tumor microenvironment, PD-L1 on EVs was found to be functionally active in signaling through PD-1 to suppress T-cell activation (27). The potential roles of PD-L1 and PD-L2 on EVs in the pathogenesis of sepsis currently warrant elucidation. They might, however, serve as alternative sources to bind to, and deliver, negative signals through PD-1 to suppress immune cell activation, as both PD-L1 and PD-L2 do on the cell surface. PD-L1 and PD-L2 interactions with PD-1 are thought to be involved in immune paralysis in SIRS and sepsis (7). It has been previously shown that sPD-L1 levels are associated with immunosuppression and infectious complications in acute pancreatitis patients (28). Our results have also shown a significant increase of EV PD-L2 expression and its correlation with reduced lymphocyte counts in the SIRS group. As compensatory anti-inflammatory response syndrome is thought to occur simultaneously with SIRS (29), this correlation might be related to a possible immune suppression in SIRS.
In our pilot animal investigation using a mouse CLP model, we were unable to observe the upregulation of EV β2 integrins and PD-L2, as opposed to our findings in sepsis patients. The reasons for these discrepancies are unclear. The major biogenesis pathways of EVs appear to be well conserved between mouse and human (2). One plausible explanation might be that the CLP model has a limited capacity to reflect the pathophysiological responses that occur in human sepsis. How well the global gene expression patterns elicited by the CLP model can mimic those by human sepsis remains a controversial issue (30).
Increased plasma sPD-L1 levels have been associated with survival in lymphoma (31) and atherosclerotic vascular pathologies in type 2 diabetes mellitus (32). sPD-L1 is biologically active in binding to PD-1, enabling it to either deliver negative signals to suppress T-cell functions (25) or to block the cellular PD-1/PD-L1 interactions that augment T-cell functions (32) depending on the pathological context. In sepsis, a study by Liu et al. (33) involving 91 sepsis patients demonstrated that sPD-L1 was increased in sepsis compared with healthy controls, was increased in nonsepsis survivors compared with sepsis survivors, and correlated with lymphocyte counts, platelet counts, and SOFA scores. In contrast, the study by Wilson et al. (34) involving 22 sepsis patients found no significant increase of sPD-L1. Consistent with Liu et al. (33), the present study showed that sPD-L1 levels in the plasma were significantly increased in the sepsis group compared with those of the control group. Furthermore, as reported by Liu et al(33), we have also shown a significant correlation of sPD-L1 levels with SOFA score in sepsis. In addition, our results revealed that sPD-L1 levels significantly correlated with impaired central nervous system function, as assessed by the GCS, as well as impaired renal function based on reduced eGFR and increased serum creatinine. These results suggest a possible link between sPD-L1 and organ failure. Increased sPD-L1 levels might be merely a consequence of aberrant activation of immune cells and nonimmune cells, resulting in the release of more sPD-L1 (6). Alternatively, increased sPD-L1 might lead to enhanced activity that blocks inhibitory signals via the PD-1/PD-L1 interaction, thereby exacerbating inflammation (32). Unlike sPD-L1, sPD-L2 did not altered in SIRS and sepsis, probably because of multiple factors such as different cellular origins and different biogenesis of sPD-L2 compared with sPD-L1 (6).
Sepsis patients are highly heterogenous with respect to their etiologies, disease progressions, and prognoses (35). A possible heterogeneity in the sizes and surface marker expression levels of EVs from sepsis and SIRS patients could make it difficult to identify the relevant underlying pathological changes underlying the disease. In an effort to examine less heterogenous samples, we subgrouped the sepsis samples depending on the EV sizes, the expression levels of the EV integrins PD-L1, and PD-L2, and on the sepsis etiologies. Subsequent subgroup analyses have produced several significant findings. For example, in the EV β2 integrin-positive subgroup, sPD-L1 levels positively correlated with neutrophil counts. A previous study found a similar correlation in lung cancer patients (33). Additionally, in the EV PD-L1-positive subgroup, EV β2 integrin expression levels correlated not only with hypotension but also with liver injury, as shown by the increased AST and ALT. These findings might point to the complexities of the deregulated immune system, intensifying inflammation both in the EV β2 integrin-positive subgroup and in the EV PD-L1-positive subgroup.
In addition, although all samples were obtained within 7 days of ICU admission and the majority of the samples were obtained within 4 days of ICU admission, differences in sampling time could lead to heterogeneity in the EV characteristics. The EV expression levels of PD-L1 and PD-L2 and sPD-L2 levels did not significantly differ at any sampling points in the SIRS and sepsis groups (Fig S14A&B, and Fig S15, http://links.lww.com/SHK/A792). By contrast, sPD-L1 levels in SIRS were higher in samples obtained at ICU day 7, compared with samples at ICU day 2 and 3 (Fig S14C&D, http://links.lww.com/SHK/A792). A future investigation would be needed to study the collecting of samples at different time points for the same patients, as such sample collection data were not available with the present patient cohorts.
The authors would like to thank Dr. Hiroshi Yonekura at the Department of Pharmaco-epidemiology, Kyoto University Graduate School of Medicine, and Dr. Takumi Imai at the Department of Clinical Biostatistics, Kyoto University School of Public Health, for advice with statistical analysis.
1. Hotchkiss RS, Monneret G, Payen D. Sepsis
-induced immunosuppression: from cellular dysfunctions to immunotherapy. Nat Rev Immunol
2. Tkach M, Théry C. Communication by extracellular vesicles
: where we are and where we need to go. Cell
3. Hoshino D, Kirkbride KC, Costello K, Clark ES, Sinha S, Grega-Larson N, Tyska MJ, Weaver AM. Exosome secretion is enhanced by invadopodia and drives invasive behavior. Cell Rep
4. Wang R, Ding Q, Yaqoob U, de Assuncao TM, Verma VK, Hirsova P, Cao S, Mukhopadhyay D, Huebert RC, Shah VH. Exosome adherence and internalization by hepatic stellate cells triggers sphingosine 1-phosphate-dependent migration. J Biol Chem
5. Francisco LM, Sage PT, Sharpe AH. The PD-1 pathway in tolerance and autoimmunity. Immunol Rev
6. Dai S, Jia R, Zhang X, Fang Q, Huang L. The PD-1/PD-Ls pathway and autoimmune diseases. Cell Immunol
7. Patera AC, Drewry AM, Chang K, Beiter ER, Osborne D, Hotchkiss RS. Frontline science: defects in immune function in patients with sepsis
are associated with PD-1 or PD-L1 expression and can be restored by antibodies targeting PD-1 or PD-L1. J Leukoc Biol
8. Lashin HM, Nadkarni S, Oggero S, Jones HR, Knight JC, Hinds CJ, Perretti M. Microvesicle subsets in sepsis
due to community acquired pneumonia compared to faecal peritonitis. Shock
9. Delabranche X, Quenot JP, Lavigne T, Mercier E, François B, Severac F, Grunebaum L, Mehdi M, Zobairi F, Toti F, et al. Early detection of disseminated intravascular coagulation during septic shock: a multicenter prospective study. Crit Care Med
10. Matsumoto H, Yamakawa K, Ogura H, Koh T, Matsumoto N, Shimazu T. Enhanced expression of cell-specific surface antigens on endothelial microparticles in sepsis
-induced disseminated intravascular coagulation. Shock
11. Lehner GF, Harler U, Haller VM, Feistritzer C, Hasslacher J, Dunzendorfer S, Bellmann R, Joannidis M. Characterization of microvesicles in septic shock using high-sensitivity flow cytometry. Shock
12. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ. Definitions for sepsis
and organ failure and guidelines for the use of innovative therapies in sepsis
. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest
13. Yuana Y, Bertina RM, Osanto S. Pre-analytical and analytical issues in the analysis of blood microparticles. Thromb Haemost
14. Crescitelli R, Lasser C, Szabo TG, Kittel A, Eldh M, Dianzani I, Buzas EI, Lotvall J. Distinct RNA profiles in subpopulations of extracellular vesicles
: apoptotic bodies, microvesicles and exosomes. J Extracell Vesicles
15. Szatanek R, Baj-Krzyworzeka M, Zimoch J, Lekka M, Siedlar M, Baran J. The methods of choice for extracellular vesicles
(EVs) characterization. Int J Mol Sci
16. Eguchi A, Lazaro RG, Wang J, Kim J, Povero D, Willliams B, Ho SB, Stärkel P, Schnabl B, Ohno-Machado L, et al. Extracellular vesicles
released by hepatocytes from gastric infusion model of alcoholic liver disease contain a MicroRNA barcode that can be detected in blood. Hepatology
17. Théry C, Amigorena S, Raposo G, Clayton A. Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Curr Protoc Cell Biol
Chapter 3:Unit 322, 2006.
18. O’Dea KP, Porter JR, Tirlapur N, Katbeh U, Singh S, Handy JM, Takata M. Circulating microvesicles are elevated acutely following major burns injury and associated with clinical severity. PLoS One
19. Guervilly C, Lacroix R, Forel JM, Roch A, Camoin-Jau L, Papazian L, Dignat-George F. High levels of circulating leukocyte microparticles are associated with better outcome in acute respiratory distress syndrome. Crit Care
20. Mostefai HA, Meziani F, Mastronardi ML, Agouni A, Heymes C, Sargentini C, Asfar P, Martinez MC, Andriantsitohaina R. Circulating microparticles from patients with septic shock exert protective role in vascular function. Am J Respir Crit Care Med
21. Shimaoka M, Springer TA. Therapeutic antagonists and conformational regulation of integrin
function. Nat Rev Drug Discov
22. Zhang H, Xie Y, Li W, Chibbar R, Xiong S, Xiang J. CD4(+) T cell-released exosomes inhibit CD8(+) cytotoxic T-lymphocyte responses and antitumor immunity. Cell Mol Immunol
23. Théry C, Regnault A, Garin J, Wolfers J, Zitvogel L, Ricciardi-Castagnoli P, Raposo G, Amigorena S. Molecular characterization of dendritic cell-derived exosomes. Selective accumulation of the heat shock protein hsc73. J Cell Biol
24. Janiszewski M, Do Carmo AO, Pedro MA, Silva E, Knobel E, Laurindo FR. Platelet-derived exosomes of septic individuals possess proapoptotic NAD(P)H oxidase activity: a novel vascular redox pathway. Crit Care Med
25. Loke P, Allison JP. PD-L1 and PD-L2 are differentially regulated by Th1 and Th2 cells. Proc Natl Acad Sci U S A
26. Ruffner MA, Kim SH, Bianco NR, Francisco LM, Sharpe AH, Robbins PD. B7-1/2, but not PD-L1/2 molecules, are required on IL-10-treated tolerogenic DC and DC-derived exosomes for in vivo function. Eur J Immunol
27. Theodoraki MN, Yerneni SS, Hoffmann TK, Gooding WE, Whiteside TL. Clinical significance of PD-L1+
exosomes in plasma of head and neck cancer patients. Clin Cancer Res
28. Chen Y, Li M, Liu J, Pan T, Zhou T, Liu Z, Tan R, Wang X, Tian L, Chen E, et al. sPD-L1 expression is associated with immunosuppression and infectious complications in patients with acute pancreatitis. Scand J Immunol
29. Binkowska AM, Michalak G, Słotwiński R. Current views on the mechanisms of immune responses to trauma and infection. Cent Eur J Immunol
30. Dejager L, Pinheiro I, Dejonckheere E, Libert C. Cecal ligation and puncture: the gold standard model for polymicrobial sepsis
? Trends Microbiol
31. Rossille D, Gressier M, Damotte D, Maucort-Boulch D, Pangault C, Semana G, Le Gouill S, Haioun C, Tarte K, Lamy T, et al. Groupe Ouest-Est des Leucémies et Autres Maladies du Sang; Groupe Ouest-Est des Leucémies et Autres Maladies du Sang. High level of soluble programmed cell death ligand 1 in blood impacts overall survival in aggressive diffuse large B-Cell lymphoma: results from a French multicenter clinical trial. Leukemia
32. Shi B, Du X, Wang Q, Chen Y, Zhang X. Increased PD-1 on CD4(+)CD28(-) T cell and soluble PD-1 ligand-1 in patients with T2DM: association with atherosclerotic macrovascular diseases. Metabolism
33. Liu M, Zhang X, Chen H, Wang G, Zhang J, Dong P, Liu Y, An S, Wang L. Serum sPD-L1, upregulated in sepsis
, may reflect disease severity and clinical outcomes in septic patients. Scand J Immunol
34. Wilson JK, Zhao Y, Singer M, Spencer J, Shankar-Hari M. Lymphocyte subset expression and serum concentrations of PD-1/PD-L1 in sepsis
: pilot study. Crit Care
35. Marshall JC. The staging of sepsis
: understanding heterogeneity in treatment efficacy. Crit Care
Extracellular vesicles; integrin; programmed death ligand-1 and ligand-2; sepsis
Supplemental Digital Content
© 2019 by the Shock Society