© 2002 American Thoracic Society
"Septic Plasma"An Immunosuppressive MilieuUP Cytokines and Inflammation Institut Pasteur Paris, France For correspondence, Dr. Cavaillon's email address is jmcavail{at}pasteur.fr When Louis Pasteur wrote about the one of seven patients with puerperal sepsis who survived, he said this: "Natura medicatrix won the victory" (1). Nature created the antiinflammatory cytokines and the specific inhibitors or antagonists to counteract the proinflammatory cytokines. One might ask this: What host features help to win the victory against sepsis? Sepsis was initially believed to be associated with an exacerbated production of proinflammatory cytokines, as assessed by their detection in the plasma of septic patients. Accordingly, "septic" plasma was considered to be an inflammatory milieu. More recently, it has been appreciated that sepsis is much more complex and that plasma from patients with sepsis also contains antiinflammatory factors. Girardin and coworkers (2) first demonstrated in plasma of patients with meningococcemia that not only was tumor necrosis factor present but also its specific inhibitor, the soluble tumor necrosis factor receptors, and that high levels were associated with a poor outcome. It became apparent that sepsis was more than a proinflammatory disorder associated with a deficit of the antiinflammatory response. This became more clear when Marchand and coworkers (3) reported for the first time that the presence of plasma interleukin-10 (IL-10), a potent antiinflammatory cytokine, correlated with the occurrence of severe sepsis or septic shock. Indeed, the antiinflammatory response takes place very early after the insult. For example, high levels of IL-10 measured in the plasma of resuscitated patients only 3 hours after a cardiac arrest were associated with a poor outcome (4). Accordingly, blood leukocytes circulate within a complex milieu, containing both proinflammatory and antiinflammatory mediators very soon after the insult. When analyzed ex vivo, blood leukocytes from septic patients were found to be downregulated in terms of cytokine production (5). Prins and coworkers (6) showed that sera from septic patients had the capacity to downregulate monocytes from healthy humans in terms of ex vivo tumor necrosis factor production. This immunosuppressive activity was extended to plasma from patients with other systemic inflammatory response syndromes (SIRSs) such as trauma (7) or after cardiac arrest and resuscitation, recently proposed as a "sepsis-like" syndrome (4). Brandtzaeg and coworkers (8) identified IL-10 as being a major monocyte deactivator in sepsis. In addition to IL-10, many other antiinflammatory/immunosuppressive molecules are found in plasma of patients with SIRSs, such as transforming growth factor-ß, prostaglandins, catecholamines, neuropeptides, cortisol, and so on. In this issue of AJRCCM (pp. 1475–1482), Fumeaux and Pugin (9) further extend this observation and elegantly demonstrate that plasma IL-10 also partially contributes to another event associated with immunosuppression found in sepsis and SIRSs: the decreased expression of human leukocyte antigen (HLA) class II antigen (HLA-DR) on monocytes. Livingston and coworkers (10) demonstrated in trauma patients that monocyte expression of HLA-DR was below values of healthy control subjects and returned to normal after 14 days. Failure to restore HLA-DR expression was associated with an episode of major infection. A reduced HLA-DR expression was demonstrated by Lin and coworkers (11) in septic shock. In this report, Fumeaux and Pugin (9) extend these observations and demonstrate that in septic shock the low surface expression of HLA-DR is a very long lasting event (at least 3 weeks). It is worth mentioning that the authors found neither a difference between healthy control subjects and patients at the HLA class II mRNA levels nor a correlation between mRNA levels and surface HLA-DR expression. Such discrepancies, already observed with toll-like receptors, illustrate how careful one should be when analyzing the currently fashionable microarray techniques. The authors enlarge the panel of immunosuppressive activities of septic plasma reported by others (4, 6–8) by establishing a direct link between the plasma content and its capacity to limit HLA-DR expression on monocytes. Most interestingly, in the presence of "septic" plasma, HLA-DR molecules were endocytosed and were retained within monocytes from healthy donors. This observation was reminiscent of what had been previously described when the effect of recombinant human IL-10 was studied on HLA-DR expression; however, there is still room for yet unidentified plasma factors that contribute to this phenomenon. First, the authors did not observe a correlation between IL-10 levels and HLA-DR expression. Second, the use of an anti–IL-10 antibody only modestly increased (32%) the HLA-DR expression on control monocytes that were exposed to septic plasma, leaving the expression still below 50% of normal value. Although the authors excluded transforming growth factor-ß, catecholamine, and prostaglandins as putative candidates, one might hypothesize that neuromediators or even cortisol may be involved, either in a positive or a negative fashion. A few questions remain: Are the parameters of immunosuppression that are measured in the blood of humans with sepsis or noninfectious SIRSs responsible for the increased susceptibility of these patients to nosocomial infections? It is worth noting that circulating leukocytes from SIRS patients that fail to respond to endotoxin may remain fully responsive to other activators such as heat-killed Staphylococci (4). Could the decreased HLA-DR expression on circulating monocytes be a major defect responsible for a decreased adaptative immune response of the host? Manjuck and coworkers (12) have reported a decreased antigen-presenting cell capacity of monocytes to tetanus toxoid in sepsis. Hensler and coworkers (13), however, found a normal monocyte capacity to sustain superantigen proliferative response in patients with major surgery. Finally, what is the relative importance of circulating monocytes in the adaptative response as compared with dendritic cells within the tissues? Is the HLA-DR expression downregulated on dendritic cells in sepsis or SIRS patients as well? The real question remains: Is this downregulation a dysregulation or homeostasis? Should we target IL-10, as the authors concluded, or should we consider that the immunosuppressive status found in the peripheral blood allows the host to avoid a systemic inflammatory process? The authors have nicely demonstrated a profound influence of the environmental milieu on circulating monocytes and established a direct link between plasma factors and HLA-DR expression. Although there is no doubt that this marker of immunosuppression correlates with severe infection and poor outcome, it remains unclear how the downregulation of this marker is associated with poor outcome. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
