Published online before print September 17, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Pathology, University of Michigan Medical School, Ann Arbor, Michigan, USA
1Correspondence: Department of Pathology, University of Michigan Medical School, 1301 Catherine Rd., Room 7520 MSRB I, Ann Arbor, MI 48109-0602, USA. E-mail: pward{at}umich.edu
ABSTRACT
At the Trauma, Shock, Inflammation and Sepsis 2007 conference, the roles of complement activation products and relevant receptors were stressed in the setting of experimental sepsis [cecal ligation and puncture (CLP)] in mice and rats. In addition, some limited evidence was presented related to humans with septic shock (requiring vasopressor support). Collectively, the data suggested that events found in CLP also occur in human sepsis. Experimental sepsis (CLP) in rodents is associated with robust complement consumption and appearance of activation products (C3a, C5a) in plasma. During sepsis, there is up-regulation of C5a receptors (C5aR, C5L2) on blood polymorphonuclear neutrophils (PMNs) and in lungs, liver, kidneys, and heart. CLP also leads to dramatic reductions of C5aRs on blood PMNs, the intensity of which correlates with lethality. Interception in vivo of C5a or C5aR dramatically improves survival after CLP, preserves innate immune functions of blood PMNs, and greatly attenuates the intensity of consumptive coagulopathy and activation of the fibrinolytic system after CLP. In humans with septic shock, there is evidence of complement activation products in plasma along with loss of C5aRs on blood PMNs. These data suggest that in septic humans, interception of C5a or C5aR might be clinically efficacious.
Key Words: mediators • C5a • C5aR • inflammation
INTRODUCTION
It is well known that humans with sepsis have evidence of systemic activation of complement, as demonstrated by the presence in plasma/serum of complement activation products (C3a, C5a, C5b-9) [1 ]. The complement anaphylatoxins, C3a and C5a, are known to have a spectrum of proinflammatory activities [2 ], suggesting that systematic formation of C3a and C5a is highly undesirable, as these complement products can function at low nM concentrations and are intensely proinflammatory. Plasma contains carboxypeptidases, which rapidly convert C3a and C5a to des arg forms, that are less biologically reactive. However, the situation is confused by the fact that plasma complement proteins, which originate predominately from hepatocytes, can also be synthesized locally {in the relative absence of carboxypeptidases, as in lung (reviewed in refs. [3 , 4 ]}. In addition, phagocytic cells, such as neutrophils and tissue macrophages, contain a C5 convertase, which can generate C5a from C5 in the absence of other complement proteins [4 ]. In addition, thrombin can generate C5a from C5 [5 ], indicating a linkage between the coagulation and complement systems [5 ].
THE C5a RECEPTOR (C5aR) AND ROLE IN SEPSIS
The C5aR (CD88) is described in Table 1 . C5aR is a 7 transmembrane-spanning receptor, which binds C5a (and C5a des arg) with high affinity (low nM range), and is linked with intracellular G proteins, necessary for cellular signaling. Interaction of C5aR with C5a results in a rapid increase in cytosolic Ca2+ as a result of release of Ca2+ from intracellular stores. The cytosolic Ca2+ increase is transient but is one of the earliest responses of cells to C5a. Although it used to be thought that the C5aR presence was restricted to myelopoietic cells, it is now clear that C5aR is expressed on a variety of other cell types, such as endothelial cells, alveolar epithelial cells, neurons, and cardiomyocytes (CMs), to name just a few examples. The expression of C5aR is constitutive and inducible. In the setting of experimental sepsis induced by cecal ligation and puncture (CLP), C5aR is up-regulated substantially in a variety of organs [6 ]. In thymocytes, up-regulation occurs within a few hours after CLP [7 ]. Up-regulation of C5aR after CLP has also been found on CMs [8 ], as well as in liver, lung, and kidney [6 ]. Such up-regulation infers that a variety of tissues may develop highly undesirable effects after encounters with C5a. In the heart, profound CM contractile defects develop after CLP [8 ]. In the thymus and perhaps in other lymphoid tissues, apoptosis develops, via the intrinsic pathway (featuring mitochondrial release of cytochrome c and caspase activation) [9 ]. Apoptosis is known to occur in lymphoid tissues of septic humans and has been postulated to be the cause of the immunosuppression of sepsis [10 ].
|
View this table: [in a new window] |
Table 1. C5aR
|
, or IL-6. In vivo, blockade of IL-6 prevents up-regulation of C5aR in CLP rodents and improves survival [11
]. In the setting of CLP, organ up-regulation of C5aR can be blocked in mice by administration of neutralizing antibody to IL-6 [2
]. Finally, interaction of C5a with C5aR on blood polymorphonuclear neutrophils (PMNs) results in internalization of the C5a•C5aR complex, with translocation to the Golgi region, where in an acidic environment, the ligand (C5a) is hydrolyzed, and at least some of the C5aR is recycled to the cell membrane. In CLP rats, the greater the loss of C5aR on the PMN surfaces, the higher the lethality [12
]. This relationship may reflect the intensity of in vivo complement activation during sepsis or the ability of PMNs to replete C5aR rapidly on cell surfaces. Currently, there are ongoing studies in human sepsis to determine if the loss of C5aR on PMNs is predictive of a lethal outcome. THE C5L2R FOR C5a AND CHANGES IN SEPSIS
The recently described second C5aR on leukocytes, C5L2, is rather enigmatic [13 14 15 16 17 18 ]. C5L2 was described several years ago [13 14 15 16 17 18 ]. C5L2 is reviewed in Table 2 . It was quickly determined that, although C5L2 binds C5a and C5a des arg with high affinity, no intracellular Ca2+ signal developed, and no functional response occurred (e.g., enzyme release, respiratory burst, MAPK signaling, etc.). Accordingly, C5L2 was judged to be a default or "scavenger" receptor, in which it would bind C5a and compete with C5aR for C5a binding. Although it has been suggested that C5L2 is a "promiscuous" receptor, which can also bind C3a and C3a des arg on myeloid cells and adipocytes, such claims have been questioned recently [15 ]. Although no cell signaling was thought to occur upon engagement of C5L2, this matter has been confused in light of two recent reports, which suggest signaling in adipocytes exposed to C3a or C3a des arg [16 ] and mouse bone marrow-derived phagocytes incubated with C5a [18 ]. In the case of bone marrow-derived macrophages, it has been suggested that in the absence of C5aR or C5L2, ERK1/2 phosphorylation is diminished. It is difficult to reconcile these contrasting claims at the present time.
|
View this table: [in a new window] |
Table 2. C5L2
|
EVIDENCE FOR THE ROLE OF C5a IN EXPERIMENTAL SEPSIS
The data described in Table 3 are the foundations for concluding that in CLP rodents, the septic syndrome is linked to generation of C5a. As in humans with sepsis, complement activation after CLP is associated with consumptive depletion of complement and the appearance in plasma/serum of C5a [21 ]. When C5a was blocked in vivo (at the time of CLP), there was a dramatic improvement in survival [21 , 22 ]. Paralysis of the ERK1/2 signaling pathway was found in blood PMNs within 12–18 h after CLP [23 ], resulting in marked defects in phagocytosis, chemotaxis, and assembly of NADPH oxidase. These functional defects in PMNs were prevented when C5a was intercepted with anti-C5a immediately after CLP. Similar restrictions in innate immune responses were found in human blood PMNs exposed in vitro to C5a. The basis for the defect in the respiratory burst was the inability of stimulated PMNs to translocate one of the critical subunits (p47phox) from the cytosol to the cell membrane, where assembly of NADPH oxidase occurs.
|
View this table: [in a new window] |
Table 3. Role of C5a in CLP-Induced Sepsis
|
ROLE FOR C5aR IN EXPERIMENTAL SEPSIS
As indicated above, C5aR is up-regulated after CLP in rats and in mice. There is considerable evidence for the participation of C5aR in the harmful consequences of experimental sepsis. Blockade of C5aR has been accomplished by the use of antibody to C5aR or by the use of a cyclic peptide, which prevents interaction of C5a with C5aR [25 ]. In vivo blockade of C5aR resulted in greatly improved survival after sepsis. In addition, interception of IL-6 (which is a potent inducer of C5aR) significantly improved survival after CLP in mice [11 ]. Collectively, these data provide direct evidence for the harmful role of C5aR in the setting of experimental sepsis. The harmful effects of C5a interaction with C5aR in the setting of experimental sepsis are described in Figure 1 , in which sepsis-induced activation of complement results in two harmful outcomes: loss of innate immune functions of blood neutrophils (PMNs) and reduced CM contractility. The appearance of both types of defects is C5a- and C5aR-dependent. In the case of PMNs, signaling paralysis results in loss of three key functions (phagocytosis, chemotaxis, assembly of NADPH oxidase), all of which results in the loss of protective functions of PMNs. In the case of CM, sepsis results in increased cell surface expression of C5aR. Upon interaction with C5a, CM contractility is reduced greatly, resulting in falling left ventricular pressures and reduced cardiac output. Development of these dysfunctions can be blocked by in vivo interception (with antibody) of C5a.
 View larger version (29K): [in a new window] |
Figure 1. Summary of C5a-C5aR interactions in sepsis, leading to loss of innate immune functions of PMNs and contractile dysfunction of CMs.
|
There is developing evidence that humans with septic shock (requiring vasopressors to maintain adequate blood pressure) show several of the features observed in septic rodents after CLP. In both situations, there is evidence of consumptive depletion of complement in serum/plasma and the presence of complement activation products (C3a, C5a). Another set of observations relates to C5L2 content on blood PMN lysates (defined by Western blots) and on blood PMNs (as determined by flow cytometric analysis) from humans with septic shock. In the case of rats, lysates from blood PMNs, 24 h after CLP, showed a dramatic reduction in C5L2 content, as defined by Western blot analysis [26 ]. The loss of C5L2 occurred as a function of time after CLP, and substantial reductions occurred at 3 h and 4 h. In addition, it was shown that the loss of PMN C5L2 after CLP was dependent on C5a, as in vivo blockade of C5a prevented the loss of C5L2 from blood PMNs. In the case of humans with septic shock, C5L2 levels fell >95% on PMNs in the aggregate group of septic patients, by 84% in septic patients without multiorgan failure (MOF), and by nearly 90% in septic patients with MOF. The loss of C5L2 on normal human PMNs could be reproduced when cells were exposed in vitro to C5a, indicating a loss of receptor by internalization or by receptor shedding. These data indicate that sepsis causes dramatic reductions in PMN surface content of C5L2 and that the same occurs on blood PMNs taken from CLP rats. The loss of C5L2 has been shown to be C5a-dependent. To what extent similar changes occur in C5aR in blood PMNs in septic humans is a matter of current study.
SUMMARY
There is accumulating evidence that complement activation occurs during experimental sepsis (CLP) and also in human sepsis. Interaction of the anaphylatoxin, C5a, with its receptor, C5aR, has highly adverse consequences. When C5a interacts with blood PMNs, signaling pathways are paralyzed, associated with internalization of C5a-C5aRs complexes. Signaling paralysis results in loss of innate immune functions of PMNs (phagocytosis, chemotaxis, assembly of NADPH oxidase), which has ominous implications for defenses against bacteremia. With respect to the well-known cardiomyopathy of sepsis, the loss of cardiac function can be linked to C5a interaction with C5aR on CMs. In vivo interception of C5a or C5aR is highly protective in the setting of experimental sepsis, greatly improving survival after CLP, preserving the innate immune functions of PMNs, and normalizing cardiac function. The data suggest that in vivo blockade of C5a or C5aR may be protective in the setting of sepsis in humans.
ACKNOWLEDGEMENTS
This work was based on presentations at the 7th World Congress on Trauma, Shock, Inflammation and Sepsis (Munich, March 2007).
Received June 9, 2007; revised August 2, 2007; accepted August 28, 2007.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  K. T. Lappegard, D. Christiansen, A. Pharo, E. B. Thorgersen, B. C. Hellerud, J. Lindstad, E. W. Nielsen, G. Bergseth, D. Fadnes, T. G. Abrahamsen, et al. Human genetic deficiencies reveal the roles of complement in the inflammatory network: Lessons from nature PNAS, September 15, 2009; 106(37): 15861 - 15866. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
