Tranexamic acid mediates proinflammatory and anti-inflammatory signaling via complement C5a regulation in a plasminogen activator-dependent manner.

BACKGROUND: Both tissue plasminogen activator (tPA) in the circulation and urokinase (uPA) in tissues cleave plasminogen (PLG) to plasmin to promote clot lysis. Tranexamic acid (TXA) blocks both the tPA-dependent generation of plasmin on blood clots as well as active plasmin binding to polymerized fibrin, and is commonly administered for bleeding in trauma to limit fibrinolysis. In addition to lysing clots, however, active plasmin also cleaves complement proteins, potentially enhancing inflammation. Because TXA does not block uPA-dependent plasmin generation from PLG and instead augments it, we hypothesized that administration of TXA could enhance or inhibit proinflammatory C5a formation in a PLG activator-dependent manner. METHODS: Citrate platelet-poor plasma (PPP) and PPP depleted of complement protein C3 or PLG were obtained from healthy donors and commercial sources. Platelet-poor plasma was treated ex vivo with or without TXA and either with or without tPA or with or without uPA. Clotting was then induced by calcium and thrombin in clotted PPP experiments, while unclotted PPP experiments were treated with vehicle controls. C5a levels were measured via enzyme-linked immunosorbent assay. Data were expressed as mean ± SEM. RESULTS: Plasmin-mediated fibrinolysis by tPA in clotted PPP led to an approximately threefold increase in C5a production (p < 0.0001), which was significantly inhibited by TXA (p < 0.001). Paradoxically, when fibrinolysis was induced by uPA, TXA treatment led to further increases in C5a production beyond uPA alone (p < 0.0001). Furthermore, clotting was not required for C5a generation from uPA + TXA. C3 depletion had no effect on C5a production, while depletion of PLG eliminated it. CONCLUSIONS: Tranexamic acid administration can have proinflammatory or anti-inflammatory effects through regulating C5a generation by plasmin, depending on the predominating PLG activator. Tranexamic acid may cause significant inflammatory C5a elevations in injured tissues by augmenting uPA-mediated plasmin generation in a fibrin-independent manner. In contrast, TXA reduces C5a generation during tPA-mediated fibrinolysis that may reduce inflammatory responses. In vivo validation of these novel ex vivo findings is warranted and may have important clinical consequences.

Complement C5a Fosters Squamous Carcinogenesis and Limits T Cell Response to Chemotherapy.

Complement is a critical component of humoral immunity implicated in cancer development; however, its biological contributions to tumorigenesis remain poorly understood. Using the K14-HPV16 transgenic mouse model of squamous carcinogenesis, we report that urokinase (uPA)+ macrophages regulate C3-independent release of C5a during premalignant progression, which in turn regulates protumorigenic properties of C5aR1+ mast cells and macrophages, including suppression of CD8+ T cell cytotoxicity. Therapeutic inhibition of C5aR1 via the peptide antagonist PMX-53 improved efficacy of paclitaxel chemotherapy associated with increased presence and cytotoxic properties of CXCR3+ effector memory CD8+ T cells in carcinomas, dependent on both macrophage transcriptional programming and IFNγ. Together, these data identify C5aR1-dependent signaling as an important immunomodulatory program in neoplastic tissue tractable for combinatorial cancer immunotherapy.

Dendritic cell function in allostimulation is modulated by C5aR signaling.

Regulation of T cell immunity by C5a has been suggested from recent studies. However, the underlying mechanisms, particularly the involved cells and biochemical basis, are not well defined. In this study, the direct modulation of dendritic cell (DC) activation and its function in T cell stimulation by C5a-C5aR interaction and the involved signaling pathways were investigated. We show that DCs from C5aR(-/-) mice and normal DCs treated with C5aR antagonist have less-activated phenotype characterized with increased IL-10 and decreased IL-12p70 production in response to LPS stimulation, lowered surface expression of MHC class II, B7.2, and consequently have reduced capacity to stimulate allospecific T cells. Conversely, C5a stimulation up-regulates DC activation and its function in allostimulation. Furthermore, stimulation of C5aR mediates the inhibition of cAMP production and protein kinase A activity and is involved in activation of PI3K/AKT and NF-kappaB signaling in DCs. These results demonstrate that C5a acts directly on C5aR expressed on DCs resulting in the cell activation and subsequently enhances its capacity for allospecific T cell stimulation. It also suggests that NF-kappaB signaling induced by down-regulation of cAMP/ protein kinase A pathway and up-regulation of PI3K/AKT pathway following C5a stimulation may contribute to up-regulation of DC function.

In vitro cleavage by asbestos fibers of the fifth component of human complement through free-radical generation and kallikrein activation.

Chrysotile and crocidolite fibers incubated in normal human plasma (NHP) generated from the C5 component of complement C5a-type fragments that stimulated polymorphonuclear leukocyte (PMN) chemotaxis. Absorption of NHP with antiserum against C5a totally abolished neutrophil chemotactic activity. Asbestos fibers also produced C5a small peptides in the presence of ethylene glycol bis(beta-aminoethyl ether) N,N,N'N'-tetraacetic acid (EGTA) but not ethylene diamine tetraacetic acid (EDTA). Activation of C5 was significantly inhibited when asbestos fibers were pretreated with iron chelators such as sodium dithionite (DTN), deferoxamine (DFX), or ascorbate (AA). Concentration-related inhibition of C5 activation was also observed when asbestos fibers were added concurrently to plasma in the presence of DFX, 1,3-dimethyl-2-thiourea (DMTU), a strong hydroxyl scavenger, or aprotinin (APR), a specific protease inhibitor. Further, chrysotile and crocidolite significantly increased plasma kallikrein activity. Data demonstrate that asbestos-induced C5 activation plays a role in inflammatory reactions characteristic of asbestosis through mechanisms involving iron ions, hydroxyl radicals, and oxidized C5-ike fragments. The ferrous ions present at the asbestos fiber surface trigger this activation and catalyze, via Fenton reaction, the production of hydroxyl radicals, which in turn convert native C5 to an oxidized C5-like form. This product is then cleaved by kallikrein, activated by the same asbestos fibers, yielding an oxidized C5a with the same functional properties as C5a.

IFN-gamma up-regulates the human C5a receptor (CD88) in myeloblastic U937 cells and related cell lines.

On human mature monocytes the immunomodulator IFN-gamma has been shown to down-regulate the receptor for the anaphylatoxic peptide C5a (CD88, C5aR). In this study, we show that in immature myelo-/monoblastic U937, HL60, and MonoMac6 cells, IFN-gamma induces C5aR-ligand binding activity. In U937 cells, this induction cannot be blocked by the protein kinase C inhibitor staurosporine. An increase in free cytosolic Ca2+ upon ligand binding indicates functional coupling of this receptor in U937 and HL-60 cells. G-Proteins involved in this C5a responsiveness after IFN-gamma induction are completely pertussis toxin sensitive. Our data suggest that an additional pertussis toxin-resistant pathway exists in U937 cells after induction by dibutyryl cAMP. However, this is not due to changes in the mRNA level of the pertussis toxin-insensitive G-protein subunit G alpha 16. Induction by dibutyryl cAMP, but not that by IFN-gamma, resulted in C5a-dependent release of N-acetyl-beta-D-glucosaminidase, further highlighting functional differences in the effects of the inducers. Our data show an IFN-gamma-dependent increase in C5aR expression and suggest a maturation-related change in signaling of the C5aR, presumably at the level of receptor coupling.

Influence of steroids on complement and cytokine generation after cardiopulmonary bypass.

BACKGROUND: It is recognized that the inflammatory mediators complement and cytokines are generated during cardiopulmonary bypass as an endogenous response to extracorporeal circulation. METHODS: Nineteen randomized patients (10 steroid/9 nonsteroid) entered an institutional review board-approved protocol to measure complement and interleukin level generation before and after elective coronary revascularization. The steroid regimen involved 1 g of methylprednisolone sodium succinate intravenously before bypass and 4 mg of dexamethasone every 6 hours for four doses during the first 24 hours of recovery. Complement and interleukin levels were measured before bypass, immediately after bypass, and at 24, 48 and 72 hours of recovery. RESULTS: In the nonsteroid group, there was a significant elevation in all inflammatory mediators relative to the steroid group. The predominant changes occurred at 24 hours after operation. CONCLUSIONS: Steroids produced a dramatic reduction in complement and interleukin levels. The number of patients was clearly too small to document a clinical consequence of steroid administration.

Complement activation by vascular sutures both alone and in combination with synthetic vascular prostheses.

Polymer surfaces activate complement pathways resulting in platelet and leucocyte deposition as well as possible release of growth factors. A consequence of these interactions may be early graft failure or intimal hyperplasia leading to late graft failure. C5a generation in human plasma by vascular sutures, both alone and in combination with synthetic vascular prostheses was measured by radioimmunoassay to determine the influence of suture materials on C5a activation. Prolene and ePTFE suture material caused significant activation of C5a (p less than 0.01), while Novafil did not. Both Dacron and ePTFE graft material caused significant activation (p less than 0.01) of C5a. The addition of the suture materials to the ePTFE did not increase the C5a levels above the ePTFE material alone. In contrast, the addition of either Prolene or Novafil suture to Dacron material elevated C5a levels significantly over Dacron alone (p less than 0.01). The combination of Dacron material with ePTFE suture did not increase C5a levels over Dacron alone. The pattern of C5a activation by Prolene, ePTFE and Novafil sutures parallels the relative degree of in-vivo platelet accumulation on these suture materials as previously reported by our group. Since these experiments demonstrate that vascular suture material influences human complement activation, it may be that this interaction contributes to either early or late graft failure by enhancing platelet reactivity or neointimal proliferation, respectively.