The role of complement in post-ischemic cerebral injury is incompletely understood. Therefore, experiments were designed to test the effect of complement depletion on cerebral infarct volume in adult rats and cerebral atrophy in neonatal rats. Cerebral infarcts were induced in adult rats by transient filamentous occlusion of the right middle cerebral artery (MCAO). Cerebral atrophy was induced by subjecting 7-day-old rats to ligation of the right common carotid artery followed by 2.5h of hypoxia (8% O2). Forty-eight hours after MCAO, coronal sections of adult brains were obtained and stained with 2,3,5-triphenyl tetrazolium chloride. The infant rat brains were removed for analysis 6 weeks after the hypoxic-ischemic insult. Volumes of infarcts and normal hemispheric parenchyma were quantified by computer-based planimetry. Twenty-four hours prior to MCAO (adults) or hypoxia-ischemia (neonates), each animal received an i.p. injection of either 1 mcg/g body weight cobra venom factor (CVF; adult n=11; neonatal n=20) or normal saline (adult n=12; neonatal n=24). In the neonates, a second dose of CVF or saline was administered 2 days after hypoxia-ischemia. The administration of CVF significantly reduced: (1) post-ischemic cerebral infarct volume in the adults and (2) post-hypoxic-ischemic cerebral atrophy in the neonates. Therefore, complement activation augmented post-ischemic cerebral injury in adult and neonatal rats. Complement depletion induced by CVF significantly reduced post-ischemic cerebral infarct volume and atrophy in adult and neonatal rats.

It is not known whether up-regulation of complement components, either circulating or locally synthesized, contributes to an increased susceptibility to neonatal hypoxic-ischemic (HI) cerebral injury. Therefore, we tested the hypothesis that in neonatal rats subjected to a unilateral HI cerebral insult, prior administration of E. coli lipopolysaccharide (LPS) augments (1) complement-mediated serum hemolytic activity, and (2) C3 mRNA and C9 mRNA levels in hepatic and cerebral tissue. Pregnant rats were injected subcutaneously with sterile normal saline (NS) or 500 μg/kg of LPS on gestational days 18 and 19. Following birth, the pups received intraperitoneal injections of NS or 250 μg/kg of LPS on postnatal days 3 and 5. On postnatal day 7, each animal was subjected to ligation of the right common carotid artery followed by 2.5 h of hypoxia (8% O2). At 3, 6,18, 24 and 48 h after hypoxia, the complement-mediated hemolytic activity of pooled serum was measured. Hepatic and cerebral C3 mRNA and C9 mRNA were quantified by qRT-PCR at 3, 6, and 18 h after HI. Serum hemolytic activity, hepatic C3 mRNA, and hepatic C9 mRNA were up-regulated after cerebral HI. LPS administration potentiated the effect of HI on serum hemolytic activity and increased cerebral C3 mRNA levels. Cerebral C9 mRNA was not detected and was not affected by HI, with or without the prior LPS administration. These observations support the theory that previously reported C9-mediated neurotoxicity following cerebral HI is induced by circulating, rather than locally synthesized C9.

The role of complement in neonatal hypoxic–ischemic brain injury is not known. Therefore, cerebral spinal fluid (CSF) and post-mortem cerebral tissue were analyzed to determine whether complement is activated and complement component 9 (C9) is deposited on neurons in the central nervous systems (CNS) of newborn infants who developed moderate to severe hypoxic–ischemic encephalopathy (HIE). Control CSF samples were obtained during routine evaluation for possible sepsis from infants who were not depressed at birth. In ELISA assays of CSF obtained from 16 infants with HIE, compared to CSF from 7 control infants, the mean concentration of terminal complement complexes was elevated and the mean C9 concentration was diminished. Immunofluorescence microscopy of post-mortem frozen brain tissue obtained from two infants who expired at 4–5 days of life after severe HIE revealed that activated C9 was deposited on cells in all lobes. Double label immunofluorescence microscopy demonstrated that nearly all of the C9-positive cells were neurons and essentially all of the neurons were C9-positive. Immunoperoxidase immunohistochemistry of formalin-fixed tissue also confirmed the presence of many C9-positive cells, particularly in the hippocampus. The C9-positive cells usually manifested morphology consistent with neurons, most of which contained fragmented nuclei. In summary, complement was activated in the CNS of newborn infants who developed moderate to severe HIE. C9 was deposited on neurons, including morphologically apoptotic neurons. Further investigations into a possible role of complement in the pathogenesis of neonatal hypoxic–ischemic cerebral injury are warranted.

BACKGROUND AND OBJECTIVE:CAP37, also known as heparin-binding protein (HBP), is neutrophil-derived protein with multifunctional properties that include monocyte chemotaxis and the enhancement of LPS-induced tumor necrosis factor (TNF-alpha), IL-1, IL-6, and PGE2production from isolated monocytes, which suggest a generalized effect on LPS-induced monocyte activation. In this study, we tested whether HBP amplifies the release of other LPS-responsive cytokines from isolated human monocytes.METHODS:Freshly isolated monocytes from 5 healthy donors were stimulated for 24 h with saline, LPS (10 ng/ml), HBP (10 microg/ml), or a combination of LPS + HBP. Cytokine levels in the supernate were measured with ELISA. ANOVA and Fisher's posthoc test were used to determine significance (p < 0.05). Differential display was used to assess cellular mRNA levels.RESULTS:HBP alone induced the production of IL-8, macrophage inhibitory protein MIP-1alpha, and TNF-alpha. HBP increased the LPS-induced production of IL-8, MIP-1alpha, TNF-alpha, IL-1beta, but HBP did not increase the significant LPS-induced release of IL-10, monocyte chemoattractant protein MCP-1, and IL- 12. Differential display demonstrated that HBP induced an mRNA pattern that was different from the mRNA pattern induced by saline, LPS, or HBP + LPS, indicating multiple and different gene activation.CONCLUSIONS:We conclude that HBP is not a general amplificator of LPS-induced monocyte activation but rather a molecule that targets the production of a distinct set of mediators including pro-inflammatory cytokines such as TNF-alpha and IL-1beta, but not the anti-inflammatory cytokine IL-10, nor IL-12 and MCP-1. The exact intracellular signaling pathways remain unknown but include mechanisms that alter gene transcription.

Endotoxin (lipopolysaccharide (LPS), 100 ng/ml) and muramyl dipeptide (MDP 100 ng/ml), two immunomodulatory bacterial cell wall products, were incubated with human whole blood, and the expression of receptors involved in antigen presentation, costimulation, and cell activation was investigated by use of flow cytometry. On monocytes, LPS and MDP increased surface expression of human leukocyte antigen-DR (HLA-DR), CD18, CD54 (intercellular adhesion molecule-1, ICAM-1), and CD86 (B7-2). On lymphocytes, LPS but not MDP increased HLA-DR expression after 18 h. The expression of CD28, CD49d/CD29, and CD106 (vascular cell adhesion molecule-1, VCAM-1) remained unchanged on both monocytes and lymphocytes. The early increase (1-6 h) of CD18 and ICAM-1 expression led us to hypothesize that CD18-dependent costimulatory signals were involved in the later (6 h) increase of monocyte HLA-DR expression. However, blocking studies using monoclonal antibodies against CD18 (IB4, 15 microg/ml) demonstrated that the LPS- and MDP-induced increase of HLA-DR and ICAM-1 expression on monocytes was not mediated through CD18. LPS induced the expression of the early activation marker CD69 by a CD14-dependent but CD18-independent mechanism, whereas MDP did not induce CD69 expression. Analysis of leukocyte subsets demonstrated that CD4(+) T-cells, CD8(+) T-cell, CD19(+) B-cells, CD56(+) natural killer (NK)-cells, and CD14(+) monocytes increased the expression of CD69 after stimulation with LPS. Collectively, these data demonstrate a stronger immunomodulatory effect of LPS compared with MDP which may, in part, explain the established difference of toxicity between these two bacterial cell wall products.

The behaviour of human leucocyte antigen-DR (HLA-DR) following injury has been extensively studied. However, the behaviour of other class II antigens following trauma has not been characterized as well, despite evidence that HLA-DQ genotype influences the response to several bacterial antigens. Our study attempts to characterize and analyse the behaviour of HLA-DQ after trauma in patients with and without infection. Twenty-five patients were studied following major injury. Fifteen of the 25 patients developed infection (men = 11, women = 4); 10 patients developed no infection (men = 9, women = 1). The mean age was 34 +/- 12 years for patients with no infection and 52 +/- 20 years for those with infection. Monocyte HLA-DQ surface expression was determined using FITC-labelled antibodies and flow cytometry. Expression was compared with a control population of 11 healthy volunteers. The percentage of monocytes expressing HLA-DQ following trauma was reduced in patients with infection and in those without infection, but returned to normal (days 8-14) only in those patients who did not develop infection. Monocyte HLA-DQ mean channel fluorescence was reduced on day 1, but quickly returned to normal in those patients who subsequently developed infection. Stimulated with lipopolysaccharide, the initial samples of 13 patients who developed infection showed that surface expression on these monocytes could be elevated into the normal range. We conclude that HLA-DQ is an additional early marker of outcome that may not function merely as an immune suppressor. The maintained ability of HLA-DQ to present self-antigens may be important in the initial stages of the host response to injury.

HYPOTHESISPrevious studies on alterations in phagocytosis and bacterial killing after trauma have yielded conflicting results. We hypothesize that these changes are variable, depending on the species of bacteria used to assay these variables.DESIGNBlood samples from patients were assayed by means of flow cytometry for phagocytosis and reactive oxygen intermediate generation. Several common clinical pathogens were used: Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus. Results were compared with those from controls.SETTINGRegional level I trauma center.PATIENTSTen consecutive patients were studied with E. coli and K. pneumoniae. Five of these were also studied with S. aureus. Patients were 18 years of age or older, with an Injury Severity Score of 16 or more. Patients who were taking corticosteroids before hospital admission or who were administered corticosteroids before blood was drawn were not studied. Isolated head injuries or limb fractures were also excluded. Controls consisted of healthy volunteers.MAIN OUTCOME MEASURESThe ingestion of bacteria by neutrophils and the generation of reactive oxygen intermediates.RESULTSAfter trauma, phagocytosis of E. coli was enhanced, whereas ingestion of K. pneumoniae was depressed. Ingestion of S aureus remained unchanged. The generation of reactive oxygen intermediates was depressed after incubation with E. coli and unchanged with K. pneumoniae, but enhanced with S. aureus.CONCLUSIONSNeutrophil response to trauma is dependent on which bacterial species the cell is attempting to kill. This may, in part, explain why only a limited number of bacterial species cause a significant proportion of early infections after trauma.

OBJECTIVE:To determine whether heparin or the low-molecular-weight heparin enoxaparin alter lipopolysaccharide (LPS)-induced monocyte activation.SUMMARY BACKGROUND DATA:Heparin is widely used in clinical practice to inhibit the coagulation cascade. However, heparin also is a naturally occurring glucosaminoglycan and a pleiotropic immunomodulator that binds to a variety of proteins. LPS is a component of gram-negative bacteria and is thought to be responsible for many of the deleterious effects seen in sepsis. The binding of LPS to CD14 induces a signaling cascade that results in the release of many inflammatory mediators, including tumor necrosis factor-alpha (TNF-alpha).METHODS:Monocytes from healthy volunteers were isolated and cultured in the presence of saline, LPS (10 ng/ml), heparin (0.1 to 1000 microg/ml), or enoxaparin (0.1 to 1000 microg/ml). In blocking experiments, cells were pretreated for 60 minutes with the monoclonal anti-CD14 antibody MY4 (10 microg/ml) or with isotype-matched control IgG2 (10 microg/ml). TNF-alpha values were measured with enzyme-linked immunosorbent assay. Significance was assessed with analysis of variance.RESULTS:Heparin (10 to 1000 microg/ml) and enoxaparin (1000 microg/ml) significantly enhanced LPS-induced TNF-alpha release. Heparin (1000 microg/ml) or enoxaparin (1000 microg/ml) did not produce TNF-alpha in the absence of LPS. Blockade of CD14 abrogated both LPS-induced TNF-alpha release and the effect of heparin or enoxaparin to enhance LPS-induced TNF-alpha release.CONCLUSIONS:The effect of heparin to enhance LPS-induced TNF-alpha release is a biologic phenomenon that reveals a novel and potentially important host defense mechanism during endotoxemia and sepsis. Binding of LPS to CD14 is necessary to induce this phenomenon, suggesting that both heparin and enoxaparin induce signaling mechanisms that are downstream from the initial binding of LPS on CD14.