Hemorrhage induces enhanced Kupffer cell cytotoxicity while decreasing peritoneal or splenic macrophage capacity. Involvement of cell-associated tumor necrosis factor and reactive nitrogen.

Studies indicate that simple hemorrhage produces a profound depression of cell-mediated immunity, thereby contributing to an enhanced susceptibility to septic challenge in the host. However, it remains unknown whether or not the macrophages' cytotoxic capacity is altered after hemorrhage. To study this, C3H/HeN mice were bled to and maintained at a blood pressure of 35 mm Hg for 60 min, and adequately resuscitated. Mice were then killed at 2 or 24 h after hemorrhage to obtain peritoneal macrophage, splenic macrophage, and Kupffer cells. Cytotoxicity was assessed by determining the capacity of these macrophages to lyse [3H]TdR labeled WEHI-164 clone 13 or P815 tumor target cells (WEHI-164, sensitive to both soluble and cell-associated TNF vs P815 cells, insensitive to soluble TNF). Peritoneal and splenic macrophages from hemorrhaged animals exhibited a significantly reduced cytotoxic capacity, whereas Kupffer cells' ability to kill the target cells was enhanced. Similarly, the Kupffer cells' capacity to release TNF and IL-1, as well as express cell-associated forms of this cytokine are significantly enhanced on macrophages isolated 2 h after hemorrhage, whereas peritoneal macrophages are not. Furthermore, antibodies directed at mouse TNF but not against murine IL-1 alpha or murine IL-6 were able to oblate the enhanced target cell lysis of unfixed, as well as paraformaldehyde fixed (metabolically inactive) Kupffer cells. Studies using inhibitors (GN-monomethyl-arginine, superoxide dismutase, catalase, and ibuprofen) of other TNF-inducible mechanisms of target cell killing indicated that only the inhibition of the release of reactive nitrogen consistently depressed the cytotoxic capacity of Kupffer cells from hemorrhaged mice. Thus, the increased Kupffer cell cytotoxicity from hemorrhaged mice is most likely mediated through the expression of cell-associated TNF and the release of reactive nitrogen.     

Differences in the altered energy metabolism of hemorrhagic shock and hypoxemia.

The effect of hemorrhagic shock, hypoxemia, and anoxia on the levels of adenine and pyridine nucleotides of liver and kidney was assessed. ATP levels in liver and kidney of animals in shock or animals subjected to 7 min of anoxia decreased by 85 and 73%, respectively. Under hypoxic conditions (arterial PO2 AT 18 MMHg), the decrease was only 62 and 48% in liver and kidney, respectively. Tissue NAD levels decreased and NADH levels increased during shock but were found to be essentially unaltered during experimental hypoxemia. Thus, shock produced greater alterations in adenine and pyridine nucleotides than did hypoxemia alone, indicating that stagnant hypoxemia due to shock is more deleterious to energy metabolism than is severe hypoxemia with an otherwise normal circulation. The results also suggest that if an anterial PO2 OF 18 MMHg represents the initial stages of tissue hypoxia, then tissue ATP levels are a more sensitive indicator of this than NAD levels.     

Hemorrhage induces an increase in serum TNF which is not associated with elevated levels of endotoxin

Although tumor necrosis factor (TNF) and interleukin 6 (IL 6) are purported to be important mediators of inflammatory responses following trauma, it is not known if the serum levels of these cytokines are altered by simple hemorrhage. The objective of this study therefore was to determine whether or not: 1) there is any elevation of TNF or IL 6, and 2) if endotoxin, an important upregulator of these cytokines, is also increased following hemorrhage. To study this, C3H/HeN mice were bled to, and maintained at a mean blood pressure of 35 mmHg for 60 min, and then resuscitated with their own shed blood and adequate fluid. Mice were sacrificed at 30 min into hemorrhage and at 2, 4 or 24 hr post-hemorrhage to obtain serum samples. IL 6 and TNF levels were measured using cytokine dependent cellular assays. Using a quantitative Limulus amebocyte lysate assay, endotoxin levels were determined. TNF levels were significantly elevated at 30 min into hemorrhage, remaining so at 2 hr after resuscitation, but absent by 4 hr. Although there was a trend toward elevated IL 6 levels at 2 hr following hemorrhage, which was sustained up to 24 hr, the values were not significantly different from sham controls. When compared to controls, no marked increase in endotoxin was seen at any time point during or following hemorrhage. These results indicate that hemorrhage, in the absence of significant tissue trauma, causes enhanced TNF release which is not the result of increased endotoxin.     

The role of lipopolysaccharide in stimulating adrenomedullin production during polymicrobial sepsis

Previous studies have shown that adrenomedullin (AM), a potent vasodilatory peptide, is upregulated during sepsis. However, it remains unknown whether the increased AM observed under such conditions is solely due to the elevated levels of circulating lipopolysaccharide (LPS). To determine this, an Alzet micro-osmotic pump, containing a low dose of Escherichia coli LPS or vehicle (sterile normal saline), was implanted in the peritoneal cavity of the normal male adult rat. At 10 h after the pump implantation, samples of blood and small intestine were harvested for the determination of AM by radioimmunoassay. In additional groups, rats were subjected to polymicrobial sepsis by cecal ligation and puncture (CLP). LPS binding agent polymyxin B was administrated intramuscularly at 1 h prior to as well as 5 h after the onset of sepsis. At 10 h after CLP or sham-operation, blood and intestinal samples were harvested and levels of AM were then determined. Plasma levels of LPS were also measured by Limulus amebocyte lysate assay. The results indicate that administration of a low dose of LPS via the peritoneal cavity in normal animals (which did not significantly alter cardiac output, blood pressure or heart rate) markedly increased plasma and intestinal levels of AM. In addition, plasma and tissue levels of AM increased significantly at 10 h after CLP. Administration of polymyxin B, however, attenuated the increase in AM levels under such conditions. Similarly, the increased plasma levels of LPS was significantly reduced by polymyxin B during sepsis. These results, taken together, suggest that the upregulated AM observed during polymicrobial sepsis is at least in part due to the increase in circulating levels of endotoxin.     

The role of Kupffer cell alpha(2)-adrenoceptors in norepinephrine-induced TNF-alpha production

Although previous studies have demonstrated that plasma levels of the proinflammatory cytokine tumor necrosis factor-alpha (TNF-alpha) increase during early sepsis, the precise mechanism responsible for its upregulation remains to be elucidated. Since recent studies have shown that the gut is an important source of norepinephrine (NE) release during early sepsis and enterectomy prior to the onset of sepsis attenuates TNF-alpha production, we hypothesized that gut-derived NE plays a major role in upregulating TNF-alpha via the activation of alpha(2)-adrenoceptors on Kupffer cells. To confirm that NE increases TNF-alpha synthesis and release, Kupffer cells were isolated from normal rats and incubated with NE (20 or 50 nM) or another alpha(2)-adrenergic agonist clonidine (50 nM) without addition of Escherichia coli endotoxin. Supernatant levels of TNF-alpha were then measured. In additional animals, intraportal infusion of NE (20 microM) with or without the specific alpha(2)-adrenergic antagonist yohimbine (1 mM) at a rate of 13 microl/min was carried out for 2 h. Plasma and Kupffer cell levels of TNF-alpha were assayed thereafter. Moreover, the effects of NE and yohimbine on TNF-alpha production was further examined using an isolated perfused liver preparation. The results indicate that both NE and clonidine increased TNF-alpha release by approximately 4-7-fold in the isolated cultured Kupffer cells. Similarly, intraportal infusion of NE in vivo or in isolated livers increased TNF-alpha synthesis and release which was inhibited by co-infusion of yohimbine. Furthermore, the increased cellular levels of TNF-alpha in Kupffer cells after in vivo administration of NE was also blocked by yohimbine. These results, taken together, suggest that gut-derived NE upregulates TNF-alpha production in Kupffer cells through an alpha(2)-adrenergic pathway, which appears to be responsible at least in part for the increased levels of circulating TNF-alpha observed during early sepsis as well as other pathophysiologic conditions such as trauma, hemorrhagic shock, or gut ischemia/reperfusion.     

Bystander suppression of collagen-induced arthritis in mice fed ovalbumin

We wanted to assess whether B-cell and/or T-cell responses to collagen and thereby the course of collagen-induced arthritis could be suppressed by regulatory mechanisms associated with oral tolerance to an unrelated protein. DBA/1 mice were fed ovalbumin (OVA)-containing pellets ad libitum for 1 week and subsequently coimmunized twice, with a mixture of bovine collagen type II (BCII) and OVA in Freund's complete adjuvant. Mice fed OVA before coimmunization with BCII and OVA had significantly lower arthritic scores than mice immunized with BCII only. Their body weight increased during the study period and their anti-BCII antibody activity was significantly IgG_2a lower. The frequency of spleen cells producing IgG anti-BCII antibody was also reduced. Coimmunization per se slightly ameliorated the development of arthritis, resulting in an early, transient reduction. It resulted in significantly higher IgG₁ anti-BCII antibody activity and increased splenocyte secretion of IFN-γ and IL-10 in response to BCII. Our findings demonstrate that OVA-specific regulatory events induced by feeding OVA, i.e. bystander suppression, reduced the severity of arthritis in animals immunized with BCII and OVA. Anti-BCII specific antibody responses and cytokine secretion by types 1 and 2 T helper cells were also decreased.     

Are the immune responses different in middle-aged and young mice following bone fracture, tissue trauma and hemorrhage?

Although immune responses following soft-tissue trauma-hemorrhage are markedly different in young (6-8 weeks) and aged (18-20 months) mice, it remains unknown if there are any differences in immune responses in middle-aged and young mice following bone fracture, soft-tissue trauma-hemorrhage (Fx-TH). To study this, young (6-8 weeks) and middle-aged (approximately 12 months) C3H/HeN male mice were subjected to sham operation or Fx-TH followed by resuscitation with Ringer's lactate. The mice were sacrificed 2 h thereafter and splenocytes, bone marrow cells (BM) and Kupffer cells (KC) were harvested, purified and stimulated with ConA (for splenocytes) or LPS (for BM and KC) in vitro. Splenocyte release of Th1 (IL-2 and IFN-gamma) cytokines was decreased and Th2 (IL-4 and IL-10) cytokine release was increased following Fx-TH in both young and middle-aged mice. However, the decrease in IL-2 and increase in IL-10 were significantly more in middle-aged mice compared to young mice (p < 0.05). Furthermore, splenocyte proliferation was decreased more in middle-aged mice compared to young mice following Fx-TH (p < 0.05). Additionally, TNF-alpha production was more in BM from middle-aged compared to BM from young mice after Fx-TH (p < 0.05). The production of IL-6 and IL-10 was also significantly higher in KC from middle-aged mice compared to young ones following Fx-TH. These results suggest that at middle age, the immune responses to Fx-TH are significantly different from those observed in young mice in different compartments of the body. Although the mechanism of the difference in various compartments in middle-aged vs. young mice following Fx-TH remains unknown, the decreased IL-2 production along with other altered T cell and macrophage functions may contribute to an increased susceptibility to sepsis in middle-aged vs. young individuals.     

Adrenomedullin binding protein-1 modulates vascular responsiveness to adrenomedullin in late sepsis

Adrenomedullin (AM), a potent vasodilatory peptide, plays an important role in initiating the hyperdynamic response during the early stage of sepsis. Moreover, the reduced vascular responsiveness to AM appears to be responsible for the transition from the early, hyperdynamic to the late, hypodynamic phase of sepsis. Although the novel specific AM binding protein-1 (AMBP-1) enhances AM-mediated action in a cultured cell line, it remains to be determined whether AMBP-1 plays any role in modulating vascular responsiveness to AM during sepsis. To study this, adult male rats were subjected to sepsis by cecal ligation and puncture (CLP). The thoracic aorta was harvested for determination of AM-induced vascular relaxation. Aortic levels of AMBP-1 were determined by Western blot analysis, and AM receptor gene expression in the aortic tissue was assessed by RT-PCR. The results indicate that AMBP-1 significantly enhanced AM-induced vascular relaxation in aortic rings from sham-operated animals. Although vascular responsiveness to AM decreased at 20 h after CLP (i.e., the late, hypodynamic stage of sepsis), addition of AMBP-1 in vitro restored the vascular relaxation induced by AM. Moreover, the aortic level of AMBP-1 decreased significantly at 20 h after CLP. In contrast, AM receptor gene expression was not altered under such conditions. These results, taken together, suggest that AMBP-1 plays an important role in modulating vascular responsiveness to AM, and the reduced AMBP-1 appears to be responsible for the vascular AM hyporesponsiveness observed during the hypodynamic phase of sepsis.