WHAT'S NEW IN SHOCK, April 2009?

The current issue of Shock contains 1 excellent review article (1), 2 clinical studies (2, 3), and 12 basic science studies that address several aspects of the pathophysiology of hemorrhage, trauma and burn injury (4-7), organ injury and dysfunction resulting from endotoxin (8-10) or bacterial sepsis (11-14), and interaction of microorganisms with the intestine (15). In their review, Novotny et al. (1) discuss the putative therapeutic use of angiopoietin 1 during ischemia and sepsis. Angiopoietin 1 is secreted by smooth muscle cells, pericytes, and astrocytes. Vascular endothelial growth factor (VEGF) and the angiopoietins function together during vascular development, with VEGF acting early during vessel formation and angiopoietin 1 acting later during vessel remodeling, maturation, and stabilization, for example, as for the maintenance of the integrity of the already existing vasculature. Thus, in 2000, Thursten et al. reported for the first time in Nature Medicine that angiopoietin 1 protects against the inflammation-induced plasma leakage. Consequently, it is tempting to speculate that angiopoietin 1 may serve as a therapeutic strategy during I/R and sepsis. The review by Novotny et al. (1) summarizes our current knowledge of the preclinical use of angiopoietin in models of endotoxin-induced lung injury and myocardial infarction, emphasizing in particular the potential benefit of strategies combining angiopoietin 1 with VEGF. The article is complemented by an observational study on the association between serum angiopoietin 2 levels and outcome in patients with sepsis (3). In a total of 73 patients (17 outpatients, 20 nonseptic intensive care unit patients, and 10, 12, and 24 patients with sepsis, severe sepsis, and septic shock, respectively), Siner et al. (3) found that the more severely diseased patients had the highest angiopoietin 2 blood levels. In the septic cohort, the plasma concentrations were directly related to the IL-6 levels, severity scores (Sepsis-Related Organ Failure Assessment, Acute Physiology and Chronic Health Evaluation III) and markers of single organ failure, for example, creatinine, bilirubin, and the Horowitz index, and in the nonsurvivors of this cohort, angiopoietin 2 levels were twice as high as in those patients who ultimately survived. In the other clinical study, Kotanidou et al. (2) investigated the prognostic value of the plasma amino-terminal pro-brain natriuretic peptide (NT-pro-BNP) levels. Amino-terminal pro-brain natriuretic peptide is the prohormone of BNP, which in turn is mainly released from cardiomyocytes as a response to pressure and volume and thus is well established as a marker of heart failure and/or guidance of therapy. Consequently, Kotanidou et al. studied a population of 233 noncardiac general, that is, medical and surgical, intensive care unit patients, the repartition of which, however, was not recorded. To avoid major confounding factors, the authors explicitly excluded patients with chronic heart failure as defined by a left ventricular ejection fraction of less than 50% at admission and those with pre-existing kidney disease (creatinine level >1.8 mg/dL). Measurements were performed at admission and on days 1 and 2. Amino-terminal pro-BNP blood concentrations were consistently higher in nonsurvivors, and moreover, they were directly related to the presence of acute renal insufficiency, thrombocytopenia, and the use of inotropes, whereas no relation was present with the Horowitz index and bilirubin concentrations. The study by Kotanidou et al. certainly adds to our knowledge of the predictive value of the NT-pro-BNP measurement, but it remains open whether decision making will be influenced beyond its current use as treatment guide in patients with congestive heart failure. Hemorrhage, trauma, and burn injury are the focus of four studies. In a well-established murine model of hemorrhage (mean blood pressure of 35 mmHg over 90 min) and resuscitation (four times the shed blood volume using Ringer's lactate), Nickel et al. investigated the impact of 17β-estradiol (estrogen) on the myocardial IL-6 formation, which is known to be a crucial mechanism of posthemorrhagic cardiac dysfunction (4). The authors hypothesized that the estradiol-related normalization of the tissue proinflammatory response is mediated by inhibition of hypoxia-inducible factor 1 (HIF-1). Estrogen and the anti-HIF-1α agent YC-1 (3-(5′-hydroxymethyl-2′-furyl)-1-benzyl indazole) comparably attenuated both the cardiac tissue nuclear factor κB (NF-κB) activation and the IL-6 release. It remains to be elucidated whether both compounds may act synergistically and/or whether the YC-1 side effects-the drug was developed as an anticoagulant-may preclude its use in other models of trauma and hemorrhage. Genovese et al. (5) addressed the question whether glycyrrhizin, a constituent of the liquorice or sweet wood root Glycyrrhiza glabra, which among other compounds contain terpenoids and phenol derivatives that are known radical scavengers, may possibly beneficially influence the tissue inflammatory response after spinal cord trauma. When treated with this compound, mice undergoing spinal cord compression showed attenuated (nuclear factor) NF-κB activation and nitrosative stress, which coincided with reduced histological tissue damage and apoptosis. It is of particular importance that these biochemical and morphological effects were associated with a markedly improved spinal cord motor function. Thus, Genovese et al. add another interesting "stone to the mosaic" of possible therapeutic interventions to treat spinal cord injury. Hoth et al. (6) studied the role of Toll-like receptors (TLRs), which are crucial for the activation of innate immunity, in a murine model of pulmonary contusion resulting from blunt chest trauma. Among the numerous TLRs, all but one (i.e., TLR-3) depends on the intracellular adapter protein MyD88 to induce NF-κB activation and the consecutive mediator expression. Moreover, because exogenous pulmonary TLR-4 expression improved acute lung injury in rats rendered septic by cecal ligation and puncture (CLP), the authors investigated wild type as well as MyD88−/− and TLR−/− strains. Impaired gas exchange, edema formation, and pulmonary tissue inflammation depended on the presence of both TLR-4 and MyD88. This observation complements the authors' report on the impact of TLR-2, but it remains open how the TLR signaling is initiated in the lung and whether there is any cell-type specificity. Finally, Costantini et al. (7) investigated the potential therapeutic manipulation of the impaired gut barrier function. It is well established that shock states may be associated with translocation of endotoxin and/or live bacteria as well as gut-derived mediators that pass through the intestinal lymphatics and aggravate the systemic response. The integrity of the gut tight junction is mandatory to prevent luminal compounds from translocating into the circulation of the intestine, and the proteins occludin and zonula occludens protein 1 are crucial for its maintenance. As a logical consequence of previous investigations on the use of hypertonic saline and pentoxifylline after hemorrhage, the authors now report on the effects of the latter after burn injury. Pentoxifylline nearly completely blunted the otherwise marked gut histological injury and the increased permeability as well as the subsequent systemic inflammation, as documented by the plasma tumor necrosis factor α levels, and this effect coincided with a near-normal expression of the occludin and zonula occludens protein 1 expression. The authors used a nonresuscitated model, and thus, it remains unclear to what extent the results observed relate to the vasodilator and, consequently, to the perfusion-enhancing and/or any other aspect of nonspecific phosphodiesterase inhibition. Another seven basic science research articles report on various aspects of the pathophysiology of systemic inflammation and sepsis either induced by endotoxin or live bacteria. Among the four articles dealing with bacterial sepsis, an interesting study was performed in mice with CLP-induced peritonitis by Peck-Palmer et al. (11), who investigated the role of the Bcl-2 family on the regulation of dendritic cells and macrophages. It is well known that there is widespread apoptosis of immune cells during sepsis, and the Bcl-2 proteins probably assume major importance in this process, because they comprise both proapoptotic and antiapoptotic proteins. Using flow cytometry of cells originating from either wild-type mice or Bim−/− and Bcl-2+/+ strains (the latter overexpressing Bcl-2 either in all hematopoietic or in major histocompatibility complex class I cells), the authors identified the different splenocyte populations and thereafter quantified the splenic cell loss. They could elegantly demonstrate the presence of differentiated "mature" as well as "immature, monocyte-like" macrophages, the latter only being susceptible to sepsis-induced cell depletion. Because this response was present neither in the Bim−/− nor in the Bcl-2+/+ mice, the authors concluded that a decreased ratio between antiapoptotic (Bcl-2) and proapoptotic (Bim) proteins assumes major importance for sepsis-induced immune cell depletion in vivo and, consequently, for the antimicrobial host defense. Overhaus et al. (12) investigated the impact of a "two-hit" model on gastrointestinal motility using a combination of mild hemorrhagic shock and either simultaneous or delayed CLP-induced peritonitis in rats. Gastrointestinal transit in vivo was assessed by measuring the fluorescent concentration in 15 equal bowel segment pieces after oral administration of fluorescein isothiocyanate-labeled dextran, whereas in vitro circular muscle contractility of the colon was quantified by its isometric force response to stimulation by the muscarinic agonist bethanechol. Cecal ligation and puncture alone markedly slowed the gastrointestinal transit and reduced the contractile force of the colonic circular muscularis. This response, however, was attenuated or completely blunted when mild hemorrhage had preceded the peritonitis induction, and this effect coincided with reduced tissue inflammation and expression of the mRNA of iNOS. The authors concluded that mild hemorrhage is able to precondition against the effects of a septic challenge. A similar phenomenon was described for heat stress preconditioning in rodent hemorrhage, and this effect was related to a reduction of NO-induced oxidative and nitrosative stress. Two further articles focused on the pathophysiology of acute lung injury resulting from bacterial sepsis. Andonegui et al. (13) characterized a new model of Streptococcus pneumonia-induced sepsis and multiple organ failure. Mice were infected by transoral injection of a bacteria aliquot directly into the trachea, and animals were observed up to 5 days after being infected. The authors determined intestinal barrier function (blood-to-lumen clearance of 51Cr-EDTA), renal function (plasma creatinine and urea levels), lung inflammation and histomorphology, and myocardial function (ex vivo isolated heart preparation working at predefined preload and afterload pressures and electric pacing-induced constant heart rate). Both lung inflammation and impairment of the gut barrier peaked at 12 h after pneumonia. Moderate kidney dysfunction was present at 24h, which coincided with reduced ejection fraction of the left ventricle. Within its limits, that is, lacking supportive therapy, for example, volume resuscitation or mechanical ventilation, this model may be used for mechanistic studies of gram-positive pneumonia and subsequent bacteremia and sepsis-induced multiorgan failure. Finally, Nin et al. (14) studied the putative therapeutic impact of a nonspecific anti-inflammatory treatment, that is, dexamethasone, in rats during high tidal volume ventilation 48h after CLP. After 75min of mechanical ventilation, the authors measured lung mechanics, pulmonary gas exchange, parameters of liver and kidney injury, and blood NO and cytokine levels, as well as lung tissue inflammation and histomorphology. Dexamethasone attenuated parameters of remote organ injury and virtually abrogated all parameters of inflammation as well as pulmonary and vascular dysfunction. Despite this overwhelming beneficial effect, it should be taken into consideration that the authors used an extreme level of tidal volume (35 mL/kg) and, moreover, did not compensate for any hemodynamic effect of the markedly increased mean airway pressure. Three additional articles used an endotoxin challenge to induce a sepsis-like systemic inflammatory state. Iida et al. (8) addressed the question whether jaundice resulting from bile duct obstruction influences endotoxin-induced liver injury. Based on the well-established fact that obstructive jaundice per se is associated with enhanced hepatocyte apoptosis, the authors subjected both wild-type and Fas ligand-deficient (gld/gld) mice to endotoxin injection at days 3 and 14 after bile duct ligation. Three hours after the endotoxin challenge, liver tissue and blood samples were obtained. At day 14, the endotoxin-induced liver apoptosis was markedly increased in the animals with obstructive jaundice, which was associated with reduced ATP content and increased cell injury as assessed by transaminase activity, and ultimately resulted in 100% mortality within 12 h after endotoxin injection. This finding coincided with significantly higher caspases 3 and 8 activity, whereas caspase 9 activity was not different between sham-operated controls and mice after bile duct ligation. Fas ligand-deficient mice presented with markedly attenuated hepatocyte apoptosis, which was affiliated with a 50% survival at 72 h after endotoxin injection. This study thus helps to understand the well-known fact that pre-existing obstructive jaundice is concomitant with increased mortality and morbidity in patients undergoing liver surgery. Martins et al. (9) tried to elucidate the pathophysiological mechanism of the well-established susceptibility to infection commonly seen in patients with diabetes. Six hours after intrabronchial injection of Escherichia coli endotoxin, the lungs from rats with alloxan-induced diabetes with and without a single insulin injection 2 h before endotoxin were harvested for mRNA of the proinflammatory cytokines TNF-α and IL-1β, the anti-inflammatory cytokine IL-10, and the adhesion molecule intercellular adhesion molecule 1, as well as for the activation of the nuclear transcription factor NF-κB. Diabetic animals showed a markedly reduced pulmonary and systemic inflammatory response to the endotoxin challenge, which did not differ from that of sham controls, and this response was concomitant with a comparable attenuation of the NF-κB activation. Albeit still hyperglycemic (mean glycemia: controls 83 mg/dL, diabetic rats 455 mg/dL, insulin treatment 259 mg/dL), insulin nearly completely restored the inflammatory response to the endotoxin injection. Thus, the authors concluded that insulin regulates the host defense in the context of an inflammatory challenge. It remains open, however, whether this phenomenon is due to the hormone itself and/or to the simultaneous regulation of the blood sugar level. The most recently discovered type of RNA editing, adenosine-to-inosine modification in pre-mRNA, with inosine acting as guanosine during translation, is a widespread type of nuclear pre-mRNA editing in higher eukaryotes. The enzyme involved, adenosine deaminase acting on RNA, has two isoforms, and it was shown to be involved in the innate immune response as well as in development of acute lung injury via its expression in pulmonary macrophages. Thus Wu et al. (10) studied the mechanisms how this enzyme modifies the secretory function of alveolar macrophages and, consequently, lung inflammation. Rat alveolar macrophages were cultured and exposed to endotoxin with or without previous transfection of siRNA targeted to knock down the expression of the larger isoform of the enzyme. Subsequently, the authors used these cells in vivo, thus creating an adoptive transfer model that allowed studying the effect of this gene silencing on inflammatory parameters in the bronchoalveolar lavage fluid as well as on pulmonary gas exchange in endotoxin-challenged animals. Knockdown of the adenosine deaminase not only reduced the macrophage inflammatory protein 1 and increased the IL-10 release after endotoxin stimulation in vitro, but also attenuated the lung inflammatory injury and the subsequent impairment of pulmonary gas exchange in vivo. Thus, the authors elegantly demonstrated the role of alveolar macrophage adenosine deaminase acting on RNA for the development of endotoxin-induced acute lung injury. A final basic science article focuses on the interaction of intestinal epithelium and intraluminal bacteria and fungi (15). Polyethylene glycol, a highly water-soluble polymer resulting from the condensation reaction between ethylene dioxide and water, is used to prepare patients before colonoscopy and/or colonic surgery as well as to treat chronic constipation. Most recently, it was reported to mimic the effects of intestinal mucus, thus reducing translocation of live bacteria, possibly acting as a physical barrier. In this context, Henry-Stanley and Wells (15) studied the interaction of E. coli, Candida albicans, and Candida glabrata with the intestinal epithelium: an in vitro approach allowed assessing the microbial adherence and internalization to Caco2-enterocytes; the cecal colonization was determined in vivo after antibiotic pretreatment and subsequent oral microbe inoculation. In a concentration-dependent manner, both low (3.35 kd) and high (15 - 20 kd) molecular-weight polyethylene glycol preparations blunted the adherence to enterocytes in vivo, and a moderate reduction of cecal microbial colonization was present in vivo. Because polyethylene glycol was also reported to reduce the adhesiveness of Staphylococcus epidermidis, Pseudomonas aeruginosa, and Candida tropicalis, its oral administration and/or addition to indwelling material may thus be a tempting approach to reduce bacteremia during systemic inflammation. In the present editorial comment, it is only possible to highlight some key aspects of each article, and thus, I invite the readers to go deeper into the subjects of their particular interest in this issue of Shock.