rBPI23 attenuates endotoxin-induced cardiovascular depression in awake rabbits

Review: Towards antibacterial strategies: studies on the mechanisms of interaction between antibacterial peptides and model membranes

Lipopolysaccharides (LPSs) play a dual role as inflammation-inducing and as membrane-forming molecules. The former role attracts significantly more attention from scientists, possibly because it is more closely related to sepsis and septic shock. This review aims to focus the reader's attention to the other role, the function of LPS as the major constituent of the outer layer of the outer membrane of Gram-negative bacteria, in particular those of enterobacterial strains. In this function, LPS is a necessary component of the cell envelope and guarantees survival of the bacterial organism. At the same time, it represents the first target for attacking molecules which may either be synthesized by the host's innate or adaptive immune system or administered to the human body. The interaction of these molecules with the outer membrane may not only directly cause the death of the bacterial organism, but may also lead to the release of LPS into the circulation. Here, we review membrane model systems and their application for the study of molecular mechanisms of interaction of peptides such as those of the human complement system, the bactericidal/permeability-increasing protein (BPI), cationic antibacterial peptide 18 kDa (CAP18) as an example of cathelicidins, defensins, and polymyxin B (PMB). Emphasis is on electrical measurements with a reconstitution system of the lipid matrix of the outer membrane which was established in the authors' laboratory as a planar asymmetric bilayer with one leaflet being composed solely of LPS and the other of the natural phospholipid mixture. The main conclusion, which can be drawn from these investigations, is that LPS and in general its negative charges are the dominant determinants for specific peptide—membrane interactions. However, the detailed mechanisms of interaction, which finally lead to bacterial killing, may involve further steps and differ for different antibacterial peptides.

Coagulation Activation by Lipopolysaccharides

Lipopolysaccharides at approximate plasma reactivities >3 ng/mL or β-glucans at >0.5-1 μg/mL are toxic for human blood; lipopolysaccharide interacts with membrane components of susceptible cells (eg, monocytes) activating phospholipase A2that destroys the cell membrane. Cell fragments (microparticles or DNA) possess polynegative niches that activate intrinsic hemostasis. Pathologic disseminated intravascular coagulation arises. Blood vessels are obstructed by disseminated thrombi, and vital organ areas become ischemic. Multiorgan failure threatens life of the patient. Diagnosis and therapy of pathologic disseminated intravascular coagulation is of extreme clinical importance. For early diagnosis of pathologic disseminated intravascular coagulation, specific activation markers of coagulation (eg, plasmatic amidolytic thrombin activity) or the plasmatic lipopolysaccharide or glucan reactivity can be measured. A new treatment target might be kallikrein or factor XIIa; 10 to 20 mM arginine is the approximate 50% inhibitory concentration against the contact phase of coagulation. The complex interaction between cell fragments and hemostasis causes pathologic disseminated intravascular coagulation in sepsis.

Bactericidal/permeability increasing protein attenuates the myocardial inflammation/dysfunction that occurs with burn complicated by subsequent infection

Intubation and mechanical ventilation after burn contribute to pneumonia-related infection. Although postburn presence or absence of endotoxin has been described, inactivation of Toll-like receptor 4 signaling has been shown to improve postburn organ function, suggesting that LPS participates in burn-related susceptibility to infection. We hypothesized that bactericidal/permeability-increasing protein (rBPI) given postburn would attenuate myocardial inflammation/dysfunction associated with postburn septic challenge given 7 days postburn. Rats were given burn over 40% total body surface area, lactated Ringer 4 ml.kg(-1).% burn(-1); burns received either vehicle or rBPI, 1 mg.kg(-1).h(-1) for 48 h postburn. Postburn day 7, subgroups of burns and shams were given intratracheal Klebsiella pneumoniae, 4 x 10(6) CFU to produce burn complicated by sepsis; additional sham and burn subgroups received intratracheal vehicle to produce sham sepsis. Vehicle-treated groups: 1) sham burn + sham sepsis 2) sham burn + sepsis, 3) burn + sham sepsis, 4) burn + sepsis. rBPI-treated groups: 5) sham burn + sham sepsis, 6) sham burn + sepsis, 7) burn + sham sepsis, 8) burn + sepsis. Cardiomyocyte cytokine secretion and myocardial function were studied 24 h after septic challenge, postburn day 8. Pneumonia-related infection 8 days after vehicle-treated burn produced myocyte cytokine secretion (pg/ml), indicated by increased myocyte TNF-alpha, 549 +/- 46; IL-1beta, 50 +/- 8; IL-6, 286 +/- 3 levels compared with levels in sham myocytes (TNF-alpha, 88 +/- 11; IL-1beta, 7 +/- 1; IL-6, 74 +/- 10; P < 0.05). Contractile dysfunction was evident from lower left ventricular pressure +/-dP/dt values in this group compared with sham. rBPI attenuated myocyte cytokine responses to septic challenge and improved contractile function, suggesting that burn-related mobilization of microbial-like products contribute to postburn susceptibility to infection.

Receptors, mediators, and mechanisms involved in bacterial sepsis and septic shock

Bacterial sepsis and septic shock result from the overproduction of inflammatory mediators as a consequence of the interaction of the immune system with bacteria and bacterial wall constituents in the body. Bacterial cell wall constituents such as lipopolysaccharide, peptidoglycans, and lipoteichoic acid are particularly responsible for the deleterious effects of bacteria. These constituents interact in the body with a large number of proteins and receptors, and this interaction determines the eventual inflammatory effect of the compounds. Within the circulation bacterial constituents interact with proteins such as plasma lipoproteins and lipopolysaccharide binding protein. The interaction of the bacterial constituents with receptors on the surface of mononuclear cells is mainly responsible for the induction of proinflammatory mediators by the bacterial constituents. The role of individual receptors such as the toll-like receptors and CD14 in the induction of proinflammatory cytokines and adhesion molecules is discussed in detail. In addition, the roles of a number of other receptors that bind bacterial compounds such as scavenger receptors and their modulating role in inflammation are described. Finally, the therapies for the treatment of bacterial sepsis and septic shock are discussed in relation to the action of the aforementioned receptors and proteins.

In vivo interaction of endotoxin and recombinant bactericidal/permeability-increasing protein (rBPI23): hemodynamic effects in a human endotoxemia model

The cardiovascular derangement that results from the administration of endotoxin in healthy subjects is qualitatively similar to what is observed in patients in septic shock. The biological response to endotoxin is attributed in part to cytokine release. In experimental endotoxemia, recombinant bactericidal/permeability increasing protein (rBPI(23)) has shown a protective effect by binding endotoxin with the subsequent inhibition of the endotoxin-induced cytokine release and of neutrophil activation. In a controlled, blinded crossover study the early cardiovascular effects of rBPI(23) were investigated in an experimental endotoxemia model in humans. The beat-to-beat changes in arterial pressure and cardiac output following infusion of endotoxin (40 EU/kg body weight) and rBPI(23) (1 mg/kg) or placebo (human serum albumin, 0.2 mg/kg) were studied for 2 hours in 8 healthy male adults. Endotoxin or rBPI(23) alone did not induce significant cardiovascular changes. Endotoxin following rBPI(23) infusion elicited a fall in total peripheral resistance with its nadir after 4 minutes to 40% (range 16-53; P <.001) of control level. Mean arterial pressure showed little change, and the fall in total peripheral resistance was associated with a reflex increase in heart rate and cardiac output (32%; range 43-106). Changes in cardiovascular variables in the subsequent 2 hours were not significant. In vitro activation of the contact system by, respectively, rBPI(23), LPS, and LPS-rBPI(23) complexes was assessed. Following incubation with rBPI(23), LPS, and LPS-rBPI(23) complexes, complex levels were generated at levels comparable to those observed in the buffer control. The rapid vasodilatation by endotoxin administered concomitantly with rBPI(23) is not mediated by complement or contact system activation. The early vasodilatation is compensated by an increase in cardiac output, which therefore does not result in arterial hypotension. The monitoring of continuous cardiac output allows for the detection of rapid effects on systemic flow and conductance that go unnoticed in a recording of arterial pressure.

Organ blood flow after partial hepatectomy in rats: modification by endotoxin-neutralizing bactericidal/permeability-increasing protein (rBPI23)

BACKGROUND/AIM

Both maintenance of adequate perfusion and regeneration of the remnant liver are important in the recovery of liver function after partial hepatectomy. In previous experiments, we have shown that profound hypotension and liver injury can be attenuated by neutralizing endotoxins. The relative contribution of endotoxemia to changes in liver blood flow and blood flow to other major organs after partial hepatectomy is not known. The aim of this study was to examine the effect of endotoxin neutralization on individual organ blood flows including hepatic artery and splanchnic blood flow after experimental partial hepatectomy and its relation to liver cell proliferation.

METHODS

Male Wistar rats underwent either two-thirds partial hepatectomy or sham operation. Treatment consisted of continuous infusion of recombinant N-terminal bactericidal/permeability-increasing protein (rBPI23) or control protein. At 4 h after surgery, organ blood flows were measured using the radiolabeled microsphere technique, and at 24 h, proliferation index in liver tissue was calculated.

RESULTS

After partial hepatectomy, blood flows to virtually all organs were significantly lower as compared to values obtained in sham-operated rats. rBPI23 greatly improved hepatic artery flow (p<0.001) but not portal venous flow. These effects of rBPI23 on liver flow preceded an equally enhanced liver cell proliferation (p<0.01). Endotoxin neutralization led to significantly higher flows to some but not all splanchnic organs. Lung perfusion was significantly improved by rBPI23.

CONCLUSIONS

Neutralization of endogenous endotoxins improves liver blood flow after partial hepatectomy and also periportal and pericentral liver cell proliferation. This proliferation effect may result from an increased hepatic artery flow. Lung, colon, spleen and pancreas flow but not kidney flow was greatly enhanced by rBPI23.

Mechanisms of action of bactericidal/permeability-increasing protein BPI on reconstituted outer membranes of gram-negative bacteria

The mechanisms of interaction of the recombinant N-terminal portion of bactericidal/permeability-increasing protein, rBPI21, with various planar asymmetric and symmetric bilayer membranes, including the lipid matrix of the outer membrane of Gram-negative bacteria, were investigated via electrical measurements. For the lipopolysaccharide (LPS) leaflet of the outer membrane, isolated deep rough mutant LPS of Escherichia coli strain F515 (F515 LPS) and Proteus mirabilis strain R45 (R45 LPS) were used. The addition of rBPI21 to the LPS side of asymmetric LPS/phospholipid membranes, as well as to black lipid membranes made from dioleoylphosphatidylglycerol (DOPG), led to membrane rupture. The innermembrane potential difference resulted in a slight increase from 0 to 5 mV for symmetric DOPG membranes but changed for asymmetric F515 LPS/PL membranes from -36 to +8 mV and for R45 LPS/PL membranes from -37 to -5 mV following the addition of rBPI21. In all cases, the addition of rBPI21 led to an increase in membrane current. The effect of rBPI21 on the innermembrane potential difference of LPS/PL membranes was significantly reduced in the presence of 40 mM MgCl2 (shift from -36 to -31 mV for F515 LPS). On the basis of these results and from our studies on the interaction of rBPI21 with lipid monolayers and aggregates [Wiese, A., et al. (1997) Biochemistry 36, 10301-10310], a model is discussed explaining how the observed membrane rupture, increase of membrane current, and change of transmembrane potential as induced by rBPI21 may contribute to bacterial dysfunction.

Mechanisms of action of the bactericidal/permeability-increasing protein BPI on endotoxin and phospholipid monolayers and aggregates

We have investigated the mechanisms of interaction of the recombinant N-terminal portion of bactericidal/permeability-increasing protein, rBPI21, with lipopolysaccharide (LPS) isolated from enterobacterial deep rough mutant strains. Experimentally, the ability of rBPI21 to form monolayers at the air/water interface and its action on lipid monolayers were analyzed. We have further studied the interaction of rBPI21 with aggregates from phospholipids and Re mutant LPS by infrared and resonance energy transfer spectroscopy and laser Doppler velocimetry. From monolayer experiments, the molecular area of a single rBPI21 molecule was estimated to be about 12 nm2. At lateral pressures of </=25 mN/m, rBPI21 incorporated into monolayers from negatively charged LPS and phosphatidylglycerol (PG) but not into those from neutral phosphatidylcholine. rBPI21 incorporated not only into monolayers but also into liposomes made from or containing negatively charged phospholipids, reducing the absolute value of the zeta-potential of LPS and PG aggregates. Furthermore, due to intercalation, rBPI21 caused the rigidification of the acyl chains of lipids in the gel as well as in the fluid phase and significantly immobilized their phosphate groups. High concentrations of Mg2+ ions were found to have a protective effect against the action of rBPI21. On the basis of these results, the biophysical characteristics of rBPI21 are discussed and a model is proposed as to how the rBPI21-induced influence on lipid monolayers and bilayers could explain rBPI21-mediated effects on the bacterial membrane.

Effects of recombinant bactericidal/permeability-increasing protein (rBPI23) on neutrophil activity in burned rats

Bactericidal/permeability-increasing protein (BPI) is a neutrophil granule protein with potent bactericidal and lipopolysaccharide (LPS)-neutralizing activities. The purpose of this study was to determine if a human recombinant BPI product, rBPI23, would influence neutrophil (PMN) sequestration into various tissues in a rat burn injury model. Leukosequestration may produce local tissue injury from proteases and high-energy oxygen species released from PMNs. Rats received tracheostomy and venous cannulation, then received 17 to 20% total body surface area full-thickness contact burns and resuscitation with 20 ml, of intraperitoneal saline. Ten mg/kg body weight rBPI23 in saline was given by intravenous injection immediately after burn injury, followed by intravenous doses of 2 mg/kg at 2 and 4 hours. Control animals received intravenous saline only. PMN retention in lung, liver, spleen, gut, skin, muscle, kidney, and brain tissues was determined by removing (before burn injury) and differentially radiolabeling PMNs (111In) and erythrocytes (51Cr), reinfusing cells 4.5 hours after burn injury, and measuring tissue radioactivity 30 minutes later. Edema was estimated by measuring extravasated 125I-labeled albumin in the various tissues, 30 minutes after injection. Peripheral blood PMNS were analyzed for intracellular H2O2 content by flow cytometry using a fluorescent dye that reacts with H2O2. Radioisotope studies demonstrated significant (p < 0.05) leukosequestration into lung, liver, gut, kidney, and skin tissues at 5 hours after burn injury. Tissue edema, manifested by radiolabeled albumin retention, was not observed in any tissues. Postburn PMN deposition in lungs and skin was decreased (p < 0.05) by the immediate administration of rBPI23 after burn injury. Flow cytometry showed increased intracellular H2O2 content in peripheral blood PMNs 5 hours after burn injury (p < 0.05), which was unaffected by administration of rBPI23. Since sequestration of metabolically active PMNs may induce tissue injury, therapies that block leukosequestration after burn injury may improve clinical outcomes by limiting remote tissue injury.