Differential antibiotic-induced release of endotoxin from gram-negative bacteria

Endotoxin Acts Synergistically With Clostridioides difficile Toxin B to Increase Interleukin 1β Production: A Potential Role for the Intestinal Biome in Modifying the Severity of C. difficile Colitis

BACKGROUND

Inflammation is a crucial driver of host damage in patients with Clostridioides difficile colitis. We examined the potential for the intestinal microbiome to modify inflammation in patients with C. difficile colitis via the effects of gut-derived endotoxin on cytokine production.

METHODS

Endotoxin from Escherichia coli and Pseudomonas aeruginosa as well as stool-derived endotoxin were tested for their ability to enhance interleukin 1β (IL-1β) and tumor necrosis factor alpha (TNF-α) production by toxin B-stimulated peripheral blood mononuclear cells. Inflammasome and Toll-like receptor 4 (TLR4) blocking studies were done to discern the importance of these pathways, while metagenomic studies were done to characterize predominant organisms from stool samples.

RESULTS

Endotoxin significantly enhanced the ability of C. difficile toxin B to promote IL-1β production but not TNF-α. The magnitude of this effect varied by endotoxin type and was dependent on combined inflammasome and TLR4 activation. Stool-derived endotoxin exhibited a similar synergistic effect on IL-1β production with less synergy observed for stools that contained a high proportion of γ-proteobacteria.

CONCLUSIONS

The ability of endotoxin to enhance IL-1β production highlights a manner by which the microbiome can modify inflammation and severity of C. difficile disease. This information may be useful in devising new therapies for severe C. difficile colitis.

Antibiotic-induced endotoxin release is organism-dependent in experimental Gram-negative sepsis

The majority of in vitro and animal experiments that have been performed to assess antibiotic-induced endotoxin release (AIER) have employed a single test isolate, usually Escherichia coli. To determine the influence of microorganism type on AIER and interleukin-6 (IL-6) response, CF-1 mice were made septic following a 12-15% total body surface area nonlethal burn and subeschar challenge (LD90) with Klebsiella pneumoniae K2 (∼103 cfu), Proteus mirabilis 4552 (∼101 cfu) and Pseudomonas aeruginosa SBI-N (∼102 cfu). Three intraperitoneal (i.p.) doses, given every 4 h, of ceftazidime (TAZ, 200 mg/kg), imipenem (IMI, 100 mg/kg), ciprofloxacin (CIP, 25 mg/kg) and gentamicin (GEN, 25 mg/kg) were administered post burn and infection beginning when mice were septic with organ dysfunction. Free endotoxin concentrations were significantly ( P < 0.001) higher following all antibiotics for treatment of K. pneumoniae as compared to Ps. aeruginosa (intermediate) and P. mirabilis infections. Differential AIER was highest for TAZ and IMI, intermediate for CIP and lowest for GEN, for the treatment of K. pneumoniae and Ps. aeruginosa infections. There was a strong positive correlation between endotoxin release and IL-6 production for K. pneumoniae-treated animals, however increased endotoxin levels for Pseudomonas were accompanied by decreases in IL-6 levels. For P. mirabilis infection endotoxin levels were comparatively low, but highest for GEN and IMI. However, corresponding IL-6 levels increased only 3.2-fold for IMI and actually decreased by 50% for GEN following the first dose. Interestingly, CIP resulted in only modest endotoxin release and TAZ caused no appreciable release, however IL-6 concentrations dramatically increased 39.9-fold (TAZ) and 32.6-fold (CIP). This suggests that other pro-inflammatory mediators released from the bacterium, and not endotoxin, were more important determinants in the overall host response to antibiotic exposure. In conclusion, these data provide supportive evidence that absolute and differential AIER and production of IL-6 is organism-dependent in experimental Gram-negative sepsis. As a result, general conclusions concerning differential AIER for infection caused by E. coli or K. pneumoniae cannot necessarily be extrapolated to other species of Gram-negative bacilli. Furthermore, these study results strongly indicate that the microorganism and other pertinent pro-inflammatory factors (i.e. exotoxins, proteases), must be taken into account in the study design and data analysis of any experimental or clinical trial that is conducted to determine the significance of differential AIER.

Review: Bacterial endotoxin and the human monoclonal antibody HA-IA: specificity, potential mechanisms of action, and limits to its effectiveness

Bacterial endotoxins are lipopolysaccharides present in the outer membrane of all Gram-negative bacteria (GNB). Endotoxins consist of a lipid moiety, lipid A, that is covalently linked to highly variable, serotype O-specific polysaccharide lateral chains. In contrast, the endotoxin core, which includes lipid A, is better conserved and can be recognized by antibodies showing cross-reactivity among various GNB. Such polyclonal and monoclonal antibodies have been developed in an attempt to neutralize the biological and dele. terious effects of endotoxin, thus preventing lipid A from binding to macrophages. In fact, almost all the biological activities of endotoxin are elicited by lipid A, and there is substantial evidence to the effect that the monocyte-macrophage is the principal mediator of endotoxicity. Antiserum against LPS isolated from rough mutants of GNB (expressing virtually only the central core-lipid A), has been shown to counteract the lethal effects of endotoxin in animals and humans. However, such serum or plasma contains antibodies of different specificities and isotypes which represent different effector functions, insofar as LPS is a very complex and highly heterogenous macromolecule. Because of the difficulties encountered in investigating the nature and specificity of the protection afforded by these antisera, and their limited capacity of production for therapeutic use, specific anti-lipid A monoclonal antibodies have been produced in their stead. A variety of mouse and human monoclonal antibodies against LPS have been generated and selected for their ability to cross-react with many GNB species. The most recent clinical trials involving the treatment of septic patients with human HA-IA (Centoxin) or with murine (E5) anti-lipid A monoclonal antibody showed no difference in survival rates, as compared to treatment with a placebo. However, statistical significance was demonstrated in subsets of patients suffering from documented Gram-negative septicemia or Gram-negative sepsis without refractory shock. The usefulness of anti-lipid A antibodies will undoubtedly remain controversial, since they appear to benefit only a minority of all patients treated, and also because no consensus exists regarding their specificity and modes of action. The aim of this review is to describe results which demonstrate the requirements for, difficulties in and limits to, elucidating the ability of certain antibodies to recognize structural elements present in the lipid A domain of LPS. A clear demonstration of antibody cross-reactivity was obtained only when rough LPS bacteria were used, and was markedly enhanced when smooth bacteria had been pretreated with cell wall active antibiotics. Further, new data have recently demonstrated the specific involvement of HA-IA in the immunocytoadherence assay in the presence of human complement and human red blood cells. Such phenomena may form part of the potential role for natural or monoclonal human IgM anti-lipid A antibodies, which will be to remove IgM-lipid A immune complexes through transhepatic clearance via C3b binding to the CR1 present on circulating human erythrocytes. Insofar as immunocytoadherence is a multiparameter phenomenon, various limiting factors probably interfere with its mechanism of clearance. These factors may be absent in various subsets of septic patients under treatment, thus explaining therapeutic failures with HA-IA in humans or preclinical animal studies. Several clinical settings involving defects to CR1 expression, C3b production, LPS recognition and hepatic clearance dysfunction are described. Long term, however, it will impossible to specify the patient subsets suitable for monoclonal therapy without first defining their characteristics. HA-IA may be able to inhibit one of the earliest stages in activation of the cytokine cascade by sequestrating and eliminating biologically active lipid A. The major problem today in terms of using anti-lipid A antibodies is an efficiently early detection of specific pathway defects which detract from or nullify the HA-IA therapeutic effect.

The clinical significance of endotoxin released by antibiotics: what is the evidence?

Basic investigations indicate variation among antibiotics in their propensity to release endotoxin from bacterial cells. Evidence for the clinical significance of this phenomenon includes: anecdotal reports; studies evaluating changes in endotoxin levels in response to antibiotics in a variety of infectious diseases; two studies which have prospectively evaluated changes in endotoxin levels in response to different antibiotics and correlated the changes with the patients' clinical course; and one study which evaluated differences in mortality in injured patients based on the endotoxin-releasing potential of antibiotics they received. These studies suggest: (i) different types of antibiotics do induce differing levels of endotoxin release during treatment of bacterial infections in humans; (ii) such antibiotic-associated endotoxin release probably does have a biologic consequence; and (iii) this biologic consequence is possibly of clinical relevance. However, the data are not sufficient to advocate changes in current antibiotic prescribing or dosing decisions. Rather, a variety of in vitro and in vivo experiments as well as clinical studies are advocated.

Differences in mode of action of (β-lactam antibiotics influence morphology, LPS release and in vivo antibiotic efficacy

Antibiotic mediated release of endotoxin (lipopolysaccharide, LPS) from Gram-negative bacteria is implicated in septic shock. The β-lactam (cell wall active) class, in particular, has been deemed responsible for release of greater quantities of LPS than other classes of antibiotics. However, it is becoming increasingly recognized that variations in the ability to liberate free LPS exist within the β-lactam subclasses. Until recently, LPS-release studies have primarily included the cephalosporin, monobactam, and penicillin β-lactams, but not the carbapenem subclass. We document here that carbapenems significantly liberate less LPS than other β-lactam subclasses, and that disparity in LPS release also occurs within the carbapenems as well (i.e. imipenem vs meropenem). The propensity to release LPS correlates with the cumulative PBP binding affinity of each β-lactam antibiotic regardless of subclass. A direct correlation has been established between antibiotic concentration, frequency of antibiotic exposure, differential LPS release, PBP specificity (morphology, rate of bacterial lysis or cell mass) and antibiotic efficacy in two mouse models of infection. In the first, differences in the pathogenic (endotoxic) potential of potently biologically active smooth LPS (S-LPS) and poorly biologically active rough chemotype LPS (R-LPS) expressing bacteria underscores the in vivo importance of endotoxin in mediating lethality following antibiotic chemotherapy. Endotoxin is, however, not innately toxic and thus its lethal effects are mediated through LPS stimulation of host cells (i.e. macrophages) to produce deleterious levels of cytokines. In the second experimental model, therefore, the toxic effects of antibiotic-liberated endotoxin have been abrogated by pretreatment of mice with carrageenan (CGN) prior to challenge with endotoxin or Gram-negative bacteria. CGN treatment eliminates or markedly reduces the numbers of LPS-responsive mediator cells thus affecting mortality. In both animal models, an in vivo role for LPS is demonstrated through differential changes in antibiotic efficacy (ED 50) in response to differential effects of LPS via modulation of the endotoxic sensitivity/responsiveness of the host. We conclude that antibiotic class, concentration, dosing intervals (timing) and perhaps, method of treatment (i.e. bolus vs infusion) may each be important in the survival of experimental animals severely infected with Gram-negative bacteria.

Effect of empiric antibiotic treatment on plasma endotoxin activity in septic patients

OBJECTIVES

Efficient empiric antibiotic therapy remains the cornerstone of sepsis treatment. However, antibiotics could be responsible for the transient clinical deterioration provoked by the release of bacterial cell-wall constituents, such as endotoxin, into the blood stream. The aim of this study was to evaluate if a transient elevation of endotoxin level occurred in septic patients following antibiotic administration.

METHODS

Thirty-three septic intensive care unit (ICU) patients were enrolled in this prospective trial. Four blood samples were collected from each of these patients during a 24-h period, and endotoxin activity was measured in these samples by the chemiluminescence technique. Fifteen ICU non-septic patients and 15 healthy volunteers were also observed for possible daily fluctuations in endotoxin activity.

RESULTS

There was no significant increase in endotoxin levels following the initiation of empiric antibiotic therapy in septic patients. A clinical deterioration in the 4 h following antibiotic administration was observed in 14 septic patients (42 %). These patients had significantly higher endotoxin levels than stable septic patients.

CONCLUSIONS

Although endotoxin levels failed to increase after the administration of antibiotic(s) to critically ill patients, they were higher in the septic patients presenting a transient deterioration than in the other patients. This observation suggests that a possible release of endotoxin due to bacteria lysis by antibiotics could be responsible for the observed clinical deterioration.

Development of small-molecule endotoxin sequestering agents

Sepsis, otherwise referred to as "blood poisoning" is a serious clinical problem, the incidence of which continues to rise in the US and worldwide despite advances in antimicrobial chemotherapy. The primary trigger in Gram-negative sepsis is endotoxin, a lipopolysaccharide (LPS) constituent of the outer membrane of all Gram-negative bacteria. The structurally highly conserved glycolipid called lipid A is the active moiety of LPS. Lipid A is composed of a hydrophilic, bis-phosphorylated di-glucosamine backbone, and a hydrophobic polyacyl domain. The bis-anionic, amphiphilic nature of lipid A enables it to interact with a variety of cationic hydrophobic ligands, including polymyxin B, a toxic peptide antibiotic which binds to lipid A and neutralizes endotoxicity. Having determined the structural basis of the interaction of polymyxin B with lipid A, our long-term goal has been to rationally design non-peptidic, nontoxic, small-molecule LPS-sequestrants. Our efforts began with defining the central pharmacophore that determined LPS-recognition and -neutralization properties in small molecules, which led to the discovery of a novel lipopolyamine lead, DS-96. DS-96 is an effective LPS-neutralizer, rivaling polymyxin B in a panel of vitro assays, as well as in protecting animals against endotoxicosis. Structure-activity relationships in our effort to rationally design endotoxin sequestering agents, preclinical assessment of hits and leads, and approaches to overcoming issues with toxicity are described in this chapter.

Intra-abdominal infections: considerations for the use of the carbapenems

Intra-abdominal infection remains a common and frequently severe medical condition, carrying with it significant morbidity and mortality. These infections are almost always polymicrobial in nature as they are caused by mixed aerobic/anaerobic intestinal flora. Despite substantial improvements in both the medical and surgical management of these infections over the last several decades, there remains an opportunity to further enhance the utilization of adjunctive antibiotic therapy. As a result of the epidemiology and the current resistance profile of the infecting pathogens, the carbapenems represent a class of antibiotics that are considered appropriate for the treatment of severe intra-abdominal infections. This review will discuss the classification and microbiology of these infections and emerging resistance in the pathogens of interest. The review also and focuses on the role of the carbapenems in the management of the constellation of diseases known as intra-abdominal infection.

Pharmacokinetics of amoxicillin administered in drinking water to recently weaned 3- to 4-week-old pigs with diarrhea experimentally induced byEscherichia coliO149:F4

OBJECTIVE

To measure effects of Escherichia coli O149:F4-induced diarrhea on water consumption and pharmacokinetics of amoxicillin after administration in drinking water.

ANIMALS

24 recently weaned 24- to 28-day-old crossbred pigs.

PROCEDURE

10 pigs were inoculated with E. coli O149:F4; all 10 pigs subsequently developed diarrhea. Pigs were medicated by administration of amoxicillin in the drinking water (0.75 mg/mL) for a 4-hour period on 2 consecutive days. Fourteen age-matched noninfected healthy pigs (control group) were medicated in a similar manner. Blood samples were obtained from both groups daily, and plasma concentrations of amoxicillin were analyzed by use of high-performance liquid chromatography.

RESULTS

Diarrhea reduced the area under the plasma concentration-versus-time curve (AUC) and maximum plasma concentration (C(max)) of amoxicillin on the first day of medication by 56% and 63%, respectively. The AUC of amoxicillin on the second day of medication for diarrheic pigs did not differ significantly from that of control pigs on the first day of medication.

CONCLUSIONS AND CLINICAL RELEVANCE

E. coli-induced diarrhea reduced the AUC of amoxicillin and time that plasma concentration of amoxicillin was > 0.025 microg/mL and, hence, less likely to have a therapeutic effect on the first day of administration in drinking water. On the assumption that plasma concentrations may indirectly reflect concentrations at the site of infection, analysis of our results suggests that higher doses of amoxicillin may be appropriate for administration in drinking water during a 4-hour period on the first day that pigs have diarrhea attributable to E. coli O149:F4.

Modified virulence of antibiotic-induced Burkholderia pseudomallei filaments

Melioidosis is a life-threatening bacterial infection caused by Burkholderia pseudomallei. Some antibiotics used to treat melioidosis can induce filamentation in B. pseudomallei. Despite studies on the mechanism of virulence of the bacteria, the properties of B. pseudomallei filaments and their impact on virulence have not been investigated before. To understand the characteristics of antibiotic-induced filaments, we performed in vitro assays to compare several aspects of virulence between normal, nonfilamentous and filamentous B. pseudomallei. Normal, nonfilamentous B. pseudomallei could cause the lysis of monocytic cells, while filaments induced by sublethal concentrations of ceftazidime, ofloxacin, or trimethoprim show decreased lysis of monocytic cells, especially after prolonged antibiotic exposure. The motility of the filamentous bacteria was reduced compared to that of nonfilamentous bacteria. However, the filamentation was reversible when the antibiotics were removed, and the revertant bacteria recovered their motility and ability to lyse monocytic cells. Meanwhile, antibiotic resistance developed in revertant bacteria exposed to ceftazidime at the MIC. Our study highlights the danger of letting antibiotic concentration drop to the MIC or sub-MICs during antibiotic treatment of melioidosis. This could potentially give rise to a temporary reduction of bacterial virulence, only to result in bacteria that are equally virulent but more resistant to antibiotics, should the antibiotics be reduced or removed.

Whole blood-mediated endothelial permeability and adhesion molecule expression: a model study into the effects of bacteria and antibiotics

AIM

To investigate whether the inflammatory response of cultured endothelial cells, as induced by conditioned plasma, depends on the bacterial species or type of antibiotic used for incubation with whole blood.

MATERIALS AND METHODS

Blood from healthy volunteers was stimulated ex vivo with different microorganisms, and with bacteria killed with different antibiotics. The resultant plasmas were incubated on monolayers of cultured human endothelial cells, followed by measurement of their permeability to albumin and expression of E-selectin and intercellular adhesion molecule-1.

RESULTS

Incubation of Escherichia coli in blood yielded plasmas that induced a marked increase in endothelial permeability and E-selectin expression. The response to Bacteroides fragilis or Enterococcus faecalis was generally weaker. Similar effects were observed after incubation of whole blood with lipopolysaccharide (LPS). Much of the permeability and adhesion molecule response to E. coli remained after removal of intact microorganisms from the culture. Whereas antibiotic treatment of E. coli with imipenem or cefuroxime resulted in a divergent production of tumour necrosis factor-alpha (TNF-alpha) in blood, no significant differences between these treatments were observed with respect to the plasma-induced endothelial response.

CONCLUSION

Bacteria differ in their capacity to generate a whole blood-mediated increase of endothelial permeability and adhesion molecule expression; this response depends, at least in part, on the presence of soluble bacterial components, such as LPS. Whereas treatment with various antibiotics may generate varying amounts of TNF-alpha, these differences are not translated into differences in endothelial permeability or adhesion molecule expression.

Osmotic stability of the cell membrane of Escherichia coli from a temperature-limited fed-batch process

The temperature-limited fed-batch (TLFB) process is a technique where the oxygen consumption rate is controlled by a gradually declining temperature profile rather than a growth-limiting glucose-feeding profile. In Escherichia coli cultures, it has been proven to prevent an extensive release of endotoxins, i.e. lipopolysaccharides, that occurs in the glucose-limited fed-batch (GLFB) processes at specific growth rates below 0.1 h(-1). The TLFB and the GLFB process were compared to each other when applied to produce the periplasmic, constitutively expressed, enzyme beta-lactamase. The extraction of the enzyme was performed by osmotic shock. A higher production of beta-lactamase was achieved with the TLFB technique while no difference in the endotoxin release was found during the extraction procedure. Furthermore, it was found that growth at declining temperature, generated by the TLFB technique, gradually stabilizes the cytoplasmic membrane, resulting in a significantly increased product quality in the extract from the TLFB cultures in the osmotic shock treatment.

Control of endotoxin release in Escherichia coli fed-batch cultures

High amounts of outer membrane (OM) components were released in glucose-limited fed-batch (GLFB) cultures at 37 degrees C at specific growth rates approaching 0.05 h(-1). Endotoxin analyses from a 20 degrees C GLFB culture gave similar results. An alternative fermentation technique, the temperature-limited fed-batch (TLFB) technique, reduced the endotoxin concentration in a culture with a biomass concentration of 30 g l(-1) from the 850 mg l(-1) in traditional GLFB cultures to about 20 mg l(-1). The TLFB technique uses the temperature to regulate the dissolved oxygen tension, while all substrate components are unregulated. It appears to be severe glucose limitation that triggers the extensive release of endotoxins rather than a low growth rate. Furthermore, it is not the low temperature that stabilizes the OM when using the TLFB technique. Simulations and experimental data show that this technique results in the same biomass productivity as the GLFB technique.

Modulation of macrophage responsiveness to lipopolysaccharide by IRAK-1 manipulation

Local activation of the macrophage by endotoxin is essential for the eradication of invasive gram-negative infections. Circulating endotoxin at lower concentrations results in immune cell activation at distant sites leading to tissue injury. Although the cellular mechanisms involved in these potentially dissimilar events are incomplete, it appears that the proximal kinase IRAK-1 plays a role. Thus, sense and antisense IRAK-1 oligonucleotides were used to determine the role IRAK-1 plays in macrophage activation by systemic (1-100 ng/mL) and local (1000 ng/mL) concentration of lipopolysaccharide (LPS) within THP-1 cells. Within the sense group, 1-1000 ng/mL of LPS within the sense group resulted in cellular activation of ERK-1/2, p38, and JNK/SAPK and the nuclear activation of NF-kappaB and AP-1. This activation was associated with proinflammatory cytokine production and cellular spreading. Systemic concentrations of LPS within the antisense group were associated with significant attenuation of intracellular signaling, cytokine production, and cellular spreading compared with the sense group. Local concentrations of LPS within the antisense group, however, were associated only with a delay in intracellular signaling, with no effect on cytokine production or cell spreading compared with the sense group. Based on these results, it appears that IRAK-1 is essential to macrophage activation at systemic, but not local, concentrations of LPS. These data suggest that redundant pathways exist that are functional at higher concentrations of LPS. Therefore, IRAK-1 appears to be the central kinase involved in the activation of the macrophage at distant sites during septic shock but is not necessary for activation in areas of local infection.

Single Live Cell Imaging of Chromosomes in Chloramphenicol-Induced Filamentous Pseudomonas aeruginosa

Pseudomonas aeruginosa is a leading opportunistic pathogen in human infections, and it is renowned for its intrinsic resistance to structurally and functionally unrelated antibiotics. Filamentation induced by antibiotics appears to trigger bacteria to depart from a normal growth phase and enter a stationary growth phase. As antibiotic concentrations decline below a therapeutic range, filamentous bacteria begin to divide normally, leading to a more rapid regrowth of the bacteria. Furthermore, filamentous bacteria are associated with an increase in endotoxin release. Moreover, the immune system of a patient needs to cope with uncharacteristic filamentous bacteria. Thus, it is biologically and clinically significant to study and understand bacterial filamentation. In this study, we investigate the frequencies, conditions, and characteristics of a filamentous P. aeruginosa at single cell and single chromosome resolutions. Our results show that filamentous cells (elongated rods) contain multiple copies of the cell's chromosome. It appears that the unsuccessful segregation of replicated chromosomes in an individual cell accompanies the formation of undivided filamentous cells. The quantity of chromosomes and the length of the filamentous wild-type cells increase as the chloramphenicol concentration increases to 50 and 250 microg/mL, suggesting that chloramphenicol induces the filamentation. Filamentation in three strains of P. aeruginosa depends on the expression level of efflux pump (MexAB-OprM) and the minimum inhibitory concentration of chloramphenicol. This study also opens up the new possibility of real-time monitoring of modes of actions of antibiotics in live cells with both temporal and spatial resolution.

Antimicrobial-induced endotoxin and cytokine activity in an in vitro model of septicemia in foals

OBJECTIVE

To determine which antimicrobials that are used to treat neonatal foals with septicemia attributable to Escherichia coli will minimize endotoxin release from bacteria and subsequent activity of inflammatory mediators while maintaining bactericidal efficacy.

SAMPLE POPULATION

Blood samples from 10 healthy foals.

PROCEDURE

Escherichia coli isolates A and B were isolated from 2 septicemic foals, and minimal inhibitory concentrations (MIC) were determined for 9 antimicrobials. Five of these antimicrobials were tested in vitro at 2 and 20 times their respective MIC. Whole blood or mononuclear cells grown in tissue-culture media were incubated with 105 colony-forming units of E. coli and each antimicrobial or saline (0.9% NaCl) solution. After 6 hours, number of viable bacteria remaining was determined, and supernatant was tested for endotoxin and tumor necrosis activity.

RESULTS

Testing in whole blood was compromised by bactericidal effects of the blood itself. In mononuclear cell suspensions, each antimicrobial significantly reduced the number of viable bacteria to low or undetectable amounts. Antimicrobials did not differ significantly in efficacy of bacterial killing. Amikacin used alone or in combination with ampicillin resulted in significantly less endotoxin activity than did ampicillin, imipenem, or ceftiofur alone. There was a correlation between TNF-alpha and endotoxin activity.

CONCLUSIONS AND CLINICAL RELEVANCE

Aminoglycosides appear less likely to induce endotoxemia and TNF-alpha synthesis during bactericidal treatment of E. coli septicemia, compared with beta-lactam antimicrobials. Use of ampicillin, imipenem, or ceftiofur in the treatment of septicemic neonatal foals should be accompanied by appropriate treatment for endotoxemia.

Evaluation of endotoxin release and cytokine production induced by antibiotics in patients with Gram-negative nosocomial pneumonia

OBJECTIVE

To determine the plasma concentrations of lipopolysaccharide, tumor necrosis factor-alpha, interleukin-1 beta, and interleukin-6 in a homogeneous group of septic patients and to evaluate the effect of antibiotic treatment, imipenem or ceftazidime, on the release of lipopolysaccharide and cytokines.

DESIGN

Prospective, randomized study.

SETTING

Sixteen-bed multidisciplinary intensive care unit.

PATIENTS

Twenty-four septic patients with documented Gram-negative nosocomial pneumonia. Controls were 20 patients admitted without sepsis and 20 healthy volunteers.

INTERVENTIONS

Septic patients were randomized between imipenem and ceftazidime. Blood samples were collected before (0 hrs) and after (4 and 12 hrs) antibiotic treatment. Concentrations of lipopolysaccharide were measured by using the limulus assay, and cytokine concentrations were measured by enzyme-linked immunosorbent assay. Statistical analyses were performed by Kruskal-Wallis test, Mann-Whitney U test, and Student's t-test.

MEASUREMENTS AND MAIN RESULTS

The mean age was 48.5 +/- 19.5. The mean Acute Physiology and Chronic Health Evaluation II score was 18.4 +/- 4.5. Overall mortality rate was 45.4%. All septic patients showed significant higher concentrations of lipopolysaccharide (p <.001), tumor necrosis factor-alpha (p <.04), and interleukin-6 (p <.001) than the controls, but interleukin-1 beta was never detected. We did not find statistically significant changes in lipopolysaccharide or cytokine plasma concentrations over time within any of the two arms of the study (ceftazidime vs. imipenem). There were no statistically significant differences in lipopolysaccharide and interleukin-6 plasma concentrations between the two antibiotic treatments. Although tumor necrosis factor-alpha plasma concentrations were significantly higher in the group treated with ceftazidime compared with the group treated with imipenem at the baseline and 4 hrs later, these differences were not statistically significant after 12 hrs of initiation of both treatments.

CONCLUSIONS

Patients with Gram-negative nosocomial pneumonia have high plasma concentrations of lipopolysaccharide, interleukin-6, and tumor necrosis factor-alpha, but the antibiotic therapy evaluated did not significantly modify these concentrations.

Antibiotic induced endotoxin release and clinical sepsis: a review

Sepsis and peritonitis have not lost much of their danger for patients. The mortality rate in peritonitis has only marginally decreased during the last 30 years despite aggressive surgical and sophisticated intensive care treatment. In intra-abdominal infection and peritonitis source control remains the mainstay of treatment, although general principles and denominators of successful source control need to be established. Endotoxin has been recognized as a major player in the pathogenesis of sepsis and its significance in clinical disease has been investigated in clinical studies for more than 20 years. Since the Sixties there is a growing interest in the effect of antibiotics and other compounds on the release of endotoxin. The effect of antibiotics on the release of endotoxin and inflammatory parameters, e.g., cytokines, remains to be clarified despite a growing body of in-vitro studies, animal studies and a few clinical studies. The purpose of this review is to evaluate the evidence of endotoxin release in clinical studies and the effect that antibiotic treatment may have in-vitro, in-vivo and in clinical studies on endotoxin and cytokine release. In-vitro antibiotic-induced endotoxin release may depend on antibiotic class, presence of serum, type of organism, site of antibiotic action and Gram-stain. Endotoxin release may be different in late or early lysis, proportional to the number of killed pathogens. Morphology of bacteria may have an impact on endotoxin release and phagocytosis. Antibiotic-treated animals may show higher endotoxin levels with a higher survival rate than untreated animals. Plasma endotoxin may increase despite decreasing bacteremia. There may be a similar killing rate by different antibiotics but a difference in endotoxin release. Intestinal endotoxin does not necessarily correlate to the level of gram-negative bacteria. However, the alteration of the gut content by pretreatment may be associated with reduced endotoxemia and increased survival. Antibiotic-induced endotoxin release may be different depending on the type of infection, the location of infection, the virulence of strains, Gram-stain, mode of application and dosage of antibiotic. Different antibiotics may induce the release of different forms of endotoxin which may be lethal for sensitized animals. The combination of antibiotics with inhibitors of endotoxin or the pro-inflammatory response may be responsible for increased survival by decrease of endotoxin release. The clinical significance of antibiotic-induced endotoxin release is documented only in a few clinical disorders, e.g., meningitis, urosepsis. The difference in endotoxin release by PBP 2-specific antibiotics, e.g., imipenem, and PBP 3-specific antibiotics, e.g., ceftazidime, may not be visible in each study. Patients with increased multi-organ failure (MOF) scores may profit from treatment with antibiotics known to decrease endotoxin. In conclusion, the clinical significance of antibiotic-induced endotoxin release remains to be clarified. Type of pathogen and its virulence may be more important than recently suggested. gram-positive pathogens were just recently recognized as an important factor for the development of the host response. In case of fever of unknown origin in intensive care patients either failure of treatment, e.g., failure of source control in intra-abdominal infection, or a side effect of antibiotic treatment, e.g., endotoxin release, should be considered as a cause of the fever.

Endotoxin release due to ciprofloxacin measured by three different methods

Antibiotics are known to induce the release of bioactive endotoxin (LPS) from gram-negative bacterial cells. Because varying data have been published on the influence of quinolone antibiotics on LPS liberation, we studied the effect of ciprofloxacin on a culture of Escherichia coli by determining bacterial killing and free LPS concentrations in comparison with imipenem and ceftazidime. LPS levels were measured by three different methods, namely (1) the Limulus amebocyte lysate test, (2) an ELISA method based on capture of LPS by monoclonal antibodies, and (3) indirect determination by measuring the ability of antibiotic-induced LPS to trigger TNFalpha release from a monocytic cell line. With both the Limulus and ELISA tests, a low endotoxin-releasing activity of ciprofloxacin was confirmed. In contrast to previous studies, this LPS also had low bioactivity in terms of TNFalpha induction. Limulus LPS determinations correlated more precisely with LPS bioactivity than did ELISA values, an observation which underlines the crucial role of LPS determination methods in studies of antibiotic-induced LPS release.

New criteria for selecting the proper antimicrobial chemotherapy for severe sepsis and septic shock

The mortality rate resulting from severe bacterial sepsis, particularly that associated with shock, still approaches 50% in spite of appropriate antimicrobial therapy and optimum supportive care. Bacterial endotoxins that are part of the cell wall are one of the cofactors in the pathogenesis of sepsis and septic shock and are often induced by antimicrobial chemotherapy even if it is administered rationally. Not all antimicrobial agents are equally capable of inducing septic shock; this is dependant on their mechanism of action rather than on the causative pathogen species. The quantity of endotoxin released depends on the drug dose and whether filaments or spheroplast formation predominates. Some antibiotics such as carbapenems, ceftriaxone, cefepime, glycopeptides, aminoglycosides and quinolones do not have the propensity to provoke septic shock because their rapid bactericidal activity induces mainly spheroplast or fragile spheroplast-like bacterial forms.

Lipopolyamines: novel antiendotoxin compounds that reduce mortality in experimental sepsis caused by gram-negative bacteria

The interactions of lipopolyamines, a class of structurally unique compounds currently being used as transfection (lipofection) agents, with lipopolysaccharide (LPS) have been characterized. Our studies have demonstrated that 1,3-di-oleoyloxy-2-(6-carboxyspermyl)-propylamide), available commercially as DOSPER, binds to purified LPS with an affinity of about 1/10 that of polymyxin B. This essentially nontoxic compound inhibits, in a dose-dependent manner, LPS-induced activation of the Limulus clotting cascade and the production of tumor necrosis factor alpha (TNF-alpha) interleukin-6 (IL-6), and nitric oxide from LPS-stimulated J774.A1 cells, a murine macrophage-like cell line. Cytokine inhibition is paralleled by decreased steady-state levels of TNF-alpha and IL-6 mRNA and inhibits the nuclear translocation of nuclear factor kappa B. These findings suggest that the lipopolyamine compound sequesters LPS, thereby blocking downstream cellular activation events that lead to the production of proinflammatory mediators. Administration of DOSPER to D-galactosamine-sensitized mice challenged either with LPS or with Escherichia coli organisms provided significant protection against lethality both with and without antibiotic chemotherapy. Partial protection is evident in LPS-challenged mice treated with DOSPER as late as 2 to 4 h following the endotoxin challenge. A greater degree of protection is observed in E. coli-challenged animals receiving ceftazidime than in those receiving imipenem, which is probably attributable to the higher levels of LPS released in vivo by the former antibiotic. Potent antiendotoxic activity, low toxicity, and ease of synthesis render the lipopolyamines candidate endotoxin-sequestering agents of potential significant therapeutic value.

Antibiotic-induced release of bacterial cell wall components in the pathogenesis of sepsis and septic shock: a review

This article reviews the new criteria for selecting the proper antimicrobial agent and dosage regimen for standard treatment of severe sepsis, with the intention of preventing septic shock. After introducing new concepts on the pathogenesis of sepsis and septic shock, the authors analyze the parameters of betalactam antibacterial activity, the antibiotic-induced release of bacterial endotoxin and the interrelationships between pharmacokinetics and pharmacodynamics of antibiotics in the search for an optimum dosage regimen of antimicrobial mono- or polytherapy for severely ill septic patients admitted to the intensive care unit.

Carbapenem-induced endotoxin release in gram-negative bacterial sepsis rat models

The carbapenem-induced endotoxin release was evaluated using experimental models of gram-negative bacterial sepsis in Wistar rats. Infections with Escherichia coli, Serratia marcescens, Klebsiella pneumoniae, Pseudomonas aeruginosa, Proteus vulgaris and Proteus mirabilis resulted in an increase of the plasma endotoxin concentration after treatment with ceftazidime and carbapenems including imipenem, panipenem, meropenem and biapenem. Except for P. aeruginosa, the plasma endotoxin concentrations after carbapenem treatment were significantly lower than those after ceftazidime treatment. It is noteworthy that treatment of P. aeruginosa sepsis with meropenem or biapenem induced significantly more endotoxin release than other carbapenems and the endotoxin concentrations induced by these carbapenems reached those of ceftazidime treatment. The plasma endotoxin concentrations appeared to correlate with the reduction of platelet counts and the elevation of both glutamic oxaloacetic transaminase and glutamic pyruvic transaminase values.

In vitro effect of cefodizime, imipenem/cilastatin and co-trimoxazole on dexamethasone and cyclosporin A depressed phagocytosis

Phagocytosis is an important part of the host defense against infection. Antibiotics can influence phagocytic function. In the present study, leukocyte metabolic response to phagocytic challenge by latex was assessed in relation to in vitro addition of cotrimoxazole, imipenem/cilastatin, cefodizime, dexamethasone (DXM), and/or cyclosporin A (CsA). Using latex particles as phagocytic challenge, glucose-1-14C utilization and 14CO2 production were measured by liquid scintillation counting. The phagocytic response was impaired by in vitro addition of DXM or CsA and this setup was used as an experimental model of immunodepression. The addition of co-trimoxazole to control samples (without DXM or CsA) depressed the response to latex challenge, whereas imipenem and cefodizime had a neutral effect. In the presence of DXM, co-trimoxazole induced a further decrease. The depressive effect of DXM was partially neutralized in the presence of cefodizime. With CsA depression, co-trimoxazole also induced a further decrease, imipenem had a neutral effect, while cefodizime partially restored the CsA suppressed reaction. Co-trimoxazole depressed the phagocytic response, imipenem had a neutral effect, whereas cefodizime restored the experimentally induced immunosuppression.

The significance of endotoxin release in experimental and clinical sepsis in surgical patients--evidence for antibiotic-induced endotoxin release?

Sepsis and peritonitis remain a serious challenge for surgical patients, despite improvement in surgical therapy and intensive care and the introduction of new powerful antibiotics. Recent in vitro studies revealed the potential of certain antibiotics, e.g. penicillin-binding protein (PBP) 3-specific antibiotics, to cause antibiotic-induced endotoxin release. Other types of antibiotics, e.g., PBP 2-specific antibiotics, were associated with no or less endotoxin release. Further in vitro experiments and investigations in animals support the hypothesis of antibiotic-induced endotoxin release, but there is little clinical evidence. The clinical significance of endotoxin is subject of open dispute with many pro's and contra's. Endotoxin, although an important trigger, may not be the only factor to induce cytokine release, e.g., peptidoglycans were able to stimulate cells to release cytokines. Gram-positive pathogens have gained more importance in clinical sepsis and may not be sufficiently reflected in current clinical studies. The hypothesis that neutralization of endotoxin and pro-inflammatory cytokines is beneficial in sepsis was seriously challenged by the results of recent clinical and experimental studies. The better understanding of mechanisms in endotoxin-induced cell activation and cell, cell-receptor and soluble receptor interactions led to new treatment options. Recent reports on the complex pathogenesis of peritonitis and the detection of pathogen-related factors with intraperitoneal immune response may have implications on clinical studies investigating the potential of new compounds and the effect of antibiotics on endotoxin release. However, only few reports are available on the clinical significance of antibiotic-induced endotoxin release, and association of endotoxin release with pathogens, mortality or alteration of physiological parameters were not observed. With regard to the particulars of these studies, e.g., a small study population or low mortality rate, mortality may not be an ideal outcome parameter for these studies. There is clinical evidence for antibiotic-induced endotoxin release. However, the need for well-designed and performed studies using newly developed monitoring devices in intensive care therapy is obvious.

Differential release of smooth-type lipopolysaccharide from Pseudomonas aeruginosa treated with carbapenem antibiotics and its relation to production of tumor necrosis factor alpha and nitric oxide

Endotoxin release from Pseudomonas aeruginosa treated with cell wall-active carbapenem antibiotics and its effect on the production of tumor necrosis factor alpha and nitric oxide were examined. Treatment of bacteria with imipenem induced much lower levels of endotoxin release than treatment with meropenem. The endotoxin released was demonstrated to be of the smooth type and O-specific polysaccharide-rich. The exposure of the filtrates of P. aeruginosa treated with imipenem to physiologically relevant cells caused low-level production of tumor necrosis factor alpha and nitric oxide, while similar treatment with meropenem induced high levels of production.