Endothelin-1 impairs neutrophil respiratory burst and elimination of Escherichia coli in rabbits

OBJECTIVE: During systemic inflammation, elevated levels of endothelin (ET)-1 have been reported. The aim of this study was to investigate the effects of ET-1 on neutrophil (PMN) respiratory burst, phagocytosis, and elimination of Escherichia coli from blood and tissues. DESIGN: Prospective, randomized, controlled trial. SETTING: Experimental laboratory in a university hospital. SUBJECTS: A total of 18 female chinchilla rabbits. INTERVENTIONS: To quantify the clearance process, defined numbers (10(8) colony-forming units) of E. coli were injected intravenously into anesthetized rabbits, 60 mins after onset of continuous 0.2 microg/kg/min ET-1 administration (n = 9) and after saline infusion (control group, n = 9), respectively. To evaluate potential effects of ET-1 on bacterial elimination and killing, blood clearance of E. coli and colonization of different organs were investigated. MEASUREMENTS: Variables monitored were neutrophil respiratory burst and phagocytosis activity, rates of bacterial elimination from the blood, arterial blood pressure, blood gases, serum lactate concentrations, and nitrite and nitrate levels. The animals were killed 3 hrs after bacterial injection and tissue samples of liver, kidney, spleen, and lung were collected for bacterial counts. MAIN RESULTS: Compared with the control group, ET-1 significantly impaired PMN respiratory burst (p < .05) and prolonged elimination of injected E. coli from the blood (p < .01), whereas phagocytosis functions remained unaltered. The reduced PMN burst activity after ET-1 was associated with a higher bacterial colonization of all organs (lung, p < .01; spleen, p < .05). Endothelin-1 induced increases in mean arterial pressure (p < .01) and serum lactate concentrations, whereas nitrite and nitrate levels remained unaltered. CONCLUSION: Endothelin-1 impairs respiratory burst and bacterial clearance from the blood and tissue. Thus, elevated levels of ET-1 during sepsis could induce organ hypoperfusion and cause disturbances in immune functions, increasing the risk of bacterial infections.     

Pentoxifylline improves bacterial clearance during hemorrhage and endotoxemia.

OBJECTIVE: The aim of this study was to investigate whether the methylxanthine-derivative pentoxifylline (PTX) affects bacterial clearance of the organism in states of hemorrhage and endotoxemia. DESIGN: Prospective, randomized, controlled trial. SETTING: Experimental laboratory in a university hospital. SUBJECTS: Fifty-four female chinchilla rabbits. INTERVENTIONS: To quantify the clearance process, defined numbers (10(7) CFO) of Escherichia coli bacteria were injected intravenously into anesthetized rabbits, 60 mins after induction of hemorrhage (n = 9 + 3) or infusion of endotoxin (lipopolysaccharide [LPS]; 40 microg/kg/hr; n = 9 + 3) and after saline infusion (control; n = 9), respectively. Hemorrhage was induced by bleeding, standardized by defined reduction of mean arterial pressure (30% of baseline value). To evaluate the potential effects of PTX on bacterial elimination and killing, in states of hemorrhage and endotoxemia, blood clearance of E. coli and colonization of different organs were investigated after pretreatment with PTX (30 mg/kg) as a bolus injection followed by continuous infusion of PTX (50 mg/kg/hr) in hemorrhagic (n = 9) and endotoxemic rabbits (n = 9). Three additional experiments were performed to evaluate the effects attributable to PTX itself. MEASUREMENTS AND MAIN RESULTS: Parameters monitored were rates of bacterial and LPS elimination from the blood, arterial blood pressure, blood gases, and serum lactate concentrations. Additionally, flow cytometric analysis of respiratory burst activity was performed. Three hours after bacterial injection, the animals were killed, and tissue samples of liver, kidney, spleen, and lung were collected for bacteriologic examinations. Compared with the controls, hemorrhage and endotoxemia resulted in a significantly prolonged elimination of injected E. coli from the blood. The delayed blood clearance was associated with a significantly (p < .01) higher bacterial colonization of all organs, which was most pronounced in the lung. Pretreatment with PTX slightly enhanced blood clearance of E. coli as well as of LPS, and significantly reduced (p < .05) the colonization of lung and kidney after hemorrhage and endotoxemia. Furthermore, PTX suppressed polymorphonuclear neutrophil respiratory burst activity. CONCLUSIONS: Hemorrhage and endotoxemia induce impaired bacterial clearance from blood and tissue. Treatment with PTX may reduce the risk of bacterial infections by attenuating bacterial colonization of organs in states of hemorrhage and endotoxemia.     

N(omega)-nitro-L-arginine methyl ester effects on neutrophil function and bacterial clearance

Nitric oxide synthase (NOS) inhibitors are considered promising as a therapeutic option in severe septic shock. The aim of this study was to investigate the effects of N-nitro-L-arginine methyl ester (L-NAME) application on neutrophil (PMN) respiratory burst, phagocytosis, and elimination of Escherichia coli from blood and tissue in rabbits. Twenty-eight female chinchilla rabbits were randomized to a treatment and control group. To quantify the bacterial clearance process, 10 colony forming units (CFU) of E. coli were injected intravenously into anesthetized rabbits. Animals in the L-NAME group had a significantly higher mortality compared with controls. NOS inhibition resulted in a significant delay of bacterial clearance (P < 0.001). These findings correlated with a significant augmentation of all organ E. coli findings (P = 0.002-0.035). PMN phagocytosis activity was notably reduced by L-NAME treatment during the experimental observation. Neutrophil burst, on the other hand, was amplified by NOS inhibition (P = 0.008). Our findings point to an interference with the PMN-dependent immune mechanisms after L-NAME treatment. The augmented PMN burst reaction could be a compensatory mechanism, potentially leading to tissue damage. Therefore, in this model, we find sufficient evidence pointing to a possible cause for the deleterious effect of early nonselective NOS inhibition in critically ill patients.     

Effects of IGM-enriched solution on polymorphonuclear neutrophil function, bacterial clearance, and lung histology in endotoxemia

Immunological interventions in endotoxemia and sepsis have been tested in experimental and clinical studies. Our group evaluated the effects of an immunoglobulin (Ig)M-enriched solution in an established model of Gram-negative bacteraemia. Ten New Zealand White rabbits (2-3 kg) were randomized to a treatment or control group. In both groups, LPS was infused at a rate of 40 mg kg(-1) h(-1). Immunoglobulin M-enriched solution (Pentaglobin; 2 mL kg(-1) h(-1)) was applied in the intervention group 15 min after beginning LPS infusion. 1 x 10(8) colony forming units of Escherichia coli were injected 30 min after LPS infusion was commenced. Baseline hemodynamic and respiratory parameters, blood E. coli concentration (30 min before and 1, 15, 30, 60, 90, 120, and 180 min after E. coli injection), polymorphonuclear neutrophil oxidative burst activity, and phagocytosis dead space (both 30 min before and 1, 15, 60, 120, and 180 min postinjection) were measured. Ex vivo phagocytosis activity was measured in a separate experiment and evaluated by electron microscopy. Diffuse alveolar damage (DAD) was measured. Organ colonization (kidney, lung, liver, spleen) was assessed in aseptic organ samples. Hemodynamic parameters did not differ between the two groups. Bacterial blood clearance was not influenced by application of IgM-enriched solution. Liver and spleen colonization was significantly reduced in the IgM group. Immunoglobulin M-enriched solution reduced in vitro residual phagocytosis capacity at 30, 90, and 180 min and improved respiratory burst at 180 min. Correspondingly, ex vivo phagocytosis activity as documented by electron microscopy was increased in the IgM group. The sum of all weighted DAD scores (except overdistension) was significantly better in the IgM group (23+/-5 vs. 30+/-8). Immunoglobulin M-enriched solution significantly improved six of seven DAD score parameters and reduced liver and spleen E. coli count. Residual phagocytosis capacity was significantly decreased in the IgM group, whereas burst activity was increased, pointing to an increased in vivo phagocytosis efficiency. Short-term IgM-enriched solution intervention had an especially beneficial effect on LPS-induced pulmonary histological changes.     

Modulation of host defense by hydrocortisone in stress doses during endotoxemia

Objective To investigate the effects of low-dose hydrocortisone (HC) on neutrophil respiratory burst, phagocytosis, and elimination of E. coli from blood and tissue under endotoxemic and non-endotoxemic conditions.Design Randomized, controlled trial.Setting Experimental laboratory, university hospital.Subjects Forty-eight female chinchilla rabbits (n=8 in six groups A–F).Interventions In order to quantify the bacterial clearance process, defined numbers [10⁸ colony forming units (CFU)] of Escherichia coli were injected intravenously into all anesthetized rabbits. Group A did not receive further intervention. Group B received bolus administration of HC 1.4 mg/kg and group C 14 mg/kg. Endotoxin (LPS, 40 g/kg/h) was given to groups D, E, and F. Group E received additional bolus administration of HC 1.4 mg/kg and group F 14 mg/kg. All HC groups (B, C, E, and F) were continuously infused with HC 0.18 mg/kg/h.Measurements Monitored parameters were neutrophil respiratory burst and phagocytosis activity, rates of bacterial elimination from the blood, arterial blood pressure, serum lactate and LPS concentrations, as well as nitrite and nitrate levels. Tissue samples of liver, kidney, spleen, and lung were collected for bacterial counts.Main results In controls HC significantly delayed elimination of injected E. coli from the blood (P<0.01). LPS also prolonged bacterial elimination but additional HC did not further delay removal of E. coli from the blood. Under endotoxemia HC depressed respiratory burst, whereas phagocytosis functions remained unaltered. Moreover, bacterial colonization of organs was reduced after HC in the LPS groups. Significance, however, was reached only in the liver (P<0.05). Due to HC, clearance from LPS (P<0.01) and lactate (P<0.05) were improved. Levels of nitrite and nitrate did not differ among the groups.Conclusion HC demonstrated immunomodulatory effects even in stress doses. In endotoxemic states use of low-dose HC seems to be favorable, although not in non-septic conditions.The University Hospital of Dresden is a Harvard Medical International Associated Institution to which the authors A.R. Heller, S.C. Heller, S.N. Stehr, and T. Koch belong     

Polymorphonuclear oxidative burst after Helicobacter pylori water extract stimulation is not influenced by the cytotoxic genotype but indicates infection and gastritis grade.

H. pylori-associated gastric mucosal inflammation is characterized by the presence of polymorphonuclear (PMN) leukocyte infiltrate, which is more severe when the infecting strain is cagA positive. After appropriate stimuli, such as bacterial products, PMN release large amounts of oxygen derived free radicals and proteases, to kill the bacterium. H. pylori seems to be particularly resistant to the oxidative machinery of PMN, which can in turn damage the host gastric mucosa. We evaluated peripheral PMN oxidative burst response after stimulation with water extracts from cagA positive (WEcagA+) or negative (WEcagA-) H. pylori strains in infected (n=31) and non-infected patients (n=32) in comparison with healthy controls (n=16); the influence of gastric mucosal inflammatory infiltrate and activity grade on PMN oxidative burst were also assessed. PMN oxidative burst was measured by FACS analysis. H. pylori water extracts were obtained from bacterial culture. H. pylori genotype was determined by means of the polymerase chain reaction. The PMN oxidative burst in H. pylori infected patients was significantly higher than that in H. pylori negative or healthy controls, no differences being found when the results following WEcagA+ and WEcagA- stimulation were compared. The difference in PMN oxidative burst obtained after WEcagA- and E. coli (standard stimulus for PMN oxidative burst) stimulation discriminated H. pylori infected from non-infected patients with a sensitivity of 90% and a specificity of 97%. The grade of PMN oxidative burst correlated with PMN infiltration grade of the gastric mucosa. Our findings allow to conclude that PMN oxidative burst activation by H. pyloriWE is species- but not strain-correlated. PMN priming, probably consequent to the action of soluble mediators released by mononuclear cells, makes PMN hyper-responsive to H. pylori products, thus favoring the release in the gastric mucosa of infected patients of large amounts of oxygen-derived free radicals, which are not enough to eliminate the infection, but may contribute to damaging the gastric mucosa itself. Peripheral PMN oxidative burst response to H. pyloriWE might furthermore be of help in diagnosing H. pylori infection.     

Bacterial Growth In Vivo AN IMPORTANT DETERMINANT OF THE PULMONARY CLEARANCE OF DIPLOCOCCUS PNEUMONIAE IN RATS

Lung clearance of Diplococcus pneumoniae was markedly reduced in rats with acute hemorrhagic pulmonary edema produced by instillation of hydrochloric acid. Bacterial clearance was enhanced in both control and acid-instilled animals by pretreatment with a bacteriostatic antibiotic, tetracycline, 30 mg/kg. From these data the contributions of bacterial multiplication and bacterial elimination to net lung bacterial clearance were estimated. In control animals the constant for exponential bacterial elimination was -1.4283 (fractional clearance = 76% per h), and the doubling time for the pneumococcus was 170 min. In acid-instilled rats the elimination constant was -0.5336 (fractional clearance = 41% per h), and the doubling time of the pneumococcus was 47 min, approximating the doubling time of 42 min observed with pneumococci grown in broth.These results indicate that, in the case of pneumococci, both bacterial elimination and bacterial growth contribute to lung bacterial clearance in normal animals as well as animals with damaged lungs. In the present study changes in both parameters were required to explain the observed results in acid-instilled animals. The pulmonary pathogenicity of some bacterial species may be determined by their capacity for growth in the lung, since infection of the lung occurs when bacterial multiplication exceeds the rate of elimination of viable organisms.     

Xanthine oxidase released from reperfused hind limbs mediate kupffer cell activation, neutrophil sequestration, and hepatic oxidative stress in rats subjected to tourniquet shock

We have shown previously that rats subjected to tourniquet shock develop an acute form of remote organ injury of the liver that is both Kupffer cell (KC) and polymorphonuclear (PMN) leukocyte dependent. Circulating plasma xanthine oxidase (XO) has been shown to be responsible for the development of endothelial dysfunction and for remote organ injury of the lung and intestine after ischemia-reperfusion protocols. We now hypothesize that XO is released from rat hind limbs upon reperfusion and that it is responsible for KC and PMN leukocyte activation in this shock model. Our results show that about 30% of rat gastrocnemius muscle xanthine dehydrogenase (XD) is converted to XO during the 5-h tourniquet period and that it is released into the femoral vein within 10 min of reperfusion. Total muscle xanthine oxidoreductase activity (XO + XD) decreases within 30 min of reperfusion and is paralleled by a corresponding increase in femoral vein lactic dehydrogenase. In addition, liver tissue XO increases significantly within 30 min of reperfusion without a corresponding conversion of endogenous XD. Conversion of hepatic XD becomes evident 60 min after reperfusion is initiated, as does XO, and alanine aminotransferase (ALT) release into the hepatic vein, presumably from damaged hepatocytes as a consequence of oxidative stress. Tissue myeloperoxidase activity also increases significantly after the 60-min reperfusion period. That XO mediates KC and PMN activation is supported by the following observations: a) the close relationships between plasma XO and the time courses of tumor necrosis factor-alpha TNFalpha release into the hepatic vein and colloidal carbon clearance by KCs; b) that colloidal carbon clearance, TNFalpha and ALT release, loss of tissue free thiols, lipid peroxidation (TBARS), and liver infiltration by PMN neutrophils can also be induced by the administration of exogenous XO to normal rats; and c) pretreatment of rats with allopurinol inhibits KC activation and liver leukocyte infiltration. These results suggest that XO, released from the ischemic limb on reperfusion, is taken up by the liver were it mediates KC and PMN neutrophil activation and thus contributes to the development of multiple system organ failure after hind limb reperfusion.     

Impairment of bacterial clearance induced by norepinephrine infusion in rabbits

Objective

Purpose of the study was to investigate the potential influence of norepinephrine (NE) on immune functions in terms of systemic and organ-specific bacterial clearance in rabbits.

Design

To enable quantification of the clearance process, defined numbers of exogenousEscherichia coli (1.3×10⁸ CFU) were injected intravenously 60 min after starting the NE infusion at a low dose (1 μg/kg per min,n=6), causing an increase (30 mmHg) in mean arterial pressure without affecting the oxygen uptake, and at a higher dose (7.5 μg/kg per min,n=6), resulting in a marked decrease (20%) in oxygen uptake, after infusion of NaCl solution (control,n=6). In additional experiments (n=6) NE (1 μg/kg per min) was tested in endotoxemia induced by simultaneous infusion of endotoxin (40 μg/kg per h). Parameters monitored were arterial pressure, oxygen uptake, and rates of bacterial elimination from the blood. At 180 min afterE. coli injection, the animals were sacrificed, and tissue samples of liver, kidney, spleen, and lung were collected for bacterial counts.

Results

NE infusion resulted in a dose-dependent prolonged elimination of the injectedE. coli from the blood and in significantly higher (p<0.05) numbers of CFU in liver and lung compared to the controls. Significant impairment of bacterial clearance was found after shockproducing endotoxemia, whereas simultaneous infusion of NE and endotoxin caused only a slightly delayed blood clearance of the injected bacteria.

Conclusion

NE dose dependently affected bacterial clearance, which might be due to ischemia-derived hypoxic impairment of the phagocytosis and lysis function of the reticuloendothelial system, whereas NE improved elimination of bacteria in a state of endotoxic shock.     

Separation of sublethal and lethal effects of polymorphonuclear leukocytes on Escherichia coli.

Escherichia coli ingested by PMN promptly stop growing and form no colonies in nutrient agar, but metabolize near normally for up to several hours. The bactericidal/permeability increasing protein (BPI) of PMN also inhibits E. coli growth without initial metabolic impairment. We recently showed that BPI-treated E. coli, although unable to grow in normal nutrient agar, can form colonies in this medium plus 0.1% BSA, as long as their metabolism is maintained, indicating that biochemical impairment is a better indicator of death than growth arrest (1990. J. Clin. Invest. 85:853-860). We have now reexamined the fate of ingested E. coli. Rabbit PMN ingest greater than 85% of several rough E. coli strains in 15 min, but greater than 80% of these bacteria, while unable to form colonies in conventional agar, grow normally on agar plus 0.1% BSA. Thus, the PMN under these conditions promptly stop growth of ingested E. coli without killing. Adding nonlethal concentrations of normal human serum (NHS) before, but not after ingestion, accelerates killing and, in parallel, loss of bacterial metabolism (t1/2 less than 0.5 h vs. greater than 3 h, respectively, with and without NHS). The rapid killing of both rough and smooth E. coli pretreated with NHS is lost after C7 depletion (C7-D) and restored when C7 is replenished. Similar results are obtained with human PMN. In contrast, ingested Staphylococcus epidermidis, opsonized with either NHS or C7-D serum rapidly stop metabolizing and do not form colonies in nutrient agar with or without BSA. Respiratory burst activity is the same during ingestion of E. coli (with or without NHS) and S. epidermidis. Killing of E. coli J5 (however, not of O111-B4) by BPI is also accelerated by pretreatment with NHS but not C7-D human serum. These findings indicate that late complement components are needed for efficient killing of both rough and smooth E. coli by PMN, and that BPI is the principal intracellular agent acting on ingested rough E. coli.     

Oxygen-independent intracellular and oxygen-dependent extracellular killing of Escherichia coli S15 by human polymorphonuclear leukocytes.

Effective killing of bacteria by polymorphonuclear leukocytes (PMN) is generally assumed to require intracellular sequestration and, depending on the bacterial species, can be both O2-dependent or O2-independent. Killing of several strains of Salmonella typhimurium and Escherichia coli by rabbit PMN does not require O2 and is apparently due to a granule-associated bactericidal/permeability-increasing protein (BPI) present in rabbit and human PMN. In this study we examined the O2 dependence of the killing of E. coli (S15) by human PMN. Ingested and noningested E. coli were separated by centrifugation after incubation with PMN in room air or under N2. In the presence of heat-treated serum approximately 50% of E. coli (10 bacteria/PMN) were taken up by PMN and rapidly (5-15 min) killed both in room air and under N2. The remaining extracellular bacteria (approximately 50%) were killed during 30-60 min of incubation in room air but not under N2. When uptake of E. coli by PMN was increased to approximately 80% by the use of C6-depleted serum (retaining heat-labile opsonins), bacterial survival under N2 was reduced from 54 +/- 7.6% to 13 +/- 5.5%. PMN from a patient with chronic granulomatous disease killed PMN-associated but not extracellular E. coli. BPI was detected, by indirect immunofluorescence, on the surface of PMN-associated E. coli within 5 min of incubation of E. coli with PMN both in room air and under N2. In contrast, at no time was BPI detected on the surface of extracellular E. coli, indicating that the non-PMN-associated E. coli had not been previously ingested. Thus, killing of ingested E. coli S15 by human as well as rabbit PMN does not require O2 and appears to be BPI-mediated. However, when ingestion is limited, extracellular bacteria can also be killed but principally by O2-dependent mechanisms.     

The role of lysosomal elastase in the digestion of Escherichia coli proteins by human polymorphonuclear leukocytes: experiments with living leukocytes.

Human polymorphonuclear leukocyte (PMN) elastase has been implicated in various pathological conditions. However, its physiological role remains undefined. One possible function of this enzyme may be digestion of bacterial proteins after phagocytosis. To test this hypothesis, we prepared Escherichia coli labeled with [3H]arginine and treated these bacteria with a lipid-soluble, active-site-directed chloromethyl ketone inactivator of pancreatic and granulocyte elastases (carbobenzoxy-L-glycyl-L-leucyl-L-alanine chloromethyl ketone, dissolved in methanol). Control bacteria were treated with methanol alone. When E. coli pretreated with the inactivator were incubated with solutions of porcine pancreatic elastase or with PMN granule extract, release of trichloroacetic acid-soluble radioactivity was significantly lower than in the control E. coli. Similar results were obtained when treated and control E. coli were fed to viable human PMN. In contrast, release of trichloroacetic acid-soluble radioactivity from E. coli containing [3H]thymidine was not affected by pretreatment of bacteria with elastase inactivator before feeding them to PMN, suggesting that phagocytosis of E. coli had not been inhibited by the chloromethyl ketone. When treated and control bacteria were fed to PMN, no significant difference was observed in the activity of lysosomal beta-glucuronidase recovered from post-granule supernatant fractions of homogenized leukocytes, suggesting that lysosomal degranulation had not been suppressed by the inactivator. However, elastase activity of the same fractions was depressed if the leukocytes had phagocytized chloromethyl ketone-treated E. coli, suggesting that inhibition of PMN elastase had occurred. We conclude that PMN elastase participates in digestion of E. coli proteins by human PMN.     

Role of Küpffer cells and PMN leukocytes in hepatic and systemic oxidative stress in rats subjected to tourniquet shock.

Küpffer cells (KCs) have been implicated in leukocyte recruitment and microvascular dysfunction associated with liver inflammation. The overall objective of this study was to assess the role of KCs and polymorphonuclear (PMN) leukocytes on the oxidative stress elicited in the liver as a consequence of hind limb reperfusion in rats subjected to tourniquet shock, a shock model that differs from other models in that hepatic injury is a consequence of remote organ damage. Colloidal carbon clearance from blood and its incorporation into KCs demonstrate that these cells are activated after the 2 h hind limb reperfusion period and that they are responsible for the observed oxidative stress and for PMN leukocyte recruitment and activation. Liver oxidative stress in this model is evidenced by increased liver tissue GSSG/GSH ratio, thiobarbituric acid reactive substances (TBARS), an index of lipid peroxidation, myeloperoxidase (MPO) activity, an index of tissue-associated neutrophil accumulation, and a significant loss in total tissue superoxide dismutase (SOD) activity. Mean arterial blood pressure (MAP), as well as plasma levels of alanine aminotransferase (ALT), an index of hepatic tissue injury, total SOD activity, plasma levels of alpha-tocopherol and beta-carotene, and total plasma nitrite are also affected as a consequence of KC activation after the 2 h hind limb reperfusion period. Inhibition of KC activity by gadolinium chloride (GdCl3) reverted most of the above alterations to values that do no differ from those found in control animals. These results support the hypothesis that hepatic and systemic oxidative stress elicited by hind limb reperfusion in rats subjected to tourniquet shock is both KC and PMN leukocyte dependent.     

Neutrophil killing of single microorganisms as measured by a new method

A new method of measuring polymorphonuclear leukocyte (PMN) microbicidal activity was developed in which single colony forming units of Staphylococcus aureus were isolated in microtiter wells with or without PMN, centrifuged, and incubated overnight. The percentage of bacteria killed was determined by comparing the number of PMN-containing wells showing bacterial growth to PMN-free controls. This percentage was positively correlated with the number of PMN in each well and was reduced if plates were not centrifuged or if cytochalasin B was added. Killing of either S. aureus or Escherichia coli was enhanced using fresh serum activated by prior exposure to large numbers of E. coli. There was no enhancement using heated serum that had been exposed to E. coli, nor was there enhancement when S. aureus was preopsonized by exposure to fresh serum. This assay circumvents problems in assessing microbicidal activity where aggregation of organisms might otherwise be an issue.     

Determinants of lung bacterial clearance in normal mice.

The determinants of the lung clearance of Streptococcus pneumoniae, Klebsiella pneumoniae, Escherichia coli, and Staphylococcus aureus were studied in normal mice after exposure to an aerosol of viable bacteria and 99mTc-labeled dead bacteria. The fraction of bacteria in lungs that remained viable 4 h after exposure were: S. pneumoniae, 7.3%; K. pneumoniae, 121%; E. coli, 88.5%; S. aureus, 27.6%. The rate of physical removal of bacterial particles (Kmc) was determined from the change in lung 99mTc counts with time: Kmc ranged between 7 and 12%/h and and was similar in all species. The rate of mucociliary clearance and of intrapulmonary bacterial killing (Kk + Kmc) was calculated from the change in bacterial counts with time in animals that had received tetracycline to inhibit bacterial multiplication. Kk, the rate of intrapulmonary killing, was obtained by subtraction of Kmc from (Kk + Kmc). The calculated values for Kk were: S. pneumoniae, - 87%/h; K. pneumoniae, - 17%/h; E. coli, - 18%/h; S. aureus, - 22%/h. The rate of intrapulmonary bacterial multiplication (Kg) was estimated from the relationship of bacterial counts in tetracycline and nontetracycline-treated animals, assuming that tetracycline altered only Kg. Kg, expressed as the doubling time, was: S. pneumoniae, 310 min; K. pneumoniae, 217 min; E.coli, 212 min; S. aureus, infinity (no multiplication). The data indicate that the marked differences in the clearance of these species from the normal mouse lung result from the interaction of differing rates of in vivo bacterial multiplication and killing.     

Gal-Gal pili vaccines prevent pyelonephritis by piliated Escherichia coli in a murine model. Single-component Gal-Gal pili vaccines prevent pyelonephritis by homologous and heterologous piliated E. coli strains.

The initial pathogenic step in nonobstructive Escherichia coli pyelonephritis usually involves the binding of a bacterial adhesin with host uroepithelial glycoprotein receptors containing the D-Gal p alpha 1----4 D-Gal p beta 1 (Gal-Gal) moiety. In this study, groups of mice were immunized with Gal-Gal pili and challenged 2 wk later intravesicularly with E. coli strains expressing homologous or heterologous pili. 63 of 129 pili-immunized mice (49%) were protected from subsequent E. coli renal colonization compared with 5 of 85 control mice (6%). Among mice that had E. coli cultured from their right kidney, control mice had greater bacterial colony counts than pili-immunized animals (P less than 0.05). Light microscopic examination of kidneys demonstrated less histopathology among pili immunized mice than among control mice (P less than 0.05). Protection correlated with the presence of specific IgG antibodies in the urine and serum that bind to the major pilin structural polypeptide and not to the Gal-Gal pili tip adhesin per se. These results support the concept that immunization with a bacterial surface-coat constituent can prevent mucosal infection by interfering with colonization. Also Gal-Gal pili of E. coli represent a suitable candidate for immunoprophylaxis against pyelonephritis.     

Effect of anti-CD14 monoclonal antibody on clearance of Escherichia coli bacteremia and endotoxemia

OBJECTIVE: To determine the effects of an anti-CD14 monoclonal antibody on the clearance of a bacteremic Escherichia coli challenge in the presence or absence of antimicrobial agents. DESIGN: Prospective randomized animal study. SETTING: University-affiliated research laboratory. SUBJECTS: New Zealand White rabbits weighing 1.5-2.5 kg. INTERVENTIONS: Animals were pretreated with either an anti-lapine CD14 monoclonal antibody (immunoglobulin G2a, 5 mg/kg intravenously) or an isotype control monoclonal antibody. The animals then were challenged with 1 x 10(6) E. coli 018:K1 in the presence or absence of ceftazidime (50 mg/kg intravenously). There were four groups of six animals randomized to receive either anti-CD14 monoclonal antibody without ceftazidime, isotype control monoclonal antibody without ceftazidime, anti-CD14 monoclonal antibody with ceftazidime, or isotype control antibody with ceftazidime. MEASUREMENTS AND MAIN RESULTS: Serial measurement of quantitative bacteremia and endotoxemia was performed over 24 hrs after the administration of the bacterial challenge. Animals also underwent necropsy with quantitative bacterial cultures from multiple organ tissue samples. The anti-lapine CD14 monoclonal antibody significantly impaired the bloodstream clearance of E. coli (p <.01) and increased quantitative counts of E. coli in tissue culture samples when compared with isotype control antibody in the absence of simultaneous administration of ceftazidime. No differences in quantitative bacteremia, endotoxemia, or organ tissue counts were found after anti-CD14 antibody and control antibody-treated animals in the presence of ceftazidime treatment. CONCLUSIONS: Anti-CD14 monoclonal antibody has the capacity to interfere with the innate immune response and systemic microbial clearance in experimental animals with E. coli bacteremia. The concomitant administration of effective antimicrobial therapy eliminated differences in the rate of microbial clearance between the control antibody and the CD14 monoclonal antibody. These results indicate that care should be taken in clinical trials with anti-CD14 monoclonal antibodies to ensure that adequate antimicrobial therapy is administered in the presence of systemic bacterial infection.     

Differential function of polymorphonuclear leukocytes between in vivo and in vitro in tumor-bearing mice

In the present report, we compared activities of polymorphonuclear leukocytes (PMN) such as phagocytosis and bactericidal activity in vivo with those in vitro in sarcoma 180 (S 180)-bearing mice. Mice showed a remarkable leukocytosis and in increase in PMN fraction of peripheral blood leukocytes (PBL) after intraperitoneal injection of S 180 cells. Tumor-bearing mice infected with Escherichia coli (E. coli) intravenously and intraperitoneally showed an apparent delay in the clearance of bacteria compared to the non-tumor-bearing control mice. However, PBL of tumor-bearing mice showed a high phagocytic activity against beads and a high chemiluminescence (CL) activity. Dichlorofluorescein (DCFH) oxidation capacity of peripheral blood PMN in S 180-bearing mice after stimulation with phorbol myristate acetate (PMA) was about the same or a little stronger than that in control mice. On the contrary, serum and ascites of tumor-bearing mice strongly suppressed the phagocytic and bactericidal activities of casein-induced PMN against E. coli. During the early phase of E. coli infection, serum level of complement (C3) was not depressed in tumor-bearing hosts. From these results, it is concluded that leukocytosis and activation of functions of PMN in tumor-bearing mice were observed in vitro but they were not effective for the protection in the early phase of actual E. coli infection in vivo. The delay of in vivo clearance may be accounted for by a suppressive effect of serum components in tumor-bearing mice.     

Lymph from a primate baboon trauma hemorrhagic shock model activates human neutrophils

We have reported that toxic factors in intestinal lymph are responsible for acute lung injury and bone marrow suppression and that they contribute to a systemic inflammatory state based on studies in rodent models of trauma-hemorrhagic shock. Rodent models may not completely reflect the responses of injured patients. Thus, it is important to confirm these findings in primates before applying them to injured human patients with trauma. Thus, we have recently established baboon trauma-hemorrhagic shock (T/HS) and trauma-sham shock (T/SS) models that showed that gut-derived factors carried in the lymph potentiates lung injury and causes human endothelial dysfunction and suppresses human bone marrow progenitor cell growth. Here, we further investigated the effects of these primate lymph samples on human neutrophils. We hypothesized that toxic factors in baboon lymph may prime and/or activate human polymorphonuclear leukocyte (PMN) leading to overproduction of superoxide, thereby contributing to the development of adult respiratory distress syndrome and multiple organ failure. To this effect, we have examined the priming effect of baboon T/HS and T/SS lymph on PMN respiratory burst and expression of adhesion molecule in human neutrophils. The results of these studies indicate that PMN treated with baboon T/HS lymph showed significantly induced respiratory burst responses compared with PMN treated with T/SS lymph or medium when phorbol myristate acetate PMA was applied after lymph pretreatment. Secondly, we found that the expression of CD11b adhesion molecule was increased by incubation with T/HS lymph. These results suggest that baboon lymph from T/HS models can increase respiratory burst and adhesion molecule expression in human PMN, thereby potentially contributing to PMN-mediated organ injury.     

Lomefloxacin-induced modification of the kinetics of growth of gram-negative bacteria and susceptibility to phagocytic killing by human neutrophils

The post-antibiotic effect (PAE) of lomefloxacin against Escherichia coli and Pseudomonas aeruginosa was determined and compared with various other antibiotics. All the quinolones tested, and chloramphenicol and gentamicin, possessed PAE activity. At 10 x MIC and 30 min exposure, the PAEs against E. coli were 1.6, 1.3, and 1.8 h for lomefloxacin, ciprofloxacin and norfloxacin respectively, and for P. aeruginosa the equivalent PAEs were 1.1, 1 and 0.5 h. The lomefloxacin PAE was dose-dependent and exposure for 5 min was sufficient to give optimal PAE at high concentrations of lomefloxacin. Such brief exposure rapidly blocked bacterial nucleic acid biosynthesis. Lomefloxacin pretreated bacteria were more susceptible to killing by PMN than untreated bacteria. Optimum enhancement of phagocytic killing of E. coli occurred when exposure to lomefloxacin was associated with an 80% decrease in cfu during pretreatment. Maximum PMN activity against P. aeruginosa occurred when bacteria were exposed to lomefloxacin producing change in cfu of +0.2 log10 to -0.7 log10. These results indicate that phenotypic changes of P. aeruginosa and E. coli exposed to lomefloxacin render the bacteria more susceptible to phagocytic killing.