The use of a phospholipase A-less Escherichia coli mutant to establish the action of granulocyte phospholipase A on bacterial phospholipids during killing by a highly purified granulocyte fraction

A metaproteomics approach to elucidate host and pathogen protein expression during catheter-associated urinary tract infections (CAUTIs)

Long-term catheterization inevitably leads to a catheter-associated bacteriuria caused by multispecies bacterial biofilms growing on and in the catheters. The overall goal of the presented study was (1) to unravel bacterial community structure and function of such a uropathogenic biofilm and (2) to elucidate the interplay between bacterial virulence and the human immune system within the urine. To this end, a metaproteomics approach combined with in vitro proteomics analyses was employed to investigate both, the pro- and eukaryotic protein inventory. Our proteome analyses demonstrated that the biofilm of the investigated catheter is dominated by three bacterial species, that is, Pseudomonas aeruginosa, Morganella morganii, and Bacteroides sp., and identified iron limitation as one of the major challenges in the bladder environment. In vitro proteome analysis of P. aeruginosa and M. morganii isolated from the biofilm revealed that these opportunistic pathogens are able to overcome iron restriction via the production of siderophores and high expression of corresponding receptors. Notably, a comparison of in vivo and in vitro protein profiles of P. aeruginosa and M. morganii also indicated that the bacteria employ different strategies to adapt to the urinary tract. Although P. aeruginosa seems to express secreted and surface-exposed proteases to escape the human innate immune system and metabolizes amino acids, M. morganii is able to take up sugars and to degrade urea. Most interestingly, a comparison of urine protein profiles of three long-term catheterized patients and three healthy control persons demonstrated the elevated level of proteins associated with neutrophils, macrophages, and the complement system in the patient's urine, which might point to a specific activation of the innate immune system in response to biofilm-associated urinary tract infections. We thus hypothesize that the often asymptomatic nature of catheter-associated urinary tract infections might be based on a fine-tuned balance between the expression of bacterial virulence factors and the human immune system.

OmpR-dependent and OmpR-independent responses of Escherichia coli to sublethal attack by the neutrophil bactericidal/permeability increasing protein

Bactericidal/permeability-increasing protein (BPI) of neutrophils is a lipopolysaccharide (LPS)-binding antibacterial protein with specificity for Gram negative bacteria. BPI binding to the bacterial surface rapidly triggers potentially reversible bacterial growth inhibition and alterations of the outer membrane and, later, disruption of the inner membrane and lethal injury. Initial effects include selective OmpR-dependent changes in the synthesis of outer membrane porins (OmpF and OmpC). Because OmpR is a global transcriptional regulator, we have examined its possible role in responses of E. coli to sublethal injury caused by BPI. Early (<15 min) reversible effects of BPI on bacterial colony-forming ability and outer membrane permeability were virtually identical in isogenic wild-type (wt) and ompR- E. coli. Both strains could repair the outer membrane permeability barrier after Mg2+-induced displacement of bound BPI. However, OmpR was essential for the ability of E. coli to tolerate low doses of BPI and escape the progression of sublethal to lethal damage. Scanning electron microscopy revealed that BPI treatment produced greater membrane perturbations in the ompR- strain, apparent even before lethal injury. These findings suggest that the fate of E. coli exposed to BPI depends on both OmpR-independent mechanisms engaged in outer membrane repair and OmpR- dependent processes that modulate porin synthesis and retard progression of injury from the outer to the inner membrane.

Determinants of activation by complement of group II phospholipase A2 acting against Escherichia coli

Prompt killing of many strains of Escherichia coli during phagocytosis in vitro by isolated polymorphonuclear leukocytes (PMN) requires the presence of nonlethal doses of nonimmune serum (B. A. Mannion, J. Weiss, and P. Elsbach, J. Clin. Invest. 86:631-641, 1990). Because this requirement is bypassed in a phospholipase A (PLA)-rich mutant (pldA ) of E. coli, we have examined the effect of serum on bacteria] phospholipid (PL) degradation during phagocytosis of wild-type (pldA+) and PLA-deficient (pldA) E. coli. In parallel with increased killing, nonlethal doses of serum increased the degradation of prelabeled bacterial PL during phagocytosis by two- to fivefold, to nearly the same levels (ca. 50 to 60%) as those produced during phagocytosis of E. coli pldA in the absence of serum. The effects on the E. coli pldA mutant imply that there is a serum-mediated enhancement of granule-associated group II PMN PLA2 activity. At the same doses, serum promoted action against E. coli in the presence of purified rabbit and human group II PLA2 but did not activate bacterial PLA. Related PLA2s that lack specific structural determinants needed for optimal activity against E. coli treated with the bactericidal/permeability-increasing protein (BPI) of PMN are also less active than wild-type group II PLA2 against serum-treated E. coli. Treatment of E. coli with C7- or C9-depleted serum did not enhance bacterial killing or PL degradation during phagocytosis or the action of purified PLA2. In summary, these findings suggest that (i) nonlethal assemblies of the membrane attack complex promote intracellular killing and destruction of E. coli ingested by PMN, in part by promoting the action of granule-associated PLA2 against ingested bacteria, and (ii) structural determinants first implicated in PLA2 action against BPI-treated E. coli are also important in PLA2 action in concert with other host defense systems, such as complement.

Platelet-activating factor and phospholipase A2 in patients with septic shock and trauma

OBJECTIVE

To study blood and bronchoalveolar lavage (BAL) fluid levels of platelet activating factor (PAF-acether) and phospholipase A2 (PLA2) in patients with septic shock or following severe trauma.

DESIGN

Prospective controlled clinical study.

SETTING

An intensive care unit (ICU) of a university hospital.

PATIENTS AND PARTICIPANTS

The study comprised 12 patients, 8 with septic shock and 4 with trauma, consecutively admitted to the ICU. Healthy volunteers were used as controls.

MEASUREMENTS AND RESULTS

Blood PAF-acether and plasma PLA2 levels were measured within 24 h after the patients arrival to the ICU. The Apache II score and outcome were registered. Median values for PAF-acether and PLA2 in the septic shock patients were 10.5 x 10(-10) M and 5300 units/ml, respectively, whereas corresponding values in the trauma patients were 1.3 x 10(-10) M and 770 units/ml. Normal healthy individuals had no detectable PAF-acether in the circulating blood (< 0.5 x 10(-10) M), and normal plasma PLA2 activity was < 300 units/ml. Moreover, both PLA2 and PAF-acether levels correlated well with the severity of the disease as assessed by the Apache II scoring system (p < 0.01 for PLA2 and p < 0.05 for PAF-acether). In addition, PAF-acether and PLA2 were determined in BAL fluid of patients with septic shock (n = 5) and trauma (n = 3); increased PAF-acether levels were found in four patients with septic shock and one patient with trauma.

CONCLUSION

These results demonstrate a significant increase of both PLA2 and PAF-acether in the circulation of trauma patients, and a further increase in septic shock patients. It is possible that PAF-acether and PLA2 can be used as markers for the severity of the disease in septic shock and following severe trauma.

Bacterial phospholipid hydrolysis enhances the destruction of Escherichia coli ingested by rabbit neutrophils. Role of cellular and extracellular phospholipases

Escherichia coli ingested by PMN are promptly growth arrested but undergo limited destruction. We have studied bacterial phospholipid hydrolysis as a possible limiting factor in the disassembly of ingested E. coli, comparing the fates, during phagocytosis by rabbit peritoneal exudate PMN, of three isogenic strains, differing in their content of the pldA gene encoding the principal E. coli phospholipase A (PLA), i.e., pldA-, pldA+, pldA (the latter strain bearing the pldA gene in a multicopy plasmid resulting in a 20-fold increase in PLA content). Ingestion and growth inhibition (greater than 99% within 15 min) were the same for the three strains, but phospholipid degradation differed according to bacterial PLA content: pldA up to 60%, pldA+ up to 30%, and pldA- up to 20%. Since the pldA- strain has no activatable PLA, phospholipid degradation in this strain demonstrates the action of a PMN PLA. Added PLA2-rich ascitic fluid (AF) or purified AF PLA2 increased the rate and extent of degradation of the pldA- strain, provided the enzyme was added before ingestion was complete. 125I-AF-PLA2 binds to both E. coli and PMN and thus can enter the vacuole during phagocytosis. Although up to 50-fold more AF-PLA2 than the PLA2 content of the PMN could be loaded into the PMN in this way, degradation of pldA- E. coli did not exceed 30%. Increased phospholipid degradation had no effect on the degradation of bacterial macromolecules. In contrast, bacterial disassembly manifest as structural disorganization, release of bacterial protein derived material, and inhibition of protein synthesis were markedly enhanced when greater than 50% of prelabelled bacterial phospholipids were degraded. These findings reveal a link between envelope phospholipid degradation and overall bacterial destruction, suggesting therefore that factors limiting PLA action limit the destruction of E. coli ingested by PMN.

A novel binding assay for phospholipase A2

We have devised a rapid and simple assay for estimating the binding of pancreatic phospholipase A2 to a bilayer lipid membrane. The binding was observed to be extremely rapid at 37 degrees and was absolutely dependent upon Ca2+. Amongst several drugs known to inhibit the catalytic activity of phospholipase only mepacrine at high concentrations (500 microM) and chlorpromazine (100 microM) were active. Treatment of the enzyme with p-bromophenacylbromide did not inhibit binding. Several alcohols potentiated binding whereas detergents tended to inhibit. Amongst several purified proteins tested, only the steroid-induced anti-phospholipase protein lipocortin prevented binding. The use of this assay in screening for antiphospholipase agents is discussed.

Studies of the phospholipase A2 activity of rat ileal mucosa

A rapid and simple procedure has been used to determine phospholipase A2 activity (EC 3.1.1.4) in rat ileal mucosa. We used 14C-oleate-labeled Escherichia coli as substrate for the phospholipase activity and a 0.45-micron Millipore filter to separate the product of hydrolysis--the 14C-oleic acid--from the unhydrolyzed substrate. The phospholipase A2 activity was optimal at pH 9.8 and at 2 mM Ca2+, but another peak of activity appeared at pH 7.2. In addition, cell fractionation revealed yet another phospholipase A2 activity at pH 5.0 in the absence of Ca2+. These findings suggest the presence of more than one phospholipase A2 in the ileal mucosa and points to the possible use of a simple procedure for studying their distribution and properties.

Oxygen-dependent and oxygen-independent mechanisms of microbicidal activity of neutrophils

The essential role of the phagocyte in host defense against the enormous variety of microbial predators in our environment requires the availability of a "universal" weapon effective against most microbes, or an arsenal of different agents with specificity for different classes and species of microorganisms. Both polymorphonuclear leukocytes (PMN) and mononuclear phagocytes possess, or can produce, a wide range of antimicrobial agents providing these cells with an "overkill" capacity [1] that will usually bypass microbial defenses against a given antimicrobial device of the phagocyte. In this brief review we focus on the PMN because its antimicrobial systems have been analyzed most extensively. However, the killing mechanisms of mononuclear phagocytes and PMN are sufficiently similar to permit the insights gained from the study of the PMN to be applied to all phagocytes, including macrophages.

Nonoxidative antimicrobial reactions of leukocytes

Increasingly abundant evidence supports the hypothesis that PMNs and perhaps alveolar macrophages have antimicrobial mechanisms independent of the presences of molecular oxygen for effective action against an array of bacteria and against some fungi. Eosinophils have mechanisms toxic for schistosomula and Trichinella larvae. In all instances the antimicrobial substances isolated have been cationic proteins and, in PMNs, associated with the azurophil cytoplasmic granules of the PMNs. Several of these substances have thus far demonstrated no enzymic function. Two of these substances are serine proteases but in one, chymotrypsin-like protein, the antimicrobial action depends on the cationic properties of the protein and is independent of the proteolytic action of the substance. In most instances, these proteins are cationic due to relatively large proportions of arginine. In two instances, a large proportion of lysine is present. All have high proportions (about 50%) of hydrophobic amino acid. Such proteins occur in the PMNs of man, rabbit, guinea pig, rat, cow, and chicken. The present view is that they are most active against gram-negative bacteria. At least two of them-37-kd and 57-kd proteins (Shafer and Spitznagel, 1983)-act on S. typhimurium in a manner analogous to that of polymyxin B through binding to lipid A. Currently available results shows that anaerobic PMNs have substantial antimicrobial capacity. Whether this capacity is due to the O2-independent mechanisms discussed in this chapter remains to be established with greater certainty.

Role of charge and hydrophobic interactions in the action of the bactericidal/permeability-increasing protein of neutrophils on gram-negative bacteria

We have recently provided evidence suggesting that the action of purified cationic bactericidal/permeability-increasing protein (BPI) from neutrophils on susceptible gram-negative bacteria requires saturation binding to negatively charged surface sites (Weiss, J., S. Beckerdite-Quagliata, and P. Elsbach, 1980, J. Clin. Invest., 65: 619-628.)We now show that this charge interaction is necessary but not sufficient to produce the effects of BPI on the envelope and on viability. By altering the hydrophobic properties of the bacterial (outer) membrane, it is possible to separate saturation binding from the biological action of BPI, indicating that steps beyond surface binding are needed for the antibacterial action. Outer membrane properties were modified by (a) reducing temperature during BPI-Escherichia coli interaction; (b) growing E. coli at 42 degrees C to increase the saturated fatty acid content of membrane phospholipids; and/or (c) using smooth E. coli with a natively less fluid outer membrane. Hydrophobic interaction chromatography on phenyl-Sepharose and measurement of sensitivity to the hydrophobic antibiotic rifampicin were used to monitor the changes in hydrophobic properties of the bacterial outer membrane produced by these manipulations. Nearly all BPI can be removed from the bacterial surface by 80 mM MgCl(2) or by trypsin. At 37 degrees C, removal of BPI results in repair of the envelope alterations, but viability is irreversibly lost, even when Mg(2+) is added after only 15 s of exposure of the bacteria to BPI. However, under conditions of reduced outer membrane hydrophobicity, when saturation binding still occurs within 30 s, E. coli can be rescued by addition of Mg(2+) after up to 5-min exposure to BPI, indicating retardation of postbinding steps. We conclude that after initial binding BPI must enter into a hydrophobic interaction with the outer membrane in order to produce its antibacterial effects. These postbinding events reversibly mediate the membrane perturbations and irreversibly trigger the bactericidal action of BPI.

Resistance of gram-negative bacteria to purified bactericidal leukocyte proteins: relation to binding and bacterial lipopolysaccharide structure

The sensitivity or resistance of gram-negative bacteria to antibacterial systems appears to be related to the length of the saccharide chain of the bacterial envelope lipopolysaccharides (LPS). To explore this relationship further, we made use of two bactericidal, membrane-active cationic proteins, recently purified to near homogeneity, one from human and one from rabbit polymorphonuclear leukocytes (PMN). We have studied the effects of these two closely similar proteins on strains of Salmonella typhimurium and Escherichia coli, each separate strain differing in the saccharide chain length of its outer membrane LPS. Binding of these proteins to the bacterial outer membrane is required for killing, and is accompanied by an almost immediate increase in outer membrane permeability to normally impermeant actinomycin D. Sensitivity to the bactericidal and permeability-increasing activities of the human and rabbit proteins increases with decreasing LPS-saccharide chain length (chemotype: [S < Ra < Rb(3) < Rc < Rd(1)]). S. typhimurium G-30 and E. coli J5, mutant strains lacking UDP-galactose-4-epimerase, synthesize incomplete LPS (chemotype Rc) when grown without galactose, and are then as sensitive to both PMN proteins as the S. typhimurium strains 395 R10 (Rd(1)) and R5 (Rb(3)). However, when these mutants are grown with galactose, they synthesize complete LPS (chemotype S) and exhibit nearly the same relative insensitivity as the smooth strains S. typhimurium 395 MS and E. coli 0111:B4. The differences among strains in sensitivity to the effects of the proteins on bacterial viability and permeability correspond to differences in bacterial binding of these PMN proteins. Thus, at protein concentrations that produce maximal antibacterial activity toward the rough bacteria, but little or no activity toward the smooth strains, rough bacteria bind from 3- to 10-fold more protein (S. typhimurium 395 R10; S. typhimurium G-30, and E. coli J5 [grown without galactose]) than do the smooth bacteria (S. typhimurium 395 MS; E. coli 0111:B4; S. typhimurium G-30 and E. coli J5 [grown with galactose]). These findings suggest that bacterial sensitivity or resistance to these purified bactericidal PMN proteins is determined by the binding properties of the outer membrane, which in turn depends upon the LPS-saccharide chain length.

Effect of leukocyte hydrolases on bacteria. XIII. Role played by leukocyte extracts, lysolecithin, phospholipase a2, lysozyme, cationic proteins, and detergents in the solubilization of lipids from Staphylococcus aureus and group A streptococci: relation to bactericidal and bacteriolytic reactions in inflammatory sites

The bactericidal and bacteriolytic effects of lysolecithin (LL) and egg-white lysozyme (LYZ) on Staph. aureus and group A streptococci and the solubilization of phospholipids from the bacterial membranes by these agents was studied. Low concentrations of lysolecithin (1--10 microgrames/ml) are highly bactericidal for Steph. aureus and group A streptococci, but induce neither bacteriolysis nor solubilization of a substantial amount of membrane phospholipids. On the other hand, while LL at greater than 50 micrograms/ml causes substantial lipid release, a combination of LL and LYZ is absolutely needed to solubilize lipids from streptococci. This combination is, however, not bacteriolytic for this microrganism. The solubilization of lipids from staphylococci by LL is much faster than that induced in streptococci by LL + LYZ. The solubilization of the bulk of membrane lipids from staphylococci can also be achieved by Triton X-100 and by sodium lauryl sulfate and from group A streptococci by Triton X-100 plus LYZ. A variety of other detergents (e.g., Cetavlon, sodium taurocholate, cetyl pyrdinium chloride) have no lipid-releasing properties even in the presence of LYZ. The release of lipids by LYZ (in the presence of LL) from group A streptococci is related to its enzymatic activity, on a still unknown substrate, but not to its cationic nature as this muramidase cannot be replaced by a variety of cation substances (histone, polylysin, leukocyte cationic proteins, polymyxin B, and spermidine). The release of lipids from staphylococci by LL is not inhibited by a variety of anionic and cationic polyelectrocytes (heparin, liquoid, chondroitin sulfate, DNA histone, and polylysine) which markedly inhibit the release of lipids from group A streptococci by LL and LYZ. Streptococci that had been cultivated in the presence of subinhibitory concentrations of penicillin G lose their membrane phospholipids to a larger extent and by much smaller concentrations of LL and LYZ, as compared to controls, suggesting that the interference with the synthesis of the peptidoglycan increases the accessibility of the cell membrane to the lipid-releasing agents. The mechanism by which LL collaborates with LYZ in lipid release is still not known. The possible role of bacterial lipids and lyso compounds in the control of bacterial survival in inflammatory sites is briefly discussed.

The prerequisites for local lysolecithin formation in the human gallbladder. II. Studies on the positional specificity of the phospholipase A activity

The positional specificity of the phospholipase A in human gallbladder epithelium was studied by using biosynthetically radioabeled diacylphosphoglycerides as substrates. Diacylphosphoglyceride in 14C-palmitic acid-labeled, autoclaved E.coli was hydrolyzed under the formation of monoacylphosphoglyceride and fatty acid that were both radiolabled. In contrast, diacylphosphoglyceride in 14C-oleate-labeled bacteria was hydrolyzed so as to give radiolabel in the fatty acid only. Since 14C-palmitate occupies predominantly the 1-acyl position and 14C oleate the 2-acyl position of the major E. coli diacylphosphoglycerides, these findings suggest that: 1) the phospholipase attacks and 2-position of diacylphosphoglycerides, and 2) a complete deacylation of diacylphosphoglycerides in the gallbladder wall is brought about by the combined action of phospholipase A2 and lysophospholipase, the latter being able to hydrolyze the 1-acyllysophosphoglyceride. It appears, therefore, that the biochemical preequisites for a local formation and degreadation of lysolecithin in the gallbladder itself are met by the positional specificity of theenzymes present. This finding further substantiates the hypothesis that lysolecithin is an adjustable mediator of aseptic cholecystitis.

The activity of the pneumococcal autolytic system and the fate of the bacterium during ingestion by rabbit polymorphonuclear leukocytes

The extent to which autolytic microbial enzymes are involved in the fate of microorganisms ingested by phagocytes has not been determined. It is known, however, that activation of degradative enzymes occurs during certain microbicidal events. We examined the possible role of the pneumococcal autolytic enzyme (an N-acetylmuramyl-L-alanine amidase) in the loss of viability and degradation of pneumococci during phagocytosis by rabbit polymorphonuclear leukocytes. Three bacterial systems were compared: (a) wild type pneumococci with an active autolytic system; (b) wild type bacteria grown under conditions that block the endogenous autolytic activity and (c) a mutant strain defective in the major autolytic enzyme of this bacterium. No differences could be detected between the autolysis-positive and negative bacteria in the rate of killing and in the fate of macromolecular cell constituents during ingestion by rabbit peritoneal polymorphonuclear leukocytes.