Effect of peritoneal dialysis effluent on superoxide anion production by polymorphonuclear neutrophils

Phospholipases and the Activation and Priming of Neutrophils by Peritoneal Dialysis Effluent

Objective

To investigate the role of phospholipases during the activation and priming of neutrophil nicotinamide adenine dinucleotide phosphate (NADPH) oxidase by peritoneal dialysis effluent (PDE).

Design

Examine the action of 4-hour dwell PDE upon phospholipase activation in the circulating neutrophils obtained from healthy individuals.

Results

We have previously reported that PDE stimulated superoxide release by the NADPH oxidase of human neutrophils and primed the response to the bacterial peptide, fMLP (fMetLeuPhe). To elucidate the biochemical mechanisms underlying these observations, we have examined the roles of phospholipases (PL) C, D, and A2, whose activation causes the release of a range of intracellular secondary messengers. Following fMLP stimulation, we observed a rapid activation of both PLC and PLD as well as a small but nonsignificant increase in PLA2 activity. Peritoneal dialysis effluent alone failed to stimulate either PLC or PLD, while pre-incubation with PDE had no affect upon fMLP-induced PLC and PLD activation. However, PDE caused a small but nonsignificant increase in PLA2 activity (which was comparable to that observed with fMLP) and primed the fMLP-induced response. In common with a role for PLA2 and the subsequent release of arachidonic acid (AA), we have demonstrated dose-dependent inhibition of PDE-induced superoxide release by the PLA2 inhibitor mepacrine, as well as activation and priming of the fMLP-induced superoxide generation by AA.

Conclusions

These results imply that PDE-induced NADPH-oxidase activation and priming in human neutrophils is mediated via a PLA2-dependent but PLC and PLD-independent mechanism.

Biocompatibility Studies on Peritoneal Cells

This review outlines the problems involved in assessing the biocompatibility of PD fluids. It has summarized the data available from conventional in vitro studies and highlights many of the inadequacies of this approach. In viva data are lacking both on host defense and on the clinical effect of changing conven tional PD fluids for a more “ideal” formulation. The best parameters for assessing biocompatibility need to be defined. Alternative formulation of fluids must be aimed towards (1) a system that interferes minimally with host defense, and (2) a system that maintains the integrity of the peritoneal membrane for ultrafiltration and clearance.

Cell culture studies should be designed to model the in viva situation. Ex viva studies (cells exposed within the peritoneal cavity) should be used to support in viva findings. Finally, in vitra results must be related to clinical significance, and changes in fluid composition should be followed by improvements in clinical outcome.

Methylglyoxal: a stimulus to neutrophil oxygen radical production in chronic renal failure?

BACKGROUND

Chronic renal failure is characterized by oxidant stress, resulting in part from increased reactive oxygen species production by neutrophils. Plasma concentrations of methylglyoxal are increased in uraemia. Methylglyoxal activates p38 mitogen-activated protein kinase (MAPK) in endothelial cells. Activation of p38 MAPK in neutrophils enhances reactive oxygen species production through exocytosis of intracellular storage granules. We tested the hypothesis that methylglyoxal enhances reactive oxygen species production by activating p38 MAPK in neutrophils.

METHODS

Neutrophils were exposed to methylglyoxal in vitro. Activation of p38 MAPK was determined by immunoblot analysis. Exocytosis was determined by measuring plasma membrane expression of CD35 and CD66b, specific markers of secretory vesicles and specific granules, respectively. Reactive oxygen species production was determined by measuring H(2)O(2) and O(-)(2) production.

RESULTS

Methylglyoxal activated p38 MAPK and caused dose-dependent increases in CD35 and CD66b expression, which were blocked by the methylglyoxal scavenger, aminoguanidine, or the p38 MAPK inhibitor, SB203580. Methylglyoxal caused dose-dependent increases in basal and Staphylococcus aureus-stimulated H(2)O(2) production and basal and formyl-methionyl-leucyl-phenylalanine-stimulated O(-)(2) production. Enhancement of reactive oxygen species production was blocked by aminoguanidine and SB203580.

CONCLUSIONS

Methylglyoxal enhances reactive oxygen species production in neutrophils through a process involving p38 MAPK-dependent exocytosis of intracellular storage granules. These findings, together with the observation that methylglyoxal concentrations are increased in renal failure, suggest a possible role for methylglyoxal as a uraemic toxin that contributes to the oxidant stress associated with renal failure.

A low-molecular-weight fraction of bovine colostrum and milk enhances the oxidative burst activity of polymorphonuclear leukocytes

Bovine colostrum and milk contain many immunomodulatory components. The low-molecular-weight fraction (< 10 kDa) was separated from colostrum and milk by gel filtration chromatography, and its effect on the oxidative burst of bovine polymorphonuclear leukocytes (PMNL) was investigated in vitro. The oxidative burst activity induced by Staphylococcus aureus was considerably enhanced when PMNLs were incubated with this low-molecular-weight fraction. However, phorbol 12-myristate 13-acetate did not trigger a burst after priming with this fraction. The oxidative burst activity enhanced by this fraction was reduced after heating. These results confirmed that a low-molecular-weight substance(s) of less than 10 kDa, present in bovine milk and colostrum, enhances the oxidative burst activity of PMNL.

Oxidant stress in hemodialysis patients: what are the determining factors?

Oxidant stress contributes to morbidity in hemodialysis patients. Three possible causes of oxidant stress have been suggested: the uremic state, the dialyzer membrane, and bacterial contaminants from the dialysate. Oxidant stress occurs in uremia before dialysis therapy is initiated, as evidenced by increased production of reactive oxygen species, increased levels of oxidized plasma proteins and lipids, and decreased antioxidant defenses. It has been proposed that increased production of reactive oxygen species during hemodialysis is also an important contributor to oxidant stress. Hemodialysis is associated with a transient increase in production of reactive oxygen species, particularly with cellulose membranes. In addition, surveys have shown widespread contamination of dialysate by endotoxin, which may cross membranes and prime production of reactive oxygen species by phagocytic cells. Recent studies, however, show a decrease in protein oxidation from pre- to post-dialysis and a normalization of neutrophil reactive oxygen species production. Taken together, these data suggest that uremia, per se, is the most important cause of oxidant stress in hemodialysis patients. Dialysate quality may also contribute to oxidant stress, but evidence that the dialyzer membrane plays a role is weak.

Effects of high-flux hemodialysis on oxidant stress

BACKGROUND

Neutrophil oxygen radical production is increased in end-stage renal disease (ESRD) patients and it is further enhanced during dialysis with low-flux cellulosic membranes. This increased oxygen radical production may contribute to the protein and lipid oxidation observed in ESRD patients. We tested the hypothesis that high-flux hemodialysis does not increase oxygen radical production and that it is not associated with protein oxidation.

METHODS

Neutrophil oxygen radical production was measured during dialysis with high-flux dialyzers containing polysulfone and cellulose triacetate membranes. Free sulfhydryl and carbonyl groups and advanced oxidation protein products were measured to assess plasma protein oxidation.

RESULTS

Pre-dialysis, neutrophil oxygen radical production was significantly greater than normal and increased significantly as blood passed through the dialyzer in the first 30 minutes of dialysis. Post-dialysis, however, neutrophil oxygen radical production had decreased and was not different from normal. Pre-dialysis, significant plasma protein oxidation was evident from reduced free sulfhydryl groups, increased carbonyl groups, and increased advanced oxidation protein products. Post-dialysis, plasma protein free sulfhydryl groups had increased to normal levels, while plasma protein carbonyl groups increased slightly, and advanced oxidation protein products remained unchanged.

CONCLUSIONS

The results of this study show that neutrophil oxygen radical production normalizes during high-flux dialysis, despite a transient increase early in dialysis. This decrease in oxygen radical production is associated with an improvement in some, but not all, measures of protein oxidation.

Protection by glutathione of neutrophils against the toxic effects of peritoneal dialysis fluid

The possibility of reducing the cytotoxic effect of heat-sterilized peritoneal dialysis (PD) fluid by addition of antioxidants/scavengers during incubation of titanium-adhering cells was investigated. Capillary blood from healthy donors was placed in drops on commercially available titanium pieces and incubated in a humidified chamber at 37 degrees C for 60min. After incubation the adherent polymorphonuclear leukocytes were immersed for 1-4h in PD-fluid, pH 7.4, containing 2.5% glucose with glutathione (GSH), superoxide dismutase, catalase or dithiothreitol (DTT). Luminol- or isoluminol-amplified chemiluminescence was used to measure the zymosan- and phorbol myristate acetate (PMA)-stimulated respiratory burst activity, as an indicator of the cytotoxicity of the PD-fluids. Heat sterilized PD-fluid had inhibitory effect on zymosan-induced respiratory burst and impaired both the extracellular and intracellular PMA-induced respiratory burst. Addition of GSH to the PD-fluid resulted in reduction of cytotoxical effects on the zymosan-induced and extracellular PMA-induced respiratory burst. The intracellular respiratory burst was not affected. The present results show that GSH and DTT have the ability to protect polymorphonuclear leukocytes against the cytotoxic effects of the PD-fluid by keeping the cell membrane in a reduced state.

Role of p38 in the priming of human neutrophils by peritoneal dialysis effluent

Peritoneal dialysis effluent (PDE) contains a low-molecular-weight substance that is able to prime human neutrophils for the release of arachidonic acid and superoxide anion. Conventional priming agents, such as tumor necrosis factor alpha (TNF-alpha), are known to signal via mitogen-activated protein (MAP) kinases; at least one possible substrate for MAP kinases is cytosolic phospholipase A(2) (cPLA(2)). Phosphorylation of this enzyme results in arachidonic acid release, and this fatty acid is a potent primer and activator of the human neutrophil NADPH oxidase. Because of the striking similarities between the priming of neutrophils with agents such as TNF-alpha and PDE, we have investigated the signalling pathways evoked by PDE and explored the possibility that cPLA(2) is a target for activated MAP kinases. Our results show that PDE treatment of human neutrophils results in the phosphorylation of the p38 kinase rather than the p42 and p44 kinases. Phosphorylation of p38 is transient with maximal activity being observed 1 min after exposure to PDE. We were unable to demonstrate that activation of p38 resulted in phosphorylation of cPLA(2); furthermore, translocation of this enzyme to a membrane-containing fraction was not enhanced in PDE-treated neutrophils. Taken together, these data suggest that, in a manner similar to that of TNF-alpha, PDE primes human neutrophils by the activation of the p38 kinase. However, unlike the cytokine, the activation of this protein does not result in phosphorylation or activation of cPLA(2).

Effects of r-metHuG-CSF on polymorphonuclear leucocyte kinetics and function in patients on continuous ambulatory peritoneal dialysis

End-stage renal failure (ESRF) patients undergoing continuous ambulatory peritoneal dialysis (CAPD) are immunocompromised and exhibit abnormal circulating polymorphonuclear leucocyte (PMN) function, including reduced phagocytosis and intracellular killing. Six uraemic patients on CAPD were each given 300 microg granulocyte colony stimulating factor (G-CSF) every day for 5 d and PMN function tests were performed daily. By day 5 of the study CD11b expression was significantly decreased in response to N-formylmethionylleucylphenylalanine (fMLP) and opsonized Staphylococcus epidermidis stimulation, and expression of L-selectin (CD62L) was significantly decreased in response to opsonized Staphylococcus epidermidis stimulation. Further, superoxide anion production and Fc gammaRI (CD64) expression were found to be significantly increased and Fc gammaRII (CD16) expression was lowered. Circulating white cell and PMN counts were significantly elevated in response to treatment. Administration of G-CSF did not appear to have corrected the abnormalities in phagocytosis and intracellular killing. This study suggests that G-CSF does no harm to ESRF patients and influences uraemic PMN function in a manner that is comparable to its effects on PMN in non-uraemic subjects.

Role of arachidonic acid and its metabolites in the priming of NADPH oxidase in human polymorphonuclear leukocytes by peritoneal dialysis effluent

Peritoneal dialysis effluent (PDE) contains a low-molecular-weight solute that will activate and prime the NADPH oxidase of human neutrophils via a phospholipase A2 (PLA2)-dependent mechanism. Since the products of PLA2 are known to activate and prime the oxidase we have investigated their role in the dialysis effluent-mediated activation and priming of human neutrophils. NADPH oxidase activity of PDE-primed and -unprimed neutrophils was measured by lucigenin-enhanced chemiluminescence in the presence of known inhibitors of the arachidonic acid cascade. Incubation of neutrophils with the nonselective PLA2 inhibitor quinacrine (0 to 100 microM) reduced oxidase activity in both primed and unprimed cells. Furthermore, primed cells were more sensitive to the action of quinacrine than were unprimed cells. We were unable to determine the relative roles of secretory PLA2 (sPLA2) and cytosolic PLA2 (cPLA2) since the selective sPLA2 inhibitor scalaradial (0 to 100 microM) inhibited oxidase activity in both groups of cells by similar degrees, while the specific cPLA2 inhibitor AACO-CF3 (0 to 50 microM) failed to affect activity in either group. Inhibition of platelet-activating factor (PAF), cycloxygenase, and 5-lipoxygenase-activating protein by hexanolamino-PAF (0 to 25 microM), flurbiprofen (0 to 25 microM), and MK886 (0 to 5 microM), respectively, had no effect upon oxidase activity. However, the direct inhibition of 5-lipoxygenase by caffeic acid or lipoxin A4 resulted in a similar concentration-dependent attenuation of oxidase activity in both primed and unprimed cells. Leukotriene B4 (LTB4) release from primed neutrophils was comparable to that from unprimed cells with the exception of phorbol myristate acetate-stimulated cells, which released fivefold more LTB4 than control. Taken together, these results suggest that it is arachidonic acid per se, and not its metabolites, that is important in priming of the neutrophil NADPH oxidase by dialysis effluent.

Effect of two-chambered bicarbonate lactate-buffered peritoneal dialysis fluids on peripheral blood mononuclear cell and polymorphonuclear cell function in vitro

Low pH, high osmolality, increasing glucose concentration, and glucose degradation products (GDP) formed during heat sterilization of conventional peritoneal dialysis (PD) fluids have been shown to have a detrimental effect on cells involved in peritoneal host defense. The two-chambered PD fluid bag in which glucose at pH approximately 3 is separated from a bicarbonate (25 mmol/L)-lactate (15 mmol/L) buffer during heat sterilization permits PD fluids with lower GDP to be delivered to the patient at neutral pH. To establish the possible benefit of two-chambered bag PD fluids on peripheral blood mononuclear cell (PBMC) and polymorphonuclear (PMN) cell function, we compared conventional 1.5% Dianeal (1.5%D) with 1.5% two-chambered bag bicarbonate-lactate (1.5%D-B), and conventional 4.25% Dianeal (4.25%D) with 4.25% two-chambered bag bicarbonate-lactate (4.25%D-B). Furthermore, to study the effect of the sterilization process on PBMC and PMN function, we compared filter-sterilized 4.25%D (4.25%D-F) with 4.25%D and 4.25%D-B. PBMC were harvested by Ficoll-Hypaque separation, and 2.5 x 10(6) cells in RPMI were incubated with an equal volume of the test fluids for 4 hours, pelleted, and resuspended in RPMI containing 10 ng endotoxin for a further 20 hours. Tumor necrosis factor alpha (TNF-alpha) production by endotoxin-stimulated PBMC was not significantly different (P = 0.10) between 1.5%D-B and 1.5%D, but was significantly higher (P = 0.01) with 4.25%D-B compared with 4.25%D. PBMC exposed to filter-sterilized fluid (4.25%D-F) showed significantly higher endotoxin-stimulated TNF-alpha production compared with 4.25%D (P = 0.02), but was not significantly different from 4.25%D-B (P = 0.40). PMN were harvested by Ficoll-Hypaque separation and 10 x 10(6) cells incubated with test fluids for 30 minutes. After incubation, phagocytosis (phagocytosis index) was determined by the uptake of 14C-labeled Staphylococcus aureus, oxidative burst by reduction of ferricytochrome C to ferrocytochrome C on stimulation with PMA, and enzyme release by measurement of endotoxin-stimulated bactericidal/permeability increasing protein (BPI). Bicarbonate-lactate two-chambered fluids of similar osmolality and glucose concentration conferred a significant improvement in phagocytosis (P = 0.02 for 1.5%D-B and P < 0.001 for 4.25%D-B). Oxidative burst and BPI release were significantly higher in 4.25%D-B compared with 4.25%D (P < 0.001). Filter-sterilized 4.25%D-F conferred a significant improvement in phagocytosis and oxidative burst compared with 4.25%D (P < 0.001) or 4.25%D-B (P < 0.001). Furthermore, conventional 4.25%D was associated with significantly lower BPI release compared with 4.25%D-F (P = 0.01). GDP's acetaldehyde and 5-HMF were analyzed in 4.25%D-B, 4.25%D, and 4.25%D-F. Acetaldehyde was below the lower limit (0.79 ppm) of the standard curve in 4.25%D-B and 4.25%D-F fluids but was detected (3.76 to 5.12 ppm) in all of the 4.25%D fluids. Relative levels of 5-HMF in the 4.25%D-B (0.032 to 0.041 Abs @ 284 nm) and 4.25%D (0.031 to 0.036 Abs @ 284 nm) were similar. The lowest levels (0.001 Abs @ 284 nm) were observed in the filter-sterilized 4.25%D-F. The beneficial effects of two-chambered bicarbonate lactate-buffered PD fluids on PBMC and PMN function are probably related to reduction of GDP from heat sterilization of glucose in a separate chamber at a lower pH. This improvement in biocompatibility could have a beneficial affect on peritoneal defenses.

Hydrogen peroxide generation by polymorphonuclear leukocytes exposed to peritoneal dialysis effluent

In the presence of peritoneal dialysis effluent (PDE), human polymorphonuclear leukocytes (PMN) showed reduced production of hydrogen peroxide and hypochlorous acid (H2O2 and HOCl, respectively) when at rest and when stimulated with both soluble (formylmethionyl-leucyl-phenylalanine and phorbol myristate acetate) and particulate (Staphylococcus epidermidis) agonists. This effect occurred in a concentration-dependent manner between 0 and 70%. (vol/vol) dialysis effluent. The inhibition of H2O2 and HOCl observed in resting, formy-methionylleucyphenyalanine-stimulated, and S. epidermidis-stimulated PMN was confined to a low-molecular-mass (< 10,000-Da) fraction of PDE, whereas the inhibition of the PMA response was equally dispersed throughout both low (< 10,000-Da)- and high-molecular-mass (> 10,000-Da) fractions. Human serum albumin, a major component of PDE, also inhibited H2O2 and HOCl production by PMN; however, results from cell-free systems suggested that human serum albumin was not wholly responsible for the inhibition of PMN function seen with PDE. The solute(s) responsible did not affect myloperoxidase but very rapidly scavenged H2O2 and HOCl. These data suggest that the factors capable of affecting H2O2 and HOCl production by PMN accumulate in uremia and are removed from the circulation into dialysis effluent.