Chemistry and biology of the anaphylatoxin related serum peptide system. 3. Evaluation of leucotactic activity as a property of a new peptide system with classical anaphylatoxin and cocytotaxin as components

Inability of rat anaphylatoxin to induce histamine release in rats

The role of rat anaphylatoxin in histamine release and increased vascular permeability during the first thirty minute period in zymosan-air-pouch inflammation, an experimental model of inflammation induced by zymosan in an air-pouch prepared on the back of rats, was investigated. Complement depletion by cobra venom factor did not affect the histamine release nor the increased vascular permeability in the inflammation of this type. In spite of apparent anaphylatoxin activity, zymosan activated serum (ZAS) failed to cause any significant release of histamine when infused in the air-pouch on the back. Anaphylatoxin purified from rat serum activated with zymosan in the presence of an inhibitor (epsilon-aminocaproic acid) of anaphylatoxin inactivator gave a single band in both polyacrylamide gel electrophoresis (PAGE) and SDS-PAGE. The molecular weight estimated by SDS-PAGE was approx. 7,000. The purified rat anaphylatoxin failed to induce histamine release nor increased vascular permeability even at 50 micrograms/ml, although it caused contraction of guinea pig ileum at 0.8 micrograms/ml. These results suggest that rat anaphylatoxin does not participate in histamine release and increased vascular permeability in the zymosan-air-pouch inflammation.

Alterations in the biophysical properties of the human endothelial cell plasma membrane induced by a chemotactic tripeptide: correlation with enhanced adherence of granulocytes

Human umbilical vein endothelial cells in monolayer culture were used to study the effects of the chemotactic tripeptide, N-formylmethionylleucylphenylalanine (FMLP), on structure and function of the endothelium. Endothelial cell morphology was unaffected by concentrations of 10(-8)-10(-4)M. No effect on endothelial cell proliferative capacity, as measured by the DNA content of cultures, was seen at the FMLP concentrations studied (10(-8)-10(-6)M). Using fluorescent molecular probes to investigate FMLP-induced alterations in membrane structure, it was shown using the monomer-excimer method with pyrene decanoic acid that FMLP caused a marked restructuring of the plasma membrane. This took the form of a restriction of the surface available to the lipophile reporter molecules, probably caused by a molecular reorganization of the membrane protein component. Experiments with diphenylhexatriene indicated that FMLP did not make the plasma membrane of the endothelial cell more fluid. Concomitant with these changes in the physical properties of the membrane, an FMLP-induced increase in granulocyte adherence to the endothelial cells was observed. A theoretical model is presented correlating granulocyte adherence with the lateral mobility of lipids in the endothelial cell membrane. The significance of the FMLP-induced increase in granulocyte adherence to endothelial cells for the pathogenesis of sepsis is discussed.

C5a-induced chemiluminescence of human granulocytes and its amplification by a serum factor

Purified human C5a elicits a fast chemiluminescence (CL) response from isolated human granulocytes in the presence of Lucigenin (bis-N-methylacridinium nitrate). The reaction is inhibitable to more than 90% by superoxide dismutase (SOD) - final concentration 200 micrograms/ml -, to about 60% by catalase - final concentration 10 mg/ml - and to 30% by the hydroxyl radical scavenger D-mannit - final concentration 100 mM. Therefore O2- seems to be the oxygen radical responsible for most of the CL, while OH and H2O2 are also involved. Addition of normal pool serum to the cells for 1-2 min before stimulation with C5a strongly enhances the effect in a dose and time-dependent manner. Therefore the existence of a "helper activity" in serum amplifying the C5a-induced CL of granulocytes is postulated. This "helper activity" is, however, no specific for C5a, since CL responses elicited with the chemotactic peptide f-met-phe or by phorbol-myristate-acetate (PMA) are also enhanced by preincubation with serum. In contrast, ConA-induced CL is not enhanced but decreased. Therefore, though not unique to C5a-induced CL, the "helper activity" seems not to represent a general "adjuvans" effect of serum on the granulocytes, but to be restricted to certain stimuli.

Rabbit polymorphonuclear leukocyte chemotactic factor generated in vivo by Bacteroides fragilis lipopolysaccharide. I. Isolation and physico-chemical characterization

By chromatographic separation on Sephadex gels a peptide, termed the lipopolysaccharide-induced chemotactic factor (LPS-CF), has been isolated from inflammatory exudate. The exudate was obtained from Teflon chambers implanted subcutaneously in rabbits 3 h after LPS from Bacteroides fragilis ss. fragilis had been injected. Three chemotactic peaks were eluted by fractionation of the exudate on Sephadex G-200 columns; one major peak with molecular weight of approximately 16,000 and two minor peaks with molecular weights of approximately 68,000 and 7,000. Refiltration of the major peak on G-75 showed the same elution profile as that found on G-200 columns. By addition of 8 M urea to the elution fluid only the major and the low molecular weight peaks appeared. The molecular weight of the major chemotactic peak was calculated to 16,000 on Sephadex gels, and also using SDS-polyacrylamide gel electrophoresis and equilibrium centrifugation. The chemotactic factor was quite heat-stable and was also non-dialyzable, and freezing and thawing as well as storage at 4 degrees C for several weeks did not impede its activity. This chemotactic factor is probably identical to the cytotaxic fragment split off from C5 upon interaction with LPS.

Rabbit polymorphonuclear leukocyte chemotactic factor generated in vivo by Bacteroides fragilis lipopolysaccharide. II. Antigenic and biologic properties

Preparations of the polymorphonuclear leukocyte (PMN) chemotactic factor isolated from lipopolysaccharide (LPS)-induced inflammatory exudate in rabbits were immunogenic in guinea pigs. Complete fusion of the precipitation lines produced against anti-CF by LPS-CF (molecular weigth 16,000) and material eluted on Sephadex G-200 columns with molecular weights (MW) of 68,000, 16,000 and 7,000 was found. Also, the chemotactically active material with MW of 68,000 and 7,000 eluted on G-75 columns after fractionation of the fraction of MW 16,000 from the G-200 eluate was antigenically identical to LPS-CF in double diffusion in agar. Normal rabbit serum (NRS) incubated with LPS, LPS-induced wound chamber exudate and NRS alone gave lines of precipitation against the anti-LPS-CF sera identical to that of LPS-CF. The capacity of LPS-CF to attract PMNs was significantly higher than that of LPS, and a peak in the number of PMNs in the exudate of wound chambers implanted in rabbits was found 4 h after the local injection of LPS-CF. When injected intraperitoneally in C5 deficient mice, LPS-CF stimulated a PMN migration which was only slightly below that in C5 normal mice. Antisera to LPS-CF inhibited the chemotactic activity of LPS-CF as well as that of LPS-NRS when the supernatants were tested using the Boyden's technique. Also, preincubation of PMNs with LPS-CF suppressed the migration towards a chemotactic gradient of LPS-CF molecules of these PMNs.

Binding of protein chemotactic factors to the surfaces of neutrophil leukocytes and its modification with lipid-specific bacterial toxins

The binding to neutrophil leukoyctes of human serum albumin (HSA), which is chemokinetic for leukocytes, i.e. influences their rate of locomotion, and of alkali-denatured HSA, which is chemotactic for leukocytes, i.e. influences their direction of locomotion, was studied. Native serum albumin showed low affinity binding to the neutrophil surface. Denatured serum albumin showed saturable binding with a Ka of approximately 1-(6) litres per mole to about 10(6) binding sites per cell. Another protein chemotactic factor, alpha5-casein, gave similar binding. These results exclude that chemotactic reactions to denatured proteins are mediated in a completely non-specific manner and suggest the presence on the cell of a restricted number of defined recognition sites. Binding was reduced following treatment of the cells with either of two lipid-specific bacterial toxins, perfringolysin, the theta-toxin of Clostridium perfringens, an oxygen-labile cholesterol-specific toxin, and Staphylococcus aureus Sphingomyelinase C. Both have previously been shown to reduce chemotactic reactions and both were used at doses which did not reduce cell viability. These results suggest an important, and possiblly direct, role for membrane lipid in the binding sites for chemotactic factors. Visual analysis of the behaviour of perfringolysin-treated neutrophils showed that these cells were still capable of chemotactic locomotion. The cells appeared to be less efficient than normal in detecting chemotactic gradients only when at a distance from the gradient source, a finding which is consistent with reduced binding of the chemotactic factor to the cell surface.

Distinct chemokinetic and chemotactic responses in neutrophil granulocytes

Evidence has been presented to show that chemokinesis and chemotaxis of human neutrophil granulocytes can be controlled by distinct cellular mechanisms and by different chemicals. Materials such as human serum albumin or fibrinogen had chemokinetic but no chemotactic properties. Highly purified preparations of serum-derived peptides containing classical anaphylatoxin had detectable chemotactic activity only. Chemokinetic as well as chemotactic substances were required for the expression of chemotaxis in the form of efficient directional locomotion. The roles of chemokinesis and chemotaxis in the regulation of directional locomotion have been analyzed. Further experiments showed that the chemotactic response conforms to the law of Weber-Fechner. The number of cells which had accumulated in response to a chemotactic simulus was proportional to the logarithm of the concentration of the chemical substance. Readaptation from high to low chemotactic stimulus occurred only to a limited extent.

Relation between stimulus intensity and neutrophil chemotactic response

The effect of chemotactic peptides which lack chemokinetic activity has been investigated. The neutrophil response is proportional to the logarithm of the stimulus intensity, or alternatively a power function with an exponent of 0.3. Equal responses are obtained for equal ratios between the peptide concentration in the lower compartment and the threshold concentration. The significance of Weber-Fechner's law in leucocyte chemotaxis is discussed.

The chemotaxis of selected cell types to connective tissue degradation products

Because rheumatoid inflammation is associated with the presence of large numbers of leukocytes in joint effusions, the question of whether enzymatic splitting of collagen and fibrin can lead to generation of chemotactic factors was investigated. Fibrinogen was purified from the plasma of four different species, and the homogeneity of the preparations was established by physicochemical and immunologic techniques. Fibrin was prepared and then lysed with plasmin to obtain fibrin degradation products (FDP). Similarly, purified collagenase was used to lyse collagen in vitro, and the chemotactic activity of the reaction mixtures was analyzed. The experiments presented indicate that fibrinogen, fibrin, and plasmin do not possess any intrinsic chemotactic activity. However, when fibrin was split by plasmin, FDP of human, bovine, sheep, and equine origin all proved to be strong leukotactic agents for polymorphonuclear leukocytes (PMN). Purified collagenase per se was found to be a cell type-specific chemotactic agent for PMN. Not only were collagen degradation products not chemotactic, but they also inhibited the leukotactic activity of the purified collagenase. Furthermore, this inhibition of the chemotactic activity of collagenase was independent of its enzymatic activity. The results presented suggest that there is a direct correlation between the process of fibrinolysis and the chemotactic attraction of leukocytes and between the presence of collagenase and leukotaxis. This system may serve as a model to explain the mechanisms by which cells accumulate in inflamed joints.