Apol i poprotei n A 4 decreases neut rop h i I degranulation and superoxide production

Neutrophils participate in the acute phase response and are often associated with tissue injury in a number of inflammatory disorders. The acute phase response is accompanied by alterations in the metabolism of apolipoprotein A-I and high density lipoprotein (HDL). Structural considerations led to studies investigating the effect of purified HDL and apolipoprotein A-I on neutrophil degranulation and superoxide production. Apolipoprotein A-I but not HDL inhibited IgG-induced neutrophil activation by about 60% as measured by degranulation and superoxide production. This suggests that the lipidassociating amphipathic helical domains of apolipoprotein A-I mediate this effect. In support of this was finding inhibitory effects with two synthetic model lipid-associating amphipathic helix peptide analogs. Apolipoprotein A-I, containing tandem repeating amphipathic helical domains, was approximately ten times more effective than the two peptide analogs and inhibited neutrophil activation at well below physiologic concentrations. Competitive binding studies indicate that resting neutrophils have approximately 190,000 ( K d = 1.7 x binding sites per cell for apolipoprotein A-I, consistent with a ligand-receptor interaction. 811 These observations suggest that apolipoprotein A-I may play an important role in regulating neutrophil function during the inflammatory response. Blackburn, W. D., Jr., J. G. Dohlman, Y. V. Venkatachalapathi, D. J. Pillion, W. J. Koopman, J. P. Segrest, and G. M. Anantharamaiah. Apolipoprotein A-I decreases neutrophil degranulation and superoxide production. J. Lipid Res. 1991. 32: 1911-1918. Supplementary key words density lipoprotein amphipathic peptides neutrophils apolipoprotein A-I high During the acute phase response there are profound changes in the structure and metabolism of high density lipoproteins (HDL) (1-12). The biological consequences of these dramatic changes are unknown. One major alteration in the composition of circulating acute phase HDL is the incorporation of serum amyloid A and depletion of apolipoprotein (apo) A-I (12). In vitro studies have shown that incubation of HDL with serum amyloid A displaces apoA-I from the HDL particle (6, 13). The fate of the displaced apoA-I is yet to be determined. However, incubation of acute phase HDL with neutrophils results in association of apoA-I with the neutrophil membrane (12). The functional consequence of this interaction has not been previously investigated, but it raises the possibility that during the acute phase response apoA-I may be liberated from HDL and become free to interact with inflammatory cells such as neutrophils, thereby altering their function. ApoA-I contains amphipathic a-helical domains that have an affinity for lipid and therefore might interact with cell membranes (14, 15). ApoA-I is responsible for much of the structural and functional properties attributed to HDL. It consists of a single peptide chain of 243 amino acids. The carboxyl end consists of a 200-amino acid sequence of 22mer tandem repeating amphipathic helixes (15). The polar-nonpolar faces of the helical segments are aligned and the amphipathic a-helical segments are joined by prolyl residues, potentially making them more amenable to cellular interaction (15, 16). The amphipathic a-helical domains of apoA-I, like the helical domains of other apolipoproteins, are unique in their clustering of acidic amino acids at the center of the polar face and positively charged residues at the polar-nonpolar interface (16). The basic residues at the interface, when associated with phospholipid, may extend toward and nearly perpendicular to the polar face of the helix and insert the charged residues into aqueous milieu for aqueous solvation (16, 17). Similarities in the amphipathic helical domains of apoA-I and peptides, known to interact with cells (16-19), led to studies to determine whether apoA-I interacted with neutrophils and regulated their function. These studies were undertaken to test the hypothesis that free apoA-I, possibly through its lipid-associating domains, would diminish certain neutrophil biological responses. Abbreviations: HDL, high density lipoprotein; FMLP, formylmethionyl-leucyl-phenylalanine; PMA, phorbol myristate acetate; apo, apolipoprotein. 'To whom reprint requests should be addressed at: University of Alabama at Birmingham Medical Center, Birmingham, AL 35294. Journal of Lipid Research Volume 32, 1991 1911 by gest, on D ecem er 9, 2017 w w w .j.org D ow nladed fom METHODS General methods Purified human neutrophils at concentrations specified as described were subsequently incubated with human HDL, human apoA-I, or test peptides. After incubation, cells were then incubated in microtiter wells. The neutrophils were activated by the addition of FMLP or simply incubation in the wells that had been previously coated with IgG. After the cells were activated, the cell-free supernatants were assayed for the particular neutrophil product in question. HDL and apoA-I purification HDL was isolated by density gradient centrifugation according to previously described procedures (20). The density of HDL was between 1.06 and 1.21 g/ml. HDL, and HDL3 were also isolated by density gradient centrifugation with densities of between 1.06 and 1.125 and between 1.125 and 1.21 gm/ml, respectively. Differences in the HDL subclasses were confirmed by analysis of each sample using non-denaturing PAGE (data not shown). ApoA-I was purified by reversed phase HPLC (20). The synthesis and physicochemical characterization of the amphipathic peptide analogues used in these studies have been previously described (21-23). Neutrophil functional studies Blood was drawn from normal human volunteers into a heparinized syringe containing 100 units of heparin per ml of blood, and neutrophils were purified as previously described (24). Briefly, the majority of the erythrocytes were removed by sedimenting the blood with 4.5% dextran for 20 min at room temperature. The leukocyte-rich upper layer was removed, washed three times in phosphatebuffered saline (PBS), and layered over Hypaque-Ficoll in a conical centrifuge tube. After centrifugation for 30 min at 300 g, the neutrophil-rich pellet was washed three times with PBS and residual erythrocytes were removed by hypotonic lysis. Cells were resuspended in HBSS and counted on a hemocytometer. In each experiment, greater than 95 % (usually greater than 98%) of cells as determined morphologically were neutrophils. Purified normal human neutrophils were incubated for 15 min with either apoA-I or test peptides at 37°C. Two hundred microliters of the cell suspensions (1.25 x lo6 celldml) was then either added to microtiter wells (96-well flat-bottom plates, Dynatech Laboratories, Chantilly, VA) that had been precoated with human IgG (0.1-0.5 kg/well) or incubated with FMLP. Superoxide production was measured by determining spectrophotometrically the superoxide dismutase inhibitable reduction of cytochrome C (25). After a 1-h incubation at 37OC, the cell suspension was aspirated, the cells were removed by centrifugation, and the cell-free supernatants were assayed for lactoferrin by radioimmunoassay (24). ApoA-I, HDL, and test peptides did not directly alter the sensitivity of the lactoferrin RIA when added to standard curves or the superoxide colorimetric assay. ApoA-I depletion from serum ApoA-I was depleted from normal heat-inactivated human serum by passage over a Sepharose column to which a polyclonal antibody (Chemicon, El Segundo, CA) with specificity to apoA-I had been coupled. The antibody was coupled to cyanogen bromide-activated Sepharose 4B according to the methods described in the package insert (Pharmacia Fine Chemicals). Depletion of apoA-I was assured by immunodotting the treated sera. The control sera was also placed over a Sepharose column to which antiBSA had been coupled. Immunodotting confirmed that apoA-I was not removed from this serum. Preparation of liposomes from neutrophil lipids Neutrophil lipids were extracted with chloroformmethanol 2:l and then sonicated in the presence of carboxyfluorescein dye. Dye-entrapped liposomes were separated from free dye by gel filtration using Sepharose CL-4B equilibrated with Tris-HC1, pH 7.4. Peptides were incubated with liposomes at a 1:6 (w/w) ratio. The concentration of the peptide in the mixture was 50 ,ug/ml. Interaction with liposomes was measured as a function of release of dye as detected by the increase in fluorescence intensity at 525 nm with an excitation frequency of 498 nm (26). ApoA-I binding studies Human neutrophils (2 x 106/334 pl), were incubated with 13 ng (39,000 cpm) 125I-labeled apoA-I in 0.33 ml of phosphate-buffered saline, pH 7.4, containing 0.1% bovine serum albumin with or without with addition of unlabeled apoA-I. ApoA-I was labeled with 1251 using the lactoperoxidase method as previously described (27). Unlabeled peptides were added at the concentrations indicated. ApoA-I was insoluble at concentrations above M. Samples were incubated for 1 h at 4OC and the reaction was stopped by centrifugation at 12,000 g for 3 min through a layer of oil (dibutyl-phthalate-dinonylphthalate 5:l). Radioactivity bound to the neutrophil pellet was counted after aspiration of the supernatant solution and the oil layer, followed by cutting the tip of the microfuge tube to minimize contamination with 1251labeled apoA-I bound to the tube. No significant radioactivity was noted in the oil layer indicating that apoA-I was not dissolving in the oil. Each data point was run in triplicate. The displacement curve data is expressed as percent maximal binding of lZ5I-labeled apoA-I. Maximal binding ranged between 180 to 250 pg lZ5I-labeled apoA-I per 2 x 106 cells. A Scatchard analysis was performed based on a representative displacement experiment by determining the ratio of bound to free apoA-I and plotting these data versus molar apoA-I bound using the program LIGAND (28). 1912 Journal of Lipid Research Volume 32, 1991 by gest, on D ecem er 9, 2017 w w w .j.org D ow nladed fom