Occurrence of glyceryl ethers in the phosphatidylcholine fraction of surfactant from dog lungs

Biochimica et Biophysics Acta 836 (1985) 19-26

19

Elsevier

BBA 51993

Occurrence of glyceryl ethers in the phosphatidylcholine fraction of surfactant from dog lungs Raj Kumar a, Richard J. King b and Donald J. Hanahan a Departments of a Biochemistry and b Physiology, The University of Texas Health Science Center at San Antonio, San Antonio, TX 78284 (U.S.A.)

(Received February 27th, 1985) (Revised manuscript received May 20th, 1985)

Key words: Lung surfactant; Phosphatidylcholine; Glyceryl ether; Platelet activating factor; (Dog lung)

The molecular species of ether-linked lipids in the phosphatidylcholine (PC) fraction of the pulmonary surfactant obtained from the lavage fluid of dog were characterized. A combination of base-catalyzed methanolysis, phospholipase C treatment, gas-liquid chromatography, and mass spectrometry procedures were applied. The phospholipid composition of the surfactant, obtained by phosphorus assay of lipids separated by silica gel G thin-layer chromatography (TLC), was: PC (75%), phosphatidylglycerol (Ml%), phosphatidylethanohunine (7%), plus small amounts of sphingomyelin, phosphatidylinositol, and phosphatidylserine. The major components of the PC were 1,2-diacylPC (!X%), and 1-0-aikyl-ZacylPC (5%). No detectable amounts of l-O-aikyl-l’-enyl-2-acylPC or di-alkyl-1-enylPC were observed. The acyl groups present in the diacylPC were MO (5%), 160 (68%), 161 (12%), l&O (6%), l&l (7%) and l&2 (2%). The predominant alkyl ether chains located at the carbon 1 position of the 1-0-aikyl-2-acylPC were 160 (84%), l&O (5%) and 18:l (14%). At the carbon 2 position only a 160 fatty acyl residue was detected. In three out of seven animals platelet-activating ,factor-like activity, as determined by a platelet aggregation assay, was isolated by TLC. This aggregating activity was lost upon base-catalyzed methanolysis, but was restored by functional levels after acetylation.

Introduction

Ether phospholipids are found in the membranes of a variety of cell types and tissues [1,2]. Both the 1-0-alkyl as well as the 1-0-alkylenyl glycerides have been reported in many tissues, among which are included lung tissue [?I, alveolar macrophages [4], polymorphonuclear neutrophils [5], platelets [6,7], testes [8] and a urethane-induced adenoma whose cells have some morphological similarities to alveolar epithelial type II cells [9]. The lipid composition of pulmonary surfactant, a secretory product of the type II cell, has been

extensively studied and has been found to contain over 85% phospolipids [lo]. Although these phospholipids have been separated and characterized, the occurrence of glyceryl ethers in these phospholipids has not been reported. Thus, to extend these analyses of lipid composition of lung surfactant, a study of the possible presence of these lipids was undertaken. We have quantified the amount of glyceryl ethers present in the phosphatidylcholine fraction of pulmonary surfactant obtained from the lavage fluids of dogs, and have characterized its fatty alcohol composition. Our results show that 1-0-alkyl-2-acylphosphocholine comprises about 5% of the phosphatidylcholine

0005-2760/85/$03.30 0 1985 Elsevier Science Publishers B.V. (Biomedical Division)

20

fraction, and that its fatty acid composition unique compared with glyceryl ethers found aveolar macrophages or tracheal cells.

is in

Materials and Methods Materials. l-O-Hexadecyl-2-acetyl-sn-glycero3-phosphocholine (16:0, m-3 AGEPC) and l-Ooctadecyl-2-acetyl-sn-glycero-3-phosphocholine (18:0, sn-3 AGEPC) were obtained from Bachem, Burerdorf, Switzerland. 1-monooleyl ether (18:l) was purchased from Serdary Research Laboratories, London, Ontario. Phospholipase C (Bacillus cereus, specific activity 145 units per mg protein) was purchased from Calbiochem-Behring Corp., CA. All fatty acid methyl ester standards were obtained from Applied Science Laboratories, State College, PA. All other phospholipid standards were purchased from Sigma Chemical Co., St. Louis, MO. 6-p-Toluidine-Znapththalene sulfonic acid (TNS) was prepared in the laboratory as previously described [ll]. Thin-layer chromatography plates coated with silica gel G and silica gel 60, 250 PM, were products of Analtech, Newark, NJ and were prewashed with the appropriate developing solvent and then activated by heating for 30 min at 140°C immediately before use. Ail solvents were of reagent grade. 10% SP 2330 on 100/120 mesh Chromosorb WAW was obtained from Supelco, Belefonte, PA. Unless otherwise stated, all solvent mixtures for extraction and thin-layer chromatography were prepared on a volume to volume basis. Preparation of surfactant. Pulmonary surfactant was prepared from the lavage fluid of mongrel dogs by the method of King [12], using differential and density gradient centrifugation. Broncho-alveolar fluid was obtained by the instillation of 2 1 of 330 mM sucrose in 0. 1 M Tris-HCl (pH 7.4) containing 2 mM Ca2+, repeating the procedure twice. The fluid was centrifuged at 150 X G for 10 min to remove cells and debris. The supernatant was recovered and further centrifuged at 60000 X G for 10 min to pellet the insoluble material. The pellet was gently suspended in 580 mM sucrose containing 3 mM Ca2+ using a Dou.nce homogenizer with a pestle of large clearance, and was carefully layered over a cushion of 800 mM sucrose in the Tris buffer. Equal volumes of 450 mM and 330 mM sucrose containing 3 mM Ca2+ were

layered over the surfactant in the 580 mM sucrose, and the gradient was centrifuged for 90 min at 68 000 X G. The majority of the purified surfactant was recovered at the interface between the 450 and 580 mM sucrose solution. Electron microscopy indicated that a large proportion of the material had the characteristic morphology of tubular myelin. This material,when suspended at a concentration of 20 pg lipid/ml, adsorbed within seconds to an air-liquid interface, as judged by a decrease in its surface tension. Monolayer quantities were able to lower the surface tension of this interface to less than 10 dynes/cm when the film was compressed at a temperature of 24°C. Preparation of washed rabbit platelets. Washed rabbit platelets were prepared as previously described [13]. Platelets labelled with [3H]serotonin were prepared as above except that the whole blood was incubated with 1 pCi/ml of [ 3H]serotonin for 15 min at 37°C prior to centrifugation. Synthesis of di-radyl-3-acetylglycerols. l-OHexadecyl-2-palmitoyl-3-acetylglycerol was prepared essentially by the method of Eibl [14]. Commercially available (16:0, sn-3 AGEPC) was subjected to base-catalyzed methanolysis and the resulting 1ysoGEPC was acylated with palmitic anhydride in the presence of perchloric acid. The acylated product was purified by TLC. l-OHexadecyl-2-palmitoyI-sn-glycero-3-phosphocholine was then subjected to phospholipase C treatment as described below. The lipid product ws acetylated with acetic anhydride and perchloric acid [15] and the resulting 1-0-hexadecyl-2-palmitoyl-3-acetylglycerol was purified with preparative TLC using silica gel G developed in petroleum ether/diethyl ether/acetic acid (9O:lO: 1). 1,2-Diacyl-3-acetylglycerol was prepared from egg phosphatidylcholine by the action of phospholipase C and subsequent acetylation. Only a trace amount of the fatty acid moiety was observed to migrate during the acetylation procedure when checked by TLC, using two separate solvent systems: (a) hexane/ether (1:l v/v), (b) petroleum ether/diethyl ether/acetic acid (90:10:1), (v/v), and then toluene. Lipid analysis. The lipids in purified surfactant were extracted by the method of Bligh and Dyer [16]. Lipid phosphorus was determined by the method of Bartlett [17], subsequent to perchloric

21

acid digestion of the lipids. An aliquot of lipid extract (200 pg phospholipid phosphorus) in 1 ml chloroform/methanol mixture (1: 1, v/v) was applied to pre-washed silica gel G plates (250 pm, 10 x 20 cm), and developed in a chloroform, methanol and water system (65:35:6), (v/v). After spraying the reference lanes with TNS, the lipid standard spots were located under ultraviolet light. Fractions corresponding to lysophosphatidyl(R, = 0.26) choline (R F = 0.15), AGEPC sphingomyeli~ (RF = 0.32), phosphatidylcholine (R, = 0.46), and neutral lipid (R, = 0.95), were scraped and extracted twice with chloroform, methanol and water (1:2:0.8), and the lipids were recovered in the chloroform phase after addition of 1 vol. chloroform and 1 vol. water. Neutral lipids were quantified by weight, and phospholipids by phosphorus assay. All fractions corresponding to lysophosphatidylcho~ne, sphingomyelin and AGEPC were assayed for platelet activating factor-like activity using a bioassay described below. In order to separate glycolipids, the phosphatidylcholine (PC) fraction was further purified on the silica gel 60 plates using chloroform/methanol/methylamine (20%) system (60:36:10 v/v). Characterization of the phosphatidylcholine fraction. Characterization of the molecular species of

phosphatidylcholine was achieved by chemical derivatization prior to gas-liquid chromatography-mass spectrometry (GC-MS) analyses. The purified PC fraction (100 pg P) was subjected to base-catalyzed methanolysis at 23°C for 45 min [lS]. The lipids were extracted with chloroform/methanol/water as described above and the organic phase was evaporated under nitrogen, redissolved in a mixture of chloroform/ methanol (l:l, v/v) and analyzed for total phosphorus. For identification of phospholipids, the organic phase was applied on TLC, developed in chloroform/ methanol/water system (65:25:6, v/v) and visualized under UV light. The spots corresponding to the l-~-alkyl-2-(lyso)-~~-glycero3-phosphorylcholine (R, = 0.15) and the methyl esters (RF = 0.98) were recovered by scraping and were extracted by the procedure of Bligh and Dyer [16]. Characterization of the glyceryl ether moiety in the above purified sample (lysoGEPC, R, 0.15) was accomplished by subjecting it to the acetolysis procedure, using acetic anhydride and glacial acetic

acid in the ratio of 2:3 (v/v) at 150°C for at least h [IS]. The resulting l-O-alkyl-2,3diacetylglycerol was purified on TLC, using chloroform/ acetone (96:4, v/v) as developing solvent, recovered by scraping and extraction as mentioned above, and examined by GC-MS. A Varian 3700 unit, equipped with a flame ionization detector and a 6 ft x 2 mm i.d. glass column, packed with 10% SP 2330 on 100-120 chrombsorb WAW was utilized. The column was operated isothe~~ly at 210°C for the diacetate derivatives. The detector and injector temperature were maintained at 230°C. Various mixtures of AGEPC (16:0, 18:0) and monoglyceryl ether (18:l) derivatized into diacetates, were run as standards. The diacetate derivatives from the unknown glyceryl ethers were tentatively identified by comparing the retention times. The areas of the peaks corresponding to the diacetates were obtained directly by use of a programmable computer (Varian CDS 111) attached to the GLC unit. A Finnigan Mat 212 mass spectrometer with Incas 2200 data system in combination with a Varian model 3700 was also utilized. The separation of acetylated compounds was accomplished with a 30 m SE 54 fused silica column 0.25 mm i.d. using helium as the carrier gas with a linear velocity of approx. 45 cm/s. Positive ion electron impact mass spectra were obtained from full mass scans. Selected ion retrieval traces were acquired by computer evaluation of the data. 5

~oiec~l~r species of fatty acids in the alkylacy~hosphatidylchoiine. An aliquot of the phos-

phatidylcholine fraction (50 pg phosphorus) was suspended in 500 ~1 of borate buffer (100 mM, pH 8.0), and 5.0 units of phospholipase C from B. cerem in the same buffer and then 15 ml of diethyl ether were added. The mixture was vigorously stirred at room temperature for 1.5 h. The glycerides were extracted with diethyl ether, dried under a stream of nitrogen and acetylated in the presence of 50 ~1 chloroform and 50 ~1 acetic anhydride with 1 ~1 of 70% perchloric acid acting as a catalyst. Additional chloroform was added and the extract was washed three times with water to obtain neutrality. The chloroform layer was subjected to TLC on silica gel G plates using a unidirectional double solvent system of petroleum ether/diethyl ether/acetic acid (90:10:1, v/v) fol-

22

lowed by toluene [19]. The spots corresponding to l-alkyl-2-acyl-3-acetyl glycerol (R, = 0.66) and diacyl-3-acetyl (R, = 0.58) glycerol standards were removed by scraping and were extracted as described earlier. Analysis of the composition of the fatty acyl moiety present on the above derivatized glycerides was accomplished by examination of the methyl esters derived by base-catalyzed methanolysis, using gas liquid chromatography on the same column described previously, run isothermally at 150°C. Bioassays of platelet activating factor (PAF) activity. The biological activity of various fractions was assayed by the aggregation of washed rabbit platelets, and the release of serotonin [13]. Platelet aggregation was assessed in a dual channel platelet aggregometer (Payton). 500 ~1 of platelets (2.5 . 10’ cells/ml) in Tyrodes gelatine buffer at pH 7.2 with Ca” were added to siliconized cuvettes prewarmed at 37°C and stirred at 900 rpm. Aggregation was recorded as the percentage change in the light transmission following the addition of various fractions. The serotonin-release assay was carried out by mixing 200 ~1 of prewarmed washed rabbit platelets (2.5 . 108/ml) labeled with [3H]serotonin, with 4 ~1 of the various TLC fractions dissolved in pyrogen-free 0.15 M NaCl containing 2.5 mg/ml of bovine serum albumin. 60 s later, 20 ~1 of cold (O’C) 1.5 M formaldehyde was added to stop the reaction. The tubes were cooled immediately to 0°C and centrifuged at 833 X g for 10 min. The supernatants were then assayed by beta scintillation spectrometry for the presence of [ 3H]serotonin. The percentage of serotonin released was compared with that found after the addition of 4 ~1 of 10% Triton X-100 to 200 ~1 of unstimulated rabbit platelets. Results and Discussion Phospholipids of surfactant constituted 90% of its total lipid fraction and were high in PC (75%) and phosphatidylglycerol (10%). There were small amounts of phosphatidylethanolamine (5%), sphingomyelin (3%), phosphatidylinositol (2%), phosphatidic acid and phosphatidylserine (3%) combined, plus trace amounts of lysophospholipids. These results and the fatty acid methyl ester composition of PC, given in Table I, are very

TABLE

I

FATTY ACID DYLCHOLINE SURFACTANT

COMPOSITION FRACTION OF

OF DOG

PHOSPHATIPULMONARY

Phosphatidylcholine fraction was subjected to base-catalyzed methanolysis and the percentage composition of the methyl esters was calculated from the gas chromatograms obtained from a Varian 3700 gas chromatograph with a flame ionization detector. Column 10% SP-2300, temperature 15O’C; injector and detector temperatures 230°C. Results are expressed as mean percentage of six experiments. 18:l is a mixture of 18:l(c5) and 18:l(c9). Fatty acids Composition

(%)

14:O 5

16:0 68

16:l 12

18:0 6

18:l 7

18:0 2

similar to the findings reported previously [12,20]. Approx. 4-6% of the PC fraction reacted to base-catalyzed methanolysis to form a product with the same R, value as that of standard lysoGEPC, indicating O-alkyl type bonds at the .l or 2 position, Further identification and characterization of the presumed alkyl ether residues were carried out by subjecting the above lysophospholipid compound to acetolysis and analyzing the resulting diacetate derivative by GC-MS. A reconstructed ion chromatogram of the above diacetate product is shown in Fig. 1. When the mass spectra of peaks A, B, and C obtained from gas-liquid chromatography were examined, a peak at m/z 43, attributable mainly to the acetyl ion, was observed in each instance (Figs. 2-4). Furthermore, a peak resulting from the loss of an acetyl group in addition to a molecule of acetic acid [M - (43 + 60)] was also observed at m/z 297 (16:0) (Fig. 2) and m/z 325 (18:0) (Fig. 4). The chain length of the alkyl moiety was characterized by the ions that were derived from cleavage of the bond between carbons 1 and 2 of the glycerol backbone (m/z 255, 16:O; m/z 283, 18:O). For unsaturated diacetate derivative (18:l) peaks important to structural evaluation were found at [M - 60]+ m/z 366, and [M- O-alkyl] m/z 159 (Fig. 3). These results were further confirmed by comparing to the diacetate derivatives of l-monooleylglycerol and AGEPC containing 16:0 and 18:0 alkyl chain lengths.

23

Fig. 1. Reconstructed ion chromatogram of diacetate products of lysophospholipid. Separation of the compound was accomplished with a 30 m SE 54 fused silica column 0.25 mm i.d., using helium as the carrier gas with a linear velocity of approx. 45 cm/s. Column temperature was held at 160°C for 1 min followed by rapid heating to 260°C. A 30 s splitless injection was employed.

To assess whether the surfactant fraction may have been contaminated with products of secretion from macrophages, or tracheal mucosal cells, we also analyzed these tissues for their glyceryl ether compositions. The percentages of glyceryl ethers in the PC portions of the macrophages and tracheal cells were 20 and 158, respectively. The alkyl ether compositions are shown in Table II. It is evident that the percentage composition of the alkyl ethers

in the PC fraction of macrophages is quite different from that of surfactant. These results are at some variance from those of Sugiura et al. [4], who reported the glyceryl ether composition of PC fraction of rabbit macrophages as 16:0 (83%), 18:0 (6%), and 18:l (10%). This discrepancy may be attributed either to species or analytical variations. Our study on dog trachea indicated the presence of 12-14% alkyl acyl phosphorylcholine in the PC

Fig. 2. Electron eV) of peak A.

impact

mass

spectrum

(70

JB.B-

Y-60 -36s

Fig. 3. Electron eV) of peak B.

fraction, with the major ether-linked chains as 16:0 (40%), 18:0 (34%), and 18:l (26%). In order to establish unambiguously that the glyceryl ether PC was not due to contamiantion by the glyceryl ethers from macrophages or tracheal cells, the surfactant PC was converted into l-O-alkyl-2-acyl-3-acetate and 1,2-diacyl-3acetate glycerol, using phospholipase C and acetylation procedures, as described in Materials and Methods. When the alkylacylacetate fraction was subjected

impact

mass spectrum

(70

to base-catalyzed methanolysis, and the methyl esters were analyzed by GLC, only palmitic acid (16:0) was found at the 2 position of the molecule. This is quite different from the reported values of glyceryl ethers present in the phosphatidylcholine fraction of rabbit macrophages [4] and human neutrophils [7], where the majority of fatty acids were found to be unsaturated, i.e., 18:l and 20:4. These results, coupled with the analysis of the ether-linked fatty alcohol in the 1 position, show

Fig. 4. Electron eV) of peak C.

impact

mass spectrum

(70

25 TABLE II ALKYL ETHER COMPOSITION OF PHOSPHATIDYLCHOLINE FRACTION OF DOG PULMONARY SURFACTANT AND OTHER LUNG TISSUES Phosphatidylchohne fractions after methanolic base treatment were subjected to acetolysis procedure and the percent composition of the diacetyl derivatives were caicuiated from the gas chromato~ams obtained from a varian 3700 gas chromatograph with a flame ionization detector. Column 10% SP-2300, temperature 210°C injector and detector 23O’C. Additional minor components of the I-O-alkyl chain distribution included 14:0 and 15:O. n = number of animals. Results for surfactant and macrophages are expressed as means jr SD.

indicates that these glyceryl ethers are quite different from those reported in other tissues. Contamination by glyceryl ethers in macrophages or the products of tracheal cells is probably minimal, since the compositions of the glyceryl ethers from these sources are distinct from that in surfactant. PAF activity was present in the surfactant obtained from three dogs, perhaps reflecting secretion of PAF by alveolar macrophages [4,23-241 or polymorphonuclear neutrophils [5], but the exact cellular origin of this activity was not determined.

Chain length (molW) Surfactant (n = 7) Macrophages (n = 3) Trachea (n = 2)

85*1.1 68 f 2.7 40

7+2.2 IOf 5.5 34

51t1.3 19k4.2 26

that about 90% of the O-~kyl~y~~l ethers of surfactant are saturated. These data indicate that these glyceryl ethers have quite different compositions from those described in other cells and tissues. Plateiet aggreguting actiuity. Lipid fractions from the surfactants obtained from three out of seven animals, corresponding to the AGEPC (PAF) fraction on silica gel TLC plates, induced an aggregation of washed rabbit platelets. This ability to induce aggregation was completely lost upon alkaline methanolysis, but was recovered to nearly the same level after acetylation of the product. This aggregation was not inhibited by the addition of 5 pm indomethacin, indicating that it was not due to activation of the cyclooxygenase pathway [21], nor was it induced by TLC fractions migrating with the R, values of lysophosphatidylcholine or sp~ngomyelin, Furthermore, fractions with the R, values of PAF also induced a release of more than 50% serotonin from the washed rabbit platelets. This is in accordance with the earlier report of Prevost et al. [22] where PAF activity was also observed in the rat pulmonary alveolar fluid. In conclusion, the results of the present study indicate that the phosphatidylcholine fraction of surfactant from dog lungs contains approx. 5% glyceryl ethers. About 90% of the 0-alkyl ethers are saturated, containing only a 16:0 fatty acyl residue at the 2 position of the molecule, which

This investigation was supported by NIH grants HL-16725 and HL-22555. The authors are indebted to Dr. Susan T. Weintraub, Department of Pharmacology, for performing GC-MS analysis and interpretation of the spectra, to Mr. James Weems for his technical assistance, and to Pamela Lenow for typing and editing this manuscript. References 1 Horrocks, L.A. (1972) in Ether Lipids (Snyder, F., ed.), pp. 177-262, Academic Press, New York 2 Horrocks, L.A. (1982) in Phosphol~pids (Hawthorne, J.N, and Ansell, G.B., eds.), pp. 51-93, EIsevier Biomedical Press, Amsterdam 3 Pfleger, R.C. and Thomas, H.G. (1971) Arch. Intern. Med. 127, 863-872 4 Sugiura, T., Nakajima, M., Sekiguchi, N. and Nakagawa, Y. (1983) Lipids, 18, 125-129 5 Muller, H.W., G’Fiaherty, J.T. and Wykle, R.L. (1982) Lipids 17, 72-77 6 Natrajan, V., Augustin, Z.M., Schmid, H.H.O. and Graff, C. (1982) Thrombosis Res. 30,119-125 7 Muller, H.W., Purdon, D.A., Smith, B.J. and Wykle, R.L. (1983) Lipids 18, 814-819 8 Diagne, A., Fauvel, J., Retard, M., Chap, H. and Dousta Blazy, L. (1984) B&him. Biophys. Acta 793, 221-223 9 Snyder, C., Malone, B., Nettesheim, P. and Snyder, F. (1973) Can. Res. 33, 2437-2443. 10 King, R.J. (1982) J. Appl. Physiol. 53 (1) 1-8 11 Jon, M., Keenan, R.W. and Horowitz, P. (1982) J. Chromatogr. 237, 522-524 12 King, R.J. (1984) in: (Robertson, B., Van Golde, L.M.G. and Batenburg, J.J., eds.), pp. l-15; Elsevier, Amsterdam 13 Pinekard, R.N., Farr, R.S. and Ham&an, D.J. (1979) J. Immunol. 123, 1847-1857 14 Eibl, H. (1980) Chem. Phys. Lipids, 26, 239-247 15 Kumar, R., Weintraub, S.T., McManus, L.M., Pinckard, R.N. and Hanahan, D.J. (1981) J. Lipid Res. 25, 198-207 16 Bligh, E.G. and Dyer, W.J. (1959) Can. J. B&hem. Physiol. 37, 911-918

26 17 Bartlett, G.R. (1959) J. Biochem. 234, 4666468 18 Kumar, R., Weintraub, ST. and Hanahan, D.J. (1983) J. Lipid Res. 24, 930-937 19 Sugiura, T., Masuezawa, Y. and Waku, K. (1980) Lipids 15, 475-478 20 Sanders, R.L. (1982) in Lung Development: Biological and Clinical Perspectives, Vol. I. Biochemistry and Physiology. (P.M., Farrell, ed.), pp. 193-210, Academic Press, New York

21 Cazenave, J.P., Benveniste, J. and Mustard, J.F. (1979) Lab. Invest. 41, 275-285 22 Prevost. M.C., Cariven, C., Simon, M.F.. Chap, H. and Douste-Blazy, L.D. (1984) Biochem. Biophys. Res. Commun. 119, 58-63 23 Arnoux, B.. Duval, D. and Benveniste, J. (1980) Eur. J. Clin. Invest. 10, 437-441 24 Albert, D.H. and Synder, F. (1983) J. Biol. Chem. 258, 97-102