Basal concentrations of free and esterified monohydroxylated fatty acids in human blood platelets.
Numerous approaches have been proposed for measuring these compounds, including HPLC [9-11], gas chromatography-mass spectrometry (GC-MS) [12,13], and liquid chromatography-mass spectrometry (LC-MS) [14,15]. However, none of them has been adapted to determine the basal concentrations of unesterified and esterified monohydroxylated fatty acids, but rather for their measurement in activated cells.
In the present study, unesterified and esterified monohydroxylated fatty acids have been measured in nonactivated platelets. For this purpose two anticoagulants were compared, the difference being the presence of a cocktail of complementary inhibitors, namely the strong calcium chelator EDTA, the cyclooxygenase inhibitor aspirin (ASA), and the adenylyl cyclase activator prostaglandin [E.sub.1] ([PGE.sub.1]). Platelets were then isolated from their plasma, and their lipids immediately hydrogenated to prevent further artifactual monohydroxylated fatty acid formations. Monohydroxylated fatty acids were then analyzed by GC-MS.
Materials and Methods
All chemicals were analytical grade and obtained from Sigma-Fluka-Aldrich Chemical Co. [PGE.sub.1] was from Cayman Chemical. Analytical-grade organic solvents and silica gel thin-layer chromatography (TLC) were purchased from Merck. The capillary column was from Hewlett Packard.
SYNTHESIS OF 2-RICINOLEOYL PHOSPHATIDYLCHOLINE
2-Ricinoleoyl phosphatidylcholine (2-ricinoleoyl-PC), used as an internal calibrator, was synthesized according to the method published by Isaacson et al.  and modified as follows. Both carboxyl and hydroxyl groups were protected with dimethyl-tert-butylsilyl chloride, and then a mild alkaline hydrolysis of the dimethyl-tert-butylsilyl ester according to Morton and Thompson . The resulting butyldimethylsilyl ether ricinoleic acid was converted into its anhydride as previously described by Selinger and Laidot . Soybean lysophosphatidylcholine was then acylated with this anhydride under mild conditions with 4-pyrrolidinopyridine according to Mason et al. . The 2-ricinoleoyl-PC was then deprotected as previously described  and purified by TLC by using the mixture chloroform:methanol:methylamine (60:20:5 by vol, 400 mL/L in water) and finally quantified by GC with diheptadecanoyl phosphatidylethanolamine as an internal calibrator.
Blood was taken from healthy donors onto citric acid, trisodium citrate, and dextrose as anticoagulant (ACD), with and without a mixture of [10.sup.-3] mol/L EDTA, 2 x [10.sup.-4] mol/L ASA, and [10.sup.-6] mol/L [PGE.sub.1]. All the donors gave informed consent. We called platelets isolated from blood taken onto ACD without inhibitors "unprotected platelets" and platelets isolated from blood taken on the anticoagulant containing the mixture of inhibitors "protected platelets." Blood was then centrifuged at 200g for 17 min. The supernatant (platelet-rich plasma) was acidified to pH 6.4 with 0.15 mol/L citric acid and spun down for 10 min at 900g [201. The pellets were resuspended into Tyrode HEPES buffer, pH 7.35, and platelets were counted with a thrombocoulter counter.
Lipids were extracted twice with a mixture of chloroform: ethanol (2:1 by vol) containing 2 nmol of 2-ricinoleoyl-PC and 1 nmol of ricinoleic acid used as internal calibrators, and the water phase was acidified to pH 3 before the second extraction step. Lipids were immediately hydrogenated by hydrogen bubbling for 6 min in the presence of Pt02 used as a catalyst. This step is important to avoid further polyunsaturated fatty acid oxidation that would lead to artifactual monohydroxylated fatty acid formation. The catalyst was eliminated by centrifugation and the different lipid fractions were separated by silica gel TLC with the solvent mixture hexane:diethyl ether:acetic acid (60:40:1 by vol). Monohydroxylated fatty acids and phospholipids were scraped off and extracted by diethyl ether and by the mixture chloroform:ethanol (2:1 by vol), respectively. Phospholipids were treated at 100[degrees]C with 5% KOH methanolic solution for 1 h and the free monohydroxylated fatty acids were purified by TLC as stated above. Finally, they were converted into methyl ester, trimethylsilyl ether (TMS) derivatives as previously described .
The methyl ester, hydrogenated, TMS derivatives were then injected into a gas chromatograph-mass spectrometer, a Hewlett Packard apparatus model 6890. The column was a 30 m x 0.25 mm i.d. model HP-5MS, coated with 50 mL/L phenylmethyl siloxane as stationary phase. The carrier gas was helium at a flow rate of 1 mL/min and samples were injected with a splitless injector with a head column pressure of 54.5 kPa (7.9 psi). The temperature of the injection port was 260[degrees]C and the transfer line was kept at 270[degrees]C. The temperature program was initially set at 57[degrees]C for 5 min, increased at 20[degrees]C/min until 200[degrees]C, and then increased at 4[degrees]C/min until 270[degrees]C. Electron impact (EI) ionization was done at 25 eV. Compounds were quantified with the selected ion monitoring (SIM) mode with two specific ions resulting from the breakdown of the molecule at both parts of the O-TMS group, including the O-TMS, and selected for each derivative. The ion m/z = 73, which corresponds to the TMS group, was only used as a qualifier.
Data were compared by using a Student-Fischer paired t-test.
Platelets were isolated from blood taken on classical ACD anticoagulant (unprotected platelets) and also on ACD containing a mixture of inhibitors of activation (protected platelets) as described in Materials and Methods. In the presence of these inhibitors, the recovery of platelets isolated from their plasma was significantly improved (P <0.05). The platelet count was 473 x [10.sup.9] [+ or -] 4.0 x [10.sup.9]/L in the protected group and 410 x [10.sup.9] [+ or -] 4.53 x [10.sup.9]/L in the unprotected one (mean [+ or -] SEM, n = 9).
After hydrogenation, the mass spectra of the monohydroxylated fatty acid derivatives showed two main specific fragments corresponding to the breakdown at both parts of the carbon carrying the derivatized hydroxyl group, including this group. The example of the 15-hydroxyeicosatetraenoic acid (15-HETE) derivative fragmentation is shown in Fig. 1. The peak at m/z = 73 represents the loss of the TMS group. The quantification of the methyl ester, hydrogenated, TMS derivatives was performed as described in Materials and Methods.
The total amount of free monohydroxylated fatty acids, which can be calculated by the sum of 12-OH-17:0, 12-OH-20:0, and 15-OH-20:0, was significantly lower (P <0.05) in the protected group (1.52 [+ or -] 0.84 pmol/[10.sup.9] platelets) than in the unprotected one (12.63 [+ or -] 10.52 pmol/[10.sup.9] platelets) (mean [+ or -] SEM, n = 9). The difference was also significant (P <0.05) for both the 12- and 15-OH-20:0 (Fig. 2). Although not statistically tested, the amount of 12-OH-17:0 was much higher in the unprotected group (1.74 [+ or -] 1.16 pmol/109 platelets) (mean [+ or -] SEM, n = 9) than in the protected one where it was undetectable. 13-Hydroxyoctadecanoic acid (13-HODE) and 5-HETE were also lowered in the protected group, but this decrease did not reach significance (results not shown). The same tendency can be observed for the total monohydroxylated fatty acids esterified in platelet phospholipids (23.48 [+ or -] 22.82 pmol/[10.sup.9] platelets in the protected group vs 47.74 [+ or -] 65.41 pmol/[10.sup.9] in the unprotected one, Fig. 3) (mean [+ or -] SEM, n = 9), but this does not reach significance. It is noteworthy that in protected platelets the ratio between the amount of monohydroxylated fatty acids esterified in platelet phospholipids and the amount of unesterified monohydroxylated fatty acids was about fourfold greater than that observed in unprotected platelets. This suggests that the monohydroxylated fatty acids produced in the unprotected platelets were poorly esterified in the phospholipid fraction.
[FIGURE 1 OMITTED]
In both groups, different monohydroxylated fatty acid isomers, such as 11-, 14-, and 17-OH-22:0 derived from 22-carbon polyunsaturated fatty acids, presumably from 22:on-3, were measurable. It remains that both the 12- and the 15-OH-20:0 were the major compounds detected in platelet phospholipids. In contrast, only traces of monohydroxylated fatty acids were detected in both the triglyceride and the sterol ester fractions (results not shown).
[FIGURE 2 OMITTED]
Our results indicate that platelet activation was prevented when blood was taken on anticoagulant containing a cocktail of complementary inhibitors. Indeed, the platelet recovery was significantly improved in the protected group. Such an activation occurring during venipuncture and during the platelet isolation process has already been established [211. The protecting effect is due to the different inhibitors added to the anticoagulant. EDTA chelates the calcium required for the phospholipase [A.sub.2] activation, and therefore less arachidonic acid is available for both the cyclooxygenase and lipoxygenase pathways [221. EDTA also blocks the iron-promoted peroxidation [23,24]. On the other hand, [PGE.sub.1] also inhibits the phospholipase [A.sub.2] activity via adenylyl cyclase activation and the increased cAMP concentration, which contribute to hinder platelet activation . Finally, ASA was also used to block any arachidonic acid-induced platelet activation, as it is known to inhibit cyclooxygenase  by acetylation of the enzyme . This inhibition was effective, since we were unable to detect any HHT in protected platelets. Therefore, these inhibitors minimized platelet activation during venipuncture and during their isolation and prevented the formation of microaggregates that could have been lost during the centrifugation steps.
[FIGURE 3 OMITTED]
Occasional appearance of very high concentrations of hydroxy compounds led us to suspect that autoxidation during the workup can still be a problem . The hydrogenation step introduced just after the lipid extraction reduces the double bonds and eliminates the possibility of further artifactual oxidation during the sample workup and analysis. This hydrogenation also provides a simplification of the mass spectra and increases the detection sensitivity. After hydrogenation and derivatization as methyl ester and TMS, the fragmentation pattern shows two main ions resulting from the breakdown of the molecule at both parts of the O-TMS group including the O-TMS. These fragments are specific to the hydroxyl group position on the chain and are used to measure the hydroxylated fatty acids with the SIM detection mode. The saturated hydroxylated derivatives elute from the column according to their carbon chain length and the position of the derivatized hydroxyl group. For example, the 12-hydroxy fatty acid for a given chain length has the shortest retention time (20.37 min) and the 15-hydroxy the longest (20.61 min). The El mode used for quantification of the monohydroxylated fatty acids is relatively more confident than the negative ionization mode, which is based only on a single ion at m/z = M-181. Therefore, three criteria were used to validate a proper peak recognition: (a) the chromatographic retention time of the two main ions used for the quantification must have the same retention time as that found for the commercial calibrator injected under the same conditions, (b) the ion at m/z = 73 (loss of TMS) used as a qualifier ion must be present at the same retention time, and (c) the relative abundance of these three ions must be in the same order of magnitude as that observed in the spectrum of the corresponding calibrator molecule. However, the hydrogenation step, which enhances the sensitivity and avoids artifacts as previously mentioned, does not define precisely the polyunsaturated fatty acid precursor. It remains that the 12-OH-20:0 and 15-OH-20:0 correspond mainly to 12-HETE and 15-HETE, respectively, arachidonic acid being the most abundant fatty acid with 20 carbons in blood cells. Interestingly, the amount of the presumed 12-HETE and 15-HETE in both the unesterified and esterified compartments of platelets were quite similar. Platelets produced mainly 12-HETE, suggesting that 15-HETE may be a marker of surrounding leukocyte activation occurring in whole blood  and made available to platelets during cell-cell interactions . In contrast to what we observed with such 12- and 15-hydroxy derivatives, 5-hydroxy-20:0 and 13-hydroxy-18:0 were not significantly higher in the unprotected platelets. This differs from the atherosclerotic situation where both 15-HETE and 13-HODE were enhanced in rabbit arteries .
In both protected and unprotected platelets, the total amount of unesterified monohydroxylated fatty acids was lower compared with that of monohydroxylated fatty acids esterified in phospholipids. However, assuming !400 nmol of phospholipids per [10.sup.9] platelets, we may calculate that <1 of 10 000 phospholipid acyl chains was a monohydroxylated fatty acid. Their relative abundance in membrane lipids is in good agreement with previous data from Spector et al.  who reported such an estimation. Moreover, in platelets, monohydroxylated fatty acids seem preferentially esterified in the phospholipid fraction, as only traces were detected in both the triglyceride and sterol ester fractions. In addition, the observation that monohydroxylated fatty acids produced in greater amounts in unprotected platelets were weakly esterified in phospholipids supports previous works reporting low rates of 12-HETE incorporation in cell phospholipids [7, 8].
Finally, this GC-MS procedure allowed us to measure the overall esterified and unesterified monohydroxylated fatty acids in human blood platelets at picomole concentrations. This may be a reliable and sensitive way to approach the in vivo specific oxidation of arachidonic acid as well as the result of oxidative stress.
This work was funded by INSERM and a grant from the Region Rhone-Alpes. We thank Patricia Postal for her skilled technical assistance.
Received June 23, 1997; revision accepted August 7, 1997.
[1.] Hamberg M, Samuelsson B. Prostaglandin endoperoxides. Novel transformations of arachidonic acid in human platelets. Proc Natl Acad Sci U S A 1974;71:3400-4.
[2.] Spector AA, Gordon JA, Moore SA. Hydroxyeicosatetraenoic acids (HETEs). Prog Lipid Res 1988;27:271-323.
[3.] Halliwell B. Oxidants and human disease: some new concepts. FASEB J 1987;1:358-64.
[4.] Jaurgens G, Hoff HF, Chisolm GM, Esterbauer H. Modification of human serum low density lipoprotein by oxidation-characterization and pathophysiological implications. Chem Phys Lipids 1987; 45:315-36.
[5.] Stenson WF, Parker CW. Metabolism of arachidonic acid in ionophore-stimulated neutrophils. Esterification of a hydroxylated metabolite into phospholipids. J Clin Invest 1979;64:1457-65.
[6.] Stenson WF, Nickells MW, Atkinson JP. Esterification of monohydroxy fatty acids into the lipids of a macrophage cell line. Prostaglandins 1983;26:253-64.
[7.] Legrand AB, Wang J, Sobo G, Blair IA, Brash AR, Oates JA. Incorporation of 12(S)-hydroxyeicosatetraenoic acid into phosphatidylcholine signaling pathway. Biochim Biophys Acta 1996;1301:150-60.
[8.] Vernhet L, Hichami A, Hamon L, Cochet MF, Legrand A. Incorporation of 12(S)-hydroxyeicosatetraenoic acid into phospholipids and active diacylglycerols in rat liver epithelial cells: effects on DNA synthesis. J Lipid Mediat Cell Signal 1996;13:233-48.
[9.] Croset M, Lagarde M. Stereospecific inhibition of PGHZ induced platelet aggregation by lipoxygenase products of icosaenoic acids. Biochem Biophys Res Commun 1983;112:878-83.
[10.] Brash AR, Hawkins DJ. High performance liquid chromatography for chiral analysis of eicosanoids. Methods Enzymol 1990;187: 187-95.
[11.] Borgeat P, Picard S, Vallerand P, Bourgoin S, Odeimat A, Sirois P, Poubelle PE. Automated on-line extraction and profiling of lipoxygenase products of arachidonic acid by high performance liquid chromatography. Methods Enzymol 1990;187:98-116.
[12.] Guido DM, McKenna R, Mathews WR. Quantitation of hydroperoxyeicosatetraenoic acids and hydroxy-eicosatetraenoic acids as indicators of lipid peroxidation using gas chromatography-mass spectrometry. Anal Biochem 1993;209:123-9.
[13.] Guichardant M, Lagarde M. Studies on platelet lipoxygenase specificity towards icosapolyenoic and docosapolyenoic acids. Biochim Biophys Acta 1985;836:210-4.
[14.] Ikeda M, Kusaka T. Liquid chromatography-mass spectrometry of hydroxy and nonhydroxy fatty acids as amide derivatives. J Chromatogr 1992;575:197-205.
[15.] Guichardant M, Lagarde M, Lesieur M, De Maack F. Thermospraymass spectrometric analysis of underivatized monohydroxy fatty acids: application to stimulated platelets. J Chromatogr 1988; 425:25-34.
[16.] Isaacson Y, Sherbourne CD, Gross RW, Stenson WE The synthesis and molecular dynamics of phospholipids having hydroxylated fatty acids at the sn-2 position. Chem Phys Lipids 1990;52:217-26.
[17.] Morton DR, Thompson JL. Total synthesis of 3-oxa-4,5,6-trinor3,7-inter-m-phenylene prostaglandins. 2. Conjugate addition approach. J Org Chem 1978;43:2102-6.
[18.] Selinger Z, Laidot Y. Synthesis of fatty acid anhydrides by reaction with dicyclohexylcarbodiimide. J Lipid Res 1966;7:174-95.
[19.] Mason JT, Broccoli AV, Huang C-H. A method for the synthesis of isomerically pure saturated mixed-chain phosphatidylcholines. Anal Biochem 1981;113:96-101.
[20.] Lagarde M, Bryon PA, Guichardant M, Dechavanne M. A simple and efficient method for platelet isolation from their plasma. Thromb Res 1980;17:581-8.
[21.] Samuelsson B, Granstrom E, Green K, Hamberg M, Hammarstrom S. Prostaglandins. Annu Rev Biochem 1975;44:669-95.
[22.] Rittenhouse-Simon S, Russel FA, Deykin D. Mobilization of arachidonic acid in human platelets: kinetics and calcium dependency. Biochim Biophys Acta 1977;488:370-88.
[23.] Van Kuijk FJGM, Thomas DW, Stephens RJ, Dratz EA. Gas chromatography-mass spectrometry methods for determination of phospholipid peroxides; I. Transesterification to form methyl esters. Free Radic Biol Med 1985;1:215-25.
[24.] Van Kuijk FJGM, Thomas DW, Stephens RJ, Dratz EA. Gas chromatography-mass spectrometry method for determination of phospholipid peroxides; II. Transesterification to form pentafluorobenzyl esters and detection with picogram sensitivity. Free Radic Biol Med 1985;1:387-93.
[25.] Siess W, Lapetina EG. Functional relationship between cyclic AMP-dependent protein phosphorylation and platelet inhibition. Biochem J 1990;271:815-9.
[26.] Weis HJ. Platelet physiology and abnormalities of platelet function. N Engl J Med 1975;293:531-41.
[27.] Roth GJ, Majerus PW. The of effect of aspirin on human platelets. J Clin Invest 1975;56:624-32.
[28.] Thomas DW, Van Kuijk FJGM, Dratz EA, Stephens RJ. Quantitative determination of hydroxy fatty acids as an indicator of in vivo lipid peroxidation: gas chromatography-mass spectrometry methods. Anal Biochem 1991;198:104-11.
[29.] Lagarde M, Gualde N, Rigaud M. Metabolic interactions between eicosanoids in blood and vascular cells. Biochem J 1988;257:313-20.
[30.] Wang T, Powell WS. Increased levels of monohydroxy metabolites of arachidonic acid and linoleic acid in LDL and aorta from atherosclerotic rabbits. Biochim Biophys Acta 1991;1084:129-38.
(1) Nonstandard abbreviations: HHT, 12-hydroxyheptadecatrienoic acid; GC-MS, gas chromatography mass spectrometry; ASA, aspirin; [PGE.sub.1], prostaglandin [E.sub.1]; 2-ricinoleoyl-PC, 2-ricinoleoyl phosphatidylcholine; TLC, thin-layer chromatography; ACD, citric acid, trisodium citrate, and dextrose (anticoagulant); TMS, trimethylsilyl ether; EI, electron impact; SIM, selected ion monitoring, HODE, hydroxyoctadecanoic acid; and HETE, hydroxyeicosatetraenoic acid.
MICHEL GUICHARDANT, * CHANTAL THEVENON, JEAN FRANCOIS PAGEAUX, and MICHEL LAGARDE
INSERM. U352, Biochimie et Pharmacologie INSA-Lyon, 20 Ave. Albert Einstein, 69621 Villeurbanne, France.
* Author for correspondence. Fax + 33 4 72 43 85 24; e-mail Michel.Guichardant@irisa-lyon.fr.
|Printer friendly Cite/link Email Feedback|
|Author:||Guichardant, Michel; Thevenon, Chantal; Pageaux, Jean Francois; Lagarde, Michel|
|Date:||Dec 1, 1997|
|Previous Article:||Age-specific distribution of plasma amino acid concentrations in a healthy pediatric population.|
|Next Article:||The relation between chemically measured total iron-binding capacity concentrations and immunologically measured transferring concentrations in human...|