Acetyl-CoA:1- O -alkyl- sn -glycero-3-phosphocholine acetyltransferase (lyso-PAF AT) activity in cortical and medullary human renal tissue Tzortzis N. Nomikos 1 , Christos Iatrou 2 and Constantine A. Demopoulos 1 1 National and Kapodistrian University of Athens, Faculty of Chemistry, Panepistimioupolis, and 2 Center for Nephrology ‘G. Papadakis’, General Hospital of Nikea-Pireaus, Athens, Greece Platelet-activating factor (PAF) is one of the most potent inflammatory mediators. It is biosynthesized by either the de novo biosynthesis of glyceryl ether lipids or by remodeling of membrane phospholipids. PAF is synthesized and catabo- lized by various renal cells and tissues and exerts a wide range of biological activities on renal tissue suggesting a potential role during renal injury. The aim of this study was to identify whether cortex and medulla of human kidney contain the acetyl-CoA:1-O-alkyl-sn-glycero-3-phosphocholine acetyl- transferase (lyso-PAF AT) activity which catalyses the last step of the remodeling biosynthetic route of PAF and is activated in inflammatory conditions. Cortex and medulla were obtained from nephrectomized patients with adeno- carcinoma and the enzymatic activity was determined by a trichloroacetic acid precipitation method. Lyso-PAF AT activity was detected in both cortex and medulla and dis- tributed among the membrane subcellular fractions. No statistical differences between the specific activity of cortical and medullary lyso-PAF AT was found. Both cortical and medullary microsomal lyso-PAF ATs share similar bio- chemical properties indicating common cellular sources. Keywords: platelet-activating factor; biosynthesis; remode- ling; acetyltransferase; human kidney. Introduction 1-O-Alkyl-2-acetyl-sn-glycero-3-phosphocholine (platelet- activating factor, PAF) [1] represents a class of highly active lipid mediators. It is produced and released by various cells such as leukocytes, lymphocytes, endothelial cells, neurons, myocytes, hepatic cells and it is known to elicit a variety of biological responses participating in the pathogenesis of many inflammatory and noninflammatory diseases [2]. PAF is produced by two distinct biosynthetic pathways. The first pathway, the de novo pathway, starts with the acetylation of 1-O-alkyl-sn-glycero-3-phosphate (ALPA) by the acetyl-CoA:1-O-alkyl-sn-glycero-3-phosphate acetyl- transferase (ALPA AT) (EC 2.3.1.105) followed by the sequential action of a phosphohydrolase and a dithiothre- itol-insensitive cholinephosphotransferase. The second pathway, the remodeling pathway, involves the hydrolysis of preexisting plasma membrane phospholipids to 1-O-alkyl-sn-glycero-3-phosphocholine (lyso-PAF) which is then acetylated by the acetyl-CoA:1-O-alkyl-sn-glycero- 3-phosphocholine acetyltransferase (lyso-PAF AT; EC 2.3.1.67) leading to the formation of PAF. The de novo pathway seems to be responsible for the constitutive production of PAF at basal levels in cells, while the remodeling one is thought to be responsible for the increased production of PAF by inflammatory cells upon stimulation. The latter pathway is regulated mainly by the level and the degree of lyso-PAF AT and 85 kDa cPLA 2 activation. The latter enzyme hydrolyses plasma membrane phospholipids serving the substrates for lyso-PAF AT [3]. Lyso-PAF AT is found in the microsomal fraction of cells and tissues. It has a rather broad substrate specificity, concerning the alkyl/acyl group at the sn-1 position and the polar head group at the sn-3 position of the glycerol backbone, it is Ca 2+ -dependent and is activated by phosphorylation through the action of cAMP-dependent protein kinases, calcium-calmodulin dependent protein kinases, protein kinase C [4,5] and mitogen-activated protein kinases [6,7]. According to our knowledge only partial purification of lyso-PAF AT from rat spleens, leading to a very low yield of pure enzyme, has been reported [8,9]. The possibility that the kidney could produce PAF, even under normal physiological conditions, has been proposed in view of its activity in human urine [10,11]. PAF production in the kidney is originated either by blood-born cells (neutrophils, platelets and macrophages) or by resident glomerular (messangial and endothelial) and medullary cells [12,13]. Both biosynthetic pathways have been exhibited in renal cells of different animal species and human and several Correspondence to C. A. Demopoulos, 39 Anafis Str., Athens, GR-113 64, Greece. Fax: + 32 10 7274265, Tel.: + 32 10 7274265, E-mail: demopoulos@chem.uoa.gr Abbreviations:ALPA,1-O-alkyl-sn-glycero-3-phosphate; ALPA AT, acetyl-CoA:1-O-alkyl-sn-glycero-3-phosphate acetyltransferase; ESI, electron spray ionization; [H 3 ]PAF, 1-O-hexadecyl-2-[H 3 ]acetyl- sn-glycero-3-phosphocholine; PAF, platelet-activating factor; lyso-PAF AT, acetyl-CoA:1-O-alkyl-sn-glycero-3-phosphocholine acetyltransferase; PAF-AH, platelet-activating factor acetylhydrolase. Note: a web site is available at http://www.chem.uoa.gr/Personel/ Laboratories/FoodChem/CVS/demopoulos.htm (Received 13 February 2003, revised 18 April 2003, accepted 16 May 2003) Eur. J. Biochem. 270, 2992–3000 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03676.x of their enzymes have been characterized in both cultures and tissues [14–18]. However, as in other cases, the activation of lyso-PAF AT (to wit, of the remodeling pathway) is mainly responsible for the rapid and increased synthesis and release of PAF from the kidney, which occurred after stimulation with inflammatory mediators [19,20]. Increased levels of PAF have been found in blood, urine and kidney tissue of animals and humans with renal inflammatory diseases, such as glomerulonephritis and could be or due to the enhanced biosynthesis or decreased degradation or to a combination of increased production and decreased degradation of PAF in nephritic tissue [21,22]. The latter proposal comes from our previous studies in which we have demonstrated: (a) the existence of PAF acetylhydrolase, PAF-AH, the degradative enzyme of PAF, in human kidney tissue (cortex and medulla) [23]; (b) the diminished activity of PAF-AH in renal tissue (mainly cortex) received from patients with primary glomerulo- nephritis compared to normal ones [24]; and (c) increased levels of PAF in plasma and urine as well as increased PAF- AH activity in serum in patients with primary glomerulo- nephritis in comparison to normal volunteers [24]. As our previous works demonstrated the presence of the degradative enzyme of PAF (PAF-AH) in human kidney tissue, in this work an attempt is made to investigate the presence of the biosynthetic enzyme of PAF, lyso-PAF AT, in the same kind of tissue. Additionally, we characterized and compared the main biochemical properties of the two enzymatic activities (cortical and medullary lyso-PAF AT), utilizing a modified trichloroacteic acid precipitation method for the lyso-PAF AT assay. Our results demon- strate the existence of lyso-PAF AT activity in both cortex and medulla of human kidneys and show that cortical and medullary lyso-PAF AT share similar biochemical proper- ties indicating common cellular sources. Materials and methods Materials and instrumentation All solvents were of analytical grade and supplied by Merck (Darmstadt, Germany). HPLC solvents were from Rath- burn (Walkerburn, Peebleshire, UK). The separation of the lipid products of the assays was performed at room temperature on a HP HPLC Series 1100 liquid chromato- graphy model (Hewlett Packard, Waldbronn, German) equipped with a 100-lL Rheodyne (7725 i.d.) 1 loop valve injector, a degasser G1322A, a quat gradient pump G1311A and a HP UV spectrophotometer G1314A as a detection system. The spectrophotometer was connected to a Hewlett- Packard (Hewlett-Packard, Waldbronn, German) model HP-3395 integrator-plotter. The separation of lipids was performed on a Partisil 10 lmC 18 column (250 · 4.6 mm i.d.) from Analysentechnick (Wo ¨ ehlerstrasse, Mainz, Germany) with an C 18 (20 · 4.0 mm i.d.) precolumn cart- ridge. Chromatographic material used for TLC was silica gel H-60 (Merck). The platelet aggregation was measured in a Chrono-Log (Havertown, PA, USA) aggregometer cou- pled to a Chrono-Log recorder (Chrono-Log). Electrospray ionization (ESI) mass spectrometry experiments were per- formed on a Q-Tof (Micromass UK Ltd, Manchester, UK) orthogonal acceleration quadrupole time-of-flight mass spectrometer equipped with nano-electrospray ionization. Radioactivity was measured in a 1209 RackBeta-Flexivial a-Counter (LKB-Pharmacia, Turku, Finland). A Virsonic 50 Ultrasonic Cell Disruptor was used for the homogeni- zation of our samples (Virtis Co Gardiner, USA). All centrifugations were performed in a Sorvall RC-5B refri- gerated Superspeed centrifuge (Sigma) apart from the centrifugation at 100 000 g, which was performed in a Heraeus Christ, Omega 70 000 ultracentrifuge (Hanau, Germany). For the precipitation of the BSA 2 pellet, a Sigma 201 M microcentrifuge was used (Sigma, St. Louis, MO, USA). 1-O-Hexadecyl-2-[H 3 ]acetyl-sn-glycero-3-phosphocho- line, [H 3 ]PAF (specific activity 6 CiÆmmol )1 ) was purchased from DuPont NEN (Boston, MA, USA). [H 3 ]acetyl-CoA (specific activity 200 mCiÆmmol )1 ) was obtained from ICN (CostaMesa,CA,USA).UnlabeledPAF,lyso-PAF(1-O- hexadecyl-2-lyso-sn-glycero-3-phosphocholine) and acetyl- CoA were from Sigma Chemicals Co. 2,5-Diphenyloxazole (PPO) and 1,4-bis(5-phenyl-2-oxazolyl)benzene (POPOP) were purchased from BDH Chemicals (Dorset, UK). 4-[2-Aminoethyl]benzenesulfonyl fluoride (pefabloc) was kindly offered by A. Tselepis, University of Ioannina, Greece 3 . All other reagents were from Sigma Chemicals. Human renal tissues Human renal tissues were obtained from 20 nephrectomized patients with adenocarcinoma. Immediately after the neph- rectomy the kidneys were perfused with normal saline and a speciment of the apparently normal parenchyma was separated under stereoscopic microscopy into cortex and medulla and placed immediately in cold saline. Subse- quently, all homogenization and subcellular fractionation procedures were completed in less than 3 h. Homogenization of renal tissues and preparation of subcellular fractions Homogenization of cortical and medullary samples and preparation of subcellular fraction is carried out by a modification of the method described by Lenihan and Lee [25]. Briefly, cortical and medullary tissues were rinsed with ice-cold 0.25 M sucrose, minced and homogenized with six strokes of a motor-driven Potter-Elvehjem homogenizer in 0.25 M sucrose, 10 m M EDTA, 5 m M mercaptoethanol, 50 m M NaF, 50 m M Tris/HCl (pH 7.4) (homogenization buffer). The final concentration of the tissue in the homogenization buffer was 10% w/v. Further homogeni- zation of the tissue by sonication (4 · 20 s with intervals of 1 min) was followed. The homogenates were centrifuged at 500 g for 10 min. The pellets were discarded, a small portion of the supernatants was kept for protein and lyso-PAF AT determination and the rest of them were centrifuged at 20 000 g for 15 min. The resulting pellets (mitochondria fraction) were suspended in 1 mL of 0.25 M sucrose, 1 m M dithiothreitol, 50 m M Tris/HCl (pH 7.4) (suspension buffer) (5–10 mg of proteinÆmL )1 ) while the supernatants were centrifuged at 100 000 g for 1 h. The 100 000 g pellet (microsomal fraction) was suspended in 1 mL of suspension buffer (1–5 mg of proteinÆmL )1 ). Total Ó FEBS 2003 Lyso-PAF acetyltransferase activity in human kidney (Eur. J. Biochem. 270) 2993 membranes were obtained by centrifugation of the 500 g supernatant at 100 000 g for 1 h. All fractions were aliquoted and stored at – 30 °C. All homogenization and fractionation procedures were carried out at 0 °C. Lyso-PAF AT activity assay Unless stated otherwise, subcellular fractions of cortex and medulla, containing 10–40 lg of total protein, were incu- bated with 4 nmol of lyso-PAF and 40 nmol of [H 3 ]acetyl- CoA (100 BqÆnmol )1 )for30minat37°C in a final volume of 200 lLof50m M Tris/HCl buffer (pH 7.4) containing 0.25 mg mL )1 BSA. The final volume of the suspension buffer in all incubation mixtures was 50 lL. By the end of the incubation time, 2 lLofBSA100 mgÆmL )1 were added and the reaction was stopped by addition of 64 lLofa40% cold trichloroacetic acid solution. The reaction mixture was kept in ice for 30 min and centrifuged at 10 000 g for 2 min. The supernatant was discarded and the pellet containing the [H 3 ]PAF bound to the denaturated BSA is dissolved in the scintillation cocktail (dioxane-base) and the radioactivity was determined by liquid scintillation counting. Matching controls were run in the absence of lyso-PAF in order to subtract the radioactivity of the endogenously produced [H 3 ]PAF. Characterization of the lipid products of the lyso-PAF AT assay In order to identify the lipid products of the lyso-PAF AT assay, the enzymatic reactions were terminated by lipid extraction using the method of Bligh and Dyer [26], modified by the addition of 1 M HCl to the methanol. The lipid products were separated on TLC plates precoated with Silica Gel H using a solvent system of chloroform– methanol–ethanoic acid–water (100 : 57 : 16 : 8, by vol- ume). The distribution of the label was determined by zonal scraping and measuring the radioactivity by liquid scintil- lation spectrometry. Standards of [H 3 ]PAF and [H 3 ]acetyl- CoAwererunwiththesamples. Lipid products were further analyzed by HPLC on a reversed-phase C 18 using an isocratic elution system consis- ted of methanol 90% and water 10% (v/v) and a flow rate of 1mLÆmin )1 . Fractions of 0.5 mL were collected and their radioactivity was measured by liquid scintillation counting. The retention time of authentic [H 3 ]PAF and [H 3 ]acetyl-CoA was compared with the retention time of the radioactive peaks obtained from the separation of assay products. The above procedures were also applied for the extraction and chromatographic analysis of the radioactive products bound to the BSA precipitates after the addition of trichloroacetic acid to the reaction mixture. The radioactive product, co eluted with authentic PAF at either TLC or HPLC separation, was further characterized by bioassay using the washed rabbit platelet aggregation method as described previously [1]. We estimated the biological activity of the product by measuring its aggre- gatory activity towards washed rabbit platelets and com- paring it with the aggregatory activity of known concentrations of synthetic PAF. The biological activity of the hydrolyzed product, obtained by mild alkaline hydro- lysis as well as the biological activity of the reacetylated product obtained by acetylation of the hydrolyzed product was also estimated as described previously [27]. Finally, the percentage inhibitory activity of 0.7 m M creatine phosphate/ creatine phosphate kinase, 10 l M 4,5 indomethacin and 0.1 l M 4,5 BN 52021 towards the biologically active product was compared with their respective activity towards standard PAF of the same aggregatory activity with the product. The biologically active lipid was also analyzed by ESI mass spectrometry. Samples were dissolved in a small volume of HPLC grade methanol–water (70 : 30, v/v) 0.01 M in ammonium acetate. Electrospray samples are typically introduced into the mass analyzer at a rate of 4.0 lLÆmin )1 . The positive and negative ions, generated by charged droplet evaporation, entered the analyzer through an interface plate and a 100-mm orifice, while the declus- tering potential was maintained between 60 and 100 V to control the collisional energy of the ions entering the mass analyzer. The emitter voltage was typically maintained at 4000 V. Analytical methods Protein was determined by the method of Lowry et al.[28]. Statistical analysis Unless otherwise stated, data are expressed as mean values ± SD. Differences between groups were assessed by Mann–Whitney U-test. A P-value <0.05 was considered as significant. Results Characterization of the trichloroacetic acid method for the determination of lyso-PAF AT activity Preliminary experiments were carried out in order to determine the best experimental conditions for the quanti- tative recovery of [ 3 H]-PAF, the product of the lyso-PAF AT assay, to the BSA precipitate. [ 3 H]-PAF of known specific activity (20 000 c.p.m., 20 l M ) 6 was incubated with various concentrations of BSA, ranging from 0 to 8mgmL )1 ,in50m M Tris/HCl buffer (pH 7.4) and the radioactivity of the pellet obtained by the addition of trichloroacetic acid was measured. Practically, all [ 3 H]-PAF is obtained in the protein pellet at BSA concentrations of 0.5 mgÆmL )1 and higher while only 0.6% of [ 3 H]-acetyl- CoA was coprecipitated with [ 3 H]-PAF the rest of which remained in the supernatant. This indicates an efficient separation of the substrate, [ 3 H]-acetyl-CoA and the reaction product, [ 3 H]-PAF. Incubation of [ 3 H]-PAF with different concentrations (0–0.5 mgÆmL )1 ) of various subcel- lular fractions of human kidney, such as the 100 000 g pellet, the 100 000 g supernatant, the 20 000 g pellet and the homogenates could not alter the recovery of [ 3 H]-PAF in the BSA pellet indicating that apart from BSA no other endogenous protein could affect the precipitation of PAF after the addition of trichloroacetic acid. None of the other constituents of the lyso-PAF AT assay, such as the type of buffer solution, the type and concentration of the enzymatic preparation, the temperature and the pH had any major effect on the recovery of the [ 3 H]-PAF at the BSA pellet. 2994 T. N. Nomikos et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Characterization of the assay products The products of the lyso-PAF AT assay, with either cortex or medulla microsomes, were extracted by the Bligh-Dyer method and separated by TLC. The main radioactive product comigrated with authentic PAF at R F 0.2. When the lyso-PAF AT assay was carried out in the absence of lyso-PAF, a minimal formation of [ 3 H]-PAF was observed suggesting that either endogenous lyso-PAF or lyso-PC may be used as acetyl acceptors. A small radioactive peak at R F 0.8 was also observed in the presence or absence of lyso- PAF, which is probably due to by products of the reaction between [ 3 H]-acetyl-CoA and microsomal constituents. The same profile was obtained when the products of the assay were analyzed by HPLC (Fig. 1). The distribution of the radioactivity between the BSA pellet and the supernatant, after the addition of trichloro- acetic acid to the incubation mixture, was also determined. Incubation of cortical or medullary microsomes with [ 3 H]-acetyl-CoA in the absence of lyso-PAF or incubation of [ 3 H]-acetyl-CoA with lyso-PAF in the absence of micro- somes resulted in the precipitation of 0.8% of the radio- activity to the BSA pellet. All the precipitated radioactivity was found in the water phase of the Bligh–Dyer extraction of the pellet indicating that it was due to the small fraction of [ 3 H]-acetyl-CoA that was precipitated to the BSA pellet under the assay conditions. Incubation of microsomes with both lyso-PAF and [ 3 H]-acetyl-CoA resulted in a six- to sevenfold increase of radioactivity in the pellet, most of which was distributed to the organic phase of the Bligh–Dyer extraction of the pellet and it comigrated with authentic PAF after TLC analysis. No [ 3 H]-PAF was found in the super- natant of the incubation mixture after the addition of trichloroacetic acid indicating that all [ 3 H]-PAF produced in the lyso-PAF AT assay is precipitated in the BSA pellet. The lipid product, comigrated with PAF, and could activate the aggregation of washed rabbit platelets. The biological activity of the product was proportional to its radioactivity indicating that the radioactive product and the biologically active compound are the same molecule. Moreover, inhibitors of platelet aggregation such as BN-52021, indomethacin and the enzymatic system of creatinine phosphate and creatinine phosphate kinase exerted the same inhibitory effect on both standard PAF and the lipid product of similar aggregatory activity. Mild alkaline hydrolysis of the product resulted to a complete loss of its biological activity, which is re-obtained by reacetyla- tion of the hydrolyzed product. The positive ESI spectra of the assay product showed [M + Na] + at m/z 546. The ions at m/z 184 and 147 cor- responding to fragments of the phosphocholine moiety were also present (Fig. 2). All the above findings show that the only product of the lyso-PAF AT assay is indeed C 16 -PAF. Effect of BSA on the microsomal lyso-PAF AT activity of human cortex and medulla As shown in Fig. 3, a plateau of maximum activity, for both cortical and medullary lyso-PAF ATs, was observed for BSA concentrations ranging from 0 mgÆmL )1 to 0.5 mgÆmL )1 . BSA concentrations above 0.5 mgÆmL )1 have a slight inhibitory effect on both acetyltransferases activities. We routinely used 0.25 mgÆmL )1 BSA for the determina- tions of lyso-PAF AT activity. However, this concentration is insufficient for the quantitative precipitation of [ 3 H]-PAF so an excess of BSA (final concentration 2 mg mL )1 )was added at the end of the incubation time in order to achieve the maximum recovery of PAF at the BSA precipitate. Effect of Tris concentration and pH on the microsomal lyso-PAF AT activity of human cortex and medulla We studied the dependence of the enzyme activity with pH at 20, 50, 100 m M 7 of Tris-buffered solution. The pH–activity profile was bell-shaped for both enzymes and the optimum pH was in the range of 7.4–7.7. Lyso-PAF AT activity was also dependent on Tris concentration showing maximum activity at 50 m M 8 of Tris. Subsequent experiments were carried out with 50 m M 9 Tris/HCl (pH 7.4). Kinetics of PAF formation The kinetics of PAF synthesis in relation to time and microsomal protein are shown in Fig. 4. Linearity for up to 20 min of incubation and 0.2 mgÆmL )1 of cortex micro- somalproteinandforupto20minand0.12mgÆmL )1 of medulla microsomal protein was observed. Apart from the experiments concerning the influence of substrate concen- trations on lyso-PAF AT activity, we routinely used 0.1– 0.2 mgÆmL )1 of microsomal protein for the lyso-PAF AT assays and incubated the reaction mixture for 30 min in order to achieve the maximum yield of the reaction product. Dependence of lyso-PAF AT activity on divalent cations Cortical and medullary microsomes were incubated with various concentrations of exogenous added CaCl 2 and MgCl 2 in the presence or absence of EDTA. Results are shown in Table 1. Low CaCl 2 and MgCl 2 concentrations, upto10 )5 M did not have any significant effect on lyso-PAF AT activity while higher concentration inhibited lyso-PAF Fig. 1. Separation of the lipid products produced by the incubation of cortical microsomes (0.192 mg proteinÆmL )1 )with[H 3 ]acetyl-CoA in the absence of lyso-PAF (m) or in the presence of lyso-PAF (e). The retention time of authentic [H 3 ]PAF (d)and[H 3 ]lyso-PAF (s)was compared with the retention time of the radioactive peaks obtained from the separation of the assay products. The extraction and separ- ation of the products was carried out as described in materials and methods. Ó FEBS 2003 Lyso-PAF acetyltransferase activity in human kidney (Eur. J. Biochem. 270) 2995 AT activity in a dose-dependent manner. Mg 2+ was a more potent inhibitor of both cortical and medullary lyso-PAF AT than Ca 2+ . Both divalent cations showed an increased inhibitory activity against medulla lyso-PAF AT. The chelating agent, EDTA (1 m M ) inhibited the cortex and medulla acetyltransferase activity by 80 and 90%, respect- ively, in the absence of divalent cations. Addition of 10 )3 M CaCl 2 totally reversed the inhibition caused by EDTA. This reversion was more significant for the cortical lyso-PAF AT activity than the medullary one. Influence of various compounds on lyso-PAF AT activity The influence of various compounds on the activity of both cortical and medullary lyso-PAF AT activity was tested Fig. 2. Positive ion electrospray mass spectrum of the lyso-PAF AT assay product. The isola- tion of the assay product and the electrospray analysis was performed as described in Mate- rials and methods. Fig. 3. Effect of BSA concentration on microsomal lyso-PAF AT. Microsomal fractions of human cortex (0.14 mgÆmL )1 )(s)orhuman medulla (0.12 mgÆmL )1 )(d) were incubated with various concentra- tions of BSA and the lyso-PAF AT activity was determined as described in the materials and methods section. Results are expressed in per- centage related to control (incubation inthe absence of BSA) and are the averages of two experiments with different enzyme preparations. Fig. 4. Time course of [H 3 ]PAF formation as a function of cortical and medullary microsomal protein concentration. (A)Kineticsof[H 3 ]PAF formation using 0.04 (d), 0.1 (s)and0.2(h)mgproteinÆmL )1 of human cortex microsomes. (B) Kinetics of [H 3 ]PAF formation using 0.02 (d), 0.06 (s) and 0.12 (h)mgproteinÆmL )1 of human medulla microsomes. Lyso-PAF AT activity was determined as described in Materials and methods. Results are representative of three experiments. 2996 T. N. Nomikos et al. (Eur. J. Biochem. 270) Ó FEBS 2003 (Table 2). Neither NaF, a phosphatase inhibitor, nor phenylmethylsulfonyl fluoride, a serine esterase inhibitor, had any significant effect on both enzymatic activities. The same results were obtained after the incubation of cortical and medullary microsomes with the reducing agents dithiothreitol and mercaptoethanol. On the other hand, lyso-PAF AT activity was completely abolished by the incubation of microsomes with 5,5¢-dithio-bis(2-nitroben- zoic acid) (DTNB), a potent inhibitor of -SH enzymes. Pefabloc, a very potent, irreversible inhibitor of the PAF degradative enzyme, PAF-AH, had no significant effect on lyso-PAF AT activities, indicating that the possible presence of PAF-AH in the lyso-PAF AT assay does not influence the product formation. Substrates of lyso-PAF AT When the activity of lyso-PAF ATs was determined at lyso- PAF concentrations ranging from 2 to 100 l M at a fixed concentration of acetyl-CoA (200 l M ) both cortical and medullary lyso-PAF AT exhibited classical Michaelis– Menten kinetics up to 60 l M . Higher concentrations of lyso-PAF resulted in a drop of the activity possibly due to the detergent effects of the substrate against the enzyme. Simple saturation kinetics were also observed up to 400 l M of acetyl- CoA when the activity of the lyso-PAF ATs was determined at acetyl-CoA concentrations ranging from 25 to 800 l M at a fixed concentration of lyso-PAF (20 l M ). When the concen- tration of acetyl-CoA exceeded 400 l M a reduction of the lyso-PAF AT activity was also observed (Fig. 5). The kinetic parameters derived from these experiments are summarized in Table 3. In all experiments, the K M,app and V max,app values for medullary lyso-PAF AT were higher than the respective values of cortical lyso-PAF AT. However, a statistical analysis for the comparison of the values could not be conducted because of the small number of samples. Subsequently, the specificity of the microsomal enzymes for ester/ether substrates was investigated. Microsomes of Table 1. Effect of divalent cations and EDTA on cortical and medullary lyso-PAF AT activity. Lyso-PAF AT was assayed in the presence of varying concentration of CaCl 2 ,MgCl 2 and EDTA as described in Materials and methods. Results are expressed in percent related to non added control (100%) are the averages of two determinations in different enzyme preparations. Addition Cortex microsomes Medulla microsomes CaCl 2 10 )6 M 100.0 96.9 CaCl 2 10 )5 M 109.2 88.6 CaCl 2 10 )4 M 89.9 96.4 CaCl 2 10 )3 M 94.5 61.5 CaCl 2 10 )2 M 80.3 59.8 EDTA 1 m M 24.8 10.8 EDTA 10 m M nd a nd EDTA 1 m M + CaCl 2 10 )3 M 138.4 93.3 EDTA 1 m M + CaCl 2 10 )2 M 101.5 67.2 MgCl 2 10 )6 M 83.9 77.3 MgCl 2 10 )5 M 110.9 96.8 MgCl 2 10 )4 M 107.8 88.2 MgCl 2 10 )3 M 82.2 80.5 MgCl 2 10 )2 M 79.2 39.2 a nd, no lyso-PAF AT activity was detected. Table 2. Effect of various chemicals on cortical and medullary lyso-PAF AT activity. Lyso-PAF AT was assayed in the presence of varying concentration of chemicals as described in Materials and methods. Results expressed in percent related to non-added control (100%) are the average of two determinations in different enzyme preparations. Addition Cortex microsomes Medulla microsomes NaF (5 m M ) 95.9 104.2 NaF (50 m M ) 95.5 122.0 Phenylmethylsulfonyl fluoride (1 m M ) 100.2 ND b Phenylmethylsulfonyl fluoride (5 m M ) 113.5 65.6 Dithiothreitol (1 m M ) 120.3 128.3 Dithiothreitol (5 m M ) 97.2 104.5 Mercaptoethanol (1 m M ) 90.6 117 Mercaptoethanol (5 m M ) 76.8 82.7 Pefabloc (0.1 m M ) 90.1 134.2 DTNB (0.1 m M ) 16.4 2.3 DTNB (0.5 m M ) 5.4 nd DTNB (1 m M )nd a nd a nd, not detected. b ND, not determined. Ó FEBS 2003 Lyso-PAF acetyltransferase activity in human kidney (Eur. J. Biochem. 270) 2997 either cortex or medulla were incubated with [H 3 ]-acetyl- CoA and lyso-PAF or the ester analog lyso-phosphatidyl- choline (lyso-PC) and the kinetics of the product formation was determined. The velocities of both acetyltransferases were twice as high in the presence of lyso-PAF (10–20 l M ) as in that of their ester analogs. In fact, no radioactive product was recovered in the BSA precipitate when microsomes were incubated with lyso-phosphatidylcholine concentrations exceeding 20 l M . In order to rule out the possibility of the degradation of lyso-phosphatidylcholine by non specific PLA 1 , the experiment was repeated in the presence of phenylmethylsulfonyl fluoride (1 m M and 5m M ), a well-known PLA 1 inhibitor. No increase of the radioactivity recovered in the BSA pellet was observed suggesting a low specificity of acetyltransferases for ester analogs. The ability of lyso-PAF ATs to acetylate ALPA, which is substratefortheacetyltransferaseofthede novo biosyn- thetic route, was also tested. Cortex or medulla microsomes were incubated with either lyso-PAF (20 l M ), ALPA (20 l M ) or with a mixture of lyso-PAF (20 l M )andALPA (20 l M ) and the products of the reactions were extracted by a modified Bligh–Dyer extraction and separated by TLC. The radioactivity of the areas with the same R f as PAF, AAPA and alkylacetylglycerol, was determined. The results showed that PAF was the only product of the incubation of the microsomes with lyso-PAF. No lipidic products were found when ALPA served as the substrate in the lyso-PAF ATs assay. Moreover, addition of ALPA to the lyso-PAF ATs assay had no significant effect on the production of PAF indicating that ALPA cannot serve as substrate for the lyso-PAF AT of cortex or medulla. Subcellular localization of lyso-PAF ATs activity The specific activity of lyso-PAF AT was determined in all subcellular fraction obtained by the subcellular fraction- ation of the tissues. Microsomes exhibited the higher lyso- PAF AT activity. The total activity of the enzyme in all subcellular fractions was also determined and the subcellu- lar distribution of the activity was calculated considering the total activity of the 500 g supernatant as 100% (Table 4). The acetyltransferase activity was distributed between the membrane fractions of the 20 000 g and the 100 000 g pellets (microsomes). The recoveries of lyso-PAF ATs activity for cortex and medulla were 27.8 ± 8.5% (n ¼ 5) and 28.4 ± 8.1% (n ¼ 5), respectively. No statistical difference between the cortical and medullary lyso-PAF AT specific activities of the same fractions was observed. Discussion This study demonstrates the existence of an acetyltrans- ferase activity, capable of transferring the acetyl moiety of acetyl-CoA to lyso-PAF, in both cortical and medullary human renal tissue. As far as we know this is the first study concerning with the biochemical characterization of lyso- PAF AT in human renal tissues. The determination of the acetyltransferase activity was carried out by a modified trichloroacetic acid precipitation method described previ- ously [29]. Assessment of the method indicates that it can be used for the routine determination of lyso-PAF AT activity Fig. 5. Influence of substrate concentration on microsomal lyso-PAF AT activity of human cortex and medulla. (A)Activityofcortical(s) and medullary (d) lyso-PAF AT as a function of lyso-PAF at fixed concentrations (200 l M )ofacetyl-CoA.(B)Activityofcortical(s)and medullary (d) lyso-PAF AT as a function of acetyl-CoA at fixed concentrations (20 l M ) of lyso-PAF. Lyso-PAF AT activity was determined as described in Materials and methods. Results are rep- resentative of four experiments. Table 3. Kinetic parameters of cortical and medullary microsomal lyso-PAF AT. Kinetic data were obtained from experiments shown in Fig. 3. Results are the averages of two determinations or the mean ± SD of four determinations in different enzyme preparations. Enzyme preparation Substrates Lyso-PAF Acetyl-CoA K M,app (l M ) V max (nmolÆmin )1 Æmg )1 ) K M,app (l M ) V max (nmolÆmin )1 Æmg )1 ) Cortex, microsomes 15.6 ± 2.6 (n ¼ 4) 2.1 ± 0.5 (n ¼ 4) 90.4 ± 14.7 (n ¼ 4) 1.49 ± 0.25 (n ¼ 4) Medulla, microsomes 24.9 (n ¼ 2) 3.72 (n ¼ 2) 100.6 (n ¼ 2) 2.92 (n ¼ 2) 2998 T. N. Nomikos et al. (Eur. J. Biochem. 270) Ó FEBS 2003 as [H 3 ]PAF, the only radioactive product of the acetyl- transferase reaction, is quantitatively precipitated to the BSA pellet, readily separating from [H 3 ]acetyl-CoA, which remains in the supernatant. This method is faster and more convenient than the TLC methods routinely used for the determination of lyso-PAF AT activity [30]. Both cortical and medullary lyso-PAF AT activities share similar biochemical characteristics indicating that they are originated from common cellular sources. As previous studies have shown that mesangial and endothelial cells possess the higher acetyltransferase activities in rat and human renal tissue [13,15], we can hypothesize that the lyso-PAF AT activity found, in this study, in homogenates of renal cortex and medulla is mainly derived by this kind of cells. The lyso-PAF AT activity is associated with the mem- branous fractions of the renal cells. No lyso-PAF AT activity could be detected in the cytoplasmic fraction. The higher specific activity of lyso-PAF AT is found in the microsomes. Microsomal lyso-PAF AT showed an opti- mum pH in the range of 7.4–7.7, similar to that found for the acetyltransferases of other tissues [5]. Lyso-PAF AT is also sensitive to the concentration of Tris in the buffer solution showing a maximum activity at 50 m M .This indicates that it is important to determine the concentration of Tris solution in order to achieve the best experimental conditions for the determination of acetyltransferase. BSA concentrations up to 0.5 mgÆmL )1 had no signifi- cant effect on lyso-PAF AT activity while higher concen- trations inhibited lyso-PAF activity dose-dependently. The above results are in conflict with previous studies in human polymorphonuclear neutrophils showing an activation of microsomal lyso-PAF AT at BSA concentrations of 0.5 mgÆmL )1 and higher. The activating effect of BSA was attributed to its ability to bind PAF and free the enzyme from the product of the reaction, which has inhibitory effects on lyso-PAF AT. The same researchers had also shown that microsomal fraction is much more effective than BSA in binding PAF or lyso-PAF [31]. In our assays, it seems that the microsomal proteins can bind PAF effect- ively preventing it from inhibiting lyso-PAF AT activity and the addition of BSA up to 0.5 mgÆmL )1 hasnoeffectsonthe lyso-PAF AT activity. However, higher BSA concentrations could prevent the enzyme to act on the substrate and an inhibitory action is observed. Addition of divalent cations in the incubation mixture resulted in a dose-dependent inhibition of lyso-PAF AT. A complete loss of lyso-PAF AT activity was also observed in the presence of chelating agents which was totally reversed by the addition of CaCl 2 , thus a direct inhibitory effect of EDTA on the enzyme should be ruled out. It seems that the endogenous microsomal stores of Ca 2+ are adequate for the proper function of the enzyme, which is inhibited by the chelation of the endogenous Ca 2+ by EDTA. The same mode of action has been observed for the lyso-PAF AT from rat spleen [9] and mouse macrophages [32]. The substrate specificity of lyso-PAF determines the composition and the biological activity of the molecular mixture of PAF analogs that are produced under inflam- matory conditions. Lyso-PAF AT specifically acts on ether analogs of PAF while its activity on ester analogs (acyl- PAF) is diminished by almost 70%. These results are in agreement with previous studies showing that the major PAF species synthesized by rat glomerular mesangial cells is the ether analog of PAF [33]. The inability of lyso-PAF AT to act on ALPA indicates that the lyso-PAF AT activity is distinct from the acetylating activity of the de novo biosynthetic route. In conclusion, an acetylating activity, capable of trans- ferring an acetyl group from acetyl-CoA to lyso-PAF, was demonstrated in cortical and medullary human renal tissues for the first time. The biochemical properties of both cortical and medullary acetylating activities are similar with the biochemical properties of lyso-PAF ATs characterized in other tissues or cells. The existence of a lyso-PAF AT activity in human renal tissues indicates that human kidneys are able to produce PAF through the remodeling pathway. However, the relative contribution of this pathway to the increased synthesis of PAF under pathological conditions needs further elucidation. Acknowledgements This work was supported by grants from the General Secreteriat for Research and Technology, Ministry of Development. References 1. Demopoulos, C.A. & Pinckard, R.N. & Hanahan, D.J. (1979) Platelet-activating factor: evidence for 1-O-alkyl-2-acetyl-sn- Table 4. Subcellular distribution and specific activities of lyso-PAF AT in human cortex and medulla. Lyso-PAF AT activity was determined as described in materials and methods. Results of subcellular distribution are expressed as percent related to the total activity of lyso-PAF AT in the 500 g supernatant (100%). Results are the mean ± SD of 4–14 determinations in different enzyme preparations. Fraction Cortex Medulla Specific activity (nmolÆmin )1 Æmg )1 ) Distribution of total activity (%) Specific activity (nmolÆmin )1 Æmg )1 ) Distribution of total activity (%) 500 g supernatant 0.37 ± 0.20 (n ¼ 6) 100 0.43 ± 0.31 (n ¼ 7) 100 20 000 g pellet 0.54 ± 0.33 (n ¼ 5) 15.6 ± 4.5 (n ¼ 5) 0.77 ± 0.42 (n ¼ 9) 18.3 ± 5.1 (n ¼ 5) 100 000 g supernatant nd a nd nd ND b 100 000 g pellet 1.14 ± 0.53 (n ¼ 14) 12.1 ± 3.5 (n ¼ 5) 1.45 ± 0.67 (n ¼ 13) 10.6 ± 3.8 (n ¼ 5) Total membrane fraction 0.80 ± 0.39 (n ¼ 4) ND 1.24 ± 0.28 (n ¼ 4) ND a nd, not detected. b ND, not determined. Ó FEBS 2003 Lyso-PAF acetyltransferase activity in human kidney (Eur. J. Biochem. 270) 2999 glyceryl-3-phosphorylcholine as the active component (a new class of lipid chemical mediators). J. Biol. Chem. 254, 9355–9358. 2. Prescott, S.M., Zimmerman, G.A., Stafforini, D.M. & McIntyre, T.M. (2000) Platelet-activating factor and related lipid mediators. Annu. Rev. Biochem. 69, 419–445. 3. Snyder, F. (1995) Platelet-activating factor and its analogs, metabolic pathways related intracellular processes. Biochim. Bio- phys. Acta 1254, 231–249. 4. Lee, T.C., Vallari, D.S. & Snyder, F. (1992) 1-Alkyl-2-lyso-sn- glycero-3-phosphocholine acetyltransferase. Methods Enzymol. 209, 396–401. 5. Snyder, F. (1995) Platelet-activating factor, the biosynthetic and catabolic enzymes. Biochem. J. 305, 689–705. 6. Nixon, A.B., O’Flaherty, J.T., Salyer, J.K. & Wykle, R.L. (1999) Acetyl-CoA: 1-O-alkyl-2-lyso-sn-glycero-3-phosphocholine acet- yltransferase is directly activated by p38 kinase. J. Biol. Chem. 274, 5469–5473. 7. Baker, P.R., Owen, J.S., Nixon, A.B., Thomas, L.N., Wooten, R., Daniel, L.W., O’Flaherty, J.T. & Wykle, R.L. (2002) Regulation of platelet-activating factor synthesis in human neutrophils by MAP kinases. Biochim. Biophys. Acta 1592, 175–184. 8. Gomez-Cambronero, J., Velasco, S., Sanchez-Crespo, M., Vivanco, F. & Mato, J.M. (1986) Partial purification and char- acterization of 1-O-alkyl-2-lyso-sn-glycero-3-phosphocholine: acetyl-CoA acetyltransferase from rat spleen. Biochem. J. 237, 439–445. 9. Seyama, K. & Ishibashi, T. (1987) Biochemical characterization of acetyl-CoA: 1-alkyl-2-lyso-sn-glycero-3-phosphocholine acetyl- transferase in rat spleen microsomes. Lipids 22, 185–189. 10. Billah, M.M. & Johnston, J.M. (1983) Identification of phos- pholipid platelet-activating factor (1-O-alkyl-2-acetyl-sn-glycero- 3-phosphocholine) in human amniotic fluid and urine. Biochem. Biophys. Res. Commun. 113, 51–58. 11. Sanchez-Crespo, M., Inarrea, P., Alvarez, V., Alonso, F., Egido, J. & Hernando, L. (1983) Presence in normal human urine of a hypotensive and platelet-activating phospholipid. Am.J.Physiol. 244, F706–F711. 12. Schlondorff, D., Goldwasser, P., Neuwirth, R., Satriano, J.A. & Clay, K.L. (1986) Production of platelet-activating factor in glo- meruli and cultured glomerular mesangial cells. Am.J.Physiol. 250, F1123–F1127. 13. Kester, M., Nowinski, R.J., Holthofer, H., Marsden, P.A. & Dunn M.J. (1994) Characterization of platelet-activating factor synthesis in glomerular endothelial cell lines. Kidney Int. 46, 1404–1412. 14. Neuwirth, R., Ardaillou, N. & Schlondorff, D. (1989) Extra and intracellular metabolism of platelet-activating factor by cultured mesangial cells. Am. J. Physiol. 256, F735–F741. 15. Pirotzky,E.,Ninio,E.,Bidault,J.,Pfister,A.&Benveniste,J., 1984) Biosynthesis of platelet-activating factor. VI. Precursor of platelet-activating factor and acetyltransferase activity in isolated rat kidney cells. Lab. Invest. 51, 567–572. 16. Zanglis, A. & Lianos, E.A. (1987) Synthesis of alkyl-ether gly- cerophospholipids in rat glomerular mesangial cells, evidence for alkyldihydroxyacetone phosphate synthase activity. Biochem. Biophys. Res. Commun. 144, 666–673. 17. Woodard, D.S., Lee, T.C. & Snyder, F. (1987) The final step in the de novo biosynthesis of platelet-activating factor. Properties of a unique CDP-choline, 1-alkyl-2-acetyl-sn-glycerol choline-phosphotransferase in microsomes from the renal inner medulla of rats. J. Biol. Chem. 262, 2520–2527. 18. Karasawa, K., Qiu, X.Y. & Lee, T.C. (1999) Purification and characterization from rat kidney membranes of a novel platelet- activating factor (PAF)-dependent transacetylase that catalyzes the hydrolysis of PAF, formation of PAF analogs, and C-2-cer- amide. J. Biol. Chem. 274, 8655–8661. 19. Lianos, E.A. & Zanglis, A. (1992) Effects of complement activa- tion on platelet-activating factor and eicosanoid synthesis in rat mesangial cells. J. Lab. Clin. Med. 120, 459–464. 20. Biancone, L., Tetta, C., Turello, E., Montrucchio, G., Iorio, E.L., Servillo, L., Balestrieri, C. & Camussi, G. (1992) Platelet-activat- ing factor biosynthesis by cultured mesangial cells is modulated by proteinase inhibitors. J. Am. Soc. Nephrol. 2, 1251–1261. 21. Camussi, G., Salvidio, G. & Tetta, C. (1989) Platelet-activating factor in renal diseases. Am. J. Nephrol. 9 (Suppl. 1), 23–26. 22. Lopez-Novoa, J.M. (1999) Potential role of platelet activating factor in acute renal failure. Kidney Int. 55, 1672–1682. 23. Antonopoulou, S., Demopoulos, C.A., Iatrou, C., Moustakas, G. & Zirogiannis, P. (1994) Platelet-activating factor acetylhydrolase (PAF-AH) in human kidney. Int. J. Biochem. 26, 1157–1162. 24. Iatrou, C., Moustakas, G., Antonopoulou, S., Demopoulos, C.A. & Ziroyiannis, P. (1996) PAF levels and PAF-acetylhydrolase activities in patients with primary glomerulonephritis. Nephron 72, 611–616. 25. Lenihan, D.G. & Lee, T.C. (1984) Regulation of platelet activating factor synthesis, modulation of 1-alkyl-2-lyso-sn-glycero-3-phos- phocholine, acetyl-CoA acetyltransferase by phosphorylation and dephosphorylation in rat spleen microsomes. Biochem. Biophys. Res. Commun. 120, 834–839. 26. Bligh, E.G. & Dyer, W.J. (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917. 27. Karantonis, H.C., Antonopoulou, S. & Demopoulos, C.A. (2002) Antithrombotic lipid minor constituents from vegetable oils: comparison between olive oils and others. J. Agric. Food Chem. 50, 1150–1160. 28. Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J. (1951) Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275. 29. Yamazaki, R., Sugatani, J., Fujii, I., Kuroyanagi, M., Umehara, K., Ueno, A., Suzuki, Y. & Miwa, M. (1994) Development of a novel method for determination of acetyl-CoA: 1-alkyl-2n-gly- cero-3-phosphocholine acetyltransferase activity and its applica- tion to screening for acetyltransferase inhibitors. Biochem. Pharmacol. 47, 995–1006. 30. Lee, T.C., Vallari, D.S. & Snyder, F. (1992) 1-Alkyl-2-lyso- sn-glycero-3-phosphocholine acetyltransferase. Methods Enzymol. 209, 396–401. 31. Ninio, E., Joly, F. & Bessou, G. (1988) Biosynthesis of PAF- acether. XI. Regulation of acetyltransferase by enzyme-substrate imbalance. Biochim. Biophys. Acta 963, 288–294. 32. Ninio, E., Mencia-Huerta, J.M., Heymans, F. & Benveniste, J. (1982) Biosynthesis of platelet-activating factor. I. Evidence for an acetyltransferase activity in murine macrophages. Biochim. Biophys. Acta 710, 23–31. 33. Lianos, E.A. & Zanglis, A. (1987) Biosynthesis and metabolism of 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine in rat glomerular mesangial cells. J. Biol. Chem. 262, 8990–8993. 3000 T. N. Nomikos et al. (Eur. J. Biochem. 270) Ó FEBS 2003 . Acetyl-CoA:1- O -alkyl- sn -glycero-3-phosphocholine acetyltransferase (lyso-PAF AT) activity in cortical and medullary human renal tissue Tzortzis N. Nomikos 1 , Christos Iatrou 2 and Constantine. degradative enzyme of PAF, in human kidney tissue (cortex and medulla) [23]; (b) the diminished activity of PAF-AH in renal tissue (mainly cortex) received