JBC Papers in Press. Published on September 26, 2002 as Manuscript M206812200 Serum Lysophosphatidic Acid is Produced Through Diverse Phospholipase Pathways Junken Aoki 1*, Akitsu Taira 1, Yasukazu Takanezawa 1, Yasuhiro Kishi 1, Kotaro Hama 1, Tatsuya Kishimoto 2, Koji Mizuno 2, Keijiro Saku 3, Ryo Taguchi 4, and Hiroyuki Arai 1 1 Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan 2 Diagnostic Research and Development Department, R&D Division, Nesco Company, Azwell Inc., 2-24-3 Sho, Osaka 567-0806, Ibaraki, Japan. 3 Second Department of Pathology, Fukuoka University School of Medicine, 7-45-1 Nanakuma, Jonanku, Fukuoka 814-80, Japan. 4 Faculty of Pharmaceutical Sciences, Nagoya City University, 3-1 Tanabe-dori, Mizuho-ku, Nagoya, Aichi 467-0027, Japan * Corresponding author. Running title Mechanism of LPA Production in Serum TEL: 81-3-5841-4723 FAX: 81-3-3818-3173 E-mail: jaoki@mol.f.u-tokyo.ac.jp Keywords: lysophosphatidic acid, LPA, serum, platelets, lysopld, PS-PLA 1, spla 2 -IIA, LCAT 1 Copyright 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
Footnote: 1. Taira et al. unpublished result Abbreviations The abbreviations used in this study are: PA, phosphatidic acid; LPA, lysophosphatidic acid; PC, phosphatidylcholine; LPC, lysophosphatidylcholine; LPS, lysophosphatidylserine; LPE, lysophosphatidylethanolamine; LPs, lysophospholipids; PLA 1, phospholipase A 1 ; PLA 2, phospholipase A 2 ; EDG, endothelial cell differentiated gene; lysopld, lysophospholipase D; PS-PLA 1, phosphatidylserine-specific PLA 1 ; spla 2 -IIA, secretory PLA 2 group IIA; LCAT, lecithin-cholesterol acyltransferase; mpa-pla 1, membrane-bound PA-selective PLA 1. 2
Summary Lysophosphatidic acid (LPA) is a lipid mediator with multiple biological activities that accounts for many biological properties of serum. LPA is thought to be produced during serum formation based on the fact that the LPA level is much higher in serum than in plasma. In this study, to better understand the pathways of LPA synthesis in serum, we evaluated the roles of platelets, plasma, and phospholipases by measuring LPA using a novel enzymelinked fluorometric assay. First, examination of platelet-depleted rats showed that half of the LPA in serum is produced via a platelet-dependent pathway. However, the amount of LPA secreted from isolated platelets after they are activated by thrombin or calcium ionophore accounted for only a small part of serum LPA. Most of the platelet-derived LPA was produced in a two-step process: lysophospholipids such as lysophosphatidylcholine (LPC), lysophosphatidylethanolamine, and lysophosphatidylserine, were released from activated rat platelets by the actions of two phospholipases, group IIA secretory phospholipase A 2 (spla 2 -IIA) and phosphatidylserine-specific phospholipase A 1 (PS-PLA 1 ), which were abundantly expressed in the cells. Then these lysophospholipids were converted to LPA by the action of plasma lysophospholipase D (lysopld). Second, accumulation of LPA in incubated plasma was strongly accelerated by the addition of recombinant lysopld with a concomitant decrease in LPC accumulation, indicating that the enzyme produces LPA by hydrolyzing LPC produced during the incubation. In addition, incubation of plasma isolated from human subjects who were deficient in lecithin-cholesterol acyltransferase (LCAT) did not result in increases of either LPC or LPA. The present study demonstrates multiple pathways for LPA production in serum and the involvement of several phospholipases including PS-PLA 1, spla 2 -IIA, LCAT, and lysopld. 3
Introduction Lysophosphatidic acid (1- or 2-acyl-lysophosphatidic acid; LPA) is a lipid that mediates multiple cellular processes (1, 2), including platelet aggregation, smooth muscle contraction, cell proliferation, and cytoskeletal reorganization (e.g., generation of actin stress fibers and inhibition of neurite outgrowth). LPA evokes its multiple effects through G- protein-coupled receptors (GPCR) that are specific to LPA. Recent studies have identified a new family of receptor genes for LPA (reviewed in (3, 4)). Members of this family include three GPCRs belonging to the EDG (endothelial cell differentiation gene) family, EDG2/LPA1 (5), EDG4/LPA2 (6) and EDG7/LPA3 (7), which are coupled with different G- proteins and may explain various cellular responses to LPA. Serum is known to contain micromolar concentrations of LPA, but the levels in plasma are much lower (8). For this reason it has been proposed that LPA in serum is produced as a result of blood coagulation and that platelets are involved, in part, in the production of LPA in serum. In the literature, two pathways have been postulated for the extracellular production of LPA. In the first pathway, LPA is produced by activated platelets. Mauco et al. showed that LPA is produced by human platelets after the cells were treated with exogenous phospholipase C from Clostridium welchii (9). LPA is also produced by platelets when they are activated by thrombin (10, 11). However, as discussed by Gaits et al. (12), the amount of LPA produced in the latter two studies is too low to explain the LPA level in serum. Thus, the full contribution of platelets to serum LPA production is currently unknown. In the second pathway, LPA is converted from lysophosphatidylcholine (LPC) by lysophospholipase D (lysopld) activity (1), which may occur in aged plasma or incubated plasma. The contribution of this pathway to the production of serum LPA is unclear because little is known about the enzymatic properties of lysopld. LPA in serum and produced in platelets is a mixture of various fatty acids. LPA species with both saturated fatty acids (16:0, 18:0) and unsaturated fatty acids (16:1, 18:1, 4
18:2, 20:4) have been detected in serum, plasma, and activated platelets (10, 13, 14). Interestingly, these LPA species exhibit differential biological activities (15-17) by differentially activating three LPA receptors, EDG2, EDG4, and EDG7 (18). These observations suggest that LPA species are biologically significant and are produced by diverse synthetic pathways. Serum LPA can be produced from phospholipid precursors either in membranes of blood cells or in plasma by sequential actions of phospholipases present in plasma or expressed by blood cells. In rats, two PLAs, secretory phospholipase A 2 group IIA (spla 2 - IIA) (19, 20) and phosphatidylserine-specific phospholipase A 1 (PS-PLA 1 ) (21, 22) are expressed predominantly in platelets (23). We previously showed that these two PLAs are involved in agonist-induced production of lysophospholipids such as LPC, lysophosphatidylethanolamine (LPE), and lysophosphatidylserine (LPS) in activated platelets by analyzing the phospholipid composition using specific inhibitors of spla 2 -IIA (24). In addition, other phospholipases capable of producing lysophospholipids, possibly LPC, have been identified in plasma. Two such phospholipases are lecithin-cholesterol acyltransferase (LCAT) and platelet-activating factor acetylhydrolase (PAF-AH). It has been suggested that part of the LPC present in blood is attributed to the transesterification of phosphatidylcholine (PC) and free cholesterol catalyzed by lecithin-cholesterol acyltransferase (LCAT) (25). It is also possible that the plasma platelet-activating factor acetylhydrolase (PAF-AH) (26) contributes to LPC production by hydrolyzing oxidized phosphatidylcholine (oxi-pc) (27), which has been implicated in various pathological conditions. We have recently identified the above-mentioned lysopld by purifying the enzyme (28). The identification and cloning of these phospholipases make it possible to examine their contribution to serum LPA production. In this study, to elucidate the synthetic pathway(s) for LPA in serum, we attempted to clarify the roles of blood cells, plasma, and the phospholipases in serum LPA production. 5
Experimental Procedures Materials Rabbit anti-rat platelet serum was purchased from Inter-Cell Technologies Inc. Bovine serum albumin (fatty acid free; A-6003) was purchased from Sigma. 1-oleoyl (18:1)-LPA, 1- oleoyl-lpc, 1-oleoyl-LPE, 1-oleoyl-LPS, porcine liver lysopi, sphingosylphosphocholine, dioleoyl PC were purchased from Avanti Polar Lipids Inc. (Alabaster, AL), and [ 3 H] 1- oleoyl-lpa (18:1) was from Amersham Pharmacia Biotech (Uppsala, Sweden). Monoglyceride lipase (MGL), glycero-3-phosphate oxidase (G3PO) and phosphocholine phosphodiesterase (GPCP) were kindly donated by Dr. S. Imamura (Asahi Chemical Industry Co. Ltd.). Horseradish peroxidase was purchased from Toyobo (Tokyo, Japan). Other chemicals were purchased from Wako Pure Chemical Industries (Osaka, Japan). Preparation of serum and plasma Male Wistar rats (15 weeks old, 300-350 g) were purchased from Japan SLC. The rats were fed a standard rat chow until the time of the study. Blood was collected by cardiac puncture. For preparation of serum, the drawn blood samples were incubated at 37 C for 60 minutes to allow blood coagulation and the supernatant after centrifugation at 2,300 x g for 20 minutes at 4 C was used as serum. For plasma, blood was drawn in the presence of one-sixth volume of acid citrate dextrose (ACD; 3 % citrate, 2.2 % D-glucose, ph 6.0) and was immediately centrifuged at 2,300 x g for 20 minutes at 4 C. The supernatant was used as plasma. For preparation of platelet-depleted rats, anti-rat platelet serum (200 µl) was injected into the animals intravenously and after 6 hours serum was prepared as described above. Plasma samples from patients with familial LCAT deficiency (FLD) who were admitted to the Fukuoka Medical University Hospital were drawn as describe above. Written informed consent was obtained from each patient. Control plasma samples were drawn from 6
healthy volunteers aged (22 to 45). All blood samples were immediately stored at 80 C to prevent additional formation of LPA. Isolation of platelets, erythrocytes and blood white cells Blood was drawn from rats or human volunteers in the presence of one-sixth volume of ACD and was immediately centrifuged at 800 x g for 20 minutes at 4 C. For platelet preparation, the supernatant (platelet-rich plasma) was further centrifuged at 2,300 x g for 20 minutes at 4 C, and the resulting cell pellet was washed twice with buffer A (12 mm citric acid, 15 mm sodium citrate, 113 mm NaCl, 0.4 mm NaH 2 PO 4, 12 mm NaHCO 3, ph 7.4) and finally suspended in buffer A. For erythrocyte preparation, the cells in the pellet fraction of the ACD blood were washed with phosphate buffered saline (PBS) twice after the buffy coat was removed. White blood cells were isolated using Nyco Prep 1.077 Animal (Daiichikagaku, Tokyo, Japan) according to the manufacturer s protocol. The major cell populations were lymphocytes and monocytes. For experiments to examine LPA production by these cells, the cell numbers of erythrocytes, platelets, and white blood cells were adjusted to 650, 11, and 2.0 x 10 7 cells/ml, respectively, in order to match their normal concentration in blood. To evaluate LPA production in these cells, the cells were suspended in PBS containing 0.1 % fatty acid-free bovine serum albumin (BSA). Quantification of LPA and LPC Concentrations of LPA and LPC were determined by an enzyme-linked fluorometric method established in the present study. LPA concentration was determined by fluorometry of H 2 O 2 using HPPA (3-(4-hydroxyphenyl) propionic acid, 7.5 mm, Dojin, Tokyo, Japan) as a peroxidase donor (29) generated by the reaction of LPA samples with 10 U/ml monoglyceride lipase (MGL, Asahi Chemical Industry Co. Ltd., Shizuoka, Japan) and 10 U/ml glycero-3-phosphate oxidase (G3PO, Asahi Chemical Industry Co. Ltd. Shizuoka, Japan) in a buffer containing 50 mm Tris, 2 mm CaCl 2, 0.2 % Triton X-100, 0.07 U/ml 7
peroxidase (TOYOBO, Tokyo, Japan), ph 7.4 in a total volume of 1,500 µl. The fluorescence intensity of excitation at 320 nm/ emission at 404 nm were measured with a fluorometer (Hitachi, Ibaraki), 5 minutes after mixing the samples. We detected LPA as low as 0.1 nmol and obtained linearity up to 10 nmol of LPA in this system. For determination of the LPA concentration in blood samples, LPA in samples (1 ml) was extracted by the method of Bligh and Dyer (30) under acidic conditions (by lowering the ph to 3.0 with 1 N HCl) prior to the LPA quantification. Lipids in the aqueous phase were re-extracted and pooled with the previous organic phase. The extracted lipids were dried and dissolved in phosphate buffered saline containing 0.1 % bovine serum albumin. The LPA content of this solution was quantified as described above. The recovery of lipids was monitored by the addition of trace amounts of 1-[ 3 H]-oleoyl-LPA to the samples. Based on recovery of 1-[ 3 H]-oleoyl- LPA, lipid recovery was always > 95 % under the above-described conditions. LPC concentration was determined by a similar method except that 10 U/ml phosphocholine phosphodiesterase (GPCP, Asahi Chemical Industry Co. Ltd. Shizuoka, Japan) was applied in the assay system. The concentrations of LPA were determined from a standard curve, after subtracting the tentative values obtained from the fluorescence intensities of the MGL (+)/G3PO (+) reaction from those of the MGL (-)/G3PO (+) reaction. Similarly, the concentration of LPC was determined from a standard curve, after subtracting the tentative values obtained from the MGL (+)/G3PO (+)/GPCP (+) reaction from those of the MGL (- )/G3PO (+)/GPCP (+) reaction. LysoPLD assay LysoPLD activity was assayed as described previously (28). Briefly, samples (1-50 µl) were incubated with 1 mm LPC (from egg) in the presence of 100 mm Tris-HCl (ph 9.0), 500 mm NaCl, 5 mm MgCl 2, and 0.05 % Triton X-100 for 1 hour at 37 C. The liberated choline was detected by an enzymatic photometric method using choline oxidase (Asahi Chemical, Tokyo, Japan), horseradish peroxidase (Toyobo, Osaka, Japan), and TOOS 8
reagent (N-ethyl-N-(2-hydoroxy-3-sulfoproryl)-3-methylaniline, Dojin, Tokyo, Japan) as a hydrogen donor. To examine lysopld activity against LPE, LPS, and LPI (each 1 mm), the formation of LPA in the reaction mixture was determined as described above. Preparation of phospholipases The recombinant enzymes used in this study (PS-PLA 1, and lysopld) were expressed by a baculovirus system as described previously (28, 31). The recombinant proteins were partially purified from the culture supernatant of Sf9 cells infected with each baculovirus using heparin and monoq column chromatography (21, 28) and were dialyzed in phosphate-buffered saline (-/-) before use. spla 2 -IIA was purified from cell supernatant of thrombin-activated rat platelets as described previously (28, 57). 9
Results Role of platelets in serum LPA production To evaluate the involvement of platelets in the production of serum LPA, we prepared platelet-depleted animals and determined their serum LPA level by a fluorometric method established in this study (see Experimental Procedures). The number of platelets in rats treated with rabbit anti-platelet serum was about 1 % of that in control animals treated with control rabbit serum (Fig. 1B). The treatment with rabbit anti-platelet serum did not affect the numbers of other blood cells such as erythrocytes and white blood cells (data not shown). Under these conditions, serum LPA levels were 0.94 +/- 0.11 µm in anti-platelet-treated animals and 1.95 +/- 0.14 µm in control antibody-treated animals (Fig. 1A). Plasma LPA levels in these animals were 0.17 +/- 0.02 µm (platelet-depleted) and 0.16 +/- 0.02 µm (control). These results confirmed a previous report that the LPA level is high in serum but low in plasma (8). In addition, the present result shows that half of LPA in serum is produced by a platelet-dependent pathway. We next examined the contribution of platelets themselves to the production of LPA in serum by measuring the LPA produced by isolated platelets upon their activation. As shown in Fig. 1C, rat platelets produced and released LPA after they were activated by thrombin or by a calcium ionophore, A23187. However, the levels of LPA produced by these activators (0.10 +/- 0.02 µm and 0.16 +/- 0.01 µm, respectively) were too low to account for the LPA produced by the platelet-dependent pathway of LPA production in serum. The human platelets were also found to have a low ability to produce LPA upon activation (Fig. 2B). Lysophospholipids secreted from activated platelets are converted to LPA by lysopld. 10
In view of the findings that much of the lysophospholipids, but not LPA is produced in activated platelets (24) and that lysopld activity is detected in plasma of several mammalian species including rat and human (28, 32), we hypothesized that part of the LPA in serum is produced in two steps: generation of lysophospholipids in activated platelets and their subsequent conversion to LPA by lysopld. To test this possibility, we examined whether the full amount of LPA produced in the platelet-dependent pathway is detected when isolated platelets are activated in the presence of lysopld. As was observed previously (24), a high concentration of LPC was detected in the supernatant of activated platelets stimulated with thrombin or A23187 (Fig. 2C). The addition of a physiological concentration of recombinant lysopld dramatically increased the amount of LPA in the supernatant of the activated rat or human platelets, and slightly decreased the LPC level (Fig. 2B). In the presence of recombinant lysopld, the concentration of LPA rose to 0.63 +/- 0.09 µm and 0.73 +/- 0.12 µm when isolated rat platelets were activated by thrombin and A23187, respectively (Fig. 2B). These concentrations are comparable to the concentration produced in the platelet-dependent pathway (Fig. 1). Among various lysophospholipids detected in the activated rat platelets, only LPC has been shown to be a substrate of lysopld. However, as shown in Fig. 2A, lysopld hydrolyzed other lysophospholipids, such as LPE, LPS and LPI, to produce LPA. These results clearly indicate that, in the platelet-dependent pathway for LPA synthesis, lysophospholipids produced by activated platelets are converted to LPA by the action of lysopld present in blood plasma. Production of lysophospholipids by spla 2 -IIA and PS-PLA 1 and their subsequent conversion to LPA by lysopld We further examined the roles of spla 2 -IIA and PS-PLA 1 in lysopld-enhanced LPA production in activated rat platelets, as the two PLAs are predominantly expressed in the cells and have been implicated in lysophospholipid production in activated platelets (24). As shown in Fig. 3, addition of recombinant spla 2 -IIA or PS-PLA 1 dramatically increased 11
the LPA production evoked by addition of recombinant lysopld from activated platelets stimulated with either thrombin or A23187. This suggests that the two PLAs are involved in serum LPA production through their roles in supplying lysophospholipids to lysopld. Contribution of other blood cells to serum LPA production We further determined whether erythrocytes or white blood cells are involved in serum LPA production. To examine LPA production in blood cells, we incubated erythrocytes, white blood cells, and platelets prepared from rat blood with a physiological concentration of recombinant lysopld. Of the different cell types examined, platelets were by far the most potent in producing LPA and LPC per cell (Fig. 4A and 4C). However, a low but significant concentration of LPA (0.17 +/- 0.02 µm) was detected in the culture supernatant of erythrocytes, when physiological cell numbers (6.5 x 10 9 cells/ml) were applied (Fig. 4B). LPC was also detected in the cell supernatant (Fig. 4C). Thus, lysophospholipids associated with erythrocytes, possibly LPC, can be a substrate for lysopld, which can explain part of the serum LPA production in the platelet-independent pathway. Involvement of lysopld and LCAT in the production of LPA in plasma Both LPA and LPC are produced in plasma during prolonged incubation at 37 C in rat (32). As shown in Fig. 5, we confirmed that incubation of plasma or serum both from rat and human at 37 C increased the accumulation of both LPA and LPC in a time coursedependent manner. The formation of LPA in serum is much faster than that in plasma, especially in the initial period (1-6 hours). LPC concentration in serum was slightly higher than that in plasma (Fig. 5), indicating that lysophospholipids accumulated in incubated serum can be derived both from plasma phospholipids and, as a result of blood coagulation, from platelet phospholipids. It is likely that lysopld converts these lysophospholipids to LPA. Consistent with this idea, the addition of recombinant lysopld to the plasma 12
dramatically increased the formation of LPA, and resulted in a smaller increase in LPC (Fig. 6). A similar result was obtained in incubated serum (data not shown). Addition of a divalent cation chelator such as EDTA or EGTA to the incubation medium almost completely inhibited the accumulation of LPA, but did not affect the accumulation of LPC (Fig. 6; EGTA data not shown). This observation strongly indicates that LPC present or generated during the incubation in plasma is converted to LPA by lysopld activity, and it is compatible with the fact that the enzyme requires a divalent cation for its activity (33). We further examined the mechanism of LPC formation in incubated plasma. It has been suggested that part of the LPC present in blood is due to the activities of lecithincholesterol acyltransferase (LCAT) (25) or platelet-activating factor acetylhydrolase (PAF- AH) activities (27). The formation of LPC during incubation was insensitive to EDTA (Fig. 8), which is compatible with the properties of the two plasma enzymes (34, 35). To evaluate the contribution of these enzymes to the accumulation of LPC in plasma during incubation, we examined the plasma LPC level of human subjects deficient in LCAT- (FLD) or PAF-AH. As was observed in plasma from normal subjects (Fig. 5), the LPC level increased during incubation at 37 C in the PAF-AH-deficient plasma. On the other hand, LPC did not form at all in the plasma from LCAT-deficient patients (Fig. 7), which confirmed that LCAT is responsible for the accumulation of LPC in the incubated plasma. In addition, it was revealed that LPA accumulation was significantly suppressed in LCAT-deficient plasma, but not in the control or PAF-AH-deficient plasma (Fig. 7). The LCAT-deficient plasma had a comparable level of LPC (~ 100 µm) (Fig. 7), indicating that a part of the plasma LPC can also be produced by an LCAT-independent pathway. 13
Discussion The present study was undertaken to clarify how LPA is produced in serum. There are at least four synthetic pathways for serum LPA (Fig. 8). The major two pathways identified in the present study are (1) secretion of lysophospholipids such as LPC, LPE, and LPS from platelets, followed by conversion of the lysophospholipids to LPA (Fig. 2), and (2) generation of LPC from PC in lipoprotein and its consequent conversion to LPA (Figs. 6 and 7). The first pathway requires activation of platelets. Thus, LPA is produced through this pathway under pathological conditions, in which platelets are activated. Such pathological conditions are found at sites of injury, inflammation, and atherosclerosis. In contrast, the second pathway is continuously active in blood, which makes it difficult to evaluate it as discussed below. The other two pathways are (3) production by isolated platelets upon their activation (Fig. 1) and (4) production by erythrocytes (Fig. 4). However, these two pathways make minor contributions to serum LPA production. LysoPLD is a key enzyme in serum LPA production. LysoPLD is involved in the first, second and fourth pathways, and thus contributes considerably to LPA production in serum (Fig. 8). Because lysopld is the only enzyme that exhibits lysopld activity in blood plasma or serum (28), it can be concluded that lysopld is a key enzyme in serum LPA production. Recently we have shown that lysopld is identical to autotaxin (ATX) (28). ATX/lysoPLD is an autocrine motility factor that stimulates motility and proliferation of cancer cells (28, 36). The product of lysopld, LPA, is an effective inducer of chemotaxis (37-40) and cell proliferation (41) in multiple cell lineages. In addition, serum has multiple cell effects on cell proliferation and motility. Thus, it is possible that some of these biological activities of serum are explained by its lysopld activity. LysoPLD does not appear to be activated as a result of blood coagulation, as its activity in serum and plasma were about the same (data not shown). Thus, lysopld seems to be continuously active in blood, which suggests that LPA is continuously produced in the 14
bloodstream through the second and fourth pathways, although the LPA level in fresh plasma is quite low (~ 0.1 µm). This low level can be explained by our preliminary finding that LPA does not accumulate in plasma when the plasma is incubated in the presence of blood cells footnote 1. There is some evidence that lipid phosphate phosphatases (LPPs) are involved in cellular degradation of lysophospholipid mediators including sphingosine 1-phosphate and LPA (42, 43). LPPs expressed by blood cells may be responsible for keeping the LPA level lower, resulting in negative regulation of LPA signaling. LCAT supplies LPC to lysopld in plasma. The formation of LPC and LPA was decreased in plasma from LCAT-deficient subjects (Fig. 7). Because plasma lysopld activities of LCAT-deficient and control subjects were not significantly different, LCAT appears to have a critical role in the second pathway by supplying LPC to lysopld (Figs. 7 and 8). LCAT deficiency in humans, referred to as familial LCAT deficiency (FLD), is characterized by a complete lack of plasma LCAT activity, corneal opacification, anemia, proteinuria, and kidney dysfunction (44-46). The plasma of these patients has decreased concentrations of total cholesterol, esterified cholesterol, and HDL cholesterol, and increased concentrations of triglyceride, phospholipid, free cholesterol, and VLDL cholesterol. Although most of the phenotypes of FLD patients seem to be explained by abnormal composition and shape of plasma HDL particles, some can be explained by the reduced LPA production. We observed that LPC is still present in LCATdeficient plasma, indicating that LPC originates from several metabolic pathways. It is likely that LPC is generated intracellularly and is also secreted directly by hepatocytes which are a quantitatively important source of plasma unsaturated LPC (47). Although the molecular mechanisms underlying the LCAT-independent LPC production are still unclear, they may have a role in lysopld-mediated production of LPA, especially LPA with unsaturated fatty acids. 15
Lysophospholipids generation by diverse phospholipase A s Lysophospholipids that are converted to LPA by lysopld can be generated by actions of various phospholipase As. The present study demonstrated that spla 2 -IIA and PS-PLA 1, which, in rats, are predominantly expressed in platelets, are involved in the production of serum LPA. We showed that, in rat platelets, these enzymes generate lysophospholipids (24) (Figs. 3 and 8), which are then converted to LPA by the lysopld present in plasma. Consistent with this observation, LPA formation was much faster in serum than in plasma during incubation at 37 C (Fig. 5). Because serum and plasma had almost equal lysopld activities (data not shown), the higher level of LPA in the serum during the incubation can be attributed to a higher level of lysophospholipids in the serum as a result of activities of spla 2 -IIA and PS-PLA 1. There is some disagreement over the involvement of spla 2 -IIA in serum LPA production. Balle et al. showed that serum LPA production in mice is not affected by either inhibitors of spla 2 -IIA or a genetic lack of spla 2 -IIA (48). Thus, a mechanism similar to that in rats may not occur in mice. This may be because spla 2 -IIA is highly expressed in rat (and human) platelets, but only weakly expressed in mouse platelets (49). Expression of PS-PLA 1 is also species-dependent. It is highly expressed in rat platelets, whereas it is barely detected in human and mouse platelets (22). PS-PLA 1 is detected in plasma of all three species (22). spla2 represent a growing family of enzymes that includes groups IB, IIA, IIC, IID, IIE, IIF, III, V, X, and XII. These spla 2 isozymes form a gene cluster on the same chromosome and seem to show species-specific expression patterns (49). In some species these spla 2 isozymes may have a role in the production of serum LPA by supplying lysophospholipids. Other phospholipases LPA can be produced through the hydrolysis of phosphatidic acid (PA) that is either exposed on the cell surface or in membrane microvesicles shed from Ca 2+ -loaded erythrocytes (50). spla 2 -IIA and a recently identified membrane-bound PA-selective PLA 1 16
(mpa-pla 1 ), which is predominantly expressed in human platelets (51), may contribute to LPA production in activated platelets in the absence of lysopld (see pathway (3) in Fig. 8). Although the amount of LPA produced in this pathway seems to be low, this pathway should not be overlooked because it can generate LPA rapidly (H. Sonoda et al. unpublished result) and may play a role in certain microenvironments such as sites of injury or hemostasis. In ovarian cancer cell lines, several PLA 2 isozymes (spla 2 -IB, spla 2 -IIA, and calcium-dependent and/or -independent cytosolic PLA 2 s (cpla 2 and ipla 2 )) have been implicated in constitutive and agonist-induced LPA synthesis (52, 53). Although it is not clear whether these isozymes (spla 2 -IB, cpla 2 and ipla 2 ) have a role in the production of serum LPA, because the latter two are expressed in platelets (49), they may make a minor contribution by supplying lysophospholipids to lysopld or by hydrolyzing PA. Finally, during the preparation of this manuscript, Sano et al. reported that LPA and sphingosine-1-phosphate are produced in the course of blood coagulation in human (54). Based on an analysis of the acyl composition of the LPAs generated and the fate of exogenously added fluorescent phospholipid analogs, they suggested some possible pathways and enzymatic activities that essentially agree with our results, even though their approach was quite different from ours. Using plasma from genetically-deficient patients and recombinant enzymes, we further clarified the involvement of individual phospholipases in the production of LPA. These phospholipases were not previously known to have such a role. We conclude that LPA is produced in multiple pathways by sequential reactions of phospholipases. Our next challenge is to evaluate the biological significance of each synthetic pathway. 17
Acknowledgments We thank Dr. Akira Tokumura (Tokushima University) for helpful discussions. This work was supported in part by research grants from the Ministry of Education, Culture, Sports, Science and Technology, and by the Human Frontier Special Program. 18
Figure legends Figure 1 Contribution of platelets in serum LPA production. (A) To prepare platelet-depleted rats, rabbit anti-rat platelet serum was injected into rats intravenously. Six hours later sera were prepared and LPA concentration was determined by an enzyme-linked fluorometric method (see Materials & methods). For a control experiment, preimmune rabbit serum was intravenously injected and serum was prepared. The numbers of platelet are shown in (B). These data represent the means +/- SD of three independent experiments. (C) LPA production in activated rat platelets. Activation of isolated rat platelets was induced by stimulating the cells either by thrombin or a calcium ionophore, A23187, and LPA concentration in the cell supernatant was determined. The amount of LPA in serum is also shown for comparison. These data represent the means +/- SD of three independent experiments. Figure 2 LPA is converted from lysophospholipids produced by platelets by the action of lysopld. (A) Substrate specificity of lysopld. LPC (1-oleoyl), LPE (1-oleoyl), LPS (1- oleoyl), LPI (from porcine liver), or PC (dioleoyl) (each 1 mm) was incubated at 37 C for 1 hour in the presence of recombinant lysopld. LPA concentration after the reaction was determined by an enzyme-linked fluorometric method. These data represent the means +/- SD of three independent experiments. (B, C) Isolated platelets from rat or human were stimulated with either thrombin or A23187 both in the presence or absence of recombinant lysopld. LPA (B) and LPC (C) concentrations in the culture cell supernatant were determined. The amount of lysopld added is the same as that detected in plasma (final concentration is 102 pmol/min.µl for rat and 87 pmol/min.µl for human). These data represent the means +/- SD of three independent experiments. 19
Figure 3 Recombinant spla 2 IIA or PS-PLA 1 accelerates lysopld-induced LPA formation in activated rat platelets. Isolated rat platelets were mixed with recombinant lysopld (final concentration is 0.12 ng/µl (102 pmol/min.µl)) and were stimulated with either thrombin or A23187 both in the presence or absence of recombinant spla 2 -IIA or PS-PLA 1. Then the LPA and LPC concentrations in the culture cell supernatant were determined. The amount of spla 2 -IIA or PS-PLA 1 applied was five times as much as that detected in the supernatant of activated rat platelets (21, 55) (final concentrations were 14.2 ng/µl (235 pmol/min.µl) for spla 2 IIA and 3.1 ng/µl (44.6 p mol/min.µl) for PS-PLA 1, respectively). Figure 4 LPA production in other blood cells. Isolated rat erythrocytes (RBC) and white blood cells (WBC) were incubated in the presence or absence of recombinant lysopld. LPA and LPC concentrations of the culture cell supernatant were determined. The amount of lysopld added was the same as that detected in rat plasma (102 pmol/min.µl). LPA secreted from thrombin-activated rat platelets both in the presence or absence of recombinant lysopld is shown for comparison. These data represent the means +/- SD of three independent experiments. Figure 5 Production of LPA and LPC in incubated serum and plasma Both plasma and serum from rat or human were prepared. Then the samples were incubated at 37 C for the indicated period and the amounts of LPA and LPC formed were determined. Closed circle indicate serum and open circle indicate plasma. Figure 6 LysoPLD accelerates LPA formation in incubated rat plasma and is sensitive to a divalent cation chelator. Plasma was incubated at 37 C for the indicated period either in the presence (closed circles) or absence (open circles) of lysopld, and the amounts of LPA and LPC formed were 20
determined. The same reaction was performed in the presence of 10 mm EDTA and lysopld (closed squares) and in the presence of 10 mm EDTA and in the absence of lysopld (open squares). Recombinant lysopld was added to a final concentration that was ten times as much as that detected in rat plasma. The data are representative of three different experiments. Figure 7 LPA and LPC formation in plasma from LCAT- or PAF-AH-deficient subjects. Plasma samples from a LCAT-deficient patient (FDL, closed circles), a PAF-AHdeficient donor (open squares), and a healthy volunteer (open circles) were incubated at 37 C and the time course-dependent accumulations of LPA and LPC were determined. The data are representative of two different experiments. Figure 8 Synthetic pathways for serum LPA. Serum LPA can be produced via at least four pathways. (1) Approximately half of serum LPA is formed through the generation of lysophospholipids such as LPC, LPE, and LPS from membrane phospholipids of activated platelets by spla 2 -IIA or PS-PLA 1, followed by conversion of the lysophospholipids to LPA by lysopld. (2) During formation of serum, part of the LPA can be generated by sequential actions of LCAT and lysopld. (3) In activated platelets, LPA can be produced through hydrolysis of PA by spla 2 -IIA or mpa- PLA 1, which may explain LPA production in isolated platelets upon their activation. (4) LPC present on erythrocyte membranes can be a substrate for lysopld. The amount of LPA generated in the third and fourth pathways make minor contributions to serum LPA formation. 21
References (1) Tokumura, A. (1995) Prog. Lipid Res. 34, 151-184 (2) Moolenaar, W. H. (1999) Exp. Cell Res. 253, 230-238 (3) Chun, J., Contos, J. J. and Munroe, D. (1999) Cell Biochem. Biophys. 30, 213-242 (4) Contos, J. J., Ishii, I. and Chun, J. (2000) Mol. Pharmacol 58, 1188-1196 (5) Hecht, J. H., Weiner, J. A., Post, S. R. and Chun, J. (1996) J. Cell Biol. 135, 1071-1083 (6) An, S., Bleu, T., Hallmark, O. G. and Goetzl, E. J. (1998) J. Biol. Chem. 273, 7906-7910 (7) Bandoh, K., Aoki, J., Hosono, H., Kobayashi, S., Kobayashi, T., Murakami, M. K., Tsujimoto, M., Arai, H. and Inoue, K. (1999) J. Biol. Chem. 274, 27776-27785 (8) Tigyi, G. and Miledi, R. (1992) J. Biol. Chem. 267, 21360-21367 (9) Mauco, G., Chap, H., Simon, M. F. and Douste, B. L. (1978) Biochimie 60, 653-661 (10) Gerrard, J. M. and Robinson, P. (1989) Biochim. Biophys. Acta 1001, 282-285 (11) Eichholtz, T., Jalink, K., Fahrenfort, I. and Moolenaar, W. H. (1993) Biochem. J. 677-680 (12) Gaits, F., Fourcade, O., Le, B. F., Gueguen, G., Gaige, B., Gassama, D. A., Fauvel, J., Salles, J. P., Mauco, G., Simon, M. F. and Chap, H. (1997) FEBS Lett. 410, 54-58 (13) Xiao, Y., Chen, Y., Kennedy, A. W., Belinson, J. and Xu, Y. (2000) Ann. N. Y. Acad. Sci. 905, 242-259 (14) Baker, D. L., Umstot, E. S., Desiderio, D. M. and Tigyi, G. J. (2000) Ann. N Y Acad. Sci. 905, 267-269 (15) Tokumura, A., Iimori, M., Nishioka, Y., Kitahara, M., Sakashita, M. and Tanaka, S. (1994) Am. J. Physiol. C204-210 22
(16) Jalink, K., Hengeveld, T., Mulder, S., Postma, F. R., Simon, M. F., Chap, H., van, der, Marel, Ga, van, B. J., van, B. W. and Moolenaar, W. H. (1995) Biochem. J. 307, 609-616 (17) Hayashi, K., Takahashi, M., Nishida, W., Yoshida, K., Ohkawa, Y., Kitabatake, A., Aoki, J., Arai, H. and Sobue, K. (2001) Circ. Res. 89, 251-258 (18) Bandoh, K., Aoki, J., Tsujimoto, M., Arai, H. and Inoue, K. (2000) FEBS Lett. 478, 159-165 (19) Komada, M., Kudo, I., Mizushima, H., Kitamura, N. and Inoue, K. (1989) J. Biochem. 106, 545-547 (20) Kudo, I., Murakami, M., Hara, S. and Inoue, K. (1993) Biochim. Biophys. Acta 1170, 217-231 (21) Sato, T., Aoki, J., Nagai, Y., Dohmae, N., Takio, K., Doi, T., Arai, H. and Inoue, K. (1997) J. Biol. Chem. 272, 2192-2198 (22) Aoki, J., Nagai, Y., Hosono, H., Inoue, K. and Arai, H. (2002) Biochimi. Biochem. Acta 1582, 26-32 (23) Horigome, K., Hayakawa, M., Inoue, K. and Nojima, S. (1987) J. Biochem. 101, 53-61 (24) Yokoyama, K., Kudo, I. and Inoue, K. (1995) J. Biochem. 117, 1280-1287 (25) Sekas, G., Patton, G. M., Lincoln, E. C. and Robins, S. J. (1985) J. Lab. Clin. Med. 105, 190-194 (26) Tjoelker, L. W., Wilder, C., Eberhardt, C., Stafforini, D. M., Dietsch, G., Schimpf, B., Hooper, S., Le, T. H., Cousens, L. S., Zimmerman, G. A., Yamada, Y., McIntyre, T. M., Prescott, S. M. and Gray, P. W. (1995) Nature 374, 549-553 (27) Watson, A. D., Navab, M., Hama, S. Y., Sevanian, A., Prescott, S. M., Stafforini, D. M., McIntyre, T. M., Du, B. N., Fogelman, A. M. and Berliner, J. A. (1995) J. Clin. Invest. 95, 774-782 23
(28) Goto-Umezu, M., Kishi, Y., Taira, A., Hama, K., Dohmae, N., Takio, K., Yamori, T., Mills, G. B., Inoue, K., Aoki, J. and Arai, H. (2002) J. Cell Biol. in press, (29) Tamaoku, K., Ueno, K., Akiura, K. and Ohkura, Y. (1982) Chem. Pharm. Bull 30, 2492-2497 (30) Bligh, E. C. and Dyer, W. F. (1959) Can. J. Biochem. Physiol. 37, 911-917 (31) Hosono, H., Aoki, J., Nagai, Y., Bandoh, K., Ishida, M., Taguchi, R., Arai, H. and Inoue, K. (2001) J. Biol. Chem. 276, 29664-29670 (32) Tokumura, A., Harada, K., Fukuzawa, K. and Tsukatani, H. (1986) Biochim. Biophys. Acta 875, 31-38 (33) Tokumura, A., Miyake, M., Yoshimoto, O., Shimizu, M. and Fukuzawa, K. (1998) Lipids 33, 1009-1015 (34) Dobiasova, M. and Schutzova, M. (1986) Physiol. Bohemoslov. 35, 464-472 (35) Satoh, K., Imaizumi, T., Kawamura, Y., Yoshida, H., Hiramoto, M., Takamatsu, S. and Takamatsu, M. (1991) J. Clin. Invest. 87, 476-481 (36) Stracke, M. L., Clair, T. and Liotta, L. A. (1997) Adv. Enzyme Regul. 37, 135-144 (37) Imamura, F., Horai, T., Mukai, M., Shinkai, K., Sawada, M. and Akedo, H. (1993) Biochem. Biophys. Res. Commun. 193, 497-503 (38) Manning, T. J., Parker, J. C. and Sontheimer, H. (2000) Cell Motil. Cytoskeleton 45, 185-199 (39) Sturm, A., Sudermann, T., Schulte, K. M., Goebell, H. and Dignass, A. U. (1999) Gastroenterology 117, 368-377 (40) Stam, J. C., Michiels, F., van, der, Kammen, Ra, Moolenaar, W. H. and Collard, J. G. (1998) EMBO. J. 17, 4066-4074 (41) van Corven, E. J., Groenink, A., Jalink, K., Eichholtz, T. and Moolenaar, W. H. (1989) Cell 59, 45-54 24
(42) Pilquil, C., Singh, I., Zhang, Q. X., Ling, Z. C., Buri, K., Stromberg, L. M., Dewald, J. and Brindley, D. N. (2001) Prostaglandins Other Lipid Mediat. 64, 83-92 (43) Hooks, S. B., Santos, W. L., Im, D. S., Heise, C. E., Macdonald, T. L. and Lynch, K. R. (2001) J. Biol. Chem. 276, 4611-4621 (44) Glomset, J. A., Assman, G., Gjone, E. and Norum, K. R. (1995) The Metabolic and Molecular Bases of Inherited Disease 1933-1951 (45) Jonas, A. (2000) Biochim. Biophys. Acta 1529, 245-256 (46) Santamarina-Fojo, S., Hoeg, J. M., Assman, G. and Brewer, H. B., Jr. (2001) The Metabolic and Molecular Bases of Inherited Disease 817-2833 (47) Croset, M., Brossard, N., Polette, A. and Lagarde, M. (2000) Biochem. J. 1, 61-67 (48) le Balle, F., Simon, M. F., Meijer, S., Fourcade, O. and Chap, H. (1999) Adv Enzyme Regul 39, 275-284 (49) Murakami, M., and Kudo, I. (2001) Adv. Immunol. 77, 163-194 (50) Fourcade, O., Simon, M. F., Viode, C., Rugani, N., Leballe, F., Ragab, A., Fournie, B., Sarda, L. and Chap, H. (1995) Cell 80, 919-927 (51) Sonoda, H., Aoki, J., Hiramatsu, T., Ishida, M., Bandoh, K., Nagai, Y., Taguchi, R., Inoue, K. and Arai, H. (2002) J. Biol. Chem. 277, 34254 34263 (52) Eder, A., Sasagawa, T., Mao, M., Aoki, J. and Mills, G. (2000) Clinical Cancer Research 6, 2482-2491 (53) Shen, Z., Belinson, J., Morton, R. E., Xu, Y. and Xu, Y. (1998) Gynecol. Oncol. 71, 364-368 (54) Sano, T., Baker, D. L., Virag, T., Wada, A., Yatomi, Y., Kobayashi, T., Igarashi, Y. and Tigyi, G. J. (2002) J. Biol. Chem. 277, 21197-21206 (55) Horigome, K., Hayakawa, M., Inoue, K. and Nojima, S. (1987) J. Biochem. 101, 625-631 25
8 A B LPA concentration (µm) C LPA concentration (µm) 2.5 2.0 1.5 1.0 0.5 0.0 Antirat platelets 2.4 2.0 1.6 1.2 0.8 0.4 Control Number of platelets (X 10 cells / ml) 1.5 1.0 0.5 0.0 Antirat platelets Control 0.0 serum thrombin A23187 no stimulus Aoki et al. Fig. 1
A 20 LPA formed (µm) 15 10 5 B LPA concentration (µm) 0.0 stimulus lysopld C LPC concentration (µm) 1.0 0.8 0.6 0.4 0.2 8 6 4 2 0 Rat LPC LPE LPS LPI Substrate - thrombin A23187 Rat LPA 0.4 0.3 0.2 0.1 0.0 LPC 4 3 + - + - + - 2 1 Human PC - thrombin A23187 Human + - + - + - 0 stimulus lysopld - thrombin A23187 + - + - + - 0 - thrombin A23187 + - + - + - Aoki et al. Fig. 2
3.5 3.0 LPA concentration (µm) stimulus 2.5 2.0 1.5 1.0 0.5 0.0 none thrombin A23187 lysopld + - + + + + + + + + + - - + - - + - PS-PLA1 spla2-iia - - + - - + - - + Aoki et al. Fig. 3
A LPA concentration (µm) / 10 8 cells 0.8 0.6 0.4 0.2 0.0 LPA RBC WBC platelets-thrombin lysopld - + - + - + C LPC concentration (µm) / 10 8 cells 6 5 4 3 2 1 0 LPC RBC WBC platelets-thrombin lysopld - + - + - + B LPA concentration (µm) 0.8 0.6 0.4 0.2 0.0 LPA RBC WBC platelets-thrombin lysopld - + - + - + D LPC concentration (µm) 7 6 5 4 3 2 1 0 LPC RBC WBC platelets-thrombin lysopld - + - + - + Aoki et al. Fig. 4
LPA concentration (µm) LPA concentration (µm) 50 40 30 20 10 0 0 40 30 20 10 rat LPA 10 time (hour) human LPA 20 LPC concentration (µm) LPC concentration (µm) 700 600 500 400 300 200 100 0 0 800 700 600 500 400 300 200 100 rat LPC 10 time (hour) human LPC 20 0 0 10 20 0 0 10 20 time (hour) time (hour)
600 LPC 540 480 LPC concentration (µm) 40 30 20 10 0 LPA 0 2 4 6 incubation time (hour) 420 0 2 4 6 incubation time (hour) lysopld, EDTA (+, +) lysopld, EDTA (+, -) lysopld, EDTA (-, +) lysopld, EDTA (-, -) Aoki et al. Fig. 6 LPA concentration (µm)
25 20 15 10 5 0 0 LPA LPC 0 10 20 30 40 50 0 10 20 30 40 50 Incubation time (hours) Incubation time (hours) 800 600 400 200 healthy volunteer PAF-AH-deficient donor LCAT-deficient donor Aoki et al. Fig. 7 LPA concentration (µm) LPC concentration (µm)
LPC LPE LPS lysopld (1) LPA lysopld (2) LPC LCAT PC activated platelets spla2-iia PS-PLA1 PC,PE,PS PA spla2-iia mpa-pla1 (3) LPA lysopld (4) LPC erythrocytes blood vessel Aoki et al. Fig. 8 Downloaded from http://www.jbc.org/ by guest on March 8, 2019
Serum lysophosphatidic acid is produced through diverse phospholipase pathways Junken Aoki, Akitsu Taira, Yasukazu Takanezawa, Yasuhiro Kishi, Kotaro Hama, Tatsuya Kishimoto, Koji Mizuno, Keijiro Saku, Ryo Taguchi and Hiroyuki Arai J. Biol. Chem. published online September 26, 2002 Access the most updated version of this article at doi: 10.1074/jbc.M206812200 Alerts: When this article is cited When a correction for this article is posted Click here to choose from all of JBC's e-mail alerts