- Email: [email protected]
Activation of human monocytic cells by lysophosphatidic acid and sphingosine-1-phosphate
Cellular Signalling 15 (2003) 367 – 375 www.elsevier.com/locate/cellsig
Activation of human monocytic cells by lysophosphatidic acid and sphingosine-1-phosphate Markus Fueller a, De An Wang b, Gabor Tigyi b, Wolfgang Siess a,* a
Institut fu¨r Prophylaxe und Epidemiologie der Kreislaufkrankheiten, Klinikum der Universita¨t Mu¨nchen, Pettenkoferstr. 9, D 80336 Munich, Germany b Department of Physiology, University of Tennessee Health Science Center, 894 Union Avenue, Memphis, TN 38163, USA Received 30 June 2002; accepted 26 August 2002
Abstract Lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) are serum-borne lipid mediators with potential proinflammatory and atherogenic properties. We studied the effects of LPA and S1P on [Ca2 +]i, a second messenger of cellular activation, in human monocytic Mono Mac 6 (MM6) cells. LPA and S1P induced [Ca2 +]i transients with EC50 values of 47 and 340 nM, respectively. Ca2 + signals evoked by LPA and S1P originated mainly from the stimulation of Ca2 + entry, were blocked by the phospholipase C inhibitor U73122, and were inhibited by pertussis toxin. The LPA1 and LPA3 receptor antagonist dioctylglycerol pyrophosphate inhibited the LPA-induced Ca2 + signal. Notably, serum and minimally modified LDL (mm-LDL) evoked [Ca2 +]i increases that were mediated entirely through activation of LPA receptors. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis showed the presence of the LPA and S1P receptor subtypes LPA1, LPA2, S1P1, S1P2, S1P4 in MM6 cells, human monocytes and macrophages. Together these results indicate that LPA, mm-LDL and serum induce via activation of the LPA1 receptor a Gi/phospholipase C/Ca2 + signalling pathway in monocytes. Our study is the first report showing the receptor-mediated activation of human monocytic cells by low nanomolar concentrations of LPA and S1P, and suggests a role of these lipid mediators in inflammation and atherogenesis. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Monocytes; Macrophages; Lysophosphatidic acid; Sphingosine-1-phosphate; Atherosclerosis; Calcium; Signal Transduction
1. Introduction Lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) are potent bioactive lipid mediators with specific and multiple cellular effects [1]. They bind to distinct specific G protein-coupled receptors in the plasma membrane [2,3]. For S1P, five receptor subtypes have been identified, whereas LPA binds to three specific receptor subtypes. The cellular responses are determined by the profile and the level of expression of S1P and LPA receptor subtypes that are able to couple to one or multiple heterotrimeric G proteins [3]. LPA and S1P are generated during blood clotting, and mediate physiological wound healing processes such as the stimulation of endothelial cell migration and proliferation that is important in vascular repair [1,3]. Recent experimental evidence indicates that LPA and S1P might also initiate * Corresponding author. Tel.: +49-89-51604380; fax: +49-8951604382. E-mail address: [email protected] (W. Siess).
and perpetuate pathophysiological processes such as inflammation and atherogenesis (reviewed in Ref. [4]). Indeed, LPA is formed by mild oxidation of low-density lipoprotein (LDL) and accumulates in the intima of human atherosclerotic lesions [5]. It stimulates endothelial adhesion molecule expression and monocyte adhesion to endothelial cells, a key process of inflammation and leading to the early atherosclerotic lesion [6]. Moreover, LPA has been reported to stimulate human T-lymphocytes to migrate and to secrete interleukin-2 and matrix metalloproteinases, effects that are dependent on the expression profile of selective LPA receptors [7,8]. LPA and S1P might also have proinflammatory effects on monocytes and macrophages. Indeed it has been reported that nanomolar concentrations of LPA inhibit the apoptosis of murine macrophages, and micromolar concentrations of S1P inhibit the apoptosis of human U937 monoblastic leukemia cells [9,10]. Apart from a previous study reporting the haptotactic migration of human monocytes induced by high micromolar concentrations of LPA [11], the activation of human monocytic cells by LPA and S1P has not been studied.
0898-6568/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 8 9 8 - 6 5 6 8 ( 0 2 ) 0 0 11 7 - 1
368
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
In the present study, we used the Mono Mac 6 (MM6) cell line, which has phenotypic and functional characteristics of mature blood monocytes [12], to study the effects of LPA and S1P on the cytosolic concentration of Ca2 +, which is a universal regulator of cellular functions. We found that MM6 cells, human monocytes and macrophages express a similar profile of LPA receptors and S1P receptors, and that LPA and S1P increase cytosolic Ca2 + in MM6 cells by the stimulation of Ca2 + entry, which is dependent on the activation of phospholipase C and mainly mediated by Gi protein activation. Moreover, activation of MM6 cells by LPA, mm-LDL and serum was caused by stimulation of the LPA1 receptor, suggesting that this LPA receptor subtype might mediate proinflammatory and atherogenic effects of LPA in human monocytes.
2. Material and methods 2.1. Materials Oleoyl-L-a-lysophosphatidic acid (LPA, C18:1, 1-oleoylsn-glycerol-3-phosphate), platelet-activating factor (PAF, h-acetyl-g-O-hexadecyl-a-phosphatidylcholine), N-formylmethionine-leucine-phenylalanine (fMLP), ATP, phospholipase C inhibitor U73122, lipopolysaccharide (LPS) from Escherichia coli serotype 055:B5, ionomycin from Streptomyces conglobatus, pertussis toxin (PTX) from Bordetella pertussis, probenecide, foetal calf serum (FCS), bovine albumin and low-endotoxin human albumin were from Sigma (St. Louis, MO). D-erythro-Sphingosine-1-phosphate, Fura2 acetomethoxymethyl ester (Fura2/AM), and the inhibitor of receptor-mediated calcium entry SKF96365 were obtained from Calbiochem (San Diego, CA). N-palmitoyl-serine phosphoric acid (NP-Ser-PA) was purchased from Biomol (Plymouth Meeting, PA). Dioctylglycerol pyrophosphate (DGPP 8:0) was from Avanti Polar Lipids (Alabaster, AL). N-palmitoyl-tyrosine phosphoric acid (NPTyr-PA) was synthesized as previously described [13]. LDL was isolated in the continuous presence of EDTA as described previously [14]. Minimally modified LDL (mmLDL) was obtained by spontaneous oxidation of native LDL in the presence of EDTA [5]. All LDL concentrations are given in terms of their protein content.
ultrafiltrated through a Gambro 2000 column (Gambro; Hechingen, Germany). LPS treatment of MM6 cells was carried out by adding 10 ng/ml LPS to 2 105/ml cells for 72 h, as described by Aepfelbacher et al. [15]. Under all culture conditions, cell viability was > 95%, as determined by ethidium bromide/ acridine orange staining. 2.3. Isolation of human blood monocytes and macrophage cell culture Purified human blood monocytes were obtained by magnetic cell sorting as described previously [16]. Briefly, human peripheral blood mononuclear cells were recovered from heparinized venous blood of healthy volunteers by density centrifugation in Ficoll Hypaque isolation solution (Biochrom; Berlin, Germany) and washed in ice-cold RPMI 1640 medium. Thereafter, 80 106 cells were incubated for 15 min on ice in 500 Al buffer consisting of 400 Al monocyte buffer (PBS with 5 mM EDTA and 0.2% endotoxin-free human albumin, pH 7.4) and 100 Al anti-CD14 antibodies labelled with magnetic microbeads (Miltenyi Biotec; Bergisch-Gladbach, Germany). Labelled CD14+ cells were passed through a separation column which was placed in a strong permanent magnet (Miltenyi Biotec), and were eluted with monocyte buffer after removing the
2.2. Cell culture of Mono Mac 6 cells Mono Mac 6 (MM6) cells, originally provided by Dr. Ziegler-Heitbrock (Institute of Immunology, University of Munich), were cultured in RPMI 1640 (Sigma) with 1 mM glutamine supplemented with 10% FCS, 1 mM oxalacetate, 1 mM pyruvate, 1 nonessential amino acids solution (Gibco; Carlsbad, CA), 2 penicillin/streptomycin solution (Sigma) and 0.9 Ag/ml insulin (Sigma) using 24well plates (Corning; New York, NY) in a humidified 5% CO2 atmosphere. Before addition of FCS, the medium was
Fig. 1. LPA and S1P increase cytosolic [Ca2 +] in human monocytic cells. (A) Rapid and transient increase in cytosolic [Ca2 +] after exposure of LPStreated MM6 cells to 1 AM of LPA (left tracing) and 5 AM of S1P (right tracing). Homologous desensitization (LPA/LPA and S1P/S1P), but no cross-desensitization (LPA/S1P and vice versa) of the cytosolic Ca2 + increase. (B) Concentration – response curves of LPA and S1P. Values are mean F S.D. of four experiments.
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
column from the magnet. The purity of the monocytes was about 90% as assessed by FACS using FITC-labelled antiCD14 antibodies. For differentiation into macrophages, monocytes were seeded in six-well cell culture dishes (Corning). Cells were allowed to attach to culture dishes for 2 h, and nonadherent cells were removed by gentle washing. Adherent monocytes were cultured for 7 days in RPMI 1640 containing 20% human autologous serum. Medium was changed every 3 –4 days. 2.4. [Ca2+]i measurement in MM6 cells To determine the intracellular calcium concentration [Ca2 +]i, MM6 cells were loaded with the calcium-sensitive fluorescent dye Fura2/AM as described previously [17]. Briefly, cells were washed and resuspended at a density of 2.5 106 cells/ml in cell culture medium without FCS. The
369
membrane-permeable Fura2/AM (final concentration 5 AM) in DMSO ( < 0.1% v/v) or DMSO alone (control for autofluorescence) was added, and cells were incubated at 37 jC for 45 min. Cells were washed and suspended at a density of 4 105 cells/ml in HEPES buffer (20 mM HEPES, 120 mM NaCl, 2.7 mM KCl, 1.4 mM MgSO4, 1.4 mM KH2PO4, 10 mM glucose and 25 mM NaHCO3) supplemented with 0.2% bovine albumin. To reduce the leakage of Fura2 from the cytosol into the extracellular space, 1 mM probenecide, a blocker of organic anion transport, was included [18]. Ca2 +dependent fluorescence in cells was measured using a double-excitation spectrofluorometer Deltascan Model 4000 (Photon Technologies International; South Brunswick, NJ) with an emission wavelength of 510 nm and alternating excitation wavelengths of 340 and 380 nm. The ratio (R) of emission intensities at the excitation wavelengths of 340 and 380 nm was converted into a [Ca2 +]i value using the Grynkiewicz equation [19].
Fig. 2. LPA and S1P stimulate mainly [Ca2 +] entry from the extracellular medium into human monocytic cells. (A, B) Inhibition of the LPA- (1 AM) and S1P(5 AM) induced increases in cytosolic [Ca2 +] by SKF96365 (added 120 s before the agonists). (C) Inhibition of the LPA- (1 AM) and S1P- (5 AM) induced increases in cytosolic [Ca2 +] by EGTA (4 mM) (added 50 s before the agonist). Values in the bar diagrams are mean F S.D. of three experiments. *p < 0.05; **p < 0.01.
370
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
2.5. RNA isolation and reverse transcriptase polymerase chain reaction (RT-PCR) Native and LPS-treated MM6 cells (15 106 each) and two samples of 15 106 human monocytes from two different donors were washed once with PBS and then resuspended in 1 ml of PBS. Cells were lysed by addition of 8 ml of RNAlaterR solution (Ambion; Austin, TX), and the samples were stored at 20 jC. Human macrophages cultured from monocytes from the same two donors were washed twice with PBS and lysed directly in the cell culture dish by addition of TRIzolR RNA isolation reagent (Gibco). The sample, containing 8 106 macrophages in 1 ml of TRIzolR reagent, was stored at 20 jC. Total cellular RNA was extracted using the TRIzolR reagent from suspensions of native and LPS-treated MM6 cells, human monocytes, and primary human macrophages. RT-PCR of LPA receptors and S1P receptors was performed as described previously [20].
3. Results 3.1. LPA and S1P induce an increase in [Ca2+]i in human monocytic cells Addition of LPA and S1P to untreated (native) MM6 cells leads only to a minor increase in [Ca2 +]i (data not
shown). However, in activated MM6 cells pretreated with LPS (10 ng/ml, 72 h), 1 AM LPA elevated cytosolic [Ca2 +] from a resting [Ca2 +]i of 171 F 11 to 398 F 46 nM (n = 4, mean F S.D.). S1P at a concentration of 5 AM caused an increase in [Ca2 +]i from 162 F 9 to 307 F 18 nM (n = 4, mean F S.D.). Upon addition of LPA, [Ca2 +]i showed a rapid increase, reaching a peak within 5– 15 s; thereafter, [Ca2 +]i returned to near resting levels within 1 – 2 min. The S1P-induced Ca2 + signal also started without a significant latency, but the velocity of the rise of [Ca2 +]i was less pronounced than after LPA stimulation; peak levels of [Ca2 +]i were reached as late as 10– 20 s after S1P addition and returned to baseline values after 3 –6 min (Fig. 1A). The LPA- and S1P-induced Ca2 + transients in LPS-treated MM6 cells were subject to homologous desensitization. No crossdesensitization between the [Ca2 +]i responses evoked by LPA and S1P was found (Fig. 1A). These results indicate that S1P has no agonistic activity on the LPA receptors expressed on the surface of MM6 cells and that LPA does not act on S1P receptors on MM6 cells. The concentration dependency of the LPA- and S1Pelicited [Ca2 +]i response in LPS-treated MM6 cells is shown in Fig. 1B. After exposure of the cells to 1 nM LPA, an increase in [Ca2 +]i of 27 F 3 nM was already detectable, and a 1 AM concentration of this mediator caused a maximal increase in [Ca2 +]i. In contrast, approximately 10-fold higher concentrations were required for the same increase in [Ca2 +]i after stimulation with S1P. The
Fig. 3. Inhibition of Gi proteins by PTX reduces the LPA- and S1P-elicited increases in cytosolic [Ca2 +] in human monocytic cells. The cells were pretreated with PTX (100 ng/ml) or vehicle (PBS) for 24 h before stimulation with 1 AM LPA, 5 AM S1P or 100 nM PAF. Values in the bar diagrams are mean F S.D. of three experiments; *p < 0.01.
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
EC50 values for LPA- and S1P-induced Ca2 + mobilization were 47 F 8 and 340 F 26 nM (mean F S.D., n = 4), respectively. 3.2. Characterisation of the LPA- and S1P-stimulated signalling pathways mediating [Ca2+]i transients Both LPA- and S1P-induced increases in [Ca2 +]i were predominantly due to influx of Ca2 + from the extracellular medium. Pretreatment of cells for 2 min with SKF-96365, an inhibitor of receptor-mediated Ca2 + entry [21], reduced in a dose-dependent manner the increase in [Ca2 +]i induced by LPA and S1P (Fig. 2A and B). A 50 AM concentration of SKF96365 reduced the peak level of [Ca2 +]i after LPA and S1P stimulation by 74% and 64%, respectively. Consistent with these results, EGTA (4 mM) added to nominally Ca2 +free buffer 50 s prior to addition of the agonist dramatically reduced the [Ca2 +]i signals induced by LPA and S1P (Fig. 2C). These data indicate that the LPA- and S1P-induced
371
[Ca2 +]i increases result mainly from receptor-mediated Ca2 + influx from the extracellular medium. Receptor-mediated Ca2 + influx is often secondary to 2+ Ca mobilization from internal stores (store-mediated Ca2 + entry) [22]. Ca2 + mobilization upon cell activation is mainly elicited by inositol 1,4,5-trisphosphate that is generated by phospholipase C-mediated hydrolysis of phosphatidylinositol 4,5-bisphosphate. In order to investigate whether the increase in cytosolic Ca2 + after LPA and S1P stimulation was mediated by phospholipase C activation, we used the phospholipase C inhibitor U73122. Pretreatment of the cells with 3 AM U73122 for 3 min reduced the LPA- and S1P-induced increases in [Ca2 +]i in LPS-treated MM6 cells by 75% and 68%, respectively (data not shown). Together, these results indicate that activation of phospholipase C and the subsequent store-mediated Ca2 + entry from the extracellular medium constitute the mechanism responsible for the increase in cytosolic Ca2 + after stimulation by LPA and S1P.
Fig. 4. Inhibition of the LPA- but not the S1P-elicited [Ca2 +]i signal in human monocytic cells by the LPA1 and LPA3 receptor antagonist DGPP 8:0. Cells were pretreated with DGPP 8:0 or solvent (0.2% methanol), and then stimulated with LPA (100 nM) or S1P (100 nM). Values are mean F S.D. of three experiments; *p < 0.01.
372
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
PTX-sensitive Gi proteins, as well as the PTX-insensitive Gq, are known to relay the signal from activated seventransmembrane-domain receptors to the phospholipase Ch. To assess the involvement of heterotrimeric Gi proteins in LPA- and S1P-induced Ca2 + mobilization, cells were pretreated with 100 ng/ml PTX for 24 h. As can be seen in Fig. 3, PTX reduced the LPA- and S1P-induced Ca2 + responses by 72% and 54%, respectively. PTX had no effect on increases in
[Ca2 +]i induced by 100 nM PAF, which is in agreement with previous reports [15]. 3.3. Inhibition of the LPA-induced Ca2+ response by LPA receptor antagonists The synthetic LPA analogues NP-Ser-PA and NP-Tyr-PA inhibit the LPA-stimulated chloride conductance in Xenopus
Fig. 5. mm-LDL and serum increase cytosolic [Ca2 +] in human monocytic cells through the activation of LPA receptors. (A) Cross-desensitization of the mmLDL- (200 Ag/ml) or FCS- (1%) induced Ca2 + signal with LPA (1 AM). (B) Inhibition of the mm-LDL- (200 Ag/ml) or FCS- (0.5%) induced Ca2 + signal by NP-Ser-PA (10 AM). Values are mean F S.D. of two (mm-LDL) or three (FCS) experiments.
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
laevis oocytes and the LPA-induced changes in platelet shape and aggregation [5,23,24]. When tested in human monocytic cells, these two antagonists inhibited the LPAinduced [Ca2 +]i response in a dose-dependent manner but were without any effect on the Ca2 + response elicited by S1P (data not shown). NP-Ser-PA was more potent than NPTyr-PA: NP-Ser-PA (10 AM) reduced the LPA- (100 nM) induced Ca2 + transient by 77%, and NP-Tyr-PA (20 AM) diminished the LPA-induced Ca2 + response by 45%. Both LPA receptor antagonists also showed a weak agonist effect. Dioctylglycerol pyrophosphate (DGPP 8:0) has recently been described as a potent antagonist of the LPA3 and LPA1 receptors [25]. DGPP 8:0 also inhibited the LPA-induced Ca2 + transient in MM6 cells to an extent comparable to that of NP-Ser-PA. DGPP did not affect the Ca2 + response elicited by S1P (Fig. 4). 3.4. mm-LDL and serum activate human monocytic cells through the activation of LPA receptors Minimal oxidation of LDL leads to an increase in its LPA content [5]. mm-LDL induced a small Ca2 +
373
transient in MM6 cells that was specifically desensitized by prior cell activation with LPA. Similarly, when the Ca2 + transient was first activated by the application of mm-LDL, the cells were then desensitized to subsequent exposure of LPA (Fig. 5A, left). Preincubation of MM6 cells with NP-Ser-PA completely abolished the Ca2 + signal induced by mm-LDL (Fig. 5B, left). These results suggest that minimally oxidized LDL activates human monocytic cells through the activation of LPA receptors. Serum contains high nanomolar and micromolar concentrations of S1P and LPA, respectively. LPA and S1P mediate the effects of serum on X. laevis oocytes and endothelial cells, respectively [26,27]. FCS (1%) evoked a transient rise in cytosolic Ca2 + concentration in MM6 cells that showed cross-desensitization with LPA (Fig. 5A, right), but not S1P (data not shown). Furthermore, the Ca2 + signal evoked by serum was drastically reduced by the LPA receptor antagonist NP-Ser-PA (Fig. 5B, right), indicating that serum induces the increase in cytosolic Ca2 + in human monocytic cells by the activation of LPA receptors but not S1P receptors.
Fig. 6. RT-PCR analysis of mRNAs encoding LPA receptors and S1P receptors in MM6 cells, human monocytes and macrophages. (A) LPA receptors and S1P receptors in LPS-treated (lane 1) and native (lane 2) MM6 cells. (B) LPA receptors and S1P receptors in macrophages (lane 1) and monocytes (lane 2) of donor A. (C) LPA receptors and S1P receptors in macrophages (lane 1) and monocytes (lane 2) of donor B.
374
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
3.5. RT-PCR analysis of LPA receptors and S1P receptors in MM6 cells, human monocytes and macrophages The various LPA receptors and S1P receptors expressed in MM6 cells, human monocytes and macrophages were analysed using RT-PCR. The pattern of LPA receptor and S1P receptor expression was very similar in all three types of cells studied. The cells expressed predominantly the LPA receptors LPA1 and LPA2 and the S1P receptors S1P2 and S1P4 (Fig. 6). There was no important difference in the expression of LPA receptor and S1P receptor transcripts between MM6 cells, human monocytes and macrophages. No major differences in the pattern of receptor expression were noted between donors.
4. Discussion The results of this study indicate that nanomolar concentrations of LPA and S1P elicit [Ca2 +]i transients in human monocytic cells. In monocytes, the second messenger Ca2 + regulates diverse biological processes such as the secretion of cytokines and the expression of proinflammatory genes [28,29]. So far, only a few studies have been published concerning the actions of LPA and S1P on monocytes and macrophages. In a previous study, rather high concentrations of LPA (0.1 – 0.3 mM) were reported to stimulate a rise in cytosolic Ca2 + and the haptotactic migration of human monocytes isolated from peripheral human blood [11]. Since in this study the peripheral blood mononuclear cells were collected and resuspended in medium containing FCS, which is known to contain micromolar concentrations of LPA, the cells might have been partially desensitized and refractory towards LPA stimulation [20]. In a further study, LPA was identified as a major serum survival factor for murine macrophages [9]. In this study, peritoneal macrophages of mice were studied, and low concentrations (>50 nM) of LPA were sufficient to inhibit apoptosis of the cells. Balazs et al. [30] have reported that topical application of LPA to skin wounds in rats elicited a rapid and massive accumulation of tissue macrophages and suggested that the histiocytic response was responsible in part for the accelerated wound healing. Effects of S1P on human or murine monocytes or macrophages have not previously been reported. However, it has been shown that S1P inhibits apoptosis, mobilizes intracellular calcium and activates the transcription factor NFnB in human U937 monoblastic cells [10], suggesting also a proinflammatory action of S1P on monocytes. Our results showing homologous but not heterologous desensitization of the Ca2 + signal indicate that LPA and S1P activate specifically their respective receptors on human monocytic cells. Human MM6 cells, peripheral monocytes and macrophages were found to express a similar profile of LPA receptors and S1P receptors. The MM6 cells expressed the LPA receptor subtypes LPA1 and LPA2 and the S1P
receptor subtypes S1P2 and S1P4. In heterologous expression studies, LPA receptors and S1P receptors have been found to couple to multiple G proteins [1– 3]. The LPA1, LPA2 and S1P2 receptors can each activate three G protein families: the Gi, Gq and G12/13 proteins. The G proteincoupling of the S1P4 receptor, which has been identified recently and found to be expressed in haematopoetic cells and lymphoid cells, has not yet been studied [3,31]. Activation of Gi as well as Gq proteins can lead to the generation of the second-messenger Ca2 +: hg subunits liberated after Gi activation, or Gaq could activate phospholipase Ch [32,33]. Our studies employing PTX, which inhibits the Gi protein family, indicate that the Ca2 + signal evoked by activation of the LPA receptors was mainly caused by Gi protein activation, whereas 50% of the Ca2 + signal evoked by activation of the S1P receptors was mediated by G proteins other than Gi. Our experiments using the phospholipase C inhibitor U73122 and the inhibitor of receptor-mediated Ca2 + entry, SKF-96365, which abolished the [Ca2 +]i transients in response to LPA and S1P, support the hypothesis that activation of phospholipase C and the ensuing store-mediated Ca2 + entry from the extracellular medium is the mechanism for the increase in cytosolic Ca2 + in MM6 cells. Because the short-chain phosphatidate analogue DGPP 8:0, a potent antagonist of the LPA1 and LPA3 receptors [25], inhibited the LPA-induced Ca2 + response, we suggest that the increase in cytosolic Ca2 + in MM6 cells induced by LPA is mediated through the stimulation of the LPA1 receptor, since the LPA3 receptor could not be detected in these cells. Monocytes are the main cellular players responsible for the inflammatory response underlying atherosclerosis [34]. Depending on the extent of its oxidative modification, oxidized LDL has different effects on human monocytes. Whereas heavily oxidized LDL binds to and is taken up by scavenger receptors on macrophages, and transforms macrophages to foam cells [34], mildly oxidized LDL such as mm-LDL escapes this scavenger receptor-mediated uptake. It behaves like a molecular ‘‘Trojan horse’’ bearing a cargo of phospholipid oxidation products that activate cells through the activation of physiological receptors [35,36]. mm-LDL contains an increased amount of LPA, and the LPA is apparently exposed on the surface of oxidized LDL particles in such a way that it can activate LPA receptors, thereby causing activation of platelets and endothelial cells [5]. In the present study, we found that mm-LDL also activates human monocytic cells through the activation of LPA receptors. It will be of considerable interest to find out whether other proinflammatory activities of oxidatively modified LDL on human monocytes, such as the stimulation of matrix metalloproteinase secretion and chemotaxis, are also mediated by the activation of LPA receptors [34,37]. In this context, LPA has been shown to stimulate the activity of matrix metalloproteinase of human lymphocytes [8]. Serum contains higher levels of LPA and S1P as compared to plasma, which is explained by the release of
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
each of these lipid mediators from platelets activated during clotting [38,39]. The serum concentrations of S1P and LPA are in the high nanomolar and low micromolar range [40,41]. S1P induces chemotaxis, survival and proliferation of endothelial cells, and has recently been identified as the main angiogenic factor present in serum [27]. In our study, serum activated human monocytic cells through the activation of LPA receptors but not S1P receptors. Since LPA was 10 times more potent than S1P in evoking the Ca2 + signal in human monocytic cells, the higher sensitivity of these cells toward LPA might explain our finding that LPA, and not S1P, was the monocyte-stimulating substance in serum. The stimulation of human monocytic cells by low nanomolar concentrations of LPA and S1P as shown in this study suggests that these substances are proinflammatory mediators. LPA and S1P locally released from activated platelets upon endothelial injury might attract and activate circulating human monocytes at the site of vascular lesions, thereby propagating inflammation and atherogenesis. LPA that accumulates in atherosclerotic lesions and oxidized LDL particles might further reinforce macrophage activation in the vascular intima. Since stimulation of the LPA1 receptor was found to be critical for monocyte activation by LPA, mm-LDL and serum, we suggest that the use of subtype selective LPA-receptor antagonists might be a strategy to attenuate inflammation and atherogenesis.
Acknowledgements This work was supported by the Graduate Program ‘‘Vascular Biology in Medicine’’ of the Deutsche Forschungsgemeinschaft (GRK 438) and the August-LenzStiftung, The results are part of the thesis of M. Fueller at the University of Munich.
References [1] Tigyi G. Prostaglandins Other Lipid Mediat 2001;64:47 – 62. [2] Goetzl EJ, An S. FASEB J 1998;12:1589 – 98. [3] Hla T, Lee MJ, Ancellin N, Paik JH, Kluk MJ. Science 2001; 294:1875 – 8. [4] Siess W. Biochim Biophys Acta 2002;1582:204 – 15. [5] Siess W, Zangl KJ, Essler M, Bauer M, Brandl R, Corrinth C, et al. Proc Natl Acad Sci U S A 1999;96:6931 – 6. [6] Rizza C, Leitinger N, Yue J, Fischer DJ, Wang DA, Shih PT, et al. Lab Invest 1999;79:1227 – 35. [7] Goetzl EJ, Kong Y, Voice JK. J Immunol 2000;164:4996 – 9. [8] Zheng Y, Kong Y, Goetzl EJ. J Immunol 2001;166:2317 – 22.
375
[9] Koh JS, Lieberthal W, Heydrick S, Levine JS. J Clin Invest 1998;102:716 – 27. [10] Cuvillier O, Pirianov G, Kleuser B, Vanek PG, Coso OA, Gutkind S, et al. Nature 1996;381:800 – 3. [11] Zhou D, Luini W, Bernasconi S, Diomede L, Salmona M, Mantovani A, et al. J Biol Chem 1995;270:25549 – 56. [12] Ziegler-Heitbrock HW, Thiel E, Futterer A, Herzog V, Wirtz A, Riethmuller G. Int J Cancer 1988;41:456 – 61. [13] Bittman R, Swords B, Liliom K, Tigyi G. J Lipid Res 1996;37:391 – 8. [14] Weidtmann A, Scheithe R, Hrboticky N, Pietsch A, Lorenz R, Siess W. Arterioscler Thromb Vasc Biol 1995;15:1131 – 8. [15] Aepfelbacher M, Ziegler-Heitbrock HW, Lux I, Weber PC. J Immunol 1992;148:2186 – 93. [16] Aepfelbacher M, Essler M, Huber E, Czech A, Weber PC. J Immunol 1996;157:5070 – 5. [17] Weber C, Aepfelbacher M, Haag H, Ziegler-Heitbrock HW, Weber PC. Eur J Immunol 1993;23:852 – 9. [18] Di Virgilio F, Steinberg TH, Silverstein SC. Cell Calcium 1990; 11:57 – 62. [19] Grynkiewicz G, Poenie M, Tsien RY. J Biol Chem 1985;260:3440 – 50. [20] Fischer DJ, Liliom K, Guo Z, Nusser N, Virag T, Murakami MK, et al. Mol Pharmacol 1998;54:979 – 88. [21] Merritt JE, Armstrong WP, Benham CD, Hallam TJ, Jacob R, JaxaChamiec A, et al. Biochem J 1990;271:515 – 22. [22] Parekh AB, Penner R. Physiol Rev 1997;77:901 – 30. [23] Liliom K, Bittman R, Swords B, Tigyi G. Mol Pharmacol 1996; 50:616 – 23. [24] Sugiura T, Tokumura A, Gregory L, Nouchi T, Weintraub ST, Hanahan DJ. Arch Biochem Biophys 1994;311:358 – 68. [25] Fischer DJ, Nusser N, Virag T, Yokoyama K, Wang D, Baker DL, et al. Mol Pharmacol 2001;60:776 – 84. [26] Tigyi G, Miledi R. J Biol Chem 1992;267:21360 – 7. [27] English D, Welch Z, Kovala AT, Harvey K, Volpert OV, Brindley DN, et al. FASEB J 2000;14:2255 – 65. [28] Jefferson KK, Smith MF, Bobak DA. J Immunol 1999;163:5183 – 91. [29] Kuhn H, Heydeck D, Brinckman R, Trebus F. Lipids [Suppl] 1999; 34:S273 – 9. [30] Balazs L, Okolicany J, Ferrebee M, Tolley B, Tigyi G. Am J Physiol, Regul Integr Comp Physiol 2001;280:R466 – 72. [31] Graler MH, Bernhardt G, Lipp M. Genomics 1998;53:164 – 9. [32] Camps M, Carozzi A, Schnabel P, Scheer A, Parker PJ, Gierschik P. Nature 1992;360:684 – 6. [33] Lee CH, Park D, Wu D, Rhee SG, Simon MI. J Biol Chem 1992; 267:16044 – 7. [34] Ross R. N Engl J Med 1999;340:115 – 26. [35] Subbanagounder G, Leitinger N, Schwenke DC, Wong JW, Lee H, Rizza C, et al. Arterioscler Thromb Vasc Biol 2000;20:2248 – 54. [36] Essler M, Retzer M, Bauer M, Zangl KJ, Tigyi G, Siess W. Ann NY Acad Sci 2000;905:282 – 6. [37] Xu XP, Meisel SR, Ong JM, Kaul S, Cercek B, Rajavashisth TB, et al. Circulation 1999;99:993 – 8. [38] Eichholtz T, Jalink K, Fahrenfort I, Moolenaar WH. Biochem J 1993;291:677 – 80. [39] Yatomi Y, Igarashi Y, Yang L, Hisano N, Qi R, Asazuma N, et al. J Biochem (Tokyo) 1997;121:969 – 73. [40] Ruwisch L, Schafer-Korting M, Kleuser B. Naunyn-Schmiedeberg’s Arch Pharmacol 2001;363:358 – 63. [41] Baker DL, Desiderio DM, Miller DD, Tolley B, Tigyi GJ. Anal Biochem 2001;292:287 – 95.
Activation of human monocytic cells by lysophosphatidic acid and sphingosine-1-phosphate Markus Fueller a, De An Wang b, Gabor Tigyi b, Wolfgang Siess a,* a
Institut fu¨r Prophylaxe und Epidemiologie der Kreislaufkrankheiten, Klinikum der Universita¨t Mu¨nchen, Pettenkoferstr. 9, D 80336 Munich, Germany b Department of Physiology, University of Tennessee Health Science Center, 894 Union Avenue, Memphis, TN 38163, USA Received 30 June 2002; accepted 26 August 2002
Abstract Lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) are serum-borne lipid mediators with potential proinflammatory and atherogenic properties. We studied the effects of LPA and S1P on [Ca2 +]i, a second messenger of cellular activation, in human monocytic Mono Mac 6 (MM6) cells. LPA and S1P induced [Ca2 +]i transients with EC50 values of 47 and 340 nM, respectively. Ca2 + signals evoked by LPA and S1P originated mainly from the stimulation of Ca2 + entry, were blocked by the phospholipase C inhibitor U73122, and were inhibited by pertussis toxin. The LPA1 and LPA3 receptor antagonist dioctylglycerol pyrophosphate inhibited the LPA-induced Ca2 + signal. Notably, serum and minimally modified LDL (mm-LDL) evoked [Ca2 +]i increases that were mediated entirely through activation of LPA receptors. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis showed the presence of the LPA and S1P receptor subtypes LPA1, LPA2, S1P1, S1P2, S1P4 in MM6 cells, human monocytes and macrophages. Together these results indicate that LPA, mm-LDL and serum induce via activation of the LPA1 receptor a Gi/phospholipase C/Ca2 + signalling pathway in monocytes. Our study is the first report showing the receptor-mediated activation of human monocytic cells by low nanomolar concentrations of LPA and S1P, and suggests a role of these lipid mediators in inflammation and atherogenesis. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Monocytes; Macrophages; Lysophosphatidic acid; Sphingosine-1-phosphate; Atherosclerosis; Calcium; Signal Transduction
1. Introduction Lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) are potent bioactive lipid mediators with specific and multiple cellular effects [1]. They bind to distinct specific G protein-coupled receptors in the plasma membrane [2,3]. For S1P, five receptor subtypes have been identified, whereas LPA binds to three specific receptor subtypes. The cellular responses are determined by the profile and the level of expression of S1P and LPA receptor subtypes that are able to couple to one or multiple heterotrimeric G proteins [3]. LPA and S1P are generated during blood clotting, and mediate physiological wound healing processes such as the stimulation of endothelial cell migration and proliferation that is important in vascular repair [1,3]. Recent experimental evidence indicates that LPA and S1P might also initiate * Corresponding author. Tel.: +49-89-51604380; fax: +49-8951604382. E-mail address: [email protected] (W. Siess).
and perpetuate pathophysiological processes such as inflammation and atherogenesis (reviewed in Ref. [4]). Indeed, LPA is formed by mild oxidation of low-density lipoprotein (LDL) and accumulates in the intima of human atherosclerotic lesions [5]. It stimulates endothelial adhesion molecule expression and monocyte adhesion to endothelial cells, a key process of inflammation and leading to the early atherosclerotic lesion [6]. Moreover, LPA has been reported to stimulate human T-lymphocytes to migrate and to secrete interleukin-2 and matrix metalloproteinases, effects that are dependent on the expression profile of selective LPA receptors [7,8]. LPA and S1P might also have proinflammatory effects on monocytes and macrophages. Indeed it has been reported that nanomolar concentrations of LPA inhibit the apoptosis of murine macrophages, and micromolar concentrations of S1P inhibit the apoptosis of human U937 monoblastic leukemia cells [9,10]. Apart from a previous study reporting the haptotactic migration of human monocytes induced by high micromolar concentrations of LPA [11], the activation of human monocytic cells by LPA and S1P has not been studied.
0898-6568/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 8 9 8 - 6 5 6 8 ( 0 2 ) 0 0 11 7 - 1
368
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
In the present study, we used the Mono Mac 6 (MM6) cell line, which has phenotypic and functional characteristics of mature blood monocytes [12], to study the effects of LPA and S1P on the cytosolic concentration of Ca2 +, which is a universal regulator of cellular functions. We found that MM6 cells, human monocytes and macrophages express a similar profile of LPA receptors and S1P receptors, and that LPA and S1P increase cytosolic Ca2 + in MM6 cells by the stimulation of Ca2 + entry, which is dependent on the activation of phospholipase C and mainly mediated by Gi protein activation. Moreover, activation of MM6 cells by LPA, mm-LDL and serum was caused by stimulation of the LPA1 receptor, suggesting that this LPA receptor subtype might mediate proinflammatory and atherogenic effects of LPA in human monocytes.
2. Material and methods 2.1. Materials Oleoyl-L-a-lysophosphatidic acid (LPA, C18:1, 1-oleoylsn-glycerol-3-phosphate), platelet-activating factor (PAF, h-acetyl-g-O-hexadecyl-a-phosphatidylcholine), N-formylmethionine-leucine-phenylalanine (fMLP), ATP, phospholipase C inhibitor U73122, lipopolysaccharide (LPS) from Escherichia coli serotype 055:B5, ionomycin from Streptomyces conglobatus, pertussis toxin (PTX) from Bordetella pertussis, probenecide, foetal calf serum (FCS), bovine albumin and low-endotoxin human albumin were from Sigma (St. Louis, MO). D-erythro-Sphingosine-1-phosphate, Fura2 acetomethoxymethyl ester (Fura2/AM), and the inhibitor of receptor-mediated calcium entry SKF96365 were obtained from Calbiochem (San Diego, CA). N-palmitoyl-serine phosphoric acid (NP-Ser-PA) was purchased from Biomol (Plymouth Meeting, PA). Dioctylglycerol pyrophosphate (DGPP 8:0) was from Avanti Polar Lipids (Alabaster, AL). N-palmitoyl-tyrosine phosphoric acid (NPTyr-PA) was synthesized as previously described [13]. LDL was isolated in the continuous presence of EDTA as described previously [14]. Minimally modified LDL (mmLDL) was obtained by spontaneous oxidation of native LDL in the presence of EDTA [5]. All LDL concentrations are given in terms of their protein content.
ultrafiltrated through a Gambro 2000 column (Gambro; Hechingen, Germany). LPS treatment of MM6 cells was carried out by adding 10 ng/ml LPS to 2 105/ml cells for 72 h, as described by Aepfelbacher et al. [15]. Under all culture conditions, cell viability was > 95%, as determined by ethidium bromide/ acridine orange staining. 2.3. Isolation of human blood monocytes and macrophage cell culture Purified human blood monocytes were obtained by magnetic cell sorting as described previously [16]. Briefly, human peripheral blood mononuclear cells were recovered from heparinized venous blood of healthy volunteers by density centrifugation in Ficoll Hypaque isolation solution (Biochrom; Berlin, Germany) and washed in ice-cold RPMI 1640 medium. Thereafter, 80 106 cells were incubated for 15 min on ice in 500 Al buffer consisting of 400 Al monocyte buffer (PBS with 5 mM EDTA and 0.2% endotoxin-free human albumin, pH 7.4) and 100 Al anti-CD14 antibodies labelled with magnetic microbeads (Miltenyi Biotec; Bergisch-Gladbach, Germany). Labelled CD14+ cells were passed through a separation column which was placed in a strong permanent magnet (Miltenyi Biotec), and were eluted with monocyte buffer after removing the
2.2. Cell culture of Mono Mac 6 cells Mono Mac 6 (MM6) cells, originally provided by Dr. Ziegler-Heitbrock (Institute of Immunology, University of Munich), were cultured in RPMI 1640 (Sigma) with 1 mM glutamine supplemented with 10% FCS, 1 mM oxalacetate, 1 mM pyruvate, 1 nonessential amino acids solution (Gibco; Carlsbad, CA), 2 penicillin/streptomycin solution (Sigma) and 0.9 Ag/ml insulin (Sigma) using 24well plates (Corning; New York, NY) in a humidified 5% CO2 atmosphere. Before addition of FCS, the medium was
Fig. 1. LPA and S1P increase cytosolic [Ca2 +] in human monocytic cells. (A) Rapid and transient increase in cytosolic [Ca2 +] after exposure of LPStreated MM6 cells to 1 AM of LPA (left tracing) and 5 AM of S1P (right tracing). Homologous desensitization (LPA/LPA and S1P/S1P), but no cross-desensitization (LPA/S1P and vice versa) of the cytosolic Ca2 + increase. (B) Concentration – response curves of LPA and S1P. Values are mean F S.D. of four experiments.
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
column from the magnet. The purity of the monocytes was about 90% as assessed by FACS using FITC-labelled antiCD14 antibodies. For differentiation into macrophages, monocytes were seeded in six-well cell culture dishes (Corning). Cells were allowed to attach to culture dishes for 2 h, and nonadherent cells were removed by gentle washing. Adherent monocytes were cultured for 7 days in RPMI 1640 containing 20% human autologous serum. Medium was changed every 3 –4 days. 2.4. [Ca2+]i measurement in MM6 cells To determine the intracellular calcium concentration [Ca2 +]i, MM6 cells were loaded with the calcium-sensitive fluorescent dye Fura2/AM as described previously [17]. Briefly, cells were washed and resuspended at a density of 2.5 106 cells/ml in cell culture medium without FCS. The
369
membrane-permeable Fura2/AM (final concentration 5 AM) in DMSO ( < 0.1% v/v) or DMSO alone (control for autofluorescence) was added, and cells were incubated at 37 jC for 45 min. Cells were washed and suspended at a density of 4 105 cells/ml in HEPES buffer (20 mM HEPES, 120 mM NaCl, 2.7 mM KCl, 1.4 mM MgSO4, 1.4 mM KH2PO4, 10 mM glucose and 25 mM NaHCO3) supplemented with 0.2% bovine albumin. To reduce the leakage of Fura2 from the cytosol into the extracellular space, 1 mM probenecide, a blocker of organic anion transport, was included [18]. Ca2 +dependent fluorescence in cells was measured using a double-excitation spectrofluorometer Deltascan Model 4000 (Photon Technologies International; South Brunswick, NJ) with an emission wavelength of 510 nm and alternating excitation wavelengths of 340 and 380 nm. The ratio (R) of emission intensities at the excitation wavelengths of 340 and 380 nm was converted into a [Ca2 +]i value using the Grynkiewicz equation [19].
Fig. 2. LPA and S1P stimulate mainly [Ca2 +] entry from the extracellular medium into human monocytic cells. (A, B) Inhibition of the LPA- (1 AM) and S1P(5 AM) induced increases in cytosolic [Ca2 +] by SKF96365 (added 120 s before the agonists). (C) Inhibition of the LPA- (1 AM) and S1P- (5 AM) induced increases in cytosolic [Ca2 +] by EGTA (4 mM) (added 50 s before the agonist). Values in the bar diagrams are mean F S.D. of three experiments. *p < 0.05; **p < 0.01.
370
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
2.5. RNA isolation and reverse transcriptase polymerase chain reaction (RT-PCR) Native and LPS-treated MM6 cells (15 106 each) and two samples of 15 106 human monocytes from two different donors were washed once with PBS and then resuspended in 1 ml of PBS. Cells were lysed by addition of 8 ml of RNAlaterR solution (Ambion; Austin, TX), and the samples were stored at 20 jC. Human macrophages cultured from monocytes from the same two donors were washed twice with PBS and lysed directly in the cell culture dish by addition of TRIzolR RNA isolation reagent (Gibco). The sample, containing 8 106 macrophages in 1 ml of TRIzolR reagent, was stored at 20 jC. Total cellular RNA was extracted using the TRIzolR reagent from suspensions of native and LPS-treated MM6 cells, human monocytes, and primary human macrophages. RT-PCR of LPA receptors and S1P receptors was performed as described previously [20].
3. Results 3.1. LPA and S1P induce an increase in [Ca2+]i in human monocytic cells Addition of LPA and S1P to untreated (native) MM6 cells leads only to a minor increase in [Ca2 +]i (data not
shown). However, in activated MM6 cells pretreated with LPS (10 ng/ml, 72 h), 1 AM LPA elevated cytosolic [Ca2 +] from a resting [Ca2 +]i of 171 F 11 to 398 F 46 nM (n = 4, mean F S.D.). S1P at a concentration of 5 AM caused an increase in [Ca2 +]i from 162 F 9 to 307 F 18 nM (n = 4, mean F S.D.). Upon addition of LPA, [Ca2 +]i showed a rapid increase, reaching a peak within 5– 15 s; thereafter, [Ca2 +]i returned to near resting levels within 1 – 2 min. The S1P-induced Ca2 + signal also started without a significant latency, but the velocity of the rise of [Ca2 +]i was less pronounced than after LPA stimulation; peak levels of [Ca2 +]i were reached as late as 10– 20 s after S1P addition and returned to baseline values after 3 –6 min (Fig. 1A). The LPA- and S1P-induced Ca2 + transients in LPS-treated MM6 cells were subject to homologous desensitization. No crossdesensitization between the [Ca2 +]i responses evoked by LPA and S1P was found (Fig. 1A). These results indicate that S1P has no agonistic activity on the LPA receptors expressed on the surface of MM6 cells and that LPA does not act on S1P receptors on MM6 cells. The concentration dependency of the LPA- and S1Pelicited [Ca2 +]i response in LPS-treated MM6 cells is shown in Fig. 1B. After exposure of the cells to 1 nM LPA, an increase in [Ca2 +]i of 27 F 3 nM was already detectable, and a 1 AM concentration of this mediator caused a maximal increase in [Ca2 +]i. In contrast, approximately 10-fold higher concentrations were required for the same increase in [Ca2 +]i after stimulation with S1P. The
Fig. 3. Inhibition of Gi proteins by PTX reduces the LPA- and S1P-elicited increases in cytosolic [Ca2 +] in human monocytic cells. The cells were pretreated with PTX (100 ng/ml) or vehicle (PBS) for 24 h before stimulation with 1 AM LPA, 5 AM S1P or 100 nM PAF. Values in the bar diagrams are mean F S.D. of three experiments; *p < 0.01.
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
EC50 values for LPA- and S1P-induced Ca2 + mobilization were 47 F 8 and 340 F 26 nM (mean F S.D., n = 4), respectively. 3.2. Characterisation of the LPA- and S1P-stimulated signalling pathways mediating [Ca2+]i transients Both LPA- and S1P-induced increases in [Ca2 +]i were predominantly due to influx of Ca2 + from the extracellular medium. Pretreatment of cells for 2 min with SKF-96365, an inhibitor of receptor-mediated Ca2 + entry [21], reduced in a dose-dependent manner the increase in [Ca2 +]i induced by LPA and S1P (Fig. 2A and B). A 50 AM concentration of SKF96365 reduced the peak level of [Ca2 +]i after LPA and S1P stimulation by 74% and 64%, respectively. Consistent with these results, EGTA (4 mM) added to nominally Ca2 +free buffer 50 s prior to addition of the agonist dramatically reduced the [Ca2 +]i signals induced by LPA and S1P (Fig. 2C). These data indicate that the LPA- and S1P-induced
371
[Ca2 +]i increases result mainly from receptor-mediated Ca2 + influx from the extracellular medium. Receptor-mediated Ca2 + influx is often secondary to 2+ Ca mobilization from internal stores (store-mediated Ca2 + entry) [22]. Ca2 + mobilization upon cell activation is mainly elicited by inositol 1,4,5-trisphosphate that is generated by phospholipase C-mediated hydrolysis of phosphatidylinositol 4,5-bisphosphate. In order to investigate whether the increase in cytosolic Ca2 + after LPA and S1P stimulation was mediated by phospholipase C activation, we used the phospholipase C inhibitor U73122. Pretreatment of the cells with 3 AM U73122 for 3 min reduced the LPA- and S1P-induced increases in [Ca2 +]i in LPS-treated MM6 cells by 75% and 68%, respectively (data not shown). Together, these results indicate that activation of phospholipase C and the subsequent store-mediated Ca2 + entry from the extracellular medium constitute the mechanism responsible for the increase in cytosolic Ca2 + after stimulation by LPA and S1P.
Fig. 4. Inhibition of the LPA- but not the S1P-elicited [Ca2 +]i signal in human monocytic cells by the LPA1 and LPA3 receptor antagonist DGPP 8:0. Cells were pretreated with DGPP 8:0 or solvent (0.2% methanol), and then stimulated with LPA (100 nM) or S1P (100 nM). Values are mean F S.D. of three experiments; *p < 0.01.
372
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
PTX-sensitive Gi proteins, as well as the PTX-insensitive Gq, are known to relay the signal from activated seventransmembrane-domain receptors to the phospholipase Ch. To assess the involvement of heterotrimeric Gi proteins in LPA- and S1P-induced Ca2 + mobilization, cells were pretreated with 100 ng/ml PTX for 24 h. As can be seen in Fig. 3, PTX reduced the LPA- and S1P-induced Ca2 + responses by 72% and 54%, respectively. PTX had no effect on increases in
[Ca2 +]i induced by 100 nM PAF, which is in agreement with previous reports [15]. 3.3. Inhibition of the LPA-induced Ca2+ response by LPA receptor antagonists The synthetic LPA analogues NP-Ser-PA and NP-Tyr-PA inhibit the LPA-stimulated chloride conductance in Xenopus
Fig. 5. mm-LDL and serum increase cytosolic [Ca2 +] in human monocytic cells through the activation of LPA receptors. (A) Cross-desensitization of the mmLDL- (200 Ag/ml) or FCS- (1%) induced Ca2 + signal with LPA (1 AM). (B) Inhibition of the mm-LDL- (200 Ag/ml) or FCS- (0.5%) induced Ca2 + signal by NP-Ser-PA (10 AM). Values are mean F S.D. of two (mm-LDL) or three (FCS) experiments.
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
laevis oocytes and the LPA-induced changes in platelet shape and aggregation [5,23,24]. When tested in human monocytic cells, these two antagonists inhibited the LPAinduced [Ca2 +]i response in a dose-dependent manner but were without any effect on the Ca2 + response elicited by S1P (data not shown). NP-Ser-PA was more potent than NPTyr-PA: NP-Ser-PA (10 AM) reduced the LPA- (100 nM) induced Ca2 + transient by 77%, and NP-Tyr-PA (20 AM) diminished the LPA-induced Ca2 + response by 45%. Both LPA receptor antagonists also showed a weak agonist effect. Dioctylglycerol pyrophosphate (DGPP 8:0) has recently been described as a potent antagonist of the LPA3 and LPA1 receptors [25]. DGPP 8:0 also inhibited the LPA-induced Ca2 + transient in MM6 cells to an extent comparable to that of NP-Ser-PA. DGPP did not affect the Ca2 + response elicited by S1P (Fig. 4). 3.4. mm-LDL and serum activate human monocytic cells through the activation of LPA receptors Minimal oxidation of LDL leads to an increase in its LPA content [5]. mm-LDL induced a small Ca2 +
373
transient in MM6 cells that was specifically desensitized by prior cell activation with LPA. Similarly, when the Ca2 + transient was first activated by the application of mm-LDL, the cells were then desensitized to subsequent exposure of LPA (Fig. 5A, left). Preincubation of MM6 cells with NP-Ser-PA completely abolished the Ca2 + signal induced by mm-LDL (Fig. 5B, left). These results suggest that minimally oxidized LDL activates human monocytic cells through the activation of LPA receptors. Serum contains high nanomolar and micromolar concentrations of S1P and LPA, respectively. LPA and S1P mediate the effects of serum on X. laevis oocytes and endothelial cells, respectively [26,27]. FCS (1%) evoked a transient rise in cytosolic Ca2 + concentration in MM6 cells that showed cross-desensitization with LPA (Fig. 5A, right), but not S1P (data not shown). Furthermore, the Ca2 + signal evoked by serum was drastically reduced by the LPA receptor antagonist NP-Ser-PA (Fig. 5B, right), indicating that serum induces the increase in cytosolic Ca2 + in human monocytic cells by the activation of LPA receptors but not S1P receptors.
Fig. 6. RT-PCR analysis of mRNAs encoding LPA receptors and S1P receptors in MM6 cells, human monocytes and macrophages. (A) LPA receptors and S1P receptors in LPS-treated (lane 1) and native (lane 2) MM6 cells. (B) LPA receptors and S1P receptors in macrophages (lane 1) and monocytes (lane 2) of donor A. (C) LPA receptors and S1P receptors in macrophages (lane 1) and monocytes (lane 2) of donor B.
374
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
3.5. RT-PCR analysis of LPA receptors and S1P receptors in MM6 cells, human monocytes and macrophages The various LPA receptors and S1P receptors expressed in MM6 cells, human monocytes and macrophages were analysed using RT-PCR. The pattern of LPA receptor and S1P receptor expression was very similar in all three types of cells studied. The cells expressed predominantly the LPA receptors LPA1 and LPA2 and the S1P receptors S1P2 and S1P4 (Fig. 6). There was no important difference in the expression of LPA receptor and S1P receptor transcripts between MM6 cells, human monocytes and macrophages. No major differences in the pattern of receptor expression were noted between donors.
4. Discussion The results of this study indicate that nanomolar concentrations of LPA and S1P elicit [Ca2 +]i transients in human monocytic cells. In monocytes, the second messenger Ca2 + regulates diverse biological processes such as the secretion of cytokines and the expression of proinflammatory genes [28,29]. So far, only a few studies have been published concerning the actions of LPA and S1P on monocytes and macrophages. In a previous study, rather high concentrations of LPA (0.1 – 0.3 mM) were reported to stimulate a rise in cytosolic Ca2 + and the haptotactic migration of human monocytes isolated from peripheral human blood [11]. Since in this study the peripheral blood mononuclear cells were collected and resuspended in medium containing FCS, which is known to contain micromolar concentrations of LPA, the cells might have been partially desensitized and refractory towards LPA stimulation [20]. In a further study, LPA was identified as a major serum survival factor for murine macrophages [9]. In this study, peritoneal macrophages of mice were studied, and low concentrations (>50 nM) of LPA were sufficient to inhibit apoptosis of the cells. Balazs et al. [30] have reported that topical application of LPA to skin wounds in rats elicited a rapid and massive accumulation of tissue macrophages and suggested that the histiocytic response was responsible in part for the accelerated wound healing. Effects of S1P on human or murine monocytes or macrophages have not previously been reported. However, it has been shown that S1P inhibits apoptosis, mobilizes intracellular calcium and activates the transcription factor NFnB in human U937 monoblastic cells [10], suggesting also a proinflammatory action of S1P on monocytes. Our results showing homologous but not heterologous desensitization of the Ca2 + signal indicate that LPA and S1P activate specifically their respective receptors on human monocytic cells. Human MM6 cells, peripheral monocytes and macrophages were found to express a similar profile of LPA receptors and S1P receptors. The MM6 cells expressed the LPA receptor subtypes LPA1 and LPA2 and the S1P
receptor subtypes S1P2 and S1P4. In heterologous expression studies, LPA receptors and S1P receptors have been found to couple to multiple G proteins [1– 3]. The LPA1, LPA2 and S1P2 receptors can each activate three G protein families: the Gi, Gq and G12/13 proteins. The G proteincoupling of the S1P4 receptor, which has been identified recently and found to be expressed in haematopoetic cells and lymphoid cells, has not yet been studied [3,31]. Activation of Gi as well as Gq proteins can lead to the generation of the second-messenger Ca2 +: hg subunits liberated after Gi activation, or Gaq could activate phospholipase Ch [32,33]. Our studies employing PTX, which inhibits the Gi protein family, indicate that the Ca2 + signal evoked by activation of the LPA receptors was mainly caused by Gi protein activation, whereas 50% of the Ca2 + signal evoked by activation of the S1P receptors was mediated by G proteins other than Gi. Our experiments using the phospholipase C inhibitor U73122 and the inhibitor of receptor-mediated Ca2 + entry, SKF-96365, which abolished the [Ca2 +]i transients in response to LPA and S1P, support the hypothesis that activation of phospholipase C and the ensuing store-mediated Ca2 + entry from the extracellular medium is the mechanism for the increase in cytosolic Ca2 + in MM6 cells. Because the short-chain phosphatidate analogue DGPP 8:0, a potent antagonist of the LPA1 and LPA3 receptors [25], inhibited the LPA-induced Ca2 + response, we suggest that the increase in cytosolic Ca2 + in MM6 cells induced by LPA is mediated through the stimulation of the LPA1 receptor, since the LPA3 receptor could not be detected in these cells. Monocytes are the main cellular players responsible for the inflammatory response underlying atherosclerosis [34]. Depending on the extent of its oxidative modification, oxidized LDL has different effects on human monocytes. Whereas heavily oxidized LDL binds to and is taken up by scavenger receptors on macrophages, and transforms macrophages to foam cells [34], mildly oxidized LDL such as mm-LDL escapes this scavenger receptor-mediated uptake. It behaves like a molecular ‘‘Trojan horse’’ bearing a cargo of phospholipid oxidation products that activate cells through the activation of physiological receptors [35,36]. mm-LDL contains an increased amount of LPA, and the LPA is apparently exposed on the surface of oxidized LDL particles in such a way that it can activate LPA receptors, thereby causing activation of platelets and endothelial cells [5]. In the present study, we found that mm-LDL also activates human monocytic cells through the activation of LPA receptors. It will be of considerable interest to find out whether other proinflammatory activities of oxidatively modified LDL on human monocytes, such as the stimulation of matrix metalloproteinase secretion and chemotaxis, are also mediated by the activation of LPA receptors [34,37]. In this context, LPA has been shown to stimulate the activity of matrix metalloproteinase of human lymphocytes [8]. Serum contains higher levels of LPA and S1P as compared to plasma, which is explained by the release of
M. Fueller et al. / Cellular Signalling 15 (2003) 367–375
each of these lipid mediators from platelets activated during clotting [38,39]. The serum concentrations of S1P and LPA are in the high nanomolar and low micromolar range [40,41]. S1P induces chemotaxis, survival and proliferation of endothelial cells, and has recently been identified as the main angiogenic factor present in serum [27]. In our study, serum activated human monocytic cells through the activation of LPA receptors but not S1P receptors. Since LPA was 10 times more potent than S1P in evoking the Ca2 + signal in human monocytic cells, the higher sensitivity of these cells toward LPA might explain our finding that LPA, and not S1P, was the monocyte-stimulating substance in serum. The stimulation of human monocytic cells by low nanomolar concentrations of LPA and S1P as shown in this study suggests that these substances are proinflammatory mediators. LPA and S1P locally released from activated platelets upon endothelial injury might attract and activate circulating human monocytes at the site of vascular lesions, thereby propagating inflammation and atherogenesis. LPA that accumulates in atherosclerotic lesions and oxidized LDL particles might further reinforce macrophage activation in the vascular intima. Since stimulation of the LPA1 receptor was found to be critical for monocyte activation by LPA, mm-LDL and serum, we suggest that the use of subtype selective LPA-receptor antagonists might be a strategy to attenuate inflammation and atherogenesis.
Acknowledgements This work was supported by the Graduate Program ‘‘Vascular Biology in Medicine’’ of the Deutsche Forschungsgemeinschaft (GRK 438) and the August-LenzStiftung, The results are part of the thesis of M. Fueller at the University of Munich.
References [1] Tigyi G. Prostaglandins Other Lipid Mediat 2001;64:47 – 62. [2] Goetzl EJ, An S. FASEB J 1998;12:1589 – 98. [3] Hla T, Lee MJ, Ancellin N, Paik JH, Kluk MJ. Science 2001; 294:1875 – 8. [4] Siess W. Biochim Biophys Acta 2002;1582:204 – 15. [5] Siess W, Zangl KJ, Essler M, Bauer M, Brandl R, Corrinth C, et al. Proc Natl Acad Sci U S A 1999;96:6931 – 6. [6] Rizza C, Leitinger N, Yue J, Fischer DJ, Wang DA, Shih PT, et al. Lab Invest 1999;79:1227 – 35. [7] Goetzl EJ, Kong Y, Voice JK. J Immunol 2000;164:4996 – 9. [8] Zheng Y, Kong Y, Goetzl EJ. J Immunol 2001;166:2317 – 22.
375
[9] Koh JS, Lieberthal W, Heydrick S, Levine JS. J Clin Invest 1998;102:716 – 27. [10] Cuvillier O, Pirianov G, Kleuser B, Vanek PG, Coso OA, Gutkind S, et al. Nature 1996;381:800 – 3. [11] Zhou D, Luini W, Bernasconi S, Diomede L, Salmona M, Mantovani A, et al. J Biol Chem 1995;270:25549 – 56. [12] Ziegler-Heitbrock HW, Thiel E, Futterer A, Herzog V, Wirtz A, Riethmuller G. Int J Cancer 1988;41:456 – 61. [13] Bittman R, Swords B, Liliom K, Tigyi G. J Lipid Res 1996;37:391 – 8. [14] Weidtmann A, Scheithe R, Hrboticky N, Pietsch A, Lorenz R, Siess W. Arterioscler Thromb Vasc Biol 1995;15:1131 – 8. [15] Aepfelbacher M, Ziegler-Heitbrock HW, Lux I, Weber PC. J Immunol 1992;148:2186 – 93. [16] Aepfelbacher M, Essler M, Huber E, Czech A, Weber PC. J Immunol 1996;157:5070 – 5. [17] Weber C, Aepfelbacher M, Haag H, Ziegler-Heitbrock HW, Weber PC. Eur J Immunol 1993;23:852 – 9. [18] Di Virgilio F, Steinberg TH, Silverstein SC. Cell Calcium 1990; 11:57 – 62. [19] Grynkiewicz G, Poenie M, Tsien RY. J Biol Chem 1985;260:3440 – 50. [20] Fischer DJ, Liliom K, Guo Z, Nusser N, Virag T, Murakami MK, et al. Mol Pharmacol 1998;54:979 – 88. [21] Merritt JE, Armstrong WP, Benham CD, Hallam TJ, Jacob R, JaxaChamiec A, et al. Biochem J 1990;271:515 – 22. [22] Parekh AB, Penner R. Physiol Rev 1997;77:901 – 30. [23] Liliom K, Bittman R, Swords B, Tigyi G. Mol Pharmacol 1996; 50:616 – 23. [24] Sugiura T, Tokumura A, Gregory L, Nouchi T, Weintraub ST, Hanahan DJ. Arch Biochem Biophys 1994;311:358 – 68. [25] Fischer DJ, Nusser N, Virag T, Yokoyama K, Wang D, Baker DL, et al. Mol Pharmacol 2001;60:776 – 84. [26] Tigyi G, Miledi R. J Biol Chem 1992;267:21360 – 7. [27] English D, Welch Z, Kovala AT, Harvey K, Volpert OV, Brindley DN, et al. FASEB J 2000;14:2255 – 65. [28] Jefferson KK, Smith MF, Bobak DA. J Immunol 1999;163:5183 – 91. [29] Kuhn H, Heydeck D, Brinckman R, Trebus F. Lipids [Suppl] 1999; 34:S273 – 9. [30] Balazs L, Okolicany J, Ferrebee M, Tolley B, Tigyi G. Am J Physiol, Regul Integr Comp Physiol 2001;280:R466 – 72. [31] Graler MH, Bernhardt G, Lipp M. Genomics 1998;53:164 – 9. [32] Camps M, Carozzi A, Schnabel P, Scheer A, Parker PJ, Gierschik P. Nature 1992;360:684 – 6. [33] Lee CH, Park D, Wu D, Rhee SG, Simon MI. J Biol Chem 1992; 267:16044 – 7. [34] Ross R. N Engl J Med 1999;340:115 – 26. [35] Subbanagounder G, Leitinger N, Schwenke DC, Wong JW, Lee H, Rizza C, et al. Arterioscler Thromb Vasc Biol 2000;20:2248 – 54. [36] Essler M, Retzer M, Bauer M, Zangl KJ, Tigyi G, Siess W. Ann NY Acad Sci 2000;905:282 – 6. [37] Xu XP, Meisel SR, Ong JM, Kaul S, Cercek B, Rajavashisth TB, et al. Circulation 1999;99:993 – 8. [38] Eichholtz T, Jalink K, Fahrenfort I, Moolenaar WH. Biochem J 1993;291:677 – 80. [39] Yatomi Y, Igarashi Y, Yang L, Hisano N, Qi R, Asazuma N, et al. J Biochem (Tokyo) 1997;121:969 – 73. [40] Ruwisch L, Schafer-Korting M, Kleuser B. Naunyn-Schmiedeberg’s Arch Pharmacol 2001;363:358 – 63. [41] Baker DL, Desiderio DM, Miller DD, Tolley B, Tigyi GJ. Anal Biochem 2001;292:287 – 95.