Calcium Signaling of Lysophosphatidylethanolamine through LPA1 in Human SH-SY5Y Neuroblastoma Cells.

Lysophosphatidylethanolamine (LPE), a lyso-type metabolite of phosphatidylethanolamine, has been reported to be an intercellular signaling molecule. LPE mobilizes intracellular Ca2+ through G-protein-coupled receptor (GPCR) in some cells types. However, GPCRs for lysophosphatidic acid (LPA) were not implicated in the LPE-mediated activities in LPA GPCR overexpression systems or in SK-OV3 ovarian cancer cells. In the present study, in human SH-SY5Y neuroblastoma cells, experiments with LPA1 antagonists showed LPE induced intracellular Ca2+ increases in an LPA1 GPCR-dependent manner. Furthermore, LPE increased intracellular Ca2+ through pertussis-sensitive G proteins, edelfosine-sensitive-phospholipase C, 2-APB-sensitive IP3 receptors, Ca2+ release from intracellular Ca2+ stores, and subsequent Ca2+ influx across plasma membranes, and LPA acted on LPA1 and LPA2 receptors to induce Ca2+ response in a 2-APB-sensitive and insensitive manner. These findings suggest novel involvements for LPE and LPA in calcium signaling in human SH-SY5Y neuroblastoma cells.

INTRODUCTION

Lysophosphatidylethanolamine (LPE) is a metabolic product from phosphatidylethanolamine (a minor constituent of cell membranes) by phospholipase A2. LPE has an ethanolamine head group linked to a lysophosphatidic acid. LPE is commercially used as a plant bio-regulator to delay leaf and fruit senescence, improve product shelf-life post harvest, and mitigate ethylene-induced process (Cowan, 2009). In addition, LPE appears to have certain roles in organisms other than mammals, for example, in the housefly, LPE has antimicrobial activity (Meylaers ). Furthermore, LPEs isolated from Grifola frondosa were recently reported to exhibit anti-apoptotic activity and to enhance neuronal differentiation via MAPK activation in PC-12 cells (Nishina ).

LPE has been detected in human serum at concentrations of about several hundreds nanograms per ml (Misra, 1965; Makide ), but the physiological significance of plasma LPE remains unknown. LPE has also been shown to play a role in intercellular signaling and in the activation of signaling enzymes (Park ), and has been suggested to act through putative G protein-coupled receptors (GPCRs) (Park , 2013). Furthermore, GPCRs for lysophosphatidic acid (LPA), a serum-derived lipid mediator, have been discovered and named LPA1–6 (Choi and Chun, 2013), and these discoveries resulted in intensive knock-out mouse studies and in the developments of selective agonists and antagonists (Im, 2010). However, few studies have been conducted on LPE GPCRs.

In SK-OV3 and OVCAR-3 ovarian cancer cells, LPE induces several responses, which include increasing intracellular Ca2+ concentration ([Ca2+]i) (Park ), and these responses have been proposed to be mediated through GPCR, but not through GPCRs for LPA (Park ). Actually, LPA GPCRs do not respond to LPE in LPA GPCR overexpression systems (Park ). However, LPE does induce [Ca2+]i increases through LPA1 in MDA-MB-231 breast cancer cells and PC-12 pheochromocytoma cells (Park , 2014a; Lee ). Intracellular Ca2+ signaling has crucial roles in development from fertilization through differentiation to organogenesis (Leclerc ). In the nervous system, Ca2+ signaling plays important roles in the development from neural induction to the proliferation, migration, and differentiation of neural cells (Leclerc ). In the present study, the relation between LPA-induced Ca2+ response and LPE-induced Ca2+ signaling was studied in human SH-SY5Y neuroblastoma cells.

MATERIALS AND METHODS

Materials

1-Oleoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine (18:1 LPE), 1-stearoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine (18:0 LPE), 1-octadecyl-2-hydroxy-sn-glycero-3-phosphoethanolamine (ether-linked 18:0 LPE), 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine (16:0 LPE), 1-oleoyl-2- hydroxy-sn-glycero-3-phosphate (LPA, sodium salt), and VPC32183 were purchased from Avanti Polar Lipids (Alabaster, AL, USA). Fura 2-AM, EGTA, 2-aminoethoxydiphenylborane (2-APB) and pertussis toxin (PTX) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Ki16425 and edelfosine were obtained from Cayman chemical (Ann Arbor, MI, USA). AM-095 was from Chemscene (Monmouth Junction, NJ, USA).

Cell culture

Human SH-SY5Y neuroblastoma cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were cultured at 37°C in a 5% CO2 humidified incubator, and maintained in RPMI 1640 medium (GenDEPOT, Barker, TX, USA) supplemented with 10% (v/v) fetal bovine serum, 100 units/mL penicillin, 50 μg/mL streptomycin, 2 mM glutamine, and 1 mM sodium pyruvate.

Measurement of [Ca2+]i concentrations

Cells were trypsin-digested, allowed to sediment, resuspended in HEPES-buffered medium (HBM), consisting of 20 mM HEPES (pH 7.4), 103 mM NaCl, 4.8 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 0.5 mM CaCl2, 25 mM NaHCO3, and 15 mM glucose, and then incubated for 40 min with 5 μM fura 2-AM. [Ca2+]i levels were estimated by measuring changes in fura-2 fluorescence at an emission wavelength of 510 nm and excitation wavelengths of 340 nm and 380 nm every 0.1 sec using a F4500 fluorescence spectrophotometer (Hitachi, Tokyo, Japan) (Park ). Ratios of fluorescence intensities (λ340/λ380) at these two wavelengths were used as surrogates of [Ca2+]i, as previously described (Park ).

Reverse transcriptase-PCR

To detect the expressions of LPA receptors in SH-SY5Y cells by RT-PCR, first strand cDNA was synthesized using total RNA isolated using Trizol reagent (Invitrogen, Waltham, MA, USA). Synthesized cDNA products and primers for LPA1–6 were subjected to PCR using Promega Go-Taq DNA polymerase (Madison, WI, USA). The primers used to amplify 317, 317, 321, 341, 308, and 247 bps fragments of LPA1–6 and GAPDH were as follows: LPA1 (sense 5′-CAG GAC CCA ATA CTC GGA GA-3′, antisense 5′-GTT GAA AAT GGC CCA GAA GA-3′), LPA2 (sense 5′-TTT CAC TTG AGG GCT GGT TC-3′, antisense 5′-CAT GAG CAG GAA GAC AAG CA-3′), LPA3 (sense 5′-CTC ATG GCC TTC CTC ATC AT-3′, antisense 5′-GCC ATA CAT GTC CTC GTC CT-3′), LPA4 (sense 5′-CTT CGC AAG CCT GCT ACT CT-3′, antisense 5′-GGC TTT GTG GTC AAA GGT GT-3′), LPA5 (sense 5′-TCT CCC GTG TCC TGA CTA CC-3′, antisense 5′-TGA GCA TCA GGA AGA TGC AG-3′), and LPA6 (sense 5′-TGC TCA GTA GTG GCA GCA GT-3′, antisense 5′-CAG GCA GCA GAT TCA TTG TC-3′), and GAPDH (sense 5′-GAG TCA ACG GAT TTG GTC GT-3′, antisense 5′-TTG ATT TTG GAG GGA TCT CG-3′). PCR reactions were performed over 30 cycles of 95°C for 30 s (denaturation), 57°C for 30 s (annealing) for LPA1–6, and 72°C for 30 s (elongation) for GAPDH in an Eppendorf Mastcycler gradient unit (Hamburg, Germany) (Park ). Aliquots of the PCR products (7 μl) obtained were electrophoresed in 1.2% agarose gels and stained with ethidium bromide.

Statistics

The results are expressed as means ± SEs for the indicated numbers of determinations. Significances of differences were determined using the student t test, and significance was accepted for p-values <0.05.

RESULTS

LPE increased [Ca2+]i in SH-SY5Y neuroblastoma cells

Synthetic oleoyl LPE (18:1 LPE) increased [Ca2+]i levels in SH-SY5Y neuroblastoma cells (Fig. 1A) in a concentration-dependent manner (Fig. 1C), and response to LPA was greater than to LPE, but in SH-SY5Y cells LPA and LPE had similar efficacies (Fig. 1B, 1D). Responses were also studied using structurally different LPEs, that is, stearoyl LPE (18:0 LPE), octadecanyl LPE (ether-linked 18:0 LPE), and palmitoyl LPE (16:0 LPE). As shown in Fig. 2, 18:1 LPE, 18:0 LPE, ether-linked 18:0 LPE, and 16:0 LPE induced a [Ca2+]i increase in SY-SY5Y cells, which contrasted to that observed in MDAMB-231 cells, in which oleoyl LPE (18:1 LPE) was the only active LPE. Structure-activity relationships in LPE-responsive cells are addressed in the Discussion.

Fig. 1.

Concentration-dependences of LPE- and LPA-induced [Ca2+]i increases in SH-SY5Y neuroblastoma cells. Representative [Ca2+]i traces of SH-SY5Y cells treated with various concentrations of 18:1 LPE (A) and 18:1 LPA (B). Arrows indicate when lipids were added. Concentration-response curves for LPE (C) and LPA (D) for [Ca2+]i increase in cells. Results are presented as the means ± SEs of three independent experiments.

Fig. 2.

Effects of synthetic LPEs, that is, 18:1 LPE, 18:0 LPE, 18:0 ether-linked LPE, 14:0 LPE, and 16:0 LPE in SH-SY5Y neuroblastoma cells. Representative [Ca2+]i traces of SH-SY5Y cells treated with synthetic LPEs (A). Arrows indicate when lipids were added. The results shown are representative of at least three independent experiments. Ca2+ responses are presented as the means ± SEs of three independent experiments (B).

Heterologous desensitization between LPE- and LPA-induced [Ca2+]i responses

Because previous studies have implicated LPA receptor in LPE-induced Ca2+ signaling in certain cell types, we investigated homologous and heterologous desensitizations of LPE- and LPA-induced [Ca2+]i increases in SH-SY5Y cells. In desensitization experiments, LPE or LPA were pre-treated for 1 min before adding LPE (10 μM) or LPA (10 μM). As shown in Fig. 3A, 3B, LPE pre-treatment blocked LPE-induced [Ca2+]i response by 100%, and LPA pre-treatment attenuated LPA-induced response by 90%, implying homologous desensitization. In addition, LPA pre-treatment attenuated LPE-induced [Ca2+]i response by 90%, and LPE pre-treatment attenuated LPA-induced [Ca2+]i response by 63%, implying heterologous desensitization (Fig. 3). In addition, we examined the expression levels of the six known LPA receptors by RT-PCR in human SH-SY5Y cells. LPA1 and LPA2 were found to be strongly expressed, whereas the other four LPA receptors were not detected (Fig. 3C). These results suggest that LPE acts on LPA1 and/or LPA2 receptors in SH-SY5Y cells.

Fig. 3.

Desensitization of LPE- or LPA-induced [Ca2+]i increase by LPE or LPA and expression analysis of six LPA receptors in SH-SY5Y neuroblastoma cells. [Ca2+]i levels in SH-SY5Y cells pre-treated with 10 μM of LPE or 10 μM LPA were monitored after treating them with 10 μM LPE or 10 μM LPA (A). Arrows indicate when lipids were added. Results are representative of at least three independent experiments. [Ca2+]i increases induced by 10 μM of LPE or 10 μM LPA alone and [Ca2+]i increases induced by lipids after pretreating SH-SY5Y cells with LPE or LPA. Results are the means ± SEs of three independent experiments (B). Statistical significance: ***p<0.001 vs. non-pretreated cells. (C) RT-PCR was performed using total mRNA from SH-SY5Y cells. Results are representative of three independent experiments that yielded similar results.

Effects of LPA antagonists on LPE- or LPA-induced [Ca2+]i responses

Three pharmacological tools were applied to investigate the involvements of LPA receptors in SH-SY5Y cells, that is, structurally different antagonists of LPA1 and LPA3 (Ki16425 and VPC32183) (Heise ; Ohta ) and a selective LPA1 antagonist, AM-095 (Castelino ; Swaney ). As shown in Fig. 4, Ki16425 (10 μM), VPC32183 (1 μM), and AM-095 (500 nM) completely inhibited LPE-induced [Ca2+]i response, whereas Ki16425 and AM-095 inhibited it by more than 50%, and VPC32183 had no effect (Fig. 4). Thus, it appeared LPE increased [Ca2+]i through LPA1 receptors in SH-SY5Y cells, whereas LPA increased [Ca2+]i through AM-095/Ki16425-sensitive LPA1 and AM-095/ Ki16425-insensitive LPA2 receptors.

Fig. 4.

Effects of Ki16425, VPC32183, and AM-095 on LPE- or LPA-induced [Ca2+]i increase in SH-SY5Y neuroblastoma cells. Representative [Ca2+]i traces of SH-SY5Y cells treated with 10 μM of LPE or 10 μM LPA in the presence of Ki16425, VPC32183, AM-095, or vehicle (A). Arrows indicate when lipids were added. The results shown are representative of more than three independent experiments. Increases in [Ca2+]i by LPE (10 μM) or LPA (10 μM) were observed in cells pre-treated with or without Ki16425 (10 μM), VPC32183 (1 μM), or AM-095 (500 nM). Results are presented as the means ± SEs of three independent experiments (B). Statistical significance: *p<0.05, **p<0.01, ***p<0.001 vs. non-treated cells.

Effects of PTX, edelfosine, 2-APB, EGTA or HA130 on LPE-or LPA-induced [Ca2+]i responses

To investigate cascades signaling LPE and LPA [Ca2+]i responses, SH-SY5Y cells were treated with specific inhibitors or blockers of Gi/o-type G proteins, phospholipase C, inositol 1,4,5-trisphosphate receptor (IP3R), extracellular Ca2+, or autotaxin, that is, pertussis toxin (PTX), edelfosine, 2-APB, EGTA, and HA130, respectively (Park ; Melchior and Frangos, 2012; Zhang ). As shown in Fig. 5, PTX, a specific inhibitor of Gi/o type G proteins, inhibited [Ca2+]i responses to LPE by 84% and to LPA by 67%, suggesting the involvements of Gi/o proteins in [Ca2+]i responses to LPE and LPA (Fig. 5). In addition, edelfosine (a specific inhibitor of phospholipase C) also partially inhibited responses to LPE and LPA, suggesting the involvement of phospholipase C in these responses (Fig. 5). Next, the involvement of IP3 receptor on Ca2+ release from endoplasmic reticulum (ER) was tested using 2-APB, a specific inhibitor of IP3R. Pretreatment with 2-APB inhibited completely LPE-induced [Ca2+]i increase, but only partly inhibited LPA-induced [Ca2+]i increase (Fig. 5). To investigate the possibility that Ca2+ influx across the plasma membrane contributed to Ca2+ response, we pretreated SHSY5Y cells with EGTA (an extracellular Ca2+ chelator). EGTA partially inhibited LPE- and LPA-induced [Ca2+]i increases, suggesting that Ca2+ influx across the plasma membrane contributed to observed [Ca2+]i increases. Because LPE could not induce Ca2+ increase even when Ca2+ ions were present in extracellular media in the presence of 2-APB, we supposed LPE-induced Ca2+ influx was mediated solely by IP3 receptor-mediated Ca2+ release. However, the partial inhibition of LPA-induced Ca2+ increase by 2-APB suggested another component, insensitive to 2-APB, signaled Ca2+ influx for LPA. These results suggest involvements of Gi/o-type proteins, phospholipase C, IP3 receptors, Ca2+ release from intracellular Ca2+ stores, and Ca2+ influx across the plasma membrane in LPE-and LPA-induced [Ca2+]i increases in SH-SY5Y cells.

Fig. 5.

Effects of EGTA, PTX, 2-APB, and edelfosine on LPE- and LPA-induced [Ca2+]i increases in SH-SY5Y neuroblastoma cells. [Ca2+]i levels in cells pre-treated with or without EGTA (5 mM, 1 min), PTX (100 ng/mL, 24 h), 2-APB (100 μM, 15 min), or edelfosine (10 μM, 6 h) were monitored after being treated with LPE (10 μM) or LPA (10 μM). (A, C) Representative [Ca2+]i traces of SH-SY5Y cells treated with 10 μM of LPE or 10 μM LPA in the presence of PTX, edelfosine, 2-APB, EGTA, or vehicle. Arrows indicate when lipids were added. The results shown are representative of more than three independent experiments. (B, D) Increases in [Ca2+]i by LPE (10 μM) or LPA (10 μM) were observed in cells pre-treated with or without PTX, edelfosine, 2-APB, or EGTA. Results are presented as the means ± SEs of three independent experiments (B). Statistical significance: *p<0.05, **p<0.01, ***p<0.001 vs. non-treated cells.

To determine whether LPE is converted to LPA by autotaxin (also known as lysophospholipase D), and this LPA mediates the action of LPE, we pretreated SH-SY5Y cells with HA130 (a specific inhibitor of autotaxin). However, HA130 did not inhibit LPE-induced Ca2+ increase, indicating that autotaxin was not responsible for the observed effects of LPE (Fig. 6).

Fig. 6.

Effects of HA130 on LPE- and LPA-induced Ca2+ responses. [Ca2+]i levels in SH-SY5Y cells pre-treated with or without HA130 (5 μM, 5 min; an autotaxin inhibitor) were monitored after being treated with LPE (10 μM) or LPA (10 μM) (A). Arrows indicate when lipids were added. The results shown are representative of more than three independent experiments. (B) Increases in [Ca2+]i by LPE (10 μM) or LPA (10 μM) were observed in cells pre-treated with or without HA130 (5 μM, 5 min). Results are presented as the means ± SEs of three independent experiments. NS, statistical non-significant.

DISCUSSION

In the present study, LPE-induced [Ca2+]i increase was found to be mediated through LPA1 in SH-SY5Y cells. Five results sustain this finding: 1) the observed heterologous desensitization found for LPE- and LPA-induced [Ca2+]i increases, 2) the abrogation of LPE-induced response by the LPA1 and LPA3 antagonist Ki16425 supported the involvements of LPA1 and/or LPA3, 3) the complete inhibition of LPE-induced response by the LPA1 antagonist, AM-095, 4) the observation that LPA1 was expressed in SH-SY5Y cells, and 5) the Gi/o-coupling character of LPA1 and the PTX-sensitivity of LPE-induced [Ca2+]i increase. LPE-induced [Ca2+]i increases have been previously observed in ovarian and breast cancer cells and in pheochromocytoma cells (Park , 2013, 2014a; Lee ). Table 1 summarizes the responses observed in SH-SY5Y cells, PC-12 cells, and ovarian (SKOV3) and breast cancer (MDA-MB-231) cells.

Table 1.

LPE-induced responses in SH-SY5Y, PC-12, MDA-MB-231, and SK-OV3 cells

	Inhibition by PTX	Responses to different LPEs				Inhibition by LPA1 antagonists
		18:1	18:0	16:0	14:0
SH-SY5Y	Yes	Yes	Yes	Yes	Yes	Yes
PC-12	Yes	Yes	Yes	No	Yes	Yes
MDA-MB-231	Yes	Yes	No	No	Yes	Yes
SK-OV3	Yes	Yes	Yes	No	Yes	No

In ovarian cancer cells, LPE-induced [Ca2+]i increase was not found to be mediated through Ki16425, VPC32183, or AM-095-sensitive receptors (Park , 2013), and heterologous desensitization was not observed, although homologous desensitization was observed for LPE- and LPA-induced [Ca2+]i increases (Park ). Therefore, it appears LPE-induced response in SK-OV3 ovarian cancer cells differs from that in SH-SY5Y, MDA-MB-231, and PC-12 cells. On the other hand, in MDA-MB-231 breast cancer cells and PC-12 cells, LPE-induced [Ca2+]i response was inhibited by Ki16425, VPC32183, or AM-095, and heterologous desensitization was observed, indicating intermediation of LPE-induced response in MDA-MB-231 and PC-12 cells by LPA1 (Park ). Consequently, LPE-induced [Ca2+]i response in SH-SY5Y cells is similar to that in MDA-MB-231 cells and PC-12 cells in terms of LPA1 involvement.

In the present study, LPE-induced Ca2+ responses of synthetic LPE analogues were cell type dependent. In particular, ether-linked 18:0 LPE and ester-linked 18:0 LPE produced more than 50% of the response elicited by ester-linked 18:1 LPE in SK-OV3, SH-SY5Y, and PC-12 cells, but did not produce any response in MDA-MB-231 cells (Park , 2013, 2014a; Lee ) (Table 1). It has been previously reported that 16:0 LPE does not induce [Ca2+]i response in SK-OV3 cells, MDA-MB-231 cells, or PC-12 cells (Park , 2014a; Lee ). However, in the present study, 16:0 LPE induced [Ca2+]i response in SH-SY5Y cells (Table 1). In addition, in previous studies, 14:0 LPE induced similar [Ca2+]i responses to LPA in all four cell types (Park ; Lee ). However, 14:0 LPE-induced [Ca2+]i response may be driven by a different mechanism to 18:1 LPE-induced [Ca2+]i response, because 14:0 LPE induced Ca2+ response in cells non-responsive to 18:1 LPE. These findings show that LPE-induced [Ca2+]i responses have similar and dissimilar features in these four cell types; that is, MDAMB-231, PC-12 and SH-SY5Y cells all exhibit LPA1 involvement in responses, whereas their responses to different LPE structural types differ (Park , 2013, 2014a; Lee ) (Table 1).

Therefore, in SH-SY5Y cells, LPE was found to act on LPA1 to induce [Ca2+]i increase via Gi/o proteins, phospholipase C, and IP3R, and LPA was found to use LPA1 and LPA2 to mobilize Ca2+ (Fig. 7). Significance of this study is not only LPE action on LPA1 in SH-SY5Y cells but also involvement of Gi/o proteins and phospholipase C in LPA Ca2+ signaling. In previous studies using SH-SY5Y cells, LPA-induced Ca2+ mobilization was shown to be independent on phosphoinositide signaling and not mediated through pertussis toxin-sensitive Gi/o proteins (Young , 2000). Activation of sphingosine kinase and its product sphingosine 1-phosphate was proposed as a second messenger for LPA-induced Ca2+ signaling (Young , 2000). However, in the present study, involvement of pertussis toxin-sensitive Gi/o proteins and edelfosine-sensitive phospholipase C were shown in the LPA-induced Ca2+ signaling. Further investigation of the physiological roles of LPE in neuronal cells is required.

Fig. 7.

Proposed methods of signaling by LPE and by LPA in SH-SY5Y cells.