Molecular Membrane Biology, 2013; 30(5-6): 315–326

Lysophosphatidylcholine-induced cytotoxicity in osteoblast-likeMG-63 cells: Involvement of transient receptor potentialvanilloid 2 (TRPV2) channels

ABDALLAH FALLAH, RACHEL PIERRE, ELIE ABED & ROBERT MOREAU

Laboratoire du Métabolisme Osseux, BIOMED, Département des Sciences Biologiques, Université du Québec à Montréal,Montreal, Quebec, Canada

(Received 18 February 2013; and in revised form 9 July 2013)

AbstractEpidemiological studies indicate that patients suffering from atherosclerosis are predisposed to develop osteoporosis.Accordingly, atherogenic determinants such as oxidized low density lipoprotein (OxLDL) particles have been shown toalter bone cell functions. In this work, we investigated the cytotoxicity of lysophosphatidylcholine (lysoPC), a majorphospholipid component generated upon LDL oxidation, on bone-forming MG-63 osteoblast-like cells. Cell viability wasreduced by lysoPC in a concentration-dependent manner with a LC50 of 18.7 ± 0.7 mM. LysoPC-induced cell death wasattributed to induction of both apoptosis and necrosis. Since impairment of intracellular calcium homeostasis is often involvedin mechanism of cell death, we determined the involvement of calcium in lysoPC-induced cytotoxicity. LysoPC promoted arapid and transient increase in intracellular calcium attributed to mobilization from calcium stores, followed by a sustainedinflux. Intracellular calcium mobilization was associated to phospholipase C (PLC)-dependent mobilization of calcium fromthe endoplasmic reticulum since inhibition of PLC or calcium depletion of reticulum endoplasmic with thapsigargin preventedthe calcium mobilization. The calcium influx induced by lysoPC was abolished by inhibition of transient receptor potentialvanilloid (TRPV) channels with ruthenium red whereas gadolinium, which inhibits canonical TRP (TRPC) channels, waswithout effect. Accordingly, expression of TRPV2 and TRPV4 were shown in MG-63 cells. The addition of TRPV2 inhibitorTranilast in the incubation medium prevent the calcium influx triggered by lysoPC and reduced lysoPC-induced cytotoxicitywhereas TRPV4 inhibitor RN 1734 was without effect, which confirms the involvement of TRPV2 activation in lysoPC-induced cell death.

Keywords: TRPV2 channel, osteoblast, lysophosphotidylcholine, calcium, apoptosis

Introduction

Epidemiologic studies have reported a correlationbetween atherosclerosis and the development of oste-oporosis (Uyama et al. 1997, Barengolts et al. 1998).Indeed, it was shown that patients with lower bonedensity and osteoporosis present higher lipid levels(Uyama et al. 1997, Barengolts et al. 1998). It is wellestablished that cardiovascular diseases are directlyrelated to plasma concentrations of LDL cholesterol(Castelli et al. 1977), which suggests that both pathol-ogies may shear one or several contributory factors.Accordingly, atherosclerosis-susceptible C57BL/6mice maintained under an atherogenic high-fat dietshow reduced bone formation and low bone mass(Parhami et al. 2001).

Osteoporosis results from the loss of equilibriumbetween bone cell functions, such as increasedosteoclast-mediated bone resorption over bone for-mation by osteoblasts, leading to reduced bone mass,lower mineral density, and thereby higher bone fra-gility and susceptibility to fractures. Although theimportance of cholesterol in the development of oste-oporosis has been highlighted, it remains largelyunknown how lipids can affect bone metabolism.Given that it is well established that LDL particlesbecome pro-atherogenic after undergoing oxidativemodifications (Steinberg 1997), oxidized LDL(OxLDL) would be expected to be also responsiblefor osteoporosis. Indeed, studies have indicated thatOxLDL have deleterious effects on the differentiationof cells towards an osteoblastic phenotype

Correspondence: Professor Robert Moreau, PhD, Département des sciences biologiques, Université du Québec àMontréal, CP 8888, succ Centre-ville, Montréal(Québec), H3C 3P8, Canada. E-mail: moreau.robert@uqam.ca

ISSN 0968-7688 print/ISSN 1464-5203 online � 2013 Informa UK, Ltd.DOI: 10.3109/09687688.2013.828855

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

(Parhami et al. 1997, Mody et al. 2001). OxLDL andoxidative products inhibit osteogenic differentiationof mesenchymal stem cells and preosteoblast in favourof an adipogenic phenotype. Also, studies showed thatOxLDL caused cell death of human osteoblasts(Klein et al. 2003, 2006). The possible implicationof OxLDL in the development of osteoporosis wasfurther highlighted by our studies which demon-strated that osteoblastic cells express receptors forOxLDL (Brodeur et al. 2008a) and that OxLDLparticles promote cell death of osteoblasts(Brodeur et al. 2008b, Hamel et al. 2008).Upon oxidative modifications of LDL, oxysterols are

generated such as 7-ketocholesterol and 7b-hydroxycholesterol which are found in atheroscleroticlesions (Brown and Jessup 1999). Moreover, phospha-tidylcholine is converted to lysophosphatidylcholine(lysoPC) by phospholipase A2 activity present inLDL (Parthasarathy et al. 2010). Thereby, lysoPC isa major phospholipid component of OxLDL and wasidentified as a critical factor in the atherogenic activity ofOxLDL (Schmitz and Ruebsaamen 2010). LysoPC hasbeen associated with atherosclerosis, as well as acuteand chronic inflammation by altering various functionsin a number of cell types including endothelial cells,smooth muscle cells, monocytes, macrophages, andlymphocyte T-cells (Fuchs and Schiller 2009). Amongothers, lysoPC triggers apoptosis of vascular smoothmuscle cells (Hsieh et al. 2000, Hsu et al. 2011) andendothelial cells (Takahashi et al. 2002, Matsubara andHasegawa 2005, Zhou et al. 2006) by altering intracel-lular calcium signaling (Zhou et al. 2006, Song et al.2010, Hsu et al. 2011). Mechanistic evidence was givenby studies showing that activation of transient receptorpotential channels was associated to oxidative stress-induced cell death (Miller 2006) and to OxLDL-induced apoptosis of vascular smooth muscle cells(Ingueneau et al. 2009).The transient receptor potential (TRP) channels

contain six putative transmembrane domains, whichare thought to assemble as homo- or hetero-tetramers to form channels (for a review, Pedersenet al. 2005). All TRPs are cation channels, althoughthe permeability for different mono- and divalentcations varies greatly between isoforms. They are acti-vated by a wide range of stimuli including intra- andextracellular messengers, chemical, mechanical, andosmotic stress, and some by the filling state of intra-cellular calcium stores. Based on amino acid homol-ogies, the mammalian TRP channel superfamily hasbeen divided into seven families: TRPC, TRPM,TRPV, TRPML, TRPP, TRPA and TRPN. TheTRPC (‘Canonical’) and TRPM (‘Melastatin’) sub-families consist of seven and eight different channels,respectively (i.e., TRPC1–TRPC7 and TRPM1–

TRPM8). The TRPV (‘Vanilloid’) subfamily presentlycomprises six members (TRPV1–TRPV6). TheTRPML (‘Mucolipin’) family comprises three mem-bers, and the TRPP (‘Polycystin’) family threechannel-like and five non-channel members, respec-tively. The most recently proposed subfamily, TRPA(‘Ankyrin’), has only one mammalian member,TRPA1, and finally, the TRPN (no mechanoreceptorpotential C, or NOMPC) has so far only been detectedin Caenorhabditis elegans, Drosophila and zebra fish.There has been increasing interest in first three familiesbecause of their involvement in several human diseaseswhile last four families are yet insufficiently character-ized. We have reported the expression of numerousTRP channels in osteoblastic cells (Abed and Moreau2007, Labelle et al. 2007, Abed et al. 2009).In the current study, we investigated the cytotoxicity

of lysoPC as a major phospholipid component gener-ated upon LDL oxidation, on human osteoblast-like MG-63 cells and determined the involvement ofcalcium in lysoPC-induced cytotoxicity.

Materials and methods

Cell culture

Human osteoblast-like MG-63 cells were purchasedfrom the American Type Culture Collection (ATCC,Rockville,MD,USA) andwere grown in a 1:1mixtureof phenol-free DMEM/Ham’s F12medium (DMEM/F12; Sigma, Oakville, Ontario, Canada) supplemen-ted with 10% fetal bovine serum (FBS; Cansera,Etobicoke, Ontario, Canada), L-glutamine (Invitro-gen, Burlington, Ontario, Canada) and penicillin/streptomycin (Invitrogen). Cells were cultured in5% CO2 at 37�C and were harvested weekly withTrypsin-EDTA solution (Invitrogen).

Cell viability assays

For cell viability experiments, cells were seeded in96-well plates (Sarstedt, Montréal, Québec, Canada).After five days of culture in supplemented media, cellswere incubated in culture medium without serum inthe absence or the presence of increasing concentra-tions of lysoPC (Sigma) or cannabidiol (Tocris Bio-science, UK) without or with gadolinium (Gd3+),ruthenium red (Sigma), RN 1734 (Sigma) and Tra-nilast (Sigma) for 48 h. Experiments were also per-formed with cells treated for 48 h with GSK1016790A(Sigma). For experiments with low levels of calcium inmedium, cells were incubated in DME/F12 (culturemedium without calcium and magnesium, Sigma)supplemented with 0.25 mM calcium and 0.8 mMmagnesium. Cell viability was determined by microti-ter tetrazolium reduction assays. Briefly, 1 h before the

316 A. Fallah et al.

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

end of treatment the medium was replaced withDMEM/F12 containing 0.5 mg/ml of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrasodium bromide(MTT) (Sigma). At the end of the incubation, mediawas aspired and formazan crystals generated by thecellular reduction activity were dissolved in dimetylsulf-oxide (DMSO). Absorbance was measured at 575 nmand data are expressed as the ratio of absorbance oftreated cells versus control cells. In parallel experiments,cells exposed to lysoPC were dyed with crystal violet,then solubilized in 1% SDS and quantified with aspectrophotometer (575 nm), or harvested by trypsini-zation to perform cell counts.

Cell death assays

MG-63 cells seeded in 12-well plates were incubatedin culture medium alone or with lysoPC, as well as inthe presence of ruthenium red or Tranilast. After24 h, floating cells were discarded and adherent cellswere collected by trypsinization and centrifugation at250 g for 5 min. Cells were washed with PBS and thenincubated with 2 mg of propidium iodide (PI) and10 ml of FITC-annexin V per milliter (Roche Diag-nostics, Mannheim, Germany) for 15 min at 4�C. Atthe end of incubation, 400 ml of cold PBS was addedand cellular FITC-annexin V and PI fluorescenceswere detected by using a fluorescence-activated cellsorter (FACScan, Becton Dickinson, Mississauga,Canada) and data were analyzed with Cell Questsoftware (Becton Dickinson). Cells positive forFITC-annexin V were considered apoptotic whereascells positive for PI and negative for FITC-annexin Vwere considered necrotic.

Reverse-transcription polymerase chain reaction

Total RNA from cells was extracted using TriZol (Invi-trogen) according to the manufacturer’s instructions.Reverse transcription (RT) reactions were carried outwith Omniscript RT kit (Qiagen, Mississauga, Ontario,Canada) using hexamers. The PCR amplifications wereconducted with Taq PCR core kit (Qiagen) using spe-cific primer sets for human TRPV1-6 (Abed et al.2009). Each primer was designed in distinct exons toensure specific transcript amplifications. Briefly, ampli-fications were carried out for 40 cycles according toincubation of 1 min at 94�C, 30 sec at 58�C and 1 minat 72�C. Amplification products were resolved in 2%agarose gel with ethidium bromide revelation.

Measurements of intracellular calcium

MG-63 cells were cultured in 8-wells Labtek (NalgeNunc, Napperville, IL, USA) for five days in

supplemented media. Cells were then transferred toHEPES-buffered saline solution (HBSS (mM):121 NaCl, 5.4 KCl, 25 HEPES, 1.8 CaCl2 and6.0 NaHCO3 at pH 7.3) and loaded with 2 mMFluo-3 AM with an equivalent volume of 20% Pluro-nic F127 (Invitrogen) for 45 min at room temperaturein the dark. Thereafter cells were washed with corre-sponding HBSS and the loaded dye was allowed tode-esterify for 45 min at room temperature in thedark. Following transfer to a calcium-free HBSS,treatments were made in an open chamber configu-ration at room temperature. The cells were examinedwith a laser scanning confocal (Bio-Rad) microscope(Nikon TE300) with an Apochromatic 40� N.A.1.0 objective lens. Fluorescence was excited by anargon laser at 488 nm and emission was collected witha 515 nm filter. Data were analyzed with Laser Sharp2.1T, Time Course 1.0 software for 7–8 fields perexperiments (between 10 and 20 cells per field). Allexperiments were performed for at least four individ-ual experiments.

Immunolocalization

MG-63 cells were cultured in eight-well Labtek(Nalge Nunc) dishes for five days in supplementedmedium. Cells were fixed in 4% formaldehyde/phos-phate-buffered saline (PBS) for 30 min at 4�C, per-meabilized with 0.05% Tween/PBS for 15 min andblocked in 2% bovine serum albumin/PBS for 30 min.Cells were incubated overnight with primary antibo-dies (1:500 for anti-TRPV2 and anti-TRPV4 fromAlomone, Jerusalem, Israel) or primary antibodiespreincubated with 1 mg antigen followed by an incu-bation with secondary antibodies (1:2000 for antirabbit IgG coupled to rhodamine from Chemicon,Temecula, CA) for 1 h. Cells were washed three timeswith PBS between each step. The cells were examinedwith a laser scanning confocal (Bio-Rad) microscope(Nikon TE300) with an Apochromatic 40� N.A.1.0 objective lens. Fluorescence was excited by anargon laser at 555 nm and emission was collected witha 565 filter.

Statistical analysis

Data is expressed as mean ± SEM, and the signifi-cance of differences between groups has been deter-mined using GrafPadPrism 5.0 Software by analysisof variance. Differences between groups were furtherevaluated by Student’s t-test or two-way ANOVAwith Bonferroni or Dunnett post tests. Differenceswere considered significant at p £ 0.05.

TRPV2 channels in lysoPC-induced cytotoxicity 317

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

Results

Effect of LysoPC on the viability of MG-63

Our previous studies showed that OxLDL particlescause osteoblastic cell death (Hamel et al. 2008,Brodeur et al. 2008b). Since lysoPC is a major phos-pholipid component generated upon LDL oxidation(Parthasarathy et al. 2010), we determined its effecton the viability of osteoblastic MG-63 cells. Weperformed MTT assays on cells incubated for 48 hwith increasing concentrations of lysoPC. As shownin Figure 1A, exposition of MG-63 cells to lysoPCresulted in a reduction of the MTT activity in aconcentration-dependent manner. Curve analysiswith the Boltzmann sigmoidal equation give a LC50

(concentration that cause 50% reduction of MTTactivity) of 18.7 ± 0.7 mM with a significant reductionstarting from 15 mM (Dunnett test, p < 0.05). Toascertain that this reduction of MTT activity corre-sponded to cell mortality, similar experiments wereperformed and remaining adherent cells were eitherstained with crystal violet or counted after trypsiniza-tion. In accordance with data from MTT assays,similar dose-dependent reductions of crystal violetstaining and cell counts were obtained (Figure 1B)for cells treated with increasing concentrations oflysoPC with LC50 values of 17.6 ± 1.4 and 16.5 ±0.9 mM, respectively. These different experiencesconfirmed that lysoPC reduces MG-63 cell viabilityin a dose-dependent manner.To further document lysoPC-induced cell death, we

determined the levels of apoptosis and necrosis bystaining adherent cells with FITC-annexin V and PI.As shown in Figure 2A, analysis of adherent cells afteran incubation period of 24 h showed that incubationwith lysoPC resulted in the appearance of cell popula-tions that were positive for FITC-annexin V and/or positive for PI staining. Apoptosis (early apoptosis

in low right quadrant and late apoptosis in up rightquadrant) and necrosis (up left quadrant) wereenhanced with increasing concentrations of lysoPC(Figure 2B) which indicate that lysoPC promotescellular apoptosis and necrosis. Mean values fromadherent cells for FITC-annexin V staining (combinedearly and late apoptosis) were of 16% and 28% forlysoPC 15 and 30 mM, respectively, and of 12% and22% for necrosis at 15 and 30 mM lysoPC, respectively.

LysoPC triggers intracellular calcium mobilization andinflux

Since impairment of intracellular calcium homeosta-sis associated with calcium influx is often seen as amechanism triggering cell death, we determined theinvolvement of extracellular calcium in the lysoPC-induced cell death. As shown in Figure 3A, decreas-ing the calcium level in the culture medium to0.25 mM (this concentration did no compromisecell viability over 48 h) reduced the cell death inducedby lysoPC suggesting that activation of calcium chan-nels and calcium influx are involved in lysoPC-induced cytotoxicity. Therefore, we further investi-gated the effect of lysoPC on calcium mobilization inMG-63 cells loaded with the fluorescent calciumprobe Fluo-3. Under incubation condition incalcium-free buffer, to reveal release of calciumfrom intracellular pools, lysoPC promoted a transientincrease of intracellular calcium levels (Figure 3B),which correspond to a mobilization of calcium fromintracellular pools. After that fluorescence hadreturned to the basal level, addition of calcium inthe incubation buffer revealed a sustained calciuminflux which increased with lysoPC concentrations.To note, no calcium influx was obtained after theaddition of calcium alone without treatment withlysoPC (data not shown). Thereafter, two strategies

0 5 10 20 30 500.0

0.2

0.4

0.6

0.8

1.0

1.2 Crystal violetCell counts

LysoPC (µM)

Via

bili

ty(r

elat

ive

to C

TL

)

A B

0 10 20 30 40 50

0.0

0.2

0.4

0.6

0.8

1.0

1.2

LysoPC (µM)M

TT

act

ivit

y(r

elat

ive

to C

TL

)

Figure 1. Cytotoxicity of lysoPC inMG-63 cells. (A) Dose-response curve of MTT activity (relative to control without lysoPC) following 48-hexposure to increasing concentrations of lysoPC in serum-free culture medium. Values are mean ± SEM of five independent experimentsperformed in triplicates. (B) Dose-response analysis of cell viability (relative to control without lysoPC) following 48-h exposure to increasingconcentrations of lysoPC in serum-free culture medium, as measured through crystal violet staining and cell counts. Values are mean ± SEM ofthree independent experiments in triplicates.

318 A. Fallah et al.

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

were used to further identify origin of the intracel-lular calcium mobilization and determine the signal-ing pathway involved. First, cells were treated with1 mM of Thapsigargine (Tg) to deplete the endo-plasmic reticulum from its calcium. As shownin Figure 4A, calcium mobilization by lysoPC wasabolished after two subsequent addition of Tgwhereas the calcium influx was still observed. Onthe other hand, cells were pretreated with an inhibitorof PLC, namely U73122, to determine the signaling

pathway triggered by lysoPC. As shown in Figure 4B,calcium mobilization by lysoPC was abolished in cellspretreated with U73122 whereas calcium influx wasstill observed. These results indicated that lysoPCinduces the mobilization of calcium from the endo-plasmic reticulum upon activation of PLC. To note,no calcium mobilization or influx was seen with theatherogenic lipid 7-ketocholesterol (data not shown)which has also been shown to promote cell death inMG-63 cells (Hamel et al. 2008).

100 101 102 103 104

104

103

102

101

100

PI

CTL LysoPC 15 µM

FITC-annexin V

LysoPC 20 µMA

B

Viables

0 10 15 20 300.0

0.5

1.0

1.5

***** ***

LysoPC (mM)

Rel

ativ

e ce

ll n

um

ber

(vs

CT

L)

Early apoptosis

0 10 15 20 300

1

2

3

4

5

LysoPC (mM)

* *

Rel

ativ

e ce

ll n

um

ber

(vs

CT

L)

Late apoptosis

0 10 15 20 300

2

4

6

8

10

*****

LysoPC (mM)

*

Rel

ativ

e ce

ll n

um

ber

(vs

CT

L)

Necrosis

0 10 15 20 300

1

2

3

4

5

***** ***

LysoPC (mM)

Rel

ativ

e ce

ll n

um

ber

(vs

CT

L)

2.3% 4.7% 9.5%1.0% 5.9% 10.2%

5.2%

100 101 102 103 104

104

103

102

101

100

100 101 102 103 104

104

103

102

101

10010.6% 13.0%

Figure 2. LysoPC promotes cell death by apoptosis and necrosis. MG-63 cells seeded in 12-well plates were incubated in serum-free culturemedium without lysoPC (CTL) or in the presence of increasing concentrations of lysoPC for 24 h. (A) Adherent cells were then labeled withFITC-annexin-V and propidium iodide (PI) and fluorescence was analyzed by flow cytometry. (B) Percentage of cell number relative to thecontrol without lysoPC are shown as a function of increasing concentrations of lysoPC for viable cells, early apoptosis, late apoptosis andnecrosis. Values are mean ± SEM of seven independent experiments compared to control without lysoPC using a Dunnett test: *p < 0.05,**p < 0.01, ***p < 0.001

TRPV2 channels in lysoPC-induced cytotoxicity 319

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

Activation of TRP channels by lysoPC

To identify which subclass family of TRP channelswere associated to the lysoPC-induced calcium influx,we first characterized two TRP channel inhibitors,namely gadolinium (Gd3+) and ruthenium red (RR).As shown in Figure 5A, Gd3+ efficiently inhibitedcalcium influx induced following Tg-mediated cal-cium mobilization from the endoplasmic reticulum.Under these conditions, we previously showed thatTg-induced calcium mobilization from the endoplas-mic reticulum triggers the activation of store-operated calcium entry mediated by TRPC1 channelsinMG-63 cells (Labelle et al. 2007, Abed et al. 2009).Therefore, Gd3+ was proven as a potent inhibitor ofTRPC channels under our experimental conditions.

As shown in Figure 5B, RR was without effect on thecalcium influx associated with Tg-induced calciummobilization which agrees with the fact that RRrather acts as a known TRPV channel inhibitor.Since our results indicated that, similarly to Tg,lysoPC also induced a calcium mobilization fromthe endoplasmic reticulum, we asked if thelysoPC-induced calcium influx involved the activa-tion of TRPC channels. As shown in Figure 5C,pretreatment of cells with Gd3+ did not prevent thecalcium influx induced by lysoPC, which excludesthe activation of store-operated calcium entry bylysoPC. On the other hand, the lysoPC-induced calcium influx was prevented by pretreatingcells with the inhibitor of the TRPV channels RR(Figure 5D).

0 100 200 300 400 5000

1

2

3

4

5 LysoPC 10 µMLysoPC 20 µM

LysoPC

2.5 mM Ca0 mM Ca

Time (sec)

Flu

o-3

flu

ore

scen

ce

A B

0 10 20 30 40 50

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Low CaNormal Ca***

******

Lyso PC (mM)

MT

T a

ctiv

ity

(rel

ativ

e to

CT

L)

Figure 3. Involvement of extracellular calcium in the lysoPC-induced cell death. (A) Dose-response curve of MTT activity (relative to controlwithout lysoPC) following 48-h exposure to increasing concentrations of lysoPC in serum-free medium with low 0.25 mM calciumconcentration or normal 1 mM calcium concentration. Values are mean ± SEM of five independent experiments in triplicates. Bonferronipost test: ***p < 0.001 when compared to condition with nomal 1 mM calcium. (B) Fluo-3 loaded cells were transferred to HBSS withoutcalcium and then lysoPC (10 or 20 mM) were added to the incubation medium. Thereafter, calcium was added to the incubation buffer todetermine calcium influx. Fluo-3 fluorescence values relative to basal fluorescence are showed as mean ± SEM of 3–5 experiments withanalyses of 7–8 fields per experiment (between 10 and 20 cells per field).

A B

0 100 200 300 4000

1

2

3

4

With U73122Without U73122

LysoPC

0 mM Ca 2.5 mM Ca

Time (sec)

Flu

o-3

Flu

ore

scen

ce

0 200 400 600 800 1000 12000

1

2

3

4

Tg LysoPC

2.5 mM0 mM Ca

Tg

Time (sec)

Flu

o-3

flu

ore

scen

ce

Figure 4. Characterization of calcium mobilization induced by lysoPC. (A) Cells loaded with Fluo-3 were transferred to HBSS withoutcalcium and then calcium depletion of the endolasmic reticulum was achieved by two subsequent additions of thapsigargin (Tg 5 mM).Thereafter, lysoPC (10 mM) were added to the incubation medium. Calcium was added to the incubation buffer to determine calcium influx.(B) Cells loaded with Fluo-3 were transferred to HBSS without calcium and preincubated with 10 mM U73122 for 5 min. Thereafter, lysoPC(10 mM) were added to the incubation medium. Measurements of intracellular calcium were performed with a laser scanning confocal asdescribed in the Material and methods section. Thereafter, calcium was added to the incubation buffer to determine calcium influx. Values aremean ± SEM of three independent experiments with analyses of 7–8 fields per experiment (between 10 and 20 cells per field).

320 A. Fallah et al.

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

Expression of TRPV channels by MG-63 cells

We previously reported that numerous osteoblast celllines express TRPV channels (Abed et al. 2009).However, the presence of corresponding proteins isstill to be confirmed. Therefore, we investigatedexpression of TRPV in MG-63 cells. As shownin Figure 6A, a single PCR product was obtainedfor TRPV2 and two bands were observed forTRPV4 with RNA from MG-63 cells. PCR band of686 bp corresponds to variants 2 and 3 ofTRPV4 (GI294459960 and GI294459976, respec-tively) and amplified DNA of 866 bp correspondsto TRPV4 variants 1, 4 and 5 (GI294459961,GI294459970 and GI294459964, respectively). NoPCR products were observed for the other membersof TRPV family such as TRPV1, TRPV3, TRPV5 andTRPV6 (data not shown). By immunofluorescence,we confirmed the presence of TRPV2 andTRPV4 proteins with detection mainly at the plasmamembrane of cells (Figure 6B). As our results indicateplasma membrane localization of TRPV2 andTRPV4 proteins in MG-63 cells, these channelswere considered likely candidate for lysoPC-inducedcalcium influx.

Activation of TRPV channels by lysoPC is associatedwith cell death

Since our results indicated that lysoPC promotes theactivation of TRPV channels and that calcium wasinvolved in lysoPC-induced MG-63 cell death, wefurther characterized which TRPV channels wereassociated to cell death induced by lysoPC. As shownin Figure 7A, the decrease of MTT activity inducedby lysoPC was reduced in the presence of the TRPVchannel inhibitor RR and of the TRPV2 inhibitorTranilast whereas the TRPV4 inhibitor RN1734 was without effect. To ascertain the efficiencyof RN 1734 as an inhibitor of TRPV4, we determinedits inhibition potential using GSK, a known activatorof TRPV4. As shown in Figure 7B, GSK promoted acalcium influx in MG-63 cells which is prevented bypretreating cells with RN 1734. Interestingly, theactivation of TRPV4 channels by GSK has no effectonMTT activity (Figure 7C). As shown in Figure 7D,preincubation of cells with Tranilast significantlyreduced the calcium influx induced by lysoPC to asimilar level obtained with RR. Moreover, theTRPV2 activator cannabidiol (Qin et al. 2008)induced calcium influx (Figure 7E) and promoted

0 200 400 600 800 10000

1

2

3

4

5

With RR

Tg

0 mM Ca 2.5 mM Ca

Without RR

Time (sec)

Flu

o-3

Flu

ore

scen

ce

C D

A B

0 100 200 300 400 5000

1

2

3

4 Without RRWith RR

LysoPC

0 mM Ca 2.5 mM Ca

Time (sec)

Flu

o-3

Flu

ore

scen

ce

0 200 400 600 800 1000 12000

1

2

3

4

5

0 mM Ca 2.5 mM Ca

Tg

Time (sec)

Flu

o-3

Flu

ore

scen

ce

0 100 200 300 400 5000

1

2

3

LysoPC

Without Gd3+

With Gd3+

Without Gd3+

With Gd3+

0 mM Ca 2.5 mM Ca

Time (sec)

Flu

o-3

Flu

ore

scen

ce

Figure 5. Characterization of calcium influx induced by lysoPC. Cells loaded with Fluo-3 were transferred to HBSS without calcium andpreincubated with 100 mMGd3+ (A, C) or 10 mMRR (B, D) for 5 min. Thereafter, 5 mMTg (A, B) or 10 mM lysoPC (C, D) were added to theincubation medium. Calcium was added to the incubation buffer to determine calcium influx. Values are mean ± SEM of three individualexperiments with analyses of 7–8 fields per experiment (between 10 and 20 cells per field).

TRPV2 channels in lysoPC-induced cytotoxicity 321

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

cell death in a dose-dependent manner (Figure 7F).Accordingly, RR and Tranilast were shown to reducelysoPC-induced cell apoptosis but were without effecton necrosis (Figure 8).

Discussion

Epidemiologic and in vitro studies argue for theinvolvement of atherogenic conditions and OxLDLin the development of osteoporosis (Parhami et al.1997, Uyama et al. 1997, Barengolts et al. 1998,Mody et al. 2001, Klein et al. 2003, 2006,Hamel et al. 2008, Brodeur et al. 2008b). SincelysoPC is a major phospholipid component ofOxLDL, we investigated its effect on the viability ofosteoblastic cells. Our results indicate that lysoPC

induced cell death of MG-63, an effect that wasdependent on extracellular calcium content whichsuggests that cell mortality was promoted by calciuminflux induced by lysoPC. Accordingly, we reportedthat lysoPC promoted calcium influx by the activationof TRPV2 channels. In concordance, apoptosisinduced by lysoPC was reduced by the inhibition ofTRPV2 channels. Our results suggest that lysoPCpromotes apoptosis in MG-63 cells by the activationof TRPV2 channels.Oxysterols such as 7-ketocholesterol and 7b-hydro-

xycholesterol, as well as lysoPC are found in athero-sclerotic lesions (Brown and Jessup 1999). Thesecomponents are generated upon oxidation and enzy-matic modification of LDL (Parthasarathy et al. 2010)and are implicated as critical factors in the atherogenicactivity of OxLDL (Schmitz and Ruebsaamen 2010)and probably for the development of osteoporosis.Our results indicate that lysoPC induced cell death ofosteoblast-like MG-63 cells. Cytotoxicity was moni-tored by the reduction of MTT activity, crystal violetstaining and cell counts, and experiments revealed aLC50 values between 16.5 and 18.7 mM. Hsu et al.(2011) have reported LC50 values between 5–10 mMfor cytotoxicity of lysoPC in primary culture of vas-cular smooth muscle cells. Other study has reportedaround 37 mM for the same cell type (Hsieh et al.2000). Values of LC50 for lysoPC were from 50–100 mM in endothelial cells (Takahashi et al. 2002,Matsubara and Hasegawa 2005). Therefore, ourresults agree with the cytotoxicity of lysoPC at micro-molar concentrations. Further investigations on themode of cell death as a function of the concentrationsrevealed that lysoPC promoted both apoptosis andnecrosis of MG-63 cells. Hsieh et al. (2000) alsoreported induction of apoptosis and necrosis bylysoPC in vascular smooth muscle cells. Zhou et al.(2006) report apoptosis and necrosis in endothelialcells. To note, lysoPC is an amphipathic molecule andin accordance, possesses ‘detergent-like’ propertieswhich have been studies extensively in relation to lysisof red blood cells (Weltzien 1979). Therefore, induc-tion of necrosis by lysoPC may likely be attributed tosuch membrane-perturbating property.Since dysfunctional regulation of intracellular cal-

cium homeostasis is often associated to cell death,we determined the involvement of extracellularcalcium in the lysoPC-induced cell death. LysoPC-induced cytotoxicity was reduced when cells wereincubated in low calcium medium. To note, cytotox-icity induced by the atherogenic oxysterol 7-ketocho-lesterol was not modified by the calcium content ofincubation medium (data not shown). Therefore,specific mechanism involving extracellular calciumis triggered upon induction of cell death by lysoPC.

A

B

TRPV2

TRPV4

With Ag Without Ag

L TRPV4

500 bp-

TRPV2

Figure 6. Expression of TRPV channels by MG-63 cells. (A) TotalRNA was isolated from MG-63 and used for synthesis of comple-mentary DNA. Amplifications by PCR were performed with spe-cific primers for human TRPV2 or TRPV4. Representative data areshown from RNA isolations of at least three independent cultureswith amplified DNA of 686 and 866 bp for TRPV4 (correspondingto variants 2 and 3, and to variants 1, 4 and 5, respectively) and of432 bp for TRPV2. L: ladder of 100 bp. (B) Localization ofTRPV2 and TRPV4 proteins were performed by confocal micros-copy with specific antibodies, and revealed by secondary fluores-cence antibodies. To ascertain specific fluorescence, primaryantibodies have been incubated with related antigens (Ag). Repre-sentative data are shown from at least three independent cultures.

322 A. Fallah et al.

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

CTL RR Tran0.0

0.5

1.0

1.5

*****

Cal

ciu

m in

flu

x(r

elat

ive

to ly

soP

C)

0 0.5 1 2 5 100.0

0.5

1.0

1.5

GSK (nM)

MT

T a

ctiv

ity

(rel

ativ

e to

CT

L)

0 15 20 30 40 500.0

0.5

1.0

1.5 ControlRN 1734Tran

*

**

RR*

* *

LysoPC (mM)

MT

Tac

tivi

ty(r

elat

ive

to C

TL

)

A B

C D

E F

0 50 100 1500.0

0.5

1.0

1.5

2.0

2.5 Without RRWith RR

Can

Time (sec)

Flu

o-3

Flu

ore

scen

ce

0 5 10 15 20 25

0.0

0.5

1.0

1.5

Cannabidiol (mM)

MT

T a

ctiv

ity

(rel

ativ

e to

CT

L)

0 20 40 60 800.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

With RN 1734

GSK

Without RN 1734

Time (sec)

Flu

o-3

Flo

resc

ence

Figure 7. Involvement of TRPV activation in lysoPC-induced cell death of MG-63 cells. (A) Dose-response curve of MTT activity (relative tocontrol without lysoPC) following 48-h exposure to increasing concentrations of lysoPC in culture medium alone (Control) or with 10 mMRR,30 mM RN 1734 or 100 mM Tranilast (Tran). Results are from five independent experiments in triplicates. Bonferroni post test: *p < 0.05,**p < 0.01 when compared to control condition. (B) Cells loaded with Fluo-3 were transferred to fresh HBSS with 2.5 mM calcium.Thereafter, cells were preincubated for 5 min without or with 30 mM RN 1734. After, GSK1016790A (GSK, 10 nM) was added to theincubation medium and fluorescence was measured. Values are mean ± SEM of the relative fluorescence compared to basal level of fiveindividual experiments with analyses of 7–8 fields per experiment (between 10 and 20 cells per field). (C) Dose-response curve ofMTT activity(relative to control without GSK) following 48-h exposure to increasing concentrations of GKS1016790A in serum-free culture medium.Values are mean ± SEM of three independent experiments performed in triplicates. (D) Fluo-3 loaded cells were transferred to HBSS withoutcalcium and then lysoPC (10 mM) were added to the incubation medium. Thereafter, calcium was added to the incubation buffer to determinecalcium influx. Fluorescence values corresponding to calcium influx relative to calcium influx induced by lysoPC are showed asmean ± SEMof3–5 experiments with analyses of 7–8 fields per experiment (between 10 and 20 cells per field). Dunnett post test: **p < 0.01, ***p < 0.001 whencompared to control condition with lysoPC alone. (E) Cells loaded with Fluo-3 were transferred to fresh HBSS with 2.5 mM calcium.Thereafter, cells were preincubated for 5 min without or with 10 mM RR. After, cannabidiol (Can, 10 mM) was added to the incubationmedium and fluorescence was measured. Values are mean ± SEM of three individual experiments with analyses of 7–8 fields per experiment(between 10 and 20 cells per field). (F) Dose-response curve of MTT activity (relative to control without cannabidiol) following 48-h exposureto increasing concentrations of cannabidiol in serum-free culture medium. Values are mean ± SEM of four independent experimentsperformed in triplicates. One-way ANOVA, p < 0.0001.

TRPV2 channels in lysoPC-induced cytotoxicity 323

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

Accordingly, our results indicated that lysoPCinduces PLC-mediated calcium mobilization fromendoplasmic reticulum and calcium influx. Two Gprotein-coupled receptors for lysoPC have been pro-posed, namely GPR4 and G2A (Meyer zu and Jakobs2007). To note, G2A shows high binding affinity forlysoPC and was shown to be coupled to Gq protein.Also, G2A was shown to promote apoptosis in thepresence of lysoPC (Lin and Ye 2003). Therefore, ourdata showing PLC-mediated calcium mobilization bylysoPC would be in accordance with the involvementof G2A as the lysoPC receptor in MG-63 cells. To thebest of our knowledge, no study has yet investigatedthe expression of G2A in osteoblasts.In addition to promote calcium mobilization from

endoplasmic reticulum, our results indicate thatlysoPC stimulate a calcium influx. We previouslyshowed gene expression of numerous TRP channelsby MG-63 cells (Abed et al. 2009). Moreover, wehave shown that Tg- and mitogen-induced calciummobilization from endoplasmic reticulum trigger theactivation of store-operated calcium entry mediatedby TRPC channels in MG-63 cells (Labelle et al.2007, Abed et al. 2009). Since, lysoPC trigger cal-cium mobilization from the endoplasmic reticulum,subsequent activation of TRPC channels was plausi-ble. Accordingly, lysoPC was shown to activateTRPC6 channels in corporal smooth muscle cells

(So et al. 2005), suggesting a mechanism linkingatherogenic condition to erectile dysfunction.LysoPC was shown to activate TRPC5 channelswhich confer to this type of channel the property ofa lipid ionotropic receptor (Flemming et al. 2006).LysoPC was shown to activate TRPC5 andTRPC6 which result in the inhibition of endothelialcell migration (Chaudhuri et al. 2008). Nevertheless,lysoPC-induced calcium influx in MG-63 cells wasnot prevented by gadolinium, although our resultsdemonstrate that this inhibitor efficiently blocked Tg-induced calcium influx in MG-63 cells. Therefore,our results suggest that lysoPC activates other mem-bers of the TRP channel family. We used RR as aknown inhibitor of TRPV channels. Our resultsshowed that lysoPC-induced calcium influx was fullyinhibited by RR, suggesting that lysoPC activatesTRPV channels. One study has reported activationof TRPV2 channels by lysoPC in prostate cancer cells(Monet et al. 2009), an effect that was associated withstimulation of cell migration. Therefore, our resultsfurther evidence that lysoPC activates TRPV channelsin osteoblasts. Accordingly, our results indicate thatTRPV2 and TRPV4 channels were localized to theplasma membrane of MG-63 cells, where they canassure calcium influx upon activation by lysoPC. Ofinterest, our results indicated that GSK promotesTRPV4-mediated calcium influx in MG-63 cell and

Necrosis

0 10 15 20 300

1

2

3

4

5 CTLRR

LysoPC (mM)

Tran

Cel

l nu

mb

er(r

elat

ive

to C

TL

)

C

ViablesA B

0 10 15 20 300.0

0.5

1.0

1.5 CTLRR

LysoPC (mM)

******Tran

******

Cel

l nu

mb

er(r

elat

ive

to C

TL

)

Apoptosis

0 10 15 20 300

2

4

6 CTLRR

******

LysoPC (mM)

Tran

******Cel

l nu

mb

er(r

elat

ive

to C

TL

)

Figure 8. Stimulation of TRPV channels by lysoPC promoted cell apoptosis. MG-63 cells seeded in 12-well plates were incubated in serum-free culture medium without or with 10 mM RR or 100 mM Tranilast (Tran) in the presence of increasing concentrations of lysoPC for 24 h.The cells were then labeled with annexin-V-FITC and propidium iodide (PI) and fluorescence was analyzed by flow cytometry for viable(negative for annexin and PI), apoptotic (positive for annexin and negative for PI) and necrotic (negative for annexin and positive for PI) cells.Values are mean ± SEM of nine independent experiments. Bonferroni post test: *p < 0.05, **p < 0.01, ***p < 0.001 when compared tocondition without RR or Tranilast.

324 A. Fallah et al.

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

that such activation of TRPV4 by GSK had no effectof cell viability suggesting that TRPV4-mediated cal-cium influx is not associated to cell death inMG-63 cells. On the other hand, the known blockerof TRPV2 Tranilast was shown to inhibit calciuminflux suggesting that TRPV2-mediated calciuminflux was involved in the lysoPC-induced cell death.In agreement with the reduced lysoPC-induced celldeath observed in low calcium medium, our resultsindicate that inhibition of TRPV channels by RR andinhibition of TRPV2 by Tranilast decreased the cyto-toxicity of lysoPC. Moreover, inhibition of TRPVchannels by RR and of TRPV2 by Tranilast reducedlysoPC-induced apoptosis of MG-63 cells whereasbeing without effect on necrosis. Our data agreewith previous studies which indicate that intracellularcalcium signaling was associated to lysoPC-induced apoptosis of smooth muscle and endothelialcells (Zhou et al. 2006, Song et al. 2010, Hsu et al.2011). In contrast to apoptosis which is a tightlyregulated process involving activation and enzymaticsteps with maintenance of plasma membrane integ-rity, necrosis is associated to plasma membrane dis-ruption and impairment of homeostasis. Therefore,the absence of effect of RR and Tranilast on lysoPC-induced necrosis agrees with necrosis being a disor-ganized process.Our results indicate that the specific activation of

TRPV2 channels and related calcium influx areinvolved in lysoPC-induced MG-63 cell death. Intra-cellular calcium is a common signaling mechanismwhich regulates numerous cell functions such as con-traction, migration, proliferation, secretion and apo-ptosis. Since it regulates numbers of cell functions as asecond messenger, the intracellular calcium level istightly regulated in term of amplitude, duration andsubcellular localization. Although numerous types ofcalcium channels are expressed in the plasma mem-brane and although all promote calcium influx, theirintrinsic properties will result in distinct calciuminflux which will result in specific signaling responsesand specific modulation of cell functions. In contrastto TRPV4 channels that are mainly associated withtransient calcium influx due to inactivation anddesensitization by increased intracellular calciumlevels (reviewed in Nilius et al. 2004), the activityof TRPV2 seems less sensitive to intracellular calciumlevels (Phelps et al. 2010), allowing thus a moresustained influx of calcium. In accordance, TRPV2has been associated with a calcium-overloadingpathway leading to dystrophic cardiomyopathy(Iwata et al. 2003). Therefore, such intrinsic channelcharacteristic of TRPV2 may underlay theassociation of TRPV2 activation with lysoPC-inducedMG-63 cell death.

Altogether, our results show that lysoPC inducescytotoxicity in MG-63 cells by activating TRPV2 andour data are in accordance with the role of lysoPC asan atherogenic factor contributing to the developmentof osteoporosis.

Acknowledgements

This work was supported by the Natural Sciences andEngineering Research Council of Canada (NSERC,RM).

Declaration of interest: The authors report noconflicts of interest. The authors alone are responsiblefor the content and writing of the paper.

References

Abed E, Labelle D, Martineau C, Loghin A, Moreau R. 2009.Expression of transient receptor potential (TRP) channels inhuman and murine osteoblast-like cells. Mol Membr Biol26:146–158.

Abed E, Moreau R. 2007. Importance of melastatin-like transientreceptor potential 7 and cations (magnesium, calcium) inhuman osteoblast-like cell proliferation. Cell Prolif 40:849–865.

Barengolts EI, Berman M, Kukreja SC, Kouznetsova T, Lin C,Chomka EV. 1998. Osteoporosis and coronary atherosclerosisin asymptomatic postmenopausal women. Calcif Tissue Int62:209–213.

Brodeur MR, Brissette L, Falstrault L, Luangrath V,Moreau R. 2008a. Scavenger receptor of class B expressed byosteoblastic cells are implicated in the uptake of cholesteryl esterand estradiol from LDL and HDL3. J Bone Miner Res23:326–337.

Brodeur MR, Brissette L, Falstrault L, Ouellet P,Moreau R. 2008b. Influence of oxidized low-density lipopro-teins (LDL) on the viability of osteoblastic cells. Free Radic BiolMed 44:506–517.

Brown AJ, Jessup W. 1999. Oxysterols and atherosclerosis. Ath-erosclerosis 142:1–28.

Castelli WP, Doyle JT, Gordon T, Hames CG, Hjortland MC,Hulley SB, et al. 1977. HDL cholesterol and other lipids incoronary heart disease. The cooperative lipoprotein phenotyp-ing study. Circulation 55:767–772.

Chaudhuri P, Colles SM, Bhat M, Van Wagoner DR,Birnbaumer L, Graham LM. 2008. Elucidation of aTRPC6-TRPC5 channel cascade that restricts endothelial cellmovement. Mol Biol Cell 19:3203–3211.

Flemming PK, Dedman AM, Xu SZ, Li J, Zeng F, Naylor J, et al.2006. Sensing of lysophospholipids by TRPC5 calcium channel.J Biol Chem 281:4977–4982.

Fuchs B, Schiller J. 2009. Lysophospholipids: Their generation,physiological role and detection. Are they important diseasemarkers? Mini Rev Med Chem 9:368–378.

Hamel P, Abed E, Brissette L,Moreau R. 2008. Characterization ofoxidized low density lipoproteins-induced hormesis-like effectsin osteoblastic cells. Am J Physiol Cell Physiol 294:C1021–C1033.

Hsieh CC, Yen MH, Liu HW, Lau YT. 2000. Lysophosphatidyl-choline induces apoptotic and non-apoptotic death in vascularsmooth muscle cells: In comparison with oxidized LDL. Ath-erosclerosis 151:481–491.

TRPV2 channels in lysoPC-induced cytotoxicity 325

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.

Hsu JH, Wu JR, Liou SF, Chen HM, Dai ZK, Chen IJ, et al. 2011.Labedipinedilol-A prevents lysophosphatidylcholine-inducedvascular smooth muscle cell death through reducing reactiveoxygen species production and anti-apoptosis. Atherosclerosis217:379–386.

Ingueneau C, Huynh UD, Marcheix B, Athias A, Gambert P,Negre-Salvayre A, et al. 2009. TRPC1 is regulated bycaveolin-1 and is involved in oxidized LDL-induced apoptosisof vascular smooth muscle cells. J Cell Mol Med 13:1620–1631.

Iwata Y, Katanosaka Y, Arai Y, Komamura K, Miyatake K,Shigekawa M. 2003. A novel mechanism of myocyte degener-ation involving the Ca2+-permeable growth factor-regulatedchannel. J Cell Biol 161:957–967.

Klein BY, Kerem Z, Rojansky N. 2006. LDL inducesSaos2 osteoblasts death via Akt pathways responsive to a neutralsphingomyelinase inhibitor. J Cell Biochem 98:661–671.

Klein BY, Rojansky N, Ben Yehuda A, Abou-Atta I, Abedat S,Friedman G. 2003. Cell death in cultured humanSaos2 osteoblasts exposed to low-density lipoprotein. J CellBiochem 90:42–58.

Labelle D, Jumarie C, Moreau R. 2007. Capacitative calcium entryand proliferation of human osteoblast-like MG-63 cells. CellProlif 40:866–884.

Lin P, Ye RD. 2003. The lysophospholipid receptor G2A activates aspecific combination of G proteins and promotes apoptosis.J Biol Chem 278:14379–14386.

Matsubara M, Hasegawa K. 2005. Benidipine, a dihydropyridine-calcium channel blocker, prevents lysophosphatidylcholine-in-duced injury and reactive oxygen species production in humanaortic endothelial cells. Atherosclerosis 178:57–66.

Meyer zu HD, Jakobs KH. 2007. Lysophospholipid receptors:Signalling, pharmacology and regulation by lysophospholipidmetabolism. Biochim Biophys Acta 1768:923–940.

Miller BA. 2006. The role of TRP channels in oxidative stress-in-duced cell death. J Membr Biol 209:31–41.

Mody N, Parhami F, Sarafian TA, Demer LL. 2001. Oxidativestress modulates osteoblastic differentiation of vascular andbone cells. Free Radic Biol Med 31:509–519.

Monet M, Gkika D, Lehen’kyi V, Pourtier A, Vanden Abeele F,Bidaux G, et al. 2009. Lysophospholipids stimulate prostatecancer cell migration via TRPV2 channel activation. BiochimBiophys Acta 1793:528–539.

Nilius B, Vriens J, Prenen J, Droogmans G, Voets T. 2004.TRPV4 calcium entry channel: A paradigm for gating diversity.Am J Physiol Cell Physiol 286:C195–C205.

Parhami F, Morrow AD, Balucan J, Leitinger N, Watson AD,Tintut Y, et al. 1997. Lipid oxidation products have oppositeeffects on calcifying vascular cell and bone cell differentiation.

A possible explanation for the paradox of arterial calcification inosteoporotic patients. Arterioscler Thromb Vasc Biol 17:680–687.

Parhami F, Tintut Y, Beamer WG, Gharavi N, Goodman W,Demer LL. 2001. Atherogenic high-fat diet reduces bone min-eralization in mice. J Bone Miner Res 16:182–188.

Parthasarathy S, Raghavamenon A, Garelnabi MO,Santanam N. 2010. Oxidized low-density lipoprotein. MethodsMol Biol 610:403–417.

Pedersen SF, Owsianik G, Nilius B. 2005. TRP channels: Anoverview. Cell Calcium 38:233–252.

Phelps CB, Wang RR, Choo SS, Gaudet R. 2010. Differentialregulation of TRPV1, TRPV3, and TRPV4 sensitivity through aconserved binding site on the ankyrin repeat domain. J BiolChem 285:731–740.

Qin N, Neeper MP, Liu Y, Hutchinson TL, Lubin ML,Flores CM. 2008. TRPV2 is activated by cannabidiol andmediates CGRP release in cultured rat dorsal root ganglionneurons. J Neurosci 28:6231–6238.

Schmitz G, Ruebsaamen K. 2010. Metabolism and atherogenicdisease association of lysophosphatidylcholine. Atherosclerosis208:10–18.

So I, ChaeMR, Kim SJ, Lee SW. 2005. Lysophosphatidylcholine, acomponent of atherogenic lipoproteins, induces the change ofcalcium mobilization via TRPC ion channels in culturedhuman corporal smooth muscle cells. Int J Impot Res17:475–483.

Song J, Liu K, Yi J, Zhu D, Liu G, Liu B. 2010. Luteolin inhibitslysophosphatidylcholine-induced apoptosis in endothelial cellsby a calcium/mitocondrion/caspases-dependent pathway. PlantaMed 76:433–438.

Steinberg D. 1997. Low density lipoprotein oxidation and itspathobiological significance. J Biol Chem 272:20963–20966.

Takahashi M, Okazaki H, Ogata Y, Takeuchi K, Ikeda U,Shimada K. 2002. Lysophosphatidylcholine induces apoptosisin human endothelial cells through a p38-mitogen-activatedprotein kinase-dependent mechanism. Atherosclerosis161:387–394.

Uyama O, Yoshimoto Y, Yamamoto Y, Kawai A. 1997. Bonechanges and carotid atherosclerosis in postmenopausal women.Stroke 28:1730–1732.

Weltzien HU. 1979. Cytolytic and membrane-perturbing proper-ties of lysophosphatidylcholine. Biochim Biophys Acta 559:259–287.

Zhou L, Shi M, Guo Z, Brisbon W, Hoover R, Yang H. 2006.Different cytotoxic injuries induced by lysophosphatidylcholineand 7-ketocholesterol in mouse endothelial cells. Endothelium13:213–226.

326 A. Fallah et al.

Mol

Mem

br B

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Tor

onto

on

09/1

2/13

For

pers

onal

use

onl

y.