1422

Lysophosphatidylcholine Inhibits SurfaceReceptor-Mediated Intracellular Signals inEndothelial Cells by a Pathway Involving

Protein Kinase C ActivationKiyotaka Kugiyama, Masamichi Ohgushi, Seigo Sugiyama, Toyoaki Murohara, Kohji Fukunaga,

Eishichi Miyamoto, and Hirofumi Yasue

Lysophosphatidylcholine (lysoPC) transferred from oxidatively modified low density lipoprotein (Ox-LDL) to the endothelial surface membrane has been shown to produce a selective unresponsiveness to cellsurface receptor-regulated endothelium-dependent relaxation (EDR) in the rabbit aorta. To determine itsmechanism we examined the effects of lysoPC on endothelial surface receptor-mediated transmembranesignals. Incubation for 1 minute with palmitoyl lysoPC (5-10 ,uM) decreased thrombin (Th, 2 units/ml)-or histamine (His, 0.1 mM)-stimulated inositol 1,4,5-trisphosphate (1P3) production in primary culturesof human umbilical vein endothelial cells (HUVECs). LysoPC also decreased Th- or His-inducedintracellular calcium ([Ca2"J1, fura 2) elevation. Pretreatment with protein kinase C (PKC) inhibitorsstaurosporine (100 nM) or H-7 (50 ,uM) prevented the inhibitory actions of lysoPC, but HA-1004 had noeffect. Incubation for 5 minutes with phorbol 12-myristate 13-acetate (PMA, 100 nM) produced theinhibitory actions on the Th- or His-induced intracellular signals, which closely mimic those exhibited bylysoPC. However, the inhibitory effect of lysoPC was lost in cells that were depleted of PKC bypretreatment for 24 hours with 100 nM PMA. Furthermore, incubation of the cells for 1 minute withlysoPC stimulated PKC activity in the membrane fraction. In organ chamber experiments with porcinecoronary artery rings, pretreatment with staurosporine (20 nM) attenuated lysoPC-induced impairmentof EDR in response to Th. These results indicate that lysoPC, which accumulates in Ox-LDL andatherosclerotic arterial walls, inhibits the early transmembrane signaling pathway in endothelial cells,and PKC activation could at least partially be involved in the negative regulation by lysoPC. Theseintracellular actions of lysoPC may play a role in the mechanism of the lysoPC-induced impairment ofEDR in response to cell surface receptor-mediated stimulations. (Circulation Research 1992;71:1422-1428)KEY WoRDs * lysophosphatidylcholine * inositol 1,4,5-trisphosphate * cytosolic free calcium a

endothelial cell * protein kinase C

It has been recently demonstrated that lysophos-phatidylcholine (lysoPC), which accumulates inoxidatively modified low density lipoprotein (Ox-

LDL), produces a selective unresponsiveness to recep-tor-regulated endothelium-dependent arterial relax-ation (EDR) that mimics the condition observed in theearly-onset atherosclerotic arteries.1-4 LysoPC in Ox-LDL impairs the EDR that is mediated by stimulationsof endothelial surface receptors, whereas lysoPC gener-ally has no effect on EDR induced by calcium iono-phores that bypass receptor-dependent membraneregulation. 1,2

From the Division of Cardiology (K.K., M.O., S.S., T.M., H.Y.)and the First Department of Pharmacology (K.F., E.M.), Kuma-moto University School of Medicine, Kumamoto City, Japan.

Supported in part by a grant-in-aid for Scientific Research (C)03670460 from the Ministry of Education, Science, and Culture inJapan, Tokyo.Address for correspondence: Hirofumi Yasue, MD, Division of

Cardiology, Kumamoto University School of Medicine, Honjo1-1-1, Kumamoto City 860, Japan.Received January 14, 1992; accepted August 18, 1992.

Protein kinase C (PKC) mediates intracellular signaltransduction in a variety of cellular functions thatinclude cell proliferation, morphological transforma-tion, modulation of gene expression, and many otherfunctions.5-7 PKC is also known to exert inhibitoryactions on agonist-induced inositol 1,4,5-trisphosRhate(IP3) formation and subsequent intracellular calciummobilization.5,8-10 The inhibitory actions of activatedPKC on receptor agonist-mediated early signal trans-duction have been shown in many different types ofcells, including endothelial cells.10 Furthermore, a vari-ety of lipids, including lysophospholipids, are capable ofactivating PKC.1' Therefore, the present study ad-dressed the possibility that PKC activation could beinvolved in the mechanism of lysoPC-induced unrespon-siveness to the endothelial surface receptor-mediatedstimulations.

Materials and MethodsCell Culture

Primary cultures of human umbilical vein endothelialcells (HUVECs) were obtained by collagenase diges-

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

Kugiyama et al Lysophosphatidylcholine and Intracellular Signals 1423

tion12 and grown to confluence on gelatin-coated dishesin medium 199 with 15% fetal calf serum (FCS), endo-thelial growth factor, and antibiotics. After confluencewas reached, the medium was replaced with medium199 that contained 5% FCS and antibiotics withoutgrowth factor for 48 hours before the experiments. Onlyprimary cultures of HUVECs were used for the exper-iments. Confluent cultures of HUVECs showed thetypical cobblestone morphology, and almost all of thosecells contained factor VIII-related antigen as deter-mined by the use of indirect immunofluorescence.12

Measurement of Inositol1,4,5-Trisphosphate ProductionThe monolayers of cultured HUVECs on 60-mm

Petri dishes at a density of 1 x 105 cells/cm2 were washedthree times with Hanks' balanced salt solution (HBSS)and incubated with serum-free medium 199 that con-tained 10 mM LiCl for 20 minutes. The cells were thenstimulated by thrombin (Th, 2 units/ml) or histamine(His, 0.1 mM) after preincubation with phospholipidsfor the appropriate times or with phorbol 12-myristate13-acetate (PMA) for 5 minutes (as indicated in the textand figure legends). In some experiments the cells werepretreated for 5 minutes with PKC inhibitors before theincubation with phospholipids or PMA. In the experi-ment that examined the effects of lysoPC on the agonist-induced 1P3 production in PKC-downregulated cells,confluent monolayers of cells were incubated for 24hours with 100 nM PMA in medium 199 that contained2% FCS. Then the culture medium was replaced withserum-free medium 199 that contained 10 mM LiClafter washing three times with serum-free medium 199;afterwards, the cells were incubated for 20 minutes at37°C. Cells were then stimulated by Th after preincu-bation for 1 minute with or without lysoPC.

Stimulation by Th or His was terminated with aspira-tion of the solution and addition of 2.0 ml ice-cold 15%trichloroacetic acid (TCA) and followed by incubationfor 30 minutes on ice. The cells were then removed fromthe dishes by scraping. After centrifugation of themixture, the supernatant was extracted with diethyl-ether to remove the TCA. 1P3 levels were determined bythe use of a protein binding assay system13 (AmershamIntl., code TRK 1000) after adjustment of the pH to 7.3by the addition of Tris buffer. In a preliminary experi-ment that examined the time courses of Th- or His-induced 1P3 production, peak increases of 1P3 wereobtained within 20 seconds of the addition of Th or His.On the basis of these time course observations, thestimulations were terminated at 20 seconds after theadmixture of Th or His.

Measurement of Cytosolic Free Calcium ElevationThe fluorescent calcium indicator fura 2 was used to

monitor changes in cytosolic free calcium concentrationin suspensions of primary cultured HUVECs. For thisexperiment, primary cultures of HUVECs were brieflyexposed to 0.01% trypsin-EDTA to detach the cells.Trypsin was inactivated by the addition of a trypsininhibitor and culture medium with 15% FCS immedi-ately after cell detachment. The cells were then centri-fuged, washed once with HBSS, resuspended in Krebs-Ringer HEPES solution (KRH, pH 7.4; composition[mM]: NaCl 128, KCI 5, CaCl2 2.7, MgSO4 1.2, Na2HPO4

1, glucose 10, and HEPES 20) and followed by exposureto 4 ,uM fura 2-AM for 30 minutes at 37°C. The cellswere washed three times with HBSS, resuspended inKRH (3x106 cells/ml), and then transferred into aquartz cuvette. The suspension was stirred continuouslyand maintained at 37°C. Fluorescence signals weremonitored on a Hitachi F-3000 fluorescence spectro-photometer (Hitachi, Tokyo) at excitation and emissionwavelengths of 340 and 505 nm, respectively. The cellswere then stimulated by Th (2 units/ml) or His (0.1 mM)after preincubation with phospholipids for the appro-priate times or with PMA for 5 minutes (indicated in thetext and figure legends). In some experiments, the cellswere pretreated for 5 minutes with PKC inhibitorsbefore the incubation with phospholipids or PMA.[Ca2"]i was calculated by using the formula

[Ca 2+]i=Kd(F-Fmin)/(Fmax-F)where F is the fluorescence intensity from cells, Fmax isthe maximum fluorescence from lysed cells by theaddition of 0.1% Triton X-100, and Fmin is the minimumfluorescence after Ca2+ in the buffer is chelated with 10mM EGTA. The value ofKd was 224 nM at 37°C for fura2 according to Grynkiewicz et al.14

Organ Chamber ExperimentsThe left anterior descending coronary arteries were

isolated from Yorkshire pigs within 10 minutes afterdeath. The arteries were cleaned of adherent connectivetissues and cut into ring segments of 3-mm length. Therings were suspended by stainless steel hooks in theorgan chambers, which were filled with Krebs' buffer(composition [mM]: NaCl 118, KCl 4.7, NaH2PO4 1.2,MgSO4 1.2, CaCl2 2.0, NaHCO3 25, and glucose 10).This solution was aerated with 15% 02-5% C02-80%N2 (Po2 100 mm Hg) and maintained at 37°C. Duringthis procedure care was taken not to injure the luminalsurface. The rings were then stretched to an optimumbasal tension of 4 g, and the isometric tension wasmonitored by means of a force transducer (Minebea,Tokyo) and a polygraph machine (Nihon Kohden, To-kyo). After equilibration for 120 minutes, the contrac-tile response to 60 mM KCl was first obtained and thenfollowed by repeated washing. The rings were subse-quently pretreated with 10 ,uM indomethacin for 30minutes followed by the incubation with 10 ,M palmi-toyl lysoPC or dipalmitoyl phosphatidylcholine for anadditional 30 minutes in the presence or absence ofstaurosporine (20 nM). Indomethacin remained in thebath solution during the incubations. After the incuba-tion the rings were washed repeatedly and then con-tracted with prostaglandin F2a (PGF2a, 30 ,M) andtested with increasing concentrations of various vasodi-lators as indicated. Vasorelaxation was expressed asthe maximum percent change in the PGF2a-inducedcontractions.

Treatment and Fractionation of the Cells for ProteinKinase C Assay

Confluent cultures of HUVECs were quickly frozenby liquid nitrogen after washing three times with HBSS.Cells were removed by scraping and homogenated andsuspended in the HEPES buffer with protease inhibi-tors. Then the suspension was centrifuged (100,OOOg for

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

1424 Circulation Research Vol 71, No 6 December 1992

60 minutes) for separation into cytosolic and particulatefractions. The particulate fraction was extracted with0.1% Triton X-100. In the experiments that examineddownregulation of PKC, confluent cultures of HUVECswere pretreated for 24 hours with 100 nM PMA.Thereafter, the cells were quickly frozen after washingand fractionized as described above. In the experimentsthat examined the effects of lysoPC or PMA on PKCactivity in intact cells, confluent cultures of the cellswere washed three times with HBSS, and the incubationmedium was replaced with serum-free medium 199. Thecells were then incubated with various concentrations oflysoPC or PMA for 0.5, 1, 5, 10, or 30 minutes. After theincubation the medium was aspirated, and the cellswere quickly frozen and fractionized as described above.

Assay of Protein Kinase C ActivityPKC activity in samples was assayed by measuring the

incorporation of [32p] into the synthetic peptide PKCsubstrate from [r-32P]ATP.15 The standard reagentmixture for the assay of PKC activity contained 50 mMHEPES (pH 7.5), 10 mM magnesium acetate, 500 ,uMCaCl2, 0.1 mM [r-32P]ATP, 3 ,M mastoparan, 5 ,uM ofa synthetic peptide inhibitor of cAMP kinase (PKI-tide), and 40 ,uM PKC substrate in a volume of 25 ,ul.The reactions were initiated by the addition of 5 ,ul ofthe samples from HUVECs to the reaction mixture.After incubation for 8 minutes at 30°C, 15-gul aliquotswere spotted on phosphocellulose paper squares, andthe radioactivity was counted.16 The standard reactionmixture that included 5 ,ug/ml diolein and 50 ,ug/mlphosphatidylserine was used when diolein and phos-phatidylserine-dependent phosphorylation of the sub-strate was assessed for the PKC activity in the cytosolicand soluble particulate fractions from the cells pre-treated with lysoPC or PMA. The standard reactionmixture that included 50 ,ttg/ml phosphatidylserine andvarying concentrations of lysoPC or PMA was used forthe examination of the direct effects of lysoPC or PMAon the activity of PKC purified from the cytosolicfraction of the untreated cells by means of a diethylami-noethyl-Sephacel column.17 Protein concentrationswere determined by the method of Bradford,18 withbovine serum albumin as the standard.

MaterialsAll reagents for the cell culture were obtained from

GIBCO. Radioactive materials came from AmershamIntl., Tokyo; staurosporine from Kyowa Hakko Co. Ltd.,Tokyo; H-7 and HA-1004 from Seikagaku Kogyo Co.Ltd., Tokyo; endothelial growth factor (bovine pitu-itary) from Collaborative Research, and PKC substratefrom Bachem, Torrance, Calif. Other chemicals wereobtained from Sigma Chemical Co., St. Louis, Mo.Phospholipids were used after sonication in HBSS.Staurosporine, fura 2-AM, PMA, or A23187 was dis-solved in dimethyl sulfoxide (DMSO). Final concentra-tions of DMSO were less than 0.1% in the solution.Indomethacin solution was prepared in an equimolar(1-mM) concentration of Na2CO3.Statistical Analysis

All values were expressed as mean+SEM. Statisticalevaluation of the data was performed by Student's t testfor paired or unpaired observations. When more than

c0'P

0AZa

00C tO

0,-v.5

C.c- EQ1

C-

E0-C

60r

40 *

20!.O PC* lysoPC

OL0 2 5

Phospholipid (P M)10

FIGURE 1. Line graph showing effects of lysophosphatidyl-choline (palmitoyl, lysoPC) or phosphatidylcholine (dipalmi-toyl, PC) on thrombin (Th)-induced inositol 1,4,5-trisphos-phate (IP3) production in human umbilical vein endothelialcells (HUVECs). Confluent monolayers of HUVECs wereincubated for 20 minutes with serum-free medium-199 thatcontained 10 mM LiCl and then exposed to lysoPC orPC for1 minute and followed by stimulation with Th (2 unitslml).The stimulation was terminated by aspiration of the solutionand the addition ofice-cold trichloroacetic acid at 20 secondsafter the initiation of Th-induced stimulation. Exposure tolysoPC significantly suppressed Th-induced IP3 production ina dose-dependent manner, whereas PC had no effect.*p<0.01 vs. PC (n=9-16).

two groups were compared analysis of variance wasused. A value of p<0.05 was considered significant.

ResultsMeasurements of Inositol 1,4,5,-TrisphosphateFormation and [Ca2+i ElevationAs shown in Figures 1 and 2, preincubation for 1

minute with submicellar concentration of palmitoyllysoPC decreased Th- or His-induced 1P3 formation inmonolayers of primary cultures of HUVECs in a dose-

,- 60.5 r Thrombin

o * * _Kistamine

40+ + +

ECh

2020-0

COCL + +

0~Control [ysPC IySOPC tysoPC IYSOPC

STS H7 HA1004

FIGURE 2. Bargraph showing effects oflysophosphatidylcho-line (lysoPC) with or withoutprotein kinase C (PKC) inhibitorson thrombin (Th)- or histamine (His)-induced inositol 1,4,5-trisphosphate (IP3) production in human umbilical vein endo-thelial cells (HUVECs). Monolayers of HUVECs were pre-treated for 5 minutes with staurosporine (STS, 100 nM), H7(50 ,uM), orHA-i1004 (50 pM) andfollowed by the addition oflysoPC (5 ,M). At 1 minute after the addition oflysoPC, cellswere stimulated by Th (2 unitslml) or His (0.1 mM). In cellspretreated for 5 minutes with STS orH7 but not HA-1004, theinhibitory effects oflysoPC on Th- or His-induced IP3produc-tion were attenuated. *p<0.01 vs. lysoPC alone (n=10-16),+p<0.01 vs. control (no pretreatment).

2

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

Kugiyama et al Lysophosphatidylcholine and Intracellular Signals 1425

B250 -

200 _ lysoPC150-looL

- 1~

h50L

thrombin

thror

I

staurosporine IysoPC ,

I I /_WA~ 11min

FIGURE 3. Charts showing thrombin (Th)-induced [Ca 2+]i elevations in human umbilicalvein endothelial cells (HUVECs). [Ca2+]i was

mibin monitored with fura 2 fluorescence. Panel A: Th(2 units/ml) that transiently elevates [Ca 2+1i inHUVECs. Panel B: Effects of the exposure tolysophosphatidylcholine (lysoPC) on Th-induced

_ f l[Ca 2+]i elevation. Lysophosphatidylcholine1 min (lysoPC) (2 ,tM) was admixed into the cuvette

and incubatedfor 1 minute before Th (2 units/ml)stimulation. Exposure to lysoPC inhibits Th-in-duced [Ca 2+]i elevation without altering a restinglevel of [Ca 2+]j. Panel C: The effect ofpretreat-ment for 5 minutes with staurosporine (100 nM)on the lysoPC (2 ,uM)-induced inhibition of[Ca 2+]i elevation in response to Th (2 unitslml).Staurosporine attenuates the inhibitory effect oflysoPC on Th-induced [Ca2+i elevation.

dependent manner. Preincubation for 1 minute with preincubation times with lysoPC on the Th-induced 1P3lysoPC (1-5 uM) also inhibited Th- or His-induced production and [Ca'+]i elevation, the maximum inhibi-[Ca'+]I elevation in suspensions of HUVECs, as shown tions were obtained as early as 1 minute after thein Figures 3 and 4. Incubation for 1 minute with stearoyl addition of lysoPC into the incubation mixture, and thelysoPC also showed inhibitory actions on Th-induced inhibitions were sustained with the same magnitude for1P3 formation and [Ca2"]i elevation (IP3: control [no up to 30 minutes (the longest incubation time tested).pretreatment, n= 16], 51±3 versus 5 ,uM stearoyl lysoPC Basal levels of 1P3 in monolayers of HUVECs were not

[n=6], 36±5 pmol/106 cells, p<0.01; peak [Ca'+]i: con- affected by the incubation with up to 10 ,uM palmitoyltrol [n= 12], 201 + 13 versus 2 ,uM stearoyl lysoPC [n 9], lysoPC alone (control [no treatment, n=5] 4.2+0.3136±10 nM, p<0.01), whereas dipalmitoyl phosphati- versus 10 ,uM lysoPC [n=61, 4.1+0.2 pmol/106 cells,dylcholine had no significant effect on Th-induced 1P3 p=NS). However, the resting levels of [Ca 2i], in suspen-

production and [Ca21]i elevation (IP3: 49±3 pmol/106 sions of HUVECs were increased by the incubation with

cells [10 ,uM phosphatidylcholine, n=7]; peak [Ca 21]i: lysoPC alone at higher concentrations than 5 ,uM (con-

196+11 nM [5 ,iM phosphatidylcholine, n=6]). In the trol [no treatment, n=8], 96±5 versus 5 ,uM lysoPC

experiment that examined the effects of various [n=8], 120±6 nM, p<0.01).To evaluate the role of PKC activation in the inhibi-

250 tory effects of lysoPC on agonist-induced IP3 productionrJ Thrombin and [Ca'+]i elevation, the effects of the PKC inhibitors

200 I * * Histamine staurosporine and H-7 were tested. As shown in Figuresg T _ _ g 2-4, pretreatment with staurosporine (100 nM) or H-7

*-150 g + | + + (50 gM) significantly attenuated the inhibitory effects of+ lysoPC Th- His-induced intracellular signals.

'100 | | | Neither staurosporine nor H-7 is a specific PKC inhib-itor, but it also inhibits A and G kinases at higher

concentrations.'920 However, HA-1004 (50 ,uM), whichhas strong inhibitory actions on A and G kinases but

control _ysoPc_ysoPc lyso_ only a weak action on PKC,20 did not attenuate the

Control lysoPC lysoPC lysoPC lysoPC inhibitory actions of lysoPC on the agonist-inducedSTS H7 HA1004 intracellular signals (Figures 2 and 4). Preincubation for

FIGURE 4. Bargraph showing effects oflysophosphatidylcho- 5 minutes with PMA (100 nM), a specific PKC activator,line (lysoPC) with or withoutprotein kinase C (PKC) inhibitors inhibited Th- or His-induced IP3 formation and [Ca 2]ion thrombin (Th)- or histamine (His)-induced [Ca 2+] eleva- elevation in HUVECs, and staurosporine or H-7 but nottion in human umbilical vein endothelial cells (HUVECs). HA-1004 attenuated the inhibitory effects of PMA, as

shown in Figure 5. However, preincubation for 1 minuteSuspensions of HUVECs were pretreated for 5 miutes with with lysoPC had no significant influence on Th-inducedstaurosporine (STS, 100 nM), H7 (50 uM), or HAi-1004 (50 IP3 production in the cells pretreated for 24 hours withuM) in a quartz cuvette, followed by the addition of 2uinM 100 nM PMA (incubation with lysoPC [5 ,uM, n=8],hysoPC that is a subthreshold concentration for elevating the 38±3 versus that without lysoPC [n=8], 41±2 pmol/106resting level of [Ca . Cells were then stimulated with Th (2 cells, p=NS). The pretreatment with PKC inhibitors

unitslml) or His (0.1 mM) at 1 minute after the addition of alone at the respective concentrations used in this studylysoPC. The inhibitory effects oflysoPC on Th- or His-induced did not influence Th-induced IP3 production and [Ca2]i[Ca2+i elevation were attenuated in cells pretreated with STS elevation (IP3: staurosporine, 52±3; H-7, 50±2; andor H7 but not with HA4-1004. *p<0.01 vs. lysoPC alone HA-1004, 49±4 pmol/106 cells; n=5-8, p=NS com-

(n=9-12). +p<0.01 vs. control (nopretreatment). pared with control [no pretreatment, n=16]; peak

thrombin

1

I x1 min

A250 -

2 200 -

P, 150

Q 100jI_50

C2 200 -

150[_so L

ci

5 min

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

1426 Circulation Research Vol 71, No 6 December 1992

60 r

-.

o 0.-__

o -

CLEnC

40-

201-

0

.:.:.:.:.A.:-:-:.A122

200*

150

00

0

00

Control PMA PMA PMA PMA+ + +

STS H7 HA1004FIGURE 5. Bar graphs showing effects of phorbol 12-myristate 13-acetate (PMA) with or without protein kinase C(PKC) inhibitors on thrombin (Th)-induced inositol 1,4,5-trisphosphate (IP3) production and [Ca 2+]i elevation in hu-man umbilical vein endothelial cells (HUVECs). HUVECswere pretreated for 5 minutes with staurosporine (STS, 100nM), H7 (50 ,uM), or HA-1004 (50 ,uM) followed by theaddition of PMA (100 nM) and then incubated for an

additional 5 minutes before Th (2 units/ml) stimulation. Theincubation for 5 minutes with PMA inhibited Th-induced IP3production and [Ca 2+1i elevation; however, the inhibitoryeffects ofPMA were attenuated in cells pretreated with STS or

H7 but not HA-1004. *p<0.01 vs. PMA alone (n=9-13),+p<0.01 vs. control (no treatment).

[Ca2+]i: staurosporine, 206+9; H-7, 203+11; HA-1004,202±8 nM; n=7-9, p=NS compared with control [nopretreatment, n= 12]). Basal levels of 1P3 and [Ca 2+]iwere not significantly changed by the pretreatment for 5minutes with staurosporine, H-7, HA-1004, or PMAalone at the respective concentrations used in this study(data not shown). DMSO (as a drug vehicle) at a

concentration of less than 0.1% in the final solution didnot affect the inhibitory effects of lysoPC on Th-induced1P3 production and [Ca 2+]i elevation (data not shown).

Organ Chamber ExperimentAfter incubation with lysoPC (10 ,uM), but not phos-

phatidylcholine, relaxation of a precontracted artery inresponse to Th was inhibited, whereas relaxation inresponse to calcium ionophore A23187 was completelypreserved, as shown in Table 1. Nitroglycerin, an endo-thelium-independent vasodilator, elicited complete re-

laxation after incubation with lysoPC or PC. Coincuba-tion with staurosporine (20 nM) attenuated theinhibitory effect of lysoPC on Th-induced vasorelax-ation (Table 1). Contractions elicited by PGF2a were notsignificantly influenced by the incubations (time control,89±11% of 60 mM KCl-induced contraction; lysoPC,92± 8%; lysoPC plus staurosporine, 86± 10%; n= 12-18,p=NS). Incubation for 30 minutes with staurosporinealone did not significantly affect vasorelaxation in re-

TABLE 1. Arterial Relaxations in Response toVarious Vasodilators

Thrombin A23187 Nitroglycerin

Time control (n= 18) 98+3 96±7 108+8LysoPC (n=14) 43±5* 92+5 102±6LysoPC andstaurosporine (n=12) 76±8t 94±9 106±7

Isolated rings from porcine coronary arteries were mounted forthe monitoring of isometric tension and equilibrated in oxygenatedKrebs buffer at 37°C. The coronary rings were then incubated withlysophosphatidylcholine (lysoPC) (10 ,M) for 30 minutes in thepresence or absence of staurosporine (20 nM). After the incuba-tion the rings were washed repeatedly and contracted with pros-taglandin F2a (PGF2a) (30 ,uM) and then relaxed with variousvasodilators as indicated. Relaxations are expressed as the maxi-mum percent changes in the PGF2a-induced contractions that werenot influenced by the incubations. The incubation with lysoPCinhibited arterial relaxation in response to thrombin but not toA23187 and nitroglycerin. Coincubation with staurosporine atten-uated the inhibitory effect of lysoPC on thrombin-induced arterialrelaxation.

*p<0.01 compared with time control, tp<0.01 compared withlysoPC.

sponse to Th (maximal relaxation; time control [n= 18],98±3% of precontraction versus staurosporine, [n=9]96±9%; p=NS). Incubation with DMSO alone, as adrug vehicle of staurosporine, at a concentration of lessthan 0.1% in the bath solution did not influence thelysoPC-induced inhibition of vasorelaxation in responseto Th (data not shown).

Protein Kinase C AssayAs shown in Figure 6, the incubation of HUVEC

monolayers in culture dishes with lysoPC (palmitoyl, 1,uM) caused a significant increase in diolein and phos-phatidylserine-dependent PKC activity in the particu-late fraction of the cells, with a maximum (by a 32%increase from the baseline value) at 1 minute after theaddition of lysoPC into the culture medium, and theactivity subsequently decreased to the baseline value.The PKC activity in the cytosolic fraction slightly de-creased during the increase of the particulate PKCactivity (a 5-13% decrease from the baseline value). Incomparison, the incubation of the cells with PMA (100nM) also caused an increase in PKC activity in theparticulate fraction, with a maximum (a 67% increasefrom the baseline value) at 5 minutes after the additionof PMA into the medium, and at the same time the PKCactivity in the cytosolic fraction decreased by 31% fromthe baseline value at 5 minutes.

Furthermore, lysoPC (palmitoyl, 1-5 ,M) directlystimulated the activity of the PKC purified from HU-VECs in the presence of phosphatidylserine (PKCactivity; presence of 1 ,M lysoPC [n=8], 826±48 versusabsence of lysoPC [n=8], 634±52 pmol/mg protein perminute,p<0.01), whereas lysoPC (up to 10 ,M) did notsignificantly stimulate the PKC activity in the absence ofphosphatidylserine (presence of 1 ,uM lysoPC [n=8],170±26 versus absence of lysoPC [n=8], 146±32pmol/mg protein per minute, p=NS). The maximumlysoPC (1 ,gM)-stimulated activity of the purified PKCin the presence of phosphatidylserine was 28% of themaximum PMA (100 nM)-stimulated activity.

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

Kugiyama et al Lysophosphatidylcholine and Intracellular Signals 1427

70r

4. 0

.0

0

Ey 00

;70

o O)

m

- .0

_~

501-

301-

10

0

10

hTime. min

-301-

FIGURE 6. Line graph showing time course ofprotein kinaseC (PKC) activity in the cytosolic and particulate fractions ofintact human umbilical vein endothelial cells (HUVECs) treatedby lysophosphatidylcholine (lysoPC) or phorbol 12-myristate13-acetate (PMA). Confluent monolayers of HUVECs werewashed three times with Hanks' balanced salt solution (HBSS),and the culture medium was replaced with serum-free medium199. The cells were then incubated with lysoPC (1 pM) orPMA(100 nM)for 0.5, 1, 5, 10, or30 minutes. After the incubation themedium was aspirated and the cells quicklyfrozen. PKC activityin the cytosolic and particulate fractions from the cell homoge-nates was assayed by measuring diolein and phosphatidylserine-independent phosphorylation of the substrate, as described in"Materials and Methods. " Results are mean±SEM of the twoseparate experiments, and each one consisted oftriplicate deter-minations.

The incubation of HUVECs with 100 nM PMA for 24hours resulted in an 87% loss of the measurable PKCactivity in both the cytosolic and particulate fractionsassessed by diolein and phosphatidylserine-inducedPKC activation, as shown in Table 2.

DiscussionAccording to current concepts,21-23 the binding of an

agonist to a cell surface receptor, which is coupled by aG-protein to a specific phospholipase C, leads to thehydrolysis of phosphatidylinositol 4,5 -bisphosphate.

TABLE 2. PKC Activity in HUVECs After ProlongedPMA Treatment

Control PMA p

Cytosol 605±22 86±8 <0.01Particulate 96±6 12±4 <0.01

PKC, protein kinase C; HUVECs, human umbilical vein endo-thelial cells; PMA, phorbol 12-myristate 13-acetate.HUVECs incubated for 24 hours with medium 199 contained

2% fetal calf serum in the presence or absence of PMA (100 nM).Then cells were homogenated and centrifuged for separation intocytosolic and particulate fractions. PKC activity was directlyassayed in both of the cytosolic and particulate fractions fromhomogenates ofHUVECs. Diolein and phosphatidylserine-depen-dent phosphorylation of the substrate was assayed under thestandard condition, except for the inclusion of diolein and phos-phatidylserine. Results (pmol/mg protein per minute) aremean±SEM of the three separate experiments, and each experi-ment consisted of quadruplicate determinations.

This leads to the production of 1P3 and 1,2-diacylglyc-erol. 1P3 may serve as a mediator of [Ca21]i release froman intracellular store. Elevated [Ca 2+]i leads to thegeneration of endothelium-derived relaxing factor.24Concomitant production of 1,2-diacylglycerol stimulatesPKC, which in turn inhibits continued or repeated acti-vation of this transmembrane signaling system5,8,9'23,25either at the level of the receptor, G-proteins, or phos-pholipase C. The present study demonstrated thatlysoPC inhibited Th- or His-induced IP3 production and[Ca21]i elevation in endothelial cells. These inhibitoryeffects of lysoPC on agonist-induced responses wereattenuated by the potent PKC inhibitors staurosporineand H-7. Furthermore, the incubation with PMA for 5minutes produced the inhibitory effects on the agonist-induced intracellular signals, which closely resembledthose exhibited by lysoPC. However, the inhibitory effectof lysoPC was lost in the cells that were depleted of PKCby long-term preincubation with PMA. Furthermore, thepresent study demonstrated that the incubation of thesecells with lysoPC activated PKC in the membrane frac-tion. These results obtained from the intracellular exper-iments indicate that lysoPC inhibits the early transmem-brane signaling pathway in endothelial cells, and PKCactivation could be partially involved in the negativeregulation by lysoPC.

Present intracellular responses to lysoPC are compat-ible with the results obtained from the organ chamberexperiments. It has been demonstrated in previousreports,1'2 in which rabbit aortas were used, that lysoPCtransferred from Ox-LDL to the endothelial surfacemembrane produced impairment of endothelial surfacereceptor-mediated EDR, which closely resembledthose observed in early-onset atherosclerotic arteries.In the present study, in which we used porcine coronaryarteries, we demonstrated that lysoPC impaired EDR inresponse to thrombin but not to A23187. Furthermore,staurosporine prevented the lysoPC-induced impair-ment of EDR. Thus, these results obtained from organchamber experiments also indicate that PKC activationmay be at least partially involved in the underlyingmechanism of lysoPC-induced inhibition of EDR inresponse to endothelial surface receptor-mediatedstimulations. Activated PKC has been shown to exert anegative feedback control on surface receptor-coupled1P3 formation and subsequent [Ca2+]i mobilization inresponse to numerous agonists in many different typesof cells.5'8-10 This negative regulation by activated PKCmay be responsible for the lysoPC-induced inhibitionsobserved in the present cellular and organ chamberexperiments.LysoPC is generated in several pathophysiological

situations.26-30 Concentrations of lysoPC are greatlyincreased in atherosclerotic arterial walls.26'27 There areat least two sources of lysoPC found in atheroscleroticarterial walls. One source is Ox-LDL, which accumu-lates in atherosclerotic arterial walls.31 LDL oxidation isassociated with the hydrolysis of PC to lysoPC by theaction of an intrinsic LDL-associated phospholipaseA2.29 It has been previously demonstrated that theinhibitory activity of Ox-LDL on EDR is present in itslipid component, and most of it is attributable to lysoPCin Ox-LDL.12 Another source of arterial lysoPC is theplasma in which the lecithin: cholesterol acyltransferaseenzyme catalyzes lysoPC formation.26'27 Lecithin: cho-

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

1428 Circulation Research Vol 71, No 6 December 1992

lesterol acyltransferase activity has been shown to in-crease with hypercholesterolemia.26,27 The abundantlysoPC in atherosclerotic arteries is transferable toaccessible membranous or macromolecular acceptorsthrough the aqueous phase.32 In fact, previous reports1,2have shown that lysoPC in Ox-LDL is transferred andincorporated into the endothelial surface membrane inan apoprotein-independent manner. TransferredlysoPC is slowly translocated to the inner plasma mem-brane and then partially metabolized.33 During trans-membrane movement lysoPC could access and directlyactivate PKC in the membrane fraction, as shown inpresent experiments for PKC assay. However, theprecise mechanism of lysoPC-induced PKC activationremains to be determined. Isozymes of lysoPC-induc-ible PKC in HUVECs, its endogenous substrates inHUVECs, and its activation process in the in vivoendothelium are now under investigation in our labo-ratory. It cannot be excluded that some metabolites oflysoPC may contribute to the lysoPC-induced inhibi-tions. However, our preliminary experiment, in whichwe used ['4C]-radiolabeled lysoPC, showed that a majormetabolite of the incorporated lysoPC in HUVECs wasphosphatidylcholine, which has been shown to be inac-tive in the endothelial responses to thrombin. At thismoment we do not have any evidence that lysoPC maybe metabolized to active substances, which could play amajor role in the lysoPC-induced inhibitions.PKC also plays crucial roles in the signal transduction

that involves cellular proliferation and modulation ofgene expression.5-79 Therefore, lysoPC in atheroscle-rotic arterial walls could activate PKC in a variety ofcells in the artery, which may play a role in themechanisms of their mitogenic changes and functionalalterations in atherosclerotic arteries.

In conclusion, lysoPC, which is abundant in athero-sclerotic arterial walls, inhibits surface receptor-medi-ated intracellular signals in human endothelial cells,and PKC activation could be at least partially involvedin the negative regulation by lysoPC. These intracellularactions of lysoPC may play a role in the mechanism oflysoPC-induced unresponsiveness to cell surface recep-tor-regulated EDR, which closely mimics that observedin early-onset atherosclerotic arteries.

References1. Kugiyama K, Kerns SA, Morrisett JD, Roberts R, Henry PD:

Impairment of endothelium-dependent relaxation by lysolecithinin modified low-density lipoproteins. Nature 1990;344:160-162

2. Kugiyama K, Henry PD: Transfer of lysophosphatidylcholine fromoxidized LDL impairs endothelium-dependent arterial relaxation.(abstract) JAm Coll Cardiol 1990;15(suppl A):12A

3. Habib GB, Bossaller C, Wells S, Williams C, Henry PD: Preser-vation of endothelium-dependent vascular relaxation in cholester-ol-fed rabbit by treatment with the calcium blocker PN 200110.Circ Res 1986;58:305-309

4. Bossaller C, Habib GB, Yamamoto H, Williams C, Wells S, HenryPD: Impaired muscarinic endothelium-dependent relaxation andcyclic guanosine 5'-monophosphate formation in atherosclerotichuman coronary artery and rabbit aorta. J Clin Invest 1987;79:170-174

5. Nishizuka Y: Studies and perspectives of protein kinase C. Science1986;233:305-312

6. Dicker P, Rozengurt E: Phorbol esters and vasopressin stimulateDNA synthesis by a common mechanism. Nature 1980;287:607-612

7. Moinat M, Chevey JM, Muzzin P, Giacobino JP, Kossovsky M:Effects of phorbol 12-myristate 13-acetate on triglyceride and cho-

lesteryl synthesis in cultured coronary smooth muscle cells andmacrophages. J Lipid Res 1990;31:329-334

8. Connolly TM, Lawing WJ, Majerus PW: Protein kinase C phos-phorylates human platelet inositol trisphosphate 5'-phosphomo-noesterase, increasing the phosphatase activity. Cell 1986;46:951-958

9. Nishizuka Y: The molecular heterogeneity of protein kinase C andits implications for cellular regulation. Nature 1988;334:661-665

10. Brock TA, Capasso EA: Thrombin and histamine activate phos-pholipase C in human endothelial cells via a phorbol ester-sensitivepathway. J Cell Physiol 1988;136:54-62

11. Bells RM, Burns DJ: Lipid activation of protein kinase C. J BiolChem 1991;266:4661-4664

12. Jaffe EA, Hoyer LW, Nachman RL: Synthesis of antihemophilicfactor antigen by cultured human endothelial cells. J Clin Invest1973;52:2757-2764

13. Langlands JM, Diamond J: The effect of phenylephrine on inositol1,4,5-trisphosphate levels in vascular smooth muscle measuredusing a protein binding assay system. Biochem Biophys Res Com-mun 1990;173:1258-1265

14. Grynkiewicz G, Poenie M, Tsien RY: A new generation of Ca 2+indicators with greatly improved fluorescence properties. J BiolChem 1985;260:3440-3450

15. Egan JJ, Chang MK, Londos C: A tandem chromatographic col-umn method for assaying cAMP-dependent protein kinase andprotein kinase C with synthetic peptide substrates. Anal Biochem1988;175:552-561

16. Roskoski R: Assays of protein kinase. Methods Enzymol 1983;99:3-6

17. Wooten MW, Vaudenplas M, Nel AE: Rapid purification of pro-tein kinase C from rat brain. Eur J Biochem 1987;164:461-467

18. Bradford MM: A rapid and sensitive method for the quantitationof microgram quantities of protein utilizing the principle of pro-tein-dye binding. Anal Biochem 1976;72:248-254

19. Tatsuya T, Hisayo N, Isami T, Yuzuru K, Makoto M, Fusao T:Staurosporine, a potent inhibitor of phospholipid/Ca++ dependentprotein kinase. Biochem Biophys Res Commun 1986;135:397-402

20. Hidaka H, Inagaki M, Kawamoto S, Sasaki Y: losquinolinesulf-onamides, novel and potent inhibitors of cyclic nucleotide depen-dent protein kinase and protein kinase C. Biochemistry 1984;23:5036-5041

21. Rana RS, Hokin LE: Role of phosphoinositides in transmembranesignaling. Physiol Rev 1990;70:115-164

22. Berridge MJ, Irvine RF: Inositol phosphates and cell signalling.Nature 1989;341:197-205

23. Majerus PW, Connolly TM, Deckmyn H, Ross TS, Bross TE, IshiiH, Bansal VS, Wilson DB: The metabolism of phosphoinositide-derived messenger molecules. Science 1986;234:1519-1526

24. Ignarro U: Biological actions and properties of endothelium-derived nitric oxide formed and released from artery and vein. CircRes 1989;65:1-21

25. Ryu SH, Kim UH, Wahl MI, Brown AB, Carpenter G, Huang KP,Rhee SG: Feedback regulation of phospholipase C-,8 by proteinkinase C. J Biol Chem 1990;265:17941-17945

26. Portman OW, Soltys P, Alexander M, Osuga T: Metabolism oflysolecithin in vivo: Effects of hyperlipemia and atherosclerosis insquirrel monkeys. J Lipid Res 1970;2:596-604

27. Vidaver GA, Ting A, Lee JW: Evidence that lysolecithin is animportant causal agent of atherosclerosis. J Theor Biol 1985;115:27-41

28. Sobel BE, Corr PB, Robinson AK, Goldstein RA, Witkowski FX,Klein MS: Accumulation of lysophosphoglycerides with arrhyth-mogenic properties in ischemic myocardium. J Clin Invest 1978;62:546-553

29. Parthasarathy S, Barnett J: Phospholipase A2 activity of low den-sity lipoprotein: Evidence for an intrinsic phospholipase A2 activityof apoprotein B-100. Proc NatlAcad Sci USA 1990;87:9741-9745

30. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL:Beyond cholesterol: Modification of low-density lipoprotein thatincreases its atherogenicity. N Engl J Med 1989;320:915-924

31. Yla-Herttuala S, Palinski W, Rosenfeld ME, Parthasarathy S,Carew TE, Butler S, Witztum JL, Steinberg D: Evidence for thepresence of oxidatively modified low density lipoprotein in athero-sclerotic lesions of rabbit and man. J Clin Invest 1989;84:1086-1095

32. Pownall HJ, Hickson DL, Smith LC: Transport of biologicallipophiles: Effects of lipophile structure. JAm Chem Soc 1983;105:2440-2445

33. Mohands N, Wyatt J, Mel SF, Rossi ME, Shohet SB: Lipid trans-location across the human erythrocyte membrane: Regulatory fac-tors. J Biol Chem 1982;257:6537-6543

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from

K Kugiyama, M Ohgushi, S Sugiyama, T Murohara, K Fukunaga, E Miyamoto and H Yasueendothelial cells by a pathway involving protein kinase C activation.

Lysophosphatidylcholine inhibits surface receptor-mediated intracellular signals in

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1992 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.71.6.14221992;71:1422-1428Circ Res. 

http://circres.ahajournals.org/content/71/6/1422World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on May 3, 2018

http://circres.ahajournals.org/D

ownloaded from