Directory UMM :Data Elmu:jurnal:A:Atherosclerosis:Vol151.Issue2.Aug2000:

(1)

Lysophosphatidylcholine induces apoptotic and non-apoptotic

death in vascular smooth muscle cells: in comparison with

oxidized LDL

Chien-Cheng Hsieh

_{, Mao-Hsiung Yen}

b,1

_{, Hwan-Wun Liu}

_{, Ying-Tung Lau}

_*

a_{Graduate Institute of Life Sciences,}_{National Defense Medical Center,}_Taipei, _Taiwan b_{Department of Pharmacology,}_{National Defense Medical Center,}_Taipei, _Taiwan c_{Department of Biology and Anatomy,}_{National Defense Medical Center,}_Taipei, _Taiwan

d_{Department of Physiology,}_{Chang Gung Uni}₆_{ersity College of Medicine,}₂₅₉_{Wen Hwa} ₁_Rd.,_Kwei-Shan,_Tao-Yuan, _Taiwan

Received 25 February 1999; received in revised form 8 September 1999; accepted 2 March 2000

Abstract

Oxidized low-density lipoprotein (oxLDL) plays a key role in the development of atherogenesis, partly by causing injury to vascular cells. However, different preparations of LDL, methods of oxidation, and/or active components often produce cellular effects of various degrees. To explore the quantitative relationship between dose and level of oxidation of the oxLDL utilized, we employed combinations of different levels of oxidation and concentrations of oxLDL to induce cell death in cultured vascular smooth muscle cells (VSMC). We also examined the effect of lysophosphatidylcholine (lysoPC), a putative active component of oxLDL, on VSMCs by determining, in parallel with a cytotoxicity test (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay), DNA fragmentation ([3_{H]thymidine release), and flow cytometric analyses. We found that oxLDL caused} cytotoxicity in an oxidative level- and dose-dependent manner, lysoPC also caused dose-dependent cytotoxicity with or without serum. Fragmentation of DNA was observed in both oxLDL- and lysoPC-treated VSMCs. Furthermore, lysoPC-induced DNA ladder was also demonstrated by gel electrophoresis at a concentration of 25mmol/l or higher. Flow cytometric analysis yielded similar results for oxLDL- and lysoPC-treated VSMC; namely, an accumulation in the fraction of cells in G0/G1phase with a reciprocal change in S-phase fraction. Membrane phosphatidylserine exposure, detected by annexin V staining, provided additional evidence that lysoPC induced significant apoptosis in VSMC. Taken together, the degree of oxLDL-induced cytotoxicity/apoptosis of VSMC depended on combined effects of oxLDL concentration and oxidative level. Moreover, lysoPC also elicited a dose-dependent apoptosis in addition to cytotoxicity. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords:Oxidized low-density lipoprotein; Lysophosphatidylcholine; Apoptosis; Vascular smooth muscle cell; Cytotoxicity

www.elsevier.com/locate/atherosclerosis

1. Introduction

Electron microscopic analyses revealed that vascular smooth muscle cell (VSMC) is the predominant cell type in ruptured atherosclerotic plaques [1], and accu-mulated evidence also demonstrated that cell death involves both oncosis and apoptosis of VSMCs during the progression of atherosclerosis [1 – 4]. Since vascular

remodelling during atherosclerosis could be mediated by many agents regulating both proliferation and apop-tosis of VSMC simultaneously, and yet independently, in both early and late stages of the remodelling process, apoptosis of VSMC is likely to play an important, albeit undefined, role (for a reveiw, see Ref. [4]). In early lesion of atherosclerosis, the proliferation of VSMCs and accumulation of matrix proteins synthe-sized by VSMCs lead to the thickening of intima and formation of fibrous atheroma [5]. However, this in-crease of arterial VSMC content was not observed at 12 weeks after injury [5]. Consequently, Walsh et al. re-ported a high frequency of medial VSMCs apoptosis occurring as early as 30 min after the balloon injury of * Corresponding author. Tel.: +886-03-328-3016; fax: +

886-03-328-3031.

E-mail addresses: [email protected] (M.-H. Yen), [email protected] (Y.-T. Lau1₎

1_{Both authors are corresponding authors. Tel.:} ₊

886-02-8792-3100 (M.H. Yen)

(2)

rat carotid as well as rabbit external iliac arteries, and this apoptosis was proposed to account for the lack of cell accumulation at the injury site during the period of observation [6]. In advanced lesions, both macrophages and VSMCs can accumulate lipid and have been iden-tified as foam cells. Cell debris are more prominent in lipid-rich areas, with death occurring in both macrophages and VSMCs [7 – 9]. A high rate of apop-totic VSMC death in atherosclerotic plaque may hence contribute to the destabilization of the fibrous cap, and increase the risk of plaque rupture and thrombosis [10,11]. The triggers inducing VSMC apoptosis in le-sions, however, are unknown at present.

There is evidence that oxidatively modified low-den-sity lipoprotein (oxLDL) exists in atherosclerotic le-sions of rabbit and humans [12]. OxLDL exerts several potentially atherogenic properties including cytotoxicity to vascular cells [13,14], impairment of endothelium-de-pendent relaxation of blood vessel [15,16], stimulation of monocyte recruitment and their adhesion to en-dothelial cells [17,18]. To study oxLDL-induced cyto-toxicity, the cell culture system provides a controlled environment to examine mechanism of actions and to identify active components of oxLDL with limited in-terference. OxLDL, whether oxidized in a metal-ion system or by cells, has been shown to injure endothelial cells [19], smooth muscle cells [20], fibroblasts [21,22], and macrophages [14]. However, the toxic-dose threshold for the cytotoxicity of oxLDL in VSMCs has not been determined. Furthermore, lysophosphatidyl-choline (lysoPC), produced during LDL oxidation by endogeneous phospholipase A2 [23], is also found in atherosclerotic lesions at high levels [24,25]. Although lysoPC has been proposed as one of the cytotoxins contained in oxLDL [26], quantitative studies are yet to be performed. Moreover, whether lysoPC could induce apoptosis in cultured VSMC was not clear [27,28]. A parallel study between the effects of oxLDL and lysoPC may provide additional information concerning the na-ture of oxLDL action.

We thus chose to investigate the relative importance of the amount and/or the oxidative level in determining the threshold concentration of oxLDL for cytotoxicity as well as the dose – response pattern of lysoPC. We also tested whether lysoPC, over the range of concen-trations employed in cytotoxicity studies, could induce apoptosis in cultured VSMCs.

2. Materials and methods

2.1. Cell culture

VSMCs were isolated from thoracic aorta of 20-week-old Wistar – Kyoto rats, identified and maintained in supplemented MCDB107 (J.R.Scientific, CA, USA)

as described previously [29,30]. Cells between the fourth and eighth passages were used.

2.2. Preparation and oxidation of LDL

Blood samples were obtained from the Blood Bank of Chang Gung Memorial Hospital. LDL (d=1.019 – 1.063 g/ml) and lipoprotein-deficient serum (LPDS) (d]1.21 g/ml) were isolated by sequential ultra-cen-trifugation as described by Havel et al. [31]. Briefly, saturated NaBr and phosphate-buffered saline (PBS) (NaCl, 136.75 mmol/l; KCl, 2.6 mmol/l; KH2PO4, 1.5 mmol/l; Na2HPO4·2H2O, 7.9 mmol/l; pH 7.4) were used for density adjustments. Following centrifugation (50 000 rpm for 24 h; Beckman 55.2 Ti rotor), isolated LDL and LPDS fractions were extensively dialyzed at 4°C in a cold room against 200× volume PBS in the presence of 0.1% ethylenediamine tetraacetic acid (EDTA) to remove remaining NaBr. Samples were then dialyzed against PBS overnight to remove EDTA prior to sterilization by filtration through 0.45 mm Gelman filters (Ann Arbor, MI, USA). Samples were stored at 4°C and used within 6 weeks. LDL concentration was expressed in term of its protein content [32], i.e. micro-grams of protein per millilitre of media. The LDL samples were also analyzed for contents of cholesterol and triglycerides (standard test of lipid profile, Chang Gung Memorial Hospital), and the weight ratio was 1.0:3.8:7.1 (triglyceride:protein:cholesterol). Apolipo-proteins (apo A1and apo B) were also analyzed by the method of Behring Turbitimer.

2.3. Oxidati6e modification of LDL

The oxidative modification of LDL was carried out by incubating freshly-prepared native LDL (nLDL) with ferrous sulfate (50 mmol/l in 0.9% NaCl solution; pH 7.4) for up to 24 h at 37°C [33]. The oxidation was terminated by first filtering the sample (0.45 mm) and then dialyzing oxLDL against phosphate buffer (with 0.1% EDTA; pH 7.4, 4°C) for three changes. OxLDL was sterilized by passing through a 0.22 mm filter (Millipore) and further dialyzed against the PBS for at least 48 h. The relative degree of oxidation for these preparations was measured by analyzing the presence of thiobarbituric acid-reactive substances (TBARS) and expressed as malondialdehyde (MDA) content (nmol MDA equivalent per milligram protein) as described elsewhere [34]. The levels of conjugated dienes (CD) of oxLDL (100 mg/ml) was also measured as the ab-sorbance at 234 nm using a spectrophotometer (DU-64; Beckman). These chemical changes that occur during oxidation of LDL preparations employed in this study are summarized in Fig. 1. Both TBARS and CD in-creased linearly with incubation time and thus allowed one to designate LDL with low-oxidative level (8 h,

(3)

low-oxLDL), oxidative level (16 h, medium-oxLDL), and high-oxidative level (24 h, high-medium-oxLDL), respectively. Utilizing a similar assay, untreated nLDL showed a value of TBARS less than 0.1 nmol MDA equivalent per milligram protein. Furthermore, the al-tered surface charge on the LDL protein was assessed by measuring the electrophoretic mobility, using Lipofilm kit (Sebia, France). The Lipofilm is composed of a gelified acrylic polymer divided into two zones of different concentration: 2% in the upper layer in which the sample wells are moulded, and 3% in the lower layer. Native LDL and oxLDL were stained with sudan black, and the electrophoresis was performed in 0.005% sodium azide buffer at 12 mA for 45 min. Following electrophoresis, the gel was dried at 51°C. The distance each sample travelled was measured in millimetres from the origin to the front of the band. The relative mobil-ity was expressed as folds of the migration distance of the sample to that of LDL standard. All oxLDL prepa-rations exhibited higher relative mobility than nLDL (data not shown).

2.4. Determination of cell 6iability

The cytotoxic effects of oxLDL and lysoPC (L-a -lysophosphatidylcholine, palmitoyl) on VSMCs were tested in a standard 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay as described previously [35]. Briefly, cells were dispensed in a 24-well flat-bottomed plate (Corning, NY, USA) at a density of 5×104 _cells_/_cm2 _{overnight for adequate attachment.} Cells were treated with nLDL, oxLDL, PC (D-a -phos-phatidylcholine, dipalmitoyl) or lysoPC for 24 h, and then MTT solution (0.5 mg/ml) was added into all wells. Plates were then further pre-incubated at 37°C for at least 4 h. After pre-incubation, acid – isopropanol (0.04 N HCl – isopropanol) was added to all wells and throughly mixed to dissolve the dark blue crystals for

30 min. The samples were read on a Dynatech MR710 Microelisa reader and absorbances were measured at 570 and 630 nm. The net difference of absorbances between 570 and 630 nm was used to express the viability of VSMCs as follow: relative viability=(Ae× 100)/Ac, whereAeis the absorbance of treated cells and Acis the absorbance of untreated controls.

2.4.1. Trypan blue exclusion test

The trypan blue exclusion test was performed to test for the loss of plasma membrane integrity [36]. VSMCs (5×105

cells per 35 mm dish) were cultured in 10% fetal calf serum (FCS)-supplemented MCDB107 overnight and then were treated with lysoPC. Follow-ing 24 h incubation, the supernatant and VSMCs were obtained by trypsinization and centrifugation (1000× g, 5 min). The cell pellet was resuspended in PBS and then was mixed with equal volume of 0.5% trypan blue solution (Serva Feinbiochemica, USA). Intact VSMCs excluded, the dye and cell number were counted using a hemacytometer (Cambridge Instruments, Inc., USA).

2.5. Flow cytometric analysis

VSMCs (5×105

cells) were cultured in 10% LPDS or 10% FCS-supplemented MCDB107 for 48 h and then nLDL, oxLDL, PC, or lysoPC was added. Following 24 h incubation, cells were isolated by trypsinization and centrifugation at 1000×gfor 5 min. The cell pellet was washed once, resuspended in 200 ml PBS at 4°C, and fixed in 2 ml ice-cold 70% ethanol for 30 min. The fixed cells were recovered by centrifugation for 5 min at 1000×g, washed twice in cold PBS. Cells were then treated with 0.1 mg/ml RNase A and 50 mg/ml propid-ium iodide (dissolved in PBS). Flow cytometric analysis was performed on FACScan (Becton Dickinson, USA) and tens of thousands of events were analyzed for each sample. The cell cycle distribution of each sample was calculated by MODIFIT2.0 software package. This time point (24 h) was selected because no significant apopto-sis was observed with propidium iodide before 18 h.

2.5.1. Annexin V and propidium iodide staining

VSMCs (5×105 _{cells per 35 mm dish) were cultured} in 0.5% FCS-supplemented MCDB107 overnight and then were treated with lysoPC (25mmol/l) for 3 or 24 h. The supernatant and VSMCs were obtained following trypsinization and centrifugation (1000×g, 5 min). Cells were washed once with PBS and then incubated with 0.5 mg/ml FITC-annexin V (AV) (Molecular Probes, Inc., USA) in 0.5 ml binding buffer (NaCl, 150 mmol/l; CaCl2, 2.5 mmol/l; MgCl2, 5 mmol/l; Hepes, 10 mmol/l; 20% bovine serum albumin; pH 7.4) for 15 min at room temperature. Following AV binding, cells were collected by centrifugation and resuspended in 0.5 ml binding buffer, and propidium iodide (PI) was added at Fig. 1. Effect of oxidation of LDL on levels of CD (in absorbance

units) and TBARS (in nmol MDA equivalent/mg protein). LDL was exposed to ferrous sulfate for different time intervals and analyzed for the presence of TBARS () and CD ( ).

(4)

the concentration of 0.6 mg/ml. FACS analysis was performed immediately after staining. The translocation of phosphatidylserine (PS) from the inner leaflet of the membrane outward (PS exposure), with cells remaining physically intact, represents an early event of apoptosis [37]. Apoptotic cells therefore can be stained with AV, which binds with high affinity to PS, resulting in a green fluorescence when excited at 450 – 480 nm. At the same time, PI capable of passing the plasma membrane is excluded (AV+_/_PI−_{). Necrotic cells have lost the}

physical integrity of their plasma membrane and are therefore stained with PI, which fluorescences in the red when excited at 510 – 550 nm (AV+

/PI+

or AV− /PI+

). Cells which are neither apoptotic nor necrotic did not stain with either dye (AV−

/PI−_{). The percentage of}

apoptotic or necrotic VSMCs was calculated by the CELLQUEST software package.

2.6. Gel electrophoresis analysis of fragmented DNA

VSMCs (1×106 _{cells) were treated with increasing} concentrations of lysoPC for 24 h. After treatment, the supernatant of cell culture and VSMCs were collected, and then DNA was purified by QIAamp Tissue Kit (QIAGEN GmbH, Germany). DNA samples were sep-arated on 1.8% agarose gels (90 V, 2 – 3 h) containing ethidium bromide and visualized under ultraviolet light.

2.7. Quantitation of fragmented DNA

The percentage of DNA fragmentation was measured with the [3_{H]thymidine release assay as} Baumgartner-Parzer et al. described [38]. Subconfluent cultures of VSMCs were labelled with [3_{H]thymidine (1}_m_Ci_/_{ml) for} 36 h and the cells were then treated with indicated LDL, oxLDL, lysoPC. DNA fragmentation was deter-mined as follows: 0.5 ml lysis buffer (Tris, 20 mmol/l;, EDTA, 4 mmol/l; 0.4% Triton X-100; pH 7.4) was added to each culture well and mixed by pipetting, and the cell suspension was transferred to an eppendorf tube, incubated on melting ice for 10 min and cen-trifuged at 8000×g for 5 min at 4°C. Subsequently, fragmented radiolabelled DNA was counted in the supernatant by liquid scintillation counting. Radioac-tivity of cells treated with lysis buffer and ultrasound homogenator was used as total activity. Results of fragmented DNA were expressed as a percentage of total DNA.

2.8. Statistical analyses

Results are expressed as mean9SEM. The means of VSMC viability (by MTT assay or trypan blue exclu-sion test) and DNA fragmentation (by thymidine re-lease assay) were analyzed using analysis of variance (ANOVA) for multiple comparisons. Paired analysis

between control and an individual treatment group was performed using Student’s t-test, where ANOVA indi-cated significance for the multiple comparisons. Two-tailed probability values less than 0.05 were considered significant. Linear regression and all comparisons were done by GRAPHPAD INSTAT 2.0 program (GraphPad software, CA, USA). The dose of high-oxLDL or lysoPC for causing 50% SMCs death (IC50) were calcu-lated by GRAPHPAD INPLOT 4.0 program.

3. Results

Morphological changes of VSMCs following treat-ments of oxLDL or lysoPC were observed. It was found that 50 mg/ml high-oxLDL or 25 mmol/l lysoPC could induce retraction of VSMCs [39]. With higher concentrations of oxLDL (200 mg/ml) or lysoPC (75 mmol/l), numerous rounded and floated cells were ob-served (data not shown). These parallel changes be-tween oxLDL and lysoPC were similar to previous findings, including our own [28,39].

Fig. 2A shows the dependence of cytotoxicity on the oxidative level of oxLDL. No significant cytotoxicity was demonstrable by nLDL and low-oxLDL even at protein concentrations up to 200 mg/ml. However, at the same concentration, the percentages of dead cells were 19.896.5% (PB0.05) and 93.293.8% (PB

0.001) for medium-oxLDL and high-oxLDL, respec-tively. Thus, as the oxidation level of the lipoprotein preparation increased, the cytotoxic effect also in-creased. Furthermore, VSMC viability decreased persis-tently as the concentration of high-oxLDL increased (Fig. 2B). At 200 mg/ml, cell viability was only 4.19

2.0%, similar to that in Fig. 2A. The cytotoxic effect of high-oxLDL on VSMCs was hence also dose depen-dent. The effective dose of high-oxLDL for causing 50% VSMC death (IC50) using sigmoid curve regression was calculated to be 10894.1mg/ml (correlation coeffi-cient=0.997; PB0.001). This estimate was consistent with earlier findings [40], where a low molecular weight fraction of oxLDL was shown to induce cell death at 150 mg/ml but not at 75 mg/ml. To compare the re-sponses of VSMCs to the treatments of oxLDL or lysoPC, a calculation of equivalent lysoPC content in oxLDL was estimated based on the report by Sakai et al. [41], where about 200 – 600 nmol lysoPC/mg oxLDL protein was determined. Accordingly, a range of 0 – 100 mmol/l lysoPC concentration was chosen for investiga-tion. Similarly, lysoPC also induced VSMC cytotoxicity in the absence of serum, and the effect became signifi-cant (PB0.001) at concentration ]30 mmol/l (Fig. 3A). The decrease of cell viability was also dose depen-dent and the IC50 of lysoPC was 36.691.5 mmol/l. However, at concentrations 580mmol/l, phosphatidyl-choline (PC) did not cause any significant cytotoxicity,

(5)

Fig. 2. Cytotoxic effect of oxLDL on VSMC. VSMCs were exposed to nLDL or oxLDL of various oxidative levels (concentrations all at 200mg/ml) in the presence of 10% LPDS. TBARS level of oxLDL: low-oxLDL=3.5, medium-oxLDL=12.7, high-oxLDL=20.4 nmol MDA equivalent/mg protein (A). Results were obtained from two experiments each with duplicate determinations. VSMCs were also exposed to increasing concentrations (0 – 200mg/ml) of high-oxLDL (B). Results were obtained from two experiments each with duplicate determinations. Compared with control (10% LPDS alone), *PB 0.05, **PB0.001.

Near-complete necrosis of VSMCs was found at 200 mM, the highest concentration employed. Thus, VSMCs were much less sensitive to the toxic effect of lysoPC in the presence of FCS.

Cell cycle distribution was examined by flow cytome-try analysis. Following high-oxLDL (200 mg/ml) treat-ment (24 h), cells accumulated principally in the G1

Fig. 3. Cytotoxic effect of lysoPC on VSMCs. VSMCs were treated with increasing concentrations (0 – 100mmol/l) of PC () or lysoPC ( ) for 24 h in the absence of serum (A). VSMCs were also incubated in the presence of 10% FCS (B), while treated with lysoPC ( , 0 – 300 mmol/l) or further tested with trypan blue exclusion (, 0 – 200 mmol/l). * PB0.05, ** PB0.001, compared with control (serum-free MCDB107 for A; 10% FCS only for B). Results were combined from three or four experiments each with triplicate determi-nations.

a slight cytotoxic effect (about 10%) was observed only at concentrations of 90 and 100 mmol/l; it was much less than that caused by equal concentration of lysoPC where near-complete cytotoxicity was found (PB

0.001). In the presence of 10% serum (FCS), the cyto-toxic effects of lysoPC were reduced. Fig. 3B illustrates that cytotoxicity was only 25% (assayed by MTT test) at 100mM added lysoPC. The value of IC50was 3-fold of that found before (Fig. 3B compared with Fig. 3A; note the different scale). Near-complete cytotoxicity was not achieved until concentration of lysoPC reached 250 mM; again, almost a 3-fold increase as before (cf. Fig. 3A). We also examined the effect of lysoPC on membrane intactness by trypan blue exclusion assay in the presence of 10% FCS (Fig. 3B, triangles). Signifi-cant cell lysis was observed at a lysoPC concentration of 30mM (Fig. 3B) and the effect was dose dependent.

Table 1

Cell cycle distribution of SMCs in the treatment of nLDL or oxLDL

VSMC (%) Treatment

G2/M

S G0/G1

80.4 9.7

Controla _9.9

11.1 9.8

nLDL (200mg/ml) 79.1

78.3 10.4

Low-oxLDL (200mg/ml) 11.3

88.2 1.9

High-oxLDL (200mg/ml) 9.9

(6)

Table 2

Cell cycle distribution of SMCs in the treatment of PC or lysoPC

Treatment VSMC (%)

G0/G1 G2/M

45.9

46.8 7.3

Controla

52.3

LysoPC (10mmol/l) 37.9 9.8

53.9

LysoPC (50mmol/l) 37.1 9.0

28.3

58.7 13.0

LysoPC (150mmol/l)

PC (150mmol/l) 54.8 37.4 7.8

a_{Control cells were cultured in MCDB107}₊_{10% FCS alone.}

of PC only exerted an effect similar to that of 10mmol/l lysoPC.

To quantify the extent of apoptosis, the percentage of fragmented DNA by [3_{H]thymidine release assay} fol-lowing the treatment of oxLDL or lysoPC has been measured. Although exposure to nLDL or low-oxLDL for 24 h failed to induce apoptosis (Fig. 4A), incubation in medium- and high-oxLDL caused thymidine release significantly, reaching 14.290.9 and 16.692.1%, re-spectively. Furthermore, there was a significant dose-dependent increase when treated with different concentrations of high-oxLDL (Fig. 4B). To determine whether lysoPC at concentrations parallel to effective oxLDL doses also induces apoptosis in VSMC, total DNA was isolated and analyzed. Typical DNA ladder pattern was observed from cells treated for 24 h with 25 and 50mmol/l lysoPC (Fig. 5, lanes 3 and 4). However, clear DNA laddering at lysoPC concentration of 10 mmol/l (Fig. 5, lane 2) was not observed, consistent with an earlier report [27]. In addition, Fig. 6 illustrates that lysoPC also induced apoptosis of VSMCs in a dose-de-pendent manner. Similar to the DNA ladder pattern, it was found that VSMC treated with low concentration of lysoPC (10mmol/l) was of no significant effect but at higher concentration of lysoPC (]25 mmol/l), signifi-cant DNA fragmentation was observed.

To further examine the nature of lysoPC-induced apoptosis in VSMC, phosphatidylserine (PS) exposure was employed as an early marker of apoptosis [37]. In a typical experiment, double staining of PS exposure (FITC-annexin V or AV, FL1-H) as well as membrane disruption (propidium iodide or PI, FL2-H) were per-formed (see Section 2.5.1) by flow cytometry, and the results showed that PS exposure was detectable as early as 3 h following lysoPC (25 mM) treatment (Fig. 7B, AV+_/_PI− _{cells of lower right panel). At 24 h, both}

apoptosis and necrosis (AV+_/_PI+ _{cells of upper right}

panel) increased significantly (Fig. 7C). Statistical anal-ysis of six determinations indicated that fractional apoptotic VSMCs was significantly higher in lysoPC-treated cells at 3 h (4.390.2% versus 1.490.1%) when compared with control VSMCs; the fraction reached 14% or 7-fold of untreated VSMCs (2.090.1%) at 24 h. Two more identical experiments showed similar results.

4. Discussion

Some of the chemical changes that occur during iron oxidation of LDL have been characterized, including increased negative charge (gel mobility), increased con-tent of CD, and TBARS. Highly significant correlation could be derived (e.g. from Fig. 1) between the level of TBARS of oxLDL samples with relative electrophoretic mobility (r=0.982; PB0.001, n=6) or with contents Fig. 4. Induction of apoptosis in VSMCs by oxLDL. VSMCs were

exposed to LDL (200mg/ml) or oxLDL of various oxidative levels (200mg/ml) for 24 h (A). TBARS levels of different oxLDL prepara-tions are given in Section 2 (Fig. 1). Results obtained from two experiments each with triplicate determinations. VSMCs were also exposed to increasing concentrations of high-oxLDL (B). Results were obtained from two experiments each with triplicate determina-tions. *PB0.001, compared with control (10% LPDS alone).

phase of the cell cycle with a concomitant reduction (7.8%) in the proportion of cells in the S phase. In contrast, no change in the proportion of cells in the S phase was evident in VSMC after low-oxLDL or nLDL treatment (Table 1). Similarly, at concentration of 10 or 50mmol/l, lysoPC led to a decrease of about 8% in the proportion of VSMCs in the S phase than that of control group. We also found a 17% reduction in the proportion of VSMCs in the S phase and 12% more cells accumulated in the G1 phase with 150 mmol/l lysoPC (Table 2). The same concentration (150mmol/l)

(7)

Fig. 5. Electrophoresis of DNA isolated from VSMCs cultured under serum-free conditions with 0, 10, 25 and 50mmol/l lysoPC for 24 h (lanes 1 – 4, respectively). A 100 bp DNA molecular weight marker was loaded on lane M.

high-oxLDL (middle column, Fig. 2B). The TBARS contents of these preparations were also similar (2.0 – 2.5 nmol MDA equivalents/ml). This ‘oxidative equiva-lency’ of induced cytotoxicity was further confirmed by considering the dose – response pattern of the high-oxLDL-induced cytotoxic effect (Fig. 2B). If we convert cell viability to cytotoxicity, i.e. 1-viability, a linear response between cytotoxicity and the calculated

Fig. 7. Apoptotic and necrotic death of VSMCs were distinguished using FITC-annexin V (AV) label and propidium iodide (PI) stain. VSMCs were treated as indicated above each panel and then analyzed by flow cytometry. The lower left quadrants of each panel show the viable cells, which exclude PI and are negative for AV binding (AV−_/_PI−_{). The lower right quadrants represent the apoptotic cell,}

positive for AV binding and negative for PI uptake (AV+_/_PI−_{). The}

upper quadrants (left and right) represent the necrotic cells, positive for PI uptake with or without AV fluorescence. FL1-H, Fluorescence height of AV; FL2-H, fluorescence height of PI. Data was taken from one out of six determinations in a typical experiment. Three such experiments were performed. Panels A, B, C were typical results for untreated control, lysoPC-treated for 3 h, and lysoPC-treated for 24 h, respectively.

Fig. 6. Dose-dependent induction of apoptosis in VSMCs by lysoPC. Serum-deprived VSMCs were treated with increasing concentrations of lysoPC (0 – 50mmol/l) for 24 h. Results were obtained from two experiments each with triplicate determinations. *PB0.005, **PB 0.001, compared with control (in the absence of lysoPC).

of CD (r=0.985; PB0.001, n=6). Therefore, the ex-tent of lipid peroxidation, as measured by the concen-tration of TBARS of iron-oxidized LDL, could be used as a marker for oxidative modifications of lipoprotein similar to those of the copper-oxidized LDL prepara-tions [42], lipoxygenase-modified LDL [43], and elec-tronegatively charged LDL (LDL−

) isolated from hyper-cholesterolemic plasma [19]. The TBARS content of oxLDL preparations was thus employed for quanti-tative analysis of relative cytotoxicity.

A strong correlation between the cytotoxicity of the oxLDL and the respective extent of TBARS in various preparations of oxLDL was found (r=0.984; PB

0.001, n=32). When the content of TBARS was less than 1.0 nmol MDA equivalents/ml (amount contained in about 50 mg/ml high-oxLDL), there was no signifi-cant cytotoxicity even at high concentration (e.g. 200 mg/ml low-oxLDL in Fig. 2A). However, preparations of oxLDL with stronger oxidation killed VSMC and 200mg/ml medium-oxLDL caused an extent of cytotox-icity (Fig. 2A) similar to that caused by 100 mg/ml

(8)

TBARS content can be obtained (correlation coeffi-cient=0.993). It thus appeared that the toxic-dose threshold of oxLDL for VSMCs was approximate 1.0 nmol MDA equivalents/ml in the absence of serum, irrespective of the concentration of oxLDL prepara-tions per se.

Several lines of evidence suggest that lysoPC, a major component of oxidatively modified LDL, exerts vascu-lar effects simivascu-lar to those induced by oxLDL. Both oxLDL and lysoPC can induce cytotoxicity of vascular cells [13,26], impairment of endothelium-dependent re-laxation [16,44], as well as activation of endothelial adhesiveness [17,45]. It was found that similar to oxLDL, lysoPC exhibited a dose-dependent (with IC50=36.691.5 mmol/l; r=0.999), and near-complete (at a concentration of 50mmol/l) cytotoxic effect in the absence of serum (Fig. 3A). Even in the presence of 10% FCS, lysoPC exerted a dose-dependent cytotoxic-ity (Fig. 3B). These observed dose dependency of lysoPC-induced cytotoxicity are consistent with the view that lysoPC content in oxLDL preparations in-creases not only as amount of oxLDL inin-creases [41], but also as level of oxidation increases [46].

The similarity between oxLDL- and lysoPC-induced cytotoxicity of VSMCs extends beyond the observa-tions that both exhibited dose-dependent cytotoxicity in VSMCs. Serum proteins may bind oxLDL or lysoPC and thus hinder their transfer to vascular cell [47]; however, specific protective effects of serum on cell survival may contribute also. In fact, low concentra-tions of oxLDL and lysoPC (with 0.25% serum) both stimulate VSMC to enter cell cycle partially via an autocrine or paracrine action to release endogenous basic fibroblast growth factor [48], which is also respon-sible for the observed lysoPC-induced migration of VSMC [49].

Furthermore, it has been demonstrated that both oxLDL and lysoPC exerted cytostatic effect on VSMCs using flow cytometric analyses (Tables 1 and 2). Cyto-toxicity of oxLDL and lysoPC could be due to selectiv-ity of the proportion of VSMCs in the S phase, consistent with the findings of Hodis et al. where sub-confluent proliferating-endothelial cells were more sen-sitive to the cytotoxic effects of oxLDL [19], and/or cells were inhibited from entering the S phase.

Although the mechanism responsible for the cytotox-icity of lysoPC is not well understood, Kume et al. had reported that lysoPC is a polar phospholipid that can exhibit detergent-like properties [50]. Amphiphiles in an aqueous solution can cause cell lysis above their critical micellar concentrations (CMC). Previous studies have shown that CMC of lysoPC (C16:0) is 40 – 50mmol/l in protein-free physiological salt solutions [51]. Above CMC, lysoPC may directly perturb membrane structure and impair the function of macromolecules embedded in the membrane. We found significant cytotoxicity of

lysoPC over a broad range of concentrations in the absence (Fig. 3A) or presence of serum (Fig. 3B) and that cell damage was observed at sub-CMC, suggesting that a non-detergent action of lysoPC could not be excluded. However, direct test of membrane intactness by trypan blue exclusion indicated significant damage occurred at sub-CMC concentrations (Fig. 3B), consis-tent with a detergent action of lysoPC. In addition, Ohara et al. had reported that lysoPC can activate protein kinase C in intact vessels [52], leading to an increase in O2− production. O2− also provides a source of other oxygen-centred radicals, such as H2O2and

. OH, which may cause membrane damage. It is also well recognized that lipoproteins, especially LDL and oxLDL, raises intracellular Ca2+ _level _probably

through the activation of calcium channel (for a review, see Ref. [53]). Recent report suggests that oxLDL but not nLDL activates the L-type Ca2+ _{channel in smooth}

muscle-derived cell line [54], and it is interesting to note that lysoPC may also activate such a Ca2+ _channel

[55]. Therefore, lysoPC could injure VSMCs via multi-ple mechanisms.

Our results also have shown that oxLDL or lysoPC could lead to distinct types of VSMCs death: apoptosis and necrosis. In the presence of 200 mg/ml oxLDL, 95.992.0% of VSMC died (Fig. 2B) while the mean fractional fragmented-DNA content was 17.490.5% (Fig. 4B). Moreover, 50 mmol/l lysoPC showed similar effects on VSMCs: the cytotoxicity of VSMCs was 82.191.1% (Fig. 3) and the mean fractional frag-mented-DNA content was 26.492.3% (Fig. 6). These observations were consistent with the findings of Crisby et al.. They demonstrated that the vast majority of injured VSMCs in the plaque undergoing cell death by necrosis and less by apoptosis [1]. Furthermore, lysoPC-induced apoptosis, as indexed by PS exposure, could be detected as early as 3 h and increased with time to 14% at 24 h following lysoPC treatment (Fig. 7B,C). Therefore, oxLDL and its active component, lysoPC (]25 mmol/l), both cause VSMC apoptosis. Watanabe et al. had also demonstrated that oxLDL could induced apoptosis in cultured VSMCs [27]. They further found that oxLDL induced apoptosis through activation of a CPP32-like protease and bcl-2 protein downregulation [56]. The signal transduction pathway for the induction of apoptosis by lysoPC is still un-known. Since lysoPC can bind to a receptor in the plasma membrane such as fatty acid binding protein or to the plasma membrane in a receptor-independent manner [57], the capacity to activate various signal pathways may exist.

However, the oxidation of LDL leads to the forma-tion of several cytotoxins, including lipid hydroperox-ides, alkenals, oxysterols and lysoPC [58]. Therefore, the cytotoxic effect of oxLDL to VSMCs may derive from a dominant toxin among these, but also may be

(9)

due to the combined actions of multiple agents. Several studies had demonstrated that the cytotoxicity of oxLDL could be mimicked by a lipid extract of the LDL [59,60]. Hughes et al. had identified the most toxic fraction of copper-oxidized LDL to be 7-ketocholesterol for cul-tured procine VSMCs [61]. In addition, Chisolm et al. reported that 7b-hydroperoxycholesterol was the princi-pal cytotoxin of oxLDL when examined using human and rabbit VSMC culture [62]. The current results showed that lysoPC was highly cytotoxic on VSMC as well. Apparently, methods for preparing oxLDL samples (e.g. LDL source and oxidation procedure), conditions for cytotoxicity test (e.g. incubation time and types of test), and the culturing cell systems (e.g. medium serum content and origin of cells) employed may all influence the outcome of cytotoxicity test; it is hence useful to compare the actions of oxLDL with its putative compo-nent by quantitative analysis and by determining more than one parameter, such as necrosis versus apoptosis. In conclusion, oxLDL and its active component, lysoPC, both induce VSMC death in a dose-dependent manner. However, a certain threshold oxidative level (ca. 1.0 nmol MDA/ml) could be detected such that the dose – response pattern actually reflected an ‘oxidative equivalency’. Both oxLDL and lysoPC elicited dose-de-pendent cytotoxicity and apoptosis, suggesting that lysoPC, being a major lipid component of oxidized LDL, could account for not only the cytotoxic effect, but also some of the apoptosis induced by oxLDL.

Acknowledgements

This study was supported by the National Science Council (NSC87-2314-B-182-088), Department of Health, Executive Yuan (Research Center DOH86-HR-610) and Chang Gung University (CMRP 736) to Y.T.L. C.C.H. also thanks the scholarship support from the Foundation of Biomedical Sciences of National Defense Medical Center. The editorial assistance of Cynthia Huang and the excellent assistance of L.Y. Chen in the preparation of the manuscript is gratefully acknowl-edged.

References

[1] Crisby M, Kallin B, Thyberg J, Zhivotovsky B, Orrenius S, Kostulas V, Nilsson J. Cell death in human atherosclerotic plaques involves both oncosis and apoptosis. Atherosclerosis 1997;130:17 – 27.

[2] Isner JM, Kearney M, Bortman S, Passeri J. Apoptosis in human atherosclerosis and restenosis. Circulation 1995;91:2703 – 11.

[3] Guyton JR, Klemp KF. Transitional features in human atherosclerosis. Intimal thickening, cholesterol clefts and cell loss in human aortic fatty streaks. Am J Pathol 1993;143:1444 – 57.

[4] Bennett MR, Boyle JJ. Apoptosis of vascular smooth muscle cells in atherosclerosis. Atherosclerosis 1998;138:3 – 9.

[5] Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular prolif-eration after arterial injury I: smooth muscle growth in the absence of endothelium. Lab Invest 1983;49:327 – 33.

[6] Perlmann H, Mailiard L, Krasinski K, Walsh K. Evidence for the rapid onset of apoptosis in medial smooth muscle cells after balloon injury. Circulation 1997;95:981 – 7.

[7] Hegyi L, Skepper JN, Cary NRB, Mitchinson MJ. Foam cell apoptosis and the development of the lipid core of human atherosclerosis. J Pathol 1996;180:423 – 9.

[8] Geng YJ, Libby P. Evidence for apoptosis in advanced human atheroma. Colocalization with interleukin-1b-converting enzyme. Am J Pathol 1995;147:251 – 66.

[9] Han DKM, Haudenschild CC, Hong MK, Tinkle BT, Leon MB, Liau G. Evidence for apoptosis in human atherogenesis and in a rat vascular injury model. Am J Pathol 1995;147:267 – 77. [10] Falk E. Why do plaques rupture? Circulation 1992;86(Suppl.

III):30 – 42.

[11] Ridolfi RL, Hutchins GM. The relationship between coronary artery lesions and myocardial infarcts: ulceration of atheroscle-rotic plaques precipitating coronary thrombosis. Am Heart J 1977;93:468 – 86.

[12] Yla¨-Herttuala S, Palinski W, Rosenfeld ME, Parthasarathy S, Carew TE, Butler S, Witztum JL, Steinberg D. Evidence for the presence of oxidatively modified low density lipoprotein in atherosclerotic lesions of rabbit and man. J Clin Invest 1989;84:1086 – 95.

[13] Hughes H, Mathews B, Lenz ML, Guyton JR. Cytotoxicity of oxidized LDL to procine aortic smooth muscle cells is associated with the oxysterols 7-ketocholesterol and 7-hydroxycholesterol. Arterioscler Thromb 1994;14:1177 – 85.

[14] Liu SX, Zhou M, Chen Y, Wen WY, Sun MJ. Lipoperoxidative injury to macrophages by oxidatively modified low density lipo-protein may play an important role in foam cell formation. Atherosclerosis 1996;121:55 – 61.

[15] Simon B, Cunningham LD, Cohen RA. Oxidized low-density lipoproteins cause contraction and inhibit endothelium-depen-dent relaxation in the pig coronary artery. J Clin Invest 1990;86:75 – 9.

[16] Galle J, Mulsch A, Busse R, Bassenge E. Effects of native and oxidized low density lipoproteins on formation and inactivation of endothelium-derived relaxing factor. Arterioscler Thromb 1991;11:198 – 203.

[17] Berliner JA, Territo MC, Sevanian A, Ramin S, Kim JA, Bamshad B, Esterson M, Fogelman AM. Minimally modified low density lipoprotein stimulates monocyte endothelial interac-tions. J Clin Invest 1990;85:1260 – 6.

[18] Liao F, Berliner JA, Mehrabian M, Navab M, Demer LL, Lusis AJ, Fogelman AM. Minimally modified low density lipoprotein is biologically active in vivo in mice. J Clin Invest 1991;87:2253 – 7.

[19] Hodis HN, Kramsch DM, Avogaro P, Bittolo-Bon G, Cazzolato G, Hwang J, Peterson H, Sevanian A. Biochemical and cytotoxic characteristics of an in vivo circulating oxidized low density lipoprotein (LDL−_{). J Lipid Res 1994;35:669 – 77.}

[20] Guyton JR, Lenz ML, Mathews B, Hughes H, Karsan D, Selinger E, Smith CV. Toxicity of oxidized low density lipo-proteins for vascular smooth muscle cells and partial protection by antioxidants. Atherosclerosis 1995;118:237 – 49.

[21] Colles SM, Irwin KC, Chisolm GM. Roles of multiple oxidized LDL lipids in cellular injury: dominance of 7b-hydroperoxyc-holesterol. J Lipid Res 1996;37:2018 – 28.

[22] Cathcart MK, McNally AK, Chisolm GM. Lipoxygenase-medi-ated transformation of human low density lipoprotein to an oxidized and cytotoxic complex. J Lipid Res 1991;32:63 – 70.

(10)

[23] Steinbrecher UP, Parthasarathy S, Leake DS, Witztum JL, Steinberg D. Modification of low density lipoprotein by endothe-lial cells involves lipid peroxidation and degradation of low density liporotein phospholipids. Proc Natl Acad Sci USA 1984;81:3883 – 8.

[24] Portman OW, Alexander M. Lysophosphatidylcholine concen-trations and metabolism in aortic intima plus inner media: effect of nutritionally induced atherosclerosis. J Lipid Res 1969;10:158 – 65.

[25] Keaney JF Jr, Xu A, Cunningham D, Jackson T, Frei B, Vita JA. Dietary probucol preserves endothelial function in choles-terol-fed rabbit by limiting vascular oxidative stress and superox-ide generation. J Clin Invest 1995;95:2520 – 9.

[26] Nilsson J, Dahlgren B, Ares M, Westman J, Nilsson AH, Cercek B, Shah PK. Lipoprotein-like phospholipid particles inhibit the smooth muscle cell cytotoxicity of lysophosphatidylcholine and platelet-activating factor. Arterioscler Thromb Vasc Biol 1998;18:13 – 9.

[27] Nishio E, Arimura S, Watanabe Y. Oxidized LDL induces apoptosis in cultured smooth muscle cells. A possible role for 7-ketocholesterol. Biochem Biophys Res Commun 1996;223:413 – 8.

[28] Hsieh CC, Yen MH, Lau YT. Oxidized low density lipoprotein (oxLDL) and lysophosphatidylcholine (lysoPC) could induce apoptosis of cultured smooth muscle cells (VSMC). Atheroscle-rosis 1998;136(Suppl.):S76.

[29] Chamley-Campbell J, Campbell GR, Ross R. The smooth mus-cle cell in culture. Physiol Rev 1979;59:1 – 61.

[30] Hsieh CC, Lau YT. Migration of vascular cells is enhanced in cultures derived from spontaneously hypertensive rat. Pflu¨g Arch Eur J Physiol 1998;435:286 – 92.

[31] Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in hu-man serum. J Clin Invest 1955;34:1345 – 53.

[32] Bradford MM. A rapid and sensitive method for the quantita-tion of microgram quantites of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248 – 54.

[33] Murgesan G, Chisolm GM, Fox PL. Oxidized low density lipoprotein inhibits the migration of aortic endothelial cells in vitro. J Cell Biol 1993;120:1011 – 9.

[34] Schuh J, Fairclough GF Jr, Haschemeyer RH. Oxygen-mediated heterogeneity of apo-low-density lipoprotein. Proc Natl Acad Sci USA 1978;75:3173 – 7.

[35] Mosmann T. Rapid colormetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55 – 63.

[36] Escargueil-Blanc I, Salvayre R, Ne`gre-Salvayre A. Necrosis and apoptosis induced by oxidized LDL occur through two calcium-dependent pathways in lymphoblastoid cells. FASEB J 1994;8:1075 – 80.

[37] Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for apoptosis: flow cytometric detection of phos-phatidylserine expression on early apoptotic cells using fluores-cein labelled annexin V. J Immunol Methods 1995;184:39 – 51. [38] Baumgartner-Parzer SM, Wanger L, Pettermann M, Grillari J,

Gessl A, Waldha¨usl W. High-glucose-triggered apoptosis in cul-tured endothelial cells. Diabetes 1995;44:1323 – 7.

[39] Auge N, Fitoussi G, Bascands JL, Pieraggi MT, Junquero D, Valet P, Girolami JP, Salvayre R, Ne`gre-Salvayre A. Mildly oxidized LDL evokes a sustained Ca2+_{-dependent retraction of}

vascular smooth muscle cells. Circ Res 1996;79:871 – 80. [40] Bjorkerud B, Bjorkerud S. Contrary effects of lightly and

strongly oxidized LDL with potent promotion of growth versus apoptosis on arterial smooth muscle cells, macrophages, and fibroblasts. Arterioscler Thromb Vasc Biol 1996;16:416 – 24. [41] Sakai M, Miyazaki A, Hakamata H, Sasaki T, Yui S, Yamazaki

M, Shichiri M, Houiuchi S. Lysophosphatidylcholine play an

essential role in the mitogenic effect of oxidized low density lipoprotein on murine macrophages. J Biol Chem 1994; 269:31430 – 5.

[42] Stiko A, Regnstrom J, Shah PK, Cercek B, Nilsson J. Active oxygen species and lysophosphatidylcholine are involved in oxi-dized low density lipoprotein activation of smooth muscle cell DNA synthesis. Arterioscler Thromb Vasc Biol 1996;16:194 – 200. [43] Cathcart MK, McNally AK, Chisolm GM. Lipoxygenase-medi-ated transformation of human low density lipoprotein to an oxidized and cytotoxic complex. J Lipid Res 1991;32:63 – 70. [44] Mangin EL, Kugiyama K, Nguy JH, Kerns SA, Henry PK.

Effects of lysolipids and oxidatively modified low density lipo-protein on endothelium-dependent relaxation of rabbit aorta. Circ Res 1993;72:161 – 6.

[45] Murohara T, Scalia R, Lefer AM. Lysophosphatidylcholine promotes P-selectin expression in platelets and endothelial cells possible involvement of protein kinase C activation and its inhibition by nitric oxide donors. Circ Res 1996;78:780 – 9. [46] Cho S, Hazama M, Urata Y, Goto S, Horiuchi S, Sumikawa K,

Kondo T. Protective role of glutathione synthesis in response to oxidized low density lipoprotein in human vascular endothelial cells. Free Radic Biol Med 1999;26:589 – 602.

[47] Stoll LL, Oskarsson HJ, Spector AA. Interaction of lysophos-phatidylcholine with aortic endothelial cells. Am J Physiol 1992;262:H1853 – 60.

[48] Chai YC, Howe PH, DiCorleto PE, Chisolm GM. Oxidized low density lipoprotein and lysophosphatidylcholine stimulate cell cycle entry in vascular smooth muscle cells. J Biol Chem 1996;271:17791 – 7.

[49] Kohno M, Yokokawa K, Yasunari K, Minami M, Kano H, Hanehira T, Yoshikawa J. Induction by lysophosphatidylcholine, a major phospholipid component of atherogenic lipoproteins, of human coronary artery smooth muscle cell migration. Circulation 1998;98:353 – 9.

[50] Kume N, Cybulsky MI, Gimbrone MA Jr. Lysophosphatidyl-choline, a component of atherogenic lipoproteins, induces mononuclear leukocyte adhesion molecules in cultured human and rabbit arterial endothelial cells. J Clin Invest 1992;90:1138 – 44.

[51] Bergamann SR, Ferguson TB Jr, Sobel BE. Effects of amphiphiles on erythrocytes, coronary arteries, and perfused hearts. Am J Physiol 1981;240:H229 – 37.

[52] Ohara Y, Peterson TE, Zheng B, Kuo JF, Harrison DG. Lysophosphatidylcholine increases vascular superoxides anions production via protein kinase C activation. Arterioscler Thromb 1994;14:1007 – 13.

[53] Massaeli H, Pierce G. Involvement of lipoproteins, free radicals, and calcium in cardiovascular disease processes. Cardiovas Res 1995;29:597 – 603.

[54] Mio Y, Suga S, Osanai T, Kanazawa T, Onodera K, Wada J, Kamimura N, Wakui M. Oxidized LDLs but not native LDLs augment Ba2+ _{currents through L-Type Ca}2+ _{channels of the}

A7r5 smooth muscle-derived cell line. Cell Signal 1997;5:367 – 72. [55] Stoll L, Spector AA. Lysophosphatidylcholine causes cGMP-de-pendent verapamil-sensitive Ca2+ _{influx in vascular smooth}

muscle cells. Cell Physiol 1993;33:C885 – 93.

[56] Nishio E, Watanabe Y. Oxysterols induced apoptosis in cultured smooth muscle cells through CPP32 protease activation and bcl-2 downregulation. Biochem Biophys Res Commun 1996;226:928 – 34.

[57] Thumser AEA, Voysey JE, Wilton DC. The binding of lysophos-pholipids to rat liver fatty acid-binding protein and albumin. Biochem J 1994;301:801 – 6.

[58] Jurgen G, Hoff HF, Chisolm GM, Esterbauer H. Modification of human serum low density lipoprotein by oxidation — character-ization and pathophysiological implications. Chem Phys Lipids 1987;45:315 – 36.

(11)

[59] Hessler JR, Morel DW, Lewis LJ, Chisolm GM. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis 1983;3:215 – 22.

[60] More DW, Hessler JR, Chisolm GM. Low density lipoprotein cytotoxicity induced by free radical peroxidation of lipid. J Lipid Res 1983;24:1070 – 6.

[61] Hughes H, Mathews B, Lenz ML, Guyton JR. Cytotoxicity of oxidized LDL to procine aortic smooth muscle cells is associ-ated with the oxysterols 7-ketocholesterol and 7-ketocholes-terol. Arterioscler Thromb 1994;14:1177 – 85.

[62] Colles SM, Irwin KC, Chisolm GM. Roles of multiple oxi-dized LDL lipids in cellular injury: dominance of 7b-hydroper-oxycholesterol. J Lipid Res 1996;37:2018 – 28.

(1)

Table 2

Cell cycle distribution of SMCs in the treatment of PC or lysoPC

Treatment VSMC (%)

G0/G1 G2/M

45.9

46.8 7.3

Controla

52.3

LysoPC (10mmol/l) 37.9 9.8

53.9

LysoPC (50mmol/l) 37.1 9.0

28.3

58.7 13.0

LysoPC (150mmol/l)

PC (150mmol/l) 54.8 37.4 7.8

a_{Control cells were cultured in MCDB107+10% FCS alone.}

of PC only exerted an effect similar to that of 10

m

mol

/

l

lysoPC.

To quantify the extent of apoptosis, the percentage of

fragmented DNA by [

_{H]thymidine release assay}

fol-lowing the treatment of oxLDL or lysoPC has been

measured. Although exposure to nLDL or low-oxLDL

for 24 h failed to induce apoptosis (Fig. 4A), incubation

in medium- and high-oxLDL caused thymidine release

significantly, reaching 14.2

9 0.9 and 16.6

9 2.1%,

re-spectively. Furthermore, there was a significant

dose-dependent

increase

when

treated

with

different

concentrations of high-oxLDL (Fig. 4B). To determine

whether lysoPC at concentrations parallel to effective

oxLDL doses also induces apoptosis in VSMC, total

DNA was isolated and analyzed. Typical DNA ladder

pattern was observed from cells treated for 24 h with 25

and 50

m

mol

/

l lysoPC (Fig. 5, lanes 3 and 4). However,

clear DNA laddering at lysoPC concentration of 10

m

mol

/

l (Fig. 5, lane 2) was not observed, consistent with

an earlier report [27]. In addition, Fig. 6 illustrates that

lysoPC also induced apoptosis of VSMCs in a

dose-de-pendent manner. Similar to the DNA ladder pattern, it

was found that VSMC treated with low concentration

of lysoPC (10

m

mol

/

l) was of no significant effect but at

higher concentration of lysoPC (

]

25 m

mol

/

l),

signifi-cant DNA fragmentation was observed.

To further examine the nature of lysoPC-induced

apoptosis in VSMC, phosphatidylserine (PS) exposure

was employed as an early marker of apoptosis [37]. In

a typical experiment, double staining of PS exposure

(FITC-annexin V or AV, FL1-H) as well as membrane

disruption (propidium iodide or PI, FL2-H) were

per-formed (see Section 2.5.1) by flow cytometry, and the

results showed that PS exposure was detectable as early

as 3 h following lysoPC (25

m

M) treatment (Fig. 7B,

AV

_/

_PI

−

_{cells of lower right panel). At 24 h, both}

apoptosis and necrosis (AV

_/

_PI

_{cells of upper right}

panel) increased significantly (Fig. 7C). Statistical

anal-ysis of six determinations indicated that fractional

apoptotic VSMCs was significantly higher in

lysoPC-treated cells at 3 h (4.3

9 0.2% versus 1.4

9 0.1%) when

compared with control VSMCs; the fraction reached

14% or 7-fold of untreated VSMCs (2.0

9 0.1%) at 24

h. Two more identical experiments showed similar

results.

4. Discussion

Some of the chemical changes that occur during iron

oxidation of LDL have been characterized, including

increased negative charge (gel mobility), increased

con-tent of CD, and TBARS. Highly significant correlation

could be derived (e.g. from Fig. 1) between the level of

TBARS of oxLDL samples with relative electrophoretic

mobility (

r

=

0.982;

P

B

0.001,

n

=

6) or with contents

Fig. 4. Induction of apoptosis in VSMCs by oxLDL. VSMCs were

phase of the cell cycle with a concomitant reduction

(7.8%) in the proportion of cells in the S phase. In

contrast, no change in the proportion of cells in the S

phase was evident in VSMC after low-oxLDL or nLDL

treatment (Table 1). Similarly, at concentration of 10 or

50 m

mol

/

l, lysoPC led to a decrease of about 8% in the

proportion of VSMCs in the S phase than that of

control group. We also found a 17% reduction in the

proportion of VSMCs in the S phase and 12% more

cells accumulated in the G

phase with 150

m

mol

/

l

(2)

high-oxLDL (middle column, Fig. 2B). The TBARS

contents of these preparations were also similar (2.0 –

2.5 nmol MDA equivalents

/

ml). This ‘oxidative

equiva-lency’ of induced cytotoxicity was further confirmed by

considering the dose – response pattern of the

high-oxLDL-induced cytotoxic effect (Fig. 2B). If we convert

cell viability to cytotoxicity, i.e. 1-viability, a linear

response between cytotoxicity and the calculated

Fig. 7. Apoptotic and necrotic death of VSMCs were distinguished using FITC-annexin V (AV) label and propidium iodide (PI) stain. VSMCs were treated as indicated above each panel and then analyzed by flow cytometry. The lower left quadrants of each panel show the viable cells, which exclude PI and are negative for AV binding (AV−_/PI−_{). The lower right quadrants represent the apoptotic cell,}

positive for AV binding and negative for PI uptake (AV+_/PI−_{). The}

of CD (

r

=

0.985;

P

B

0.001,

n

=

6). Therefore, the

ex-tent of lipid peroxidation, as measured by the

concen-tration of TBARS of iron-oxidized LDL, could be used

as a marker for oxidative modifications of lipoprotein

similar to those of the copper-oxidized LDL

prepara-tions [42], lipoxygenase-modified LDL [43], and

elec-tronegatively charged LDL (LDL

−

) isolated from

hyper-cholesterolemic plasma [19]. The TBARS content

of oxLDL preparations was thus employed for

quanti-tative analysis of relative cytotoxicity.

A strong correlation between the cytotoxicity of the

oxLDL and the respective extent of TBARS in various

preparations of oxLDL was found (

r

=

0.984;

P

B

0.001,

n

=

32). When the content of TBARS was less

than 1.0 nmol MDA equivalents

/

ml (amount contained

in about 50

m

g

/

ml high-oxLDL), there was no

signifi-cant cytotoxicity even at high concentration (e.g. 200

m

g

/

ml low-oxLDL in Fig. 2A). However, preparations

of oxLDL with stronger oxidation killed VSMC and

200 m

g

/

ml medium-oxLDL caused an extent of

cytotox-icity (Fig. 2A) similar to that caused by 100

m

g

/

ml

(3)

TBARS content can be obtained (correlation

coeffi-cient

=

0.993). It thus appeared that the toxic-dose

threshold of oxLDL for VSMCs was approximate 1.0

nmol MDA equivalents

/

ml in the absence of serum,

irrespective of the concentration of oxLDL

prepara-tions per se.

Several lines of evidence suggest that lysoPC, a major

component of oxidatively modified LDL, exerts

vascu-lar effects simivascu-lar to those induced by oxLDL. Both

oxLDL and lysoPC can induce cytotoxicity of vascular

cells [13,26], impairment of endothelium-dependent

re-laxation [16,44], as well as activation of endothelial

adhesiveness [17,45]. It was found that similar to

oxLDL, lysoPC exhibited a dose-dependent (with

IC

=

36.6

9

1.5 m

mol

/

l;

r

=

0.999), and near-complete

(at a concentration of 50

m

mol

/

l) cytotoxic effect in the

absence of serum (Fig. 3A). Even in the presence of

10% FCS, lysoPC exerted a dose-dependent

cytotoxic-ity (Fig. 3B). These observed dose dependency of

lysoPC-induced cytotoxicity are consistent with the

view that lysoPC content in oxLDL preparations

in-creases not only as amount of oxLDL inin-creases [41],

but also as level of oxidation increases [46].

The similarity between oxLDL- and lysoPC-induced

cytotoxicity of VSMCs extends beyond the

observa-tions that both exhibited dose-dependent cytotoxicity in

VSMCs. Serum proteins may bind oxLDL or lysoPC

and thus hinder their transfer to vascular cell [47];

however, specific protective effects of serum on cell

survival may contribute also. In fact, low

concentra-tions of oxLDL and lysoPC (with 0.25% serum) both

stimulate VSMC to enter cell cycle partially via an

autocrine or paracrine action to release endogenous

basic fibroblast growth factor [48], which is also

respon-sible for the observed lysoPC-induced migration of

VSMC [49].

Furthermore, it has been demonstrated that both

oxLDL and lysoPC exerted cytostatic effect on VSMCs

using flow cytometric analyses (Tables 1 and 2).

Cyto-toxicity of oxLDL and lysoPC could be due to

selectiv-ity of the proportion of VSMCs in the S phase,

consistent with the findings of Hodis et al. where

sub-confluent proliferating-endothelial cells were more

sen-sitive to the cytotoxic effects of oxLDL [19], and

/

or

cells were inhibited from entering the S phase.

Although the mechanism responsible for the

cytotox-icity of lysoPC is not well understood, Kume et al. had

reported that lysoPC is a polar phospholipid that can

exhibit detergent-like properties [50]. Amphiphiles in an

aqueous solution can cause cell lysis above their critical

micellar concentrations (CMC). Previous studies have

shown that CMC of lysoPC (C16:0) is 40 – 50

m

mol

/

l in

protein-free physiological salt solutions [51]. Above

CMC, lysoPC may directly perturb membrane structure

and impair the function of macromolecules embedded

in the membrane. We found significant cytotoxicity of

lysoPC over a broad range of concentrations in the

absence (Fig. 3A) or presence of serum (Fig. 3B) and

that cell damage was observed at sub-CMC, suggesting

that a non-detergent action of lysoPC could not be

excluded. However, direct test of membrane intactness

by trypan blue exclusion indicated significant damage

occurred at sub-CMC concentrations (Fig. 3B),

consis-tent with a detergent action of lysoPC. In addition,

Ohara et al. had reported that lysoPC can activate

protein kinase C in intact vessels [52], leading to an

increase in O

2−

production. O

2−

also provides a source

of other oxygen-centred radicals, such as H

O

and

OH,

which may cause membrane damage. It is also well

recognized that lipoproteins, especially LDL and

oxLDL,

raises

intracellular

Ca

_level

_probably

through the activation of calcium channel (for a review,

see Ref. [53]). Recent report suggests that oxLDL but

not nLDL activates the L-type Ca

_{channel in smooth}

muscle-derived cell line [54], and it is interesting to note

that lysoPC may also activate such a Ca

_channel

[55]. Therefore, lysoPC could injure VSMCs via

multi-ple mechanisms.

Our results also have shown that oxLDL or lysoPC

could lead to distinct types of VSMCs death: apoptosis

and necrosis. In the presence of 200

m

g

/

ml oxLDL,

95.9

9 2.0% of VSMC died (Fig. 2B) while the mean

fractional fragmented-DNA content was 17.4

9 0.5%

(Fig. 4B). Moreover, 50

m

mol

/

l lysoPC showed similar

effects on VSMCs: the cytotoxicity of VSMCs was

82.1

9 1.1% (Fig. 3) and the mean fractional

frag-mented-DNA content was 26.4

9 2.3% (Fig. 6). These

observations were consistent with the findings of Crisby

et al.. They demonstrated that the vast majority of

injured VSMCs in the plaque undergoing cell death by

necrosis and less by apoptosis [1]. Furthermore,

lysoPC-induced apoptosis, as indexed by PS exposure,

could be detected as early as 3 h and increased with

time to 14% at 24 h following lysoPC treatment (Fig.

7B,C). Therefore, oxLDL and its active component,

lysoPC (

]

25 m

mol

/

l), both cause VSMC apoptosis.

Watanabe et al. had also demonstrated that oxLDL

could induced apoptosis in cultured VSMCs [27]. They

further found that oxLDL induced apoptosis through

activation of a CPP32-like protease and bcl-2 protein

downregulation [56]. The signal transduction pathway

for the induction of apoptosis by lysoPC is still

un-known. Since lysoPC can bind to a receptor in the

plasma membrane such as fatty acid binding protein or

to the plasma membrane in a receptor-independent

manner [57], the capacity to activate various signal

pathways may exist.

However, the oxidation of LDL leads to the

forma-tion of several cytotoxins, including lipid

hydroperox-ides, alkenals, oxysterols and lysoPC [58]. Therefore,

the cytotoxic effect of oxLDL to VSMCs may derive

from a dominant toxin among these, but also may be

(4)

due to the combined actions of multiple agents. Several

studies had demonstrated that the cytotoxicity of oxLDL

could be mimicked by a lipid extract of the LDL [59,60].

Hughes et al. had identified the most toxic fraction of

copper-oxidized LDL to be 7-ketocholesterol for

cul-tured procine VSMCs [61]. In addition, Chisolm et al.

reported that 7

b

-hydroperoxycholesterol was the

princi-pal cytotoxin of oxLDL when examined using human

and rabbit VSMC culture [62]. The current results

showed that lysoPC was highly cytotoxic on VSMC as

well. Apparently, methods for preparing oxLDL samples

(e.g. LDL source and oxidation procedure), conditions

for cytotoxicity test (e.g. incubation time and types of

test), and the culturing cell systems (e.g. medium serum

content and origin of cells) employed may all influence

the outcome of cytotoxicity test; it is hence useful to

compare the actions of oxLDL with its putative

compo-nent by quantitative analysis and by determining more

than one parameter, such as necrosis versus apoptosis.

In conclusion, oxLDL and its active component,

lysoPC, both induce VSMC death in a dose-dependent

manner. However, a certain threshold oxidative level (ca.

1.0 nmol MDA

/

ml) could be detected such that the

dose – response pattern actually reflected an ‘oxidative

equivalency’. Both oxLDL and lysoPC elicited

dose-de-pendent cytotoxicity and apoptosis, suggesting that

lysoPC, being a major lipid component of oxidized LDL,

could account for not only the cytotoxic effect, but also

some of the apoptosis induced by oxLDL.

Acknowledgements

This study was supported by the National Science

Council

(NSC87-2314-B-182-088),

Department

of

Health, Executive Yuan (Research Center

DOH86-HR-610) and Chang Gung University (CMRP 736) to Y.T.L.

C.C.H. also thanks the scholarship support from the

Foundation of Biomedical Sciences of National Defense

Medical Center. The editorial assistance of Cynthia

Huang and the excellent assistance of L.Y. Chen in the

preparation of the manuscript is gratefully

acknowl-edged.

References