Effects of high-density lipoproteins on intracellular pH and proliferation of human vascular endothelial cells

atherosclerosis Atherosclerosis 123 (1996) 73-82

Effects of high-density lipoproteins on intracellular pH and proliferation of human vascular endothelial cells Toshiyuki Tamagakl ‘*, Shohei Sawada, Hitoshi Imamura, Yuusuke Tada, Seiki Yamasaki, Akihisa Toratani, Toshiyuki Sato, Sumio Komatsu, Naoaki Akamatsu, Mashahito Yamagami, Kyoichiro Kobayashi, Kazuharu Kato, Katsumi Yamamoto, Kaoru Shirai, Keizo Yamada, Tadashi Higaki, Katsumi Nakagawa, Hajime Tsuji, Masao Nakagawa Second Depa.rtment

of Medicine, Kyoto Prgfectural Unioersity qf Medicine, Kyoto, Japan

Received 4 August 1993; revised 9 November 1995; accepted 21 November 1995

___-

_____~___ Abstract

We investigated the effects of high-density lipoprotein (HDL) on the intracellular pH ([PHI,), and on the proliferation of human vascular endothelial cells (HUVEC), as well as on their production of prostacyclin (PGI,). The [pH], was slightly acidified when extracellular Ca’+ was chelated with EGTA. Pretreatment of HUVEC with amiloride, the Na + /H + exchange inhibitor, caused the [pH], to become strongly acidic. The addition of HDL produced a biphasic shift in [pH],, with a brief initial acidification followed by a rapid alkaline shift. The initial decrease in [pH], was abolished in the cells pretreated with EGTA, and subsequent alkalinization was inhibited. The alkalinization of [pHIi disappeared in the cells pretreated with amiloride. These results suggest that [pH], depends mainly on Na + ,‘H+ exchange and partially on the extracellular Ca2 + of the HUVEC either in the resting unstimulated state or during HDL stimulation. In contrast, the addition of LDL produced an acidification of [pHIi, which was increas;ed by LDL in the Ca z +-free condition. In the cells pretreated with amiloride, [pH], was not further acidified by LDL. As a result, HDL promoted the proliferation of cells, an action that was inhibited by pretreatment with EGTA. However LDL inhibited cell proliferation, an action unaffected by EGTA pretreatment. The addition of HDL also enhanced the generation of prostacyclin in endothelial cells, the enhancement of PGI, generation resulted from an increase in the release of Ca 2+ from storage sites, due not only to an increased production of inositol 1,4,5-trisphosphate (IP,), but also to the alkalinization of [pH],. These effects may be involved in the mechanism of HDL’s anti-atherosclerotic action. Keywords:

Intracellular

pH; Prostacyclin;

Human vascular endothelial

-~ * Corresponding author. 0021-9150/96/$15.000 1996 Elsevier Science Ireland Ltd. All rights reserved SSDI 002 1-9 150(95305774-9

cell; HDL; Atherosclerosis

14

T. Tamagaki et al. /Atherosclerosis

1. Introduction A significant shift in the intracellular pH ([pHIi) is among the many events observed in stimulated cells. The expression of certain cellular responses following stimulation has been attributed to this shift in [pH],. For example it has been proposed that, in neutrophils, a stimulus-induced alkaline shift in [pH], modulates chemotaxis, aggregation, phagocytosis, secretion, and the generation of superoxide radicals [ l-41. A convincing role for IpHI, changes in stimulating extracellular growth was recently documented in various types of cultured cells [5,6]. However, the relationship between [PHI, and the anti-atherosclerotic mechanism of vascular endothelial cells has not been reported. Epidemiologic studies support an inverse relationship between serum levels of high-density lipoprotein (HDL) and the risk of coronary artery disease[7], but the pathophysiologic mechanism is uncertain. Endothelial cells which line the inner surface of the vessel wall are constantly exposed to plasma lipoproteins, and they are thought to function as a barrier, and also to play an important role in protecting against atherosclerotic changes in the vesselwall. In the present study, we investigated the effects of HDL on [pH]i and the proliferation of vascular endothelial cells, and the generation of prostacyclin, which is thought to modulate platelet reactivity and adhesion on the damaged vessel walls. 2. Materials and methods 2.1. Culture of vascular endothelial cells

Human vascular endothelial cells (HWEC) were cultured by a previously described method [8]. In short, the umbilical cord was infused with collagenase in cord buffer (136.9 mM NaCl, 0.5 mM Na,HPO.+, 4.0 mM KCl, 0.1 mM KH,P04, 11.1 mM glucose) and placed in a water bath at 37°C for 20 min. After incubation, the 0.2% collagenase solution containing the endothelial cells was flushed from the cord by perfusion with cord buffer and centrifuged. The cell suspension was divided equally among several 35-mm Petri dishes.

123 (1996) 73-82

Endothelial cells were cultured in TCM 199 containing 20% FCS, penicillin (200 U/ml), streptomycin (200 pg/ml), and L-glutamine (2 mM) at 37°C in 5% CO,. Cells were fed twice each week with a complete change of fresh culture medium. Primary cultured cells that formed confluent monolayers were used in these experiments. Cells were verified as vascular endothelial cells by the detection of Weibel-Palade bodies by electron microscopic study [9]. 2.2. Measurement of intracellular

pH

The [pHIi in HUVEC was determined by use of a pH-sensitive dye, 2,7-bis-(carboxyethyl)-carboxyfluorescein-AM (BCECF-AM) as described by Rink et al. [lo]. Cells were harvested in HEPES buffer (153.0 mM NaCl, 5.0 mM KCl, 5.0 mM glucose, 10.0 mM HEPES, pH 7.4) and incubated for 40 min at 37°C with 3 PM BCECFAM. The ratio of fluorescence intensity was determined with a Ca2+ analyzer CAF- 100, using excitation wavelengths of 450 and 500 nm, and an emission wavelength of 540 nm. The [pH]i was calibrated by the nigericin/K+ method, as described by Thomas et al. [l I]. 2.3. Assay of prostacyclin

concentration

HUVEC were incubated with 500 ~1 of buffer A (150 mM NaCl, 5 mM KCl, 1 mM MgCl,, 1.8 mM CaCl,, 5 mM glucose, 10 mM HEPES, pH 7.4) containing several test reagents at 37°C for 15 min. An aliquot of this substance was assayed for prostacyclin concentration. The prostaglandins generated mainly by HUVEC were confirmed to consist mainly of prostacyclin by thin-layer chromatography in a preliminary experiment with [l14C]-arachidonic acid [8]. We measured the release of prostacyclin from HUVEC as the stable metabolite, 6-keto prostaglandin F,,, using a [3H]6-keto prostaglandin F,, RlA kit (New England Nuclear, Boston, MA, USA). In short, antiserum was raised in rabbits against a bovine serum albumin conjugate of 6-keto prostaglandin F,, to obtain antibodies 6-keto prostaglandin F,,. Unlabelled antigen from samples that were isolated, frozen and stored at - 20°C or the standard

T. Tamagaki et al. / Atherosclerosis

solutions and a fixed amount of tracer (labelled antigen) were incubated with a constant amount of antibody for 16 h at 4°C. Using an RIA kit for 6-keto-prostaglandin F,, (New England Nuclear, Boston, MA, USA), antibody-antigen complexes were separated from free antigen by adsorbing the free tracer onto activated charcoal. Following centrifugation, the supernatant containing the antigen-antibody complexes was decanted into a counting vial and a scintillation cocktail was added. Radioactivity in the vials was determined with a beta counter. Results obtained with the standards were used to construct a standard (dose-response) curve. 2.4. Measurement of cytosolic free Ca’+ concentration The cytosolic concentration of free Ca2+ was measured by the modified method of Grynkiewicz [12]. The endothelial cells were scraped from the dishes with a rubber spatula and collected in centrifugation tubes then rinsed with 10 ml of buffer A (without Ca2+ ) containing 1.O% bovine serum albumin. After the cells had been resuspended in 3 ml of the solution they were incubated with 3.2 PM fura-2/AM at 37°C for 45 min and diluted to 20 ml. Following resuspension in 1.0% bovine serum albumin solution (cell counts are 106/ml), they were centrifuged at 250 x g for 10 min. Fluorescence signals from the endothelial cell suspensions were recorded in UV-compatible cuvettes utilizing a Ca’ + analyzer CAF-100 (Japan Spectroscopic Co. Ltd., Tokyo, Japan). 2.5. Assay of inositol 1,4,5-trisphosphate (IP,) concentration The IP, released from HUVEC was measured by use of an IP, (D-myo-inositol 1,4,5-trisphosphate) [3H]R114kit (Amersham International Plc, Buckinghamshire, UK). After the cells (3 x lo5 cells/200 ~1) had been harvested with 500 ~1 of buffer A, the reaction was stopped by the addition of 15% trichloroacetic acid, and the sample was centrifuged (1500 x g) at 4°C for 15 min. The supernatant was obtained and TCA in the supernatant was then extracted 3 times with 4 ml of

123 (1996) 73-82

75

water-saturated diethylether. The sample was adjusted to pH 7.5 with 1 M NaHCO,, and IP, in the above-treated supernatant was then measured. 2.6. Measurement of 45Careleasefrom storage sites HUVEC were permeabilized with saponin using a previously described method as follows [13]. HUVEC were rinsed with a conditioned buffer solution (buffer B: 150 mM NaCl, 5 mM KCl, 1 mM MgCl,, 5 mM glucose, 10 mM HEPES, 0.1 mM EGTA, and 0.1% bovine serum albumin, pH 7.4). Cells were incubated for 5 min at 37°C with 30.0 nM Ca*+ -EGTA buffer solution containing 50 pgg/ml saponin, 140 mM KCl, 12 mM NaHCO,, 5 mM HEPES, 0.42 mM NaH,PO,, 1 mM MgCl,, 5.5 mM glucose, 0.1% bovine serum albumin, 5 mM creatine phosphate, 10 U creatine phosphokinase, 1 mM EGTA, and 3.3 x lop4 M CaCl,. The cells were incubated for 60 min following the addition of 1.85 x lo5 Bq/ml 45Ca2+ (7.1484 x 10’ Bq/mg CaC12, New England Nuclear, Boston, MA, USA), 5.0 mM phosphocreatine (PC), 10.0 U/ml creatine phosphokinase (CPK), 5.0 mM succinic acid disodium salt (Sue), and 3.0 mM adenosine 5’triphosphate magnesium salt (ATP). They were resuspended in 30.0 nM Ca2+-EGTA buffer solution containing 0.5 mM PC, 10.0 U/ml CPK, 5.0 mM Sue, and 3.0 mM ATP after centrifugation at 1500 x g for 10 min, and were then apportioned into several tubes (2 x lo5 cells/200 ~1). After the addition of agents, 200 ~1 of cell suspension was dripped onto a filter (Whatman glass microfiber filter, GF/C, Maidstone, England) at each specified time interval and vacuumed with a diaphragm vacuum pump (DA-30D, ULVAC Shinku-kikosha, Osaka, Japan) and a 1225 Sampling Manifold (Millipore, Bedford, MA, USA) with 30 ml of a rinsing buffer (120 mM KCl, 5 mM NaCl, 5 mM glucose, 10 mM HEPES, 1 mM MgCl,, 1 mg/ml bovine serum albumin, 1 mM EGTA; pH in the buffer changed from 6.8 to 7.4 by adding 1 mM KOH). The radioactivity of HUVEC on the vacuum pump filter was measured as residual 45Ca in the cells by means of a liquid scintillation counter. Residual 4sCa in the

16

T. Tamagaki et al. /Atherosclerosis

cells fell to minimum levels 10 seconds after the start of incubation and remained at a plateau. The Ca2+ release from storage sites induced by agents was calculated at 10 seconds as compared with the Ca2+ released by the Ca2+ ionophore A23187.

123 (1996) 73-82

be assumed were compared by the Kruskal-Wallis non-parametric method for analysis of variance. When results of chi-square testing were significant, the Mann-Whitney test was used to determine the significance of the differences between pairs of means. A level of P < 0.05 was accepted as statistically significant.

2.7. HUVEC proliferation HUVEC proliferation was measured as cell number and DNA synthesis using [6-3H]thymidine (37.0 MBq/ml, Du Pont Company, DE, USA). HUVEC was harvested at confluence and plated (2 x lo4 cells) onto 24-well dishes in TCM199 with 20% fetal calf serum, followed by incubation for 48 h. HUVEC was washed twice with phosphate buffer solution (PBS), and incubated with TCM199 for 24 h. The medium was removed, incubated for 12 h after the addition of 1 ml of sample medium, and labelled with [6-3H]thymidine (37.0 kBq/ml) for 8 h. We removed the medium and washed the cells with ice-cold PBS 3 times. We then added 1 ml of 10% trichloroacetic acid (TCA), and placed the dishes on ice for 20 min. We then washed the dishes 3 times with ice-cold 5% TCA, lysed the cells with 0.5 ml of 0.5 N NaOH, and placed the dishes on ice for 10 min. A volume of 0.25 ml of 1 N HCl was added to each well and mixed gently, followed by the addition of 0.25 ml of 40% TCA (to a final lo%), again mixed gently. Dishes were then placed on ice for 20 min. Insoluble materials were removed by filtration using a Whatman GFjC filter, and dried after washing the filter 3 times with 100% ethanol. Radiation was then measured by liquid scintillation counting. Radioactivity was therefore measured as the DNA uptake of [3H]thymidine. 2.8. Statistical analysis Data were expressed as mean &- S.E.M. and compared by analysis of variance. For F-ratios, it was determined as significant at or below the 5% level. Duncan’s multiple range test was used to determine differences between any two groups. Differences at or below the 5% level (P < 0.05) were considered statistically significant. Percent changes for which a normal distribution could not

3. Results

3.1. Eflects on intracellular pH The [PHI, became slightly alkalinized after the addition of 1.8 mM CaCl,. When extracellular Ca2+ was chelated with EGTA, the IpHI, became slightly acidic. Pretreatment with amiloride markedly acidified the [PHI,. The addition of HDL (0.2 mg/ml, Sigma Chemicals, St. Louis, MO, USA) produced a biphasic shift in IpHI,, with an initial brief acidification followed by a rapid alkaline shift. When the cells were pretreated with EGTA, the initial decrease in [pH]i was abolished and the subsequent alkalinization was inhibited. The alkalinization of bHli disappeared in the cells pretreated with amiloride (Figs. 1 and 3). However, the addition of LDL (0.2 mg/ml, Sigma) produced an acidification of [PHI,, which was further increased by adding LDL in Ca2+ -free condition. The IpHI, of cells pretreated with amiloride was not further acidified by adding LDL (Figs. 2 and 3). The addition of thrombin (10 U/ml, Green Cross Corporation, Osaka, Japan) also produced a biphasic shift in IpHI, with an initial brief acidification followed by a rapid alkaline shift. Pretreatment of cells with EGTA abolished the initial decrease in [pHIi and inhibited subsequent alkalinization. The alkalinization of [pH]i disappeared in the cells pretreated with amiloride (Figs. 2 and 3). 3.2. Generation of PGI, The addition of buffer A increased the generation of PGI, in a time-dependent manner. This increase plateaued at 15 min as previously described [8]. The generation of PGI, by HUVEC was not enhanced by HDL at a final concentra-

T. Tamagaki et al. /Atherosclerosis

I: PH Ii

123 (1996) 73-82

EGTA(10 -3M)

7.4

17

amiloride(103M) 3&c

-7.1

-

r t

7.4

r-

t amiloride

CaClz m

I7.1

-

i

t HDL(O.Pmg/ml)

t ---L-W-t HDL(O.Pmg/ml)

Fig. 1. Effect of aretreatment with EGTA (lop3 M) and amiloride (lo-’ intracellular pH ([pH],) of human vascular endothelial cells.

tion of either 0.002 mg/ml ior 0.02 mg/ml, but was enhanced by HDL at a final concentration of 0.2 mg/ml. This enhancement was inhibited by pretreatment with EGTA (1 mM) and was strongly inhibited by pretreatment of cells with acetyl salicylic acid, a cyclooxygenase inhibitor (1 mM, 30 min). Thus, this enhancement of PGI, generation was not due to the enhanced releasefrom cells but to an increase in synthesis. The addition of thrombin (10 IJ/ml) also in.creasedthe generation of PGI, and this effect was inhibited by pretreatment with EGTA (1 mM) (Fig. 4). 3.3. Change in cytosolic Ca2+ concentration The addition of HDL (0.2 mg/ml) increased [Ca2+]i which led to an increase in PG12 generation. Pretreatment with 1 mM EGTA inhibited this increase. The addition Iof thrombin (10 U/ml) also increased [Ca2+li, this effect was similarly inhibited by pretreatment with 1 mM EGTA (Fig. 5). 3.4. Generation of IP, The generation of IP, in HUVEC was enhanced by HDL at a final concentration of 0.2 mg/ml. The addition of thrombin (10 U/ml) also increased the generation of IP, (Fig. 6).

t HDL(O.Pmglml)

M) on basal and HDL (0.2 mg/ml)-induced change in

3.5. 45Careleasefrom storage sites IP, (10 - 5 M) increased the release of Ca2+ from storage sites in saponin-treated HUVEC. The increase was 35.7% of that produced by A23187 (lo-’ M). Pretreatment with lo-’ M antimycin A or 5 x 10- 6 M oligomycin did not inhibit the increase caused by IP,. This observation suggested that IP, increased the release of Ca” + from the non-mitochondrial Ca2+ pool (Fig. 7). The release of Ca2+ from storage sites that was enhanced by IP, ( 10W5 M) increased further when the LpH], was alkalinized from 6.8 to 7.4. HDL presumably increased the release of Ca2+ from storage sites, not only by increasing the generation of IP, but also by the alkalinization of IpHI,. This may explain the observed increase in cytosolic Ca2+ concentration (Fig. 8). 3.6. Effects on cell proliferation HDL (0.2 mg/ml) promoted the cell proliferation, an effect that was inhibited by pretreatment with EGTA. However, LDL (0.2 mg/ml) inhibited cell proliferation; this action was not affected by pretreatment with EGTA. The addition of thrombin (10 U/ml) also increased the cell proliferation. This increase in uptake was inhibited in cells pretreated with EGTA (Figs. 9 and 10).

78

T. Tamagaki et al. /Atherosclerosis

hW

123 (1996) 73-82

EGTA(10 %)

7.4

amiloride(lO”M) 3DkiC

7.1

r

7.4

r

7.1

* . k

t

CaCl2

amlloride

--k

LDL(O.Smg/ml)

k

k

LDL(O.Pmg/ml)

LDL(O.PmgIml)

Fig. 2. Effect of pretreatment with EGTA (10C3 M) and amiloride (lop3 M) on basal and LDL (0.2 mg/ml)- and thrombin (10 U/ml)-induced change in intracellular pH ([pHIi) of human vascular endothelial cells.

4. Discussion An exchange of Na + /H + has been detected in the cell membrane of various types of cells, including the HUVEC [14]. The importance of inhibiting this exchange on the regulation of IpHI, has been reported in some cells [I 5- 171. The 7A

alkalinization of [PHI, produced by growth factor or vasoactive agents has also been shown to promote cell proliferation [l&20] and to increase the generation of PG12, which modulates platelet function and vascular tonus, and exerts a protective effect on endothelial cells [21-231. Growth factors act by binding to specific cell surface receptors. The activated receptor then mediates a cascade of biochemical and physiological changes in the cell ultimately leading to DNA synthesis

-l-Y

7.3

100 af 4

**

I*

r

T

r

72

7.1

7.0

control

HDL (0.2mglml)

LDL (0.2mglml)

0

bufferA

&@

EGTA(lO*

m

amiloridc

thrombln (10uniWml)

H) W’

M)

Fig. 3. Effect of pretreatment with EGTA (lo- 3 M) and amiloride (10V3 M) on basal and HDL (0.2 mg/ml)-, LDL (0.2 mg/ml)- and thrombin (10 U/ml)-induced changes in intracellular pH ([pHIi) of human vascular endothelial cells. Data are mean k S.E.M. (n = 6; *P i 0.05, “P < 0.01).

HDL(mg/ml)

.L

thrombin (10unlts/ml)

Fig. 4. Effect of HDL (0.002 mg/ml, 0.02 mg/ml and 0.2 mg/ml) and thrombin (10 U/ml) on generation of PGI, by human vascular endothelial cells. Data are mean k S.E.M. (n = 6; “P < 0.01).

T. Tamagaki et al. /Atherosclerosis

19

123 (1996) 73-82

II

HDL (&2mg/ml)

ttlromMn (iounnetml)

Fig 5. Effect of pretreatment with EGTA (IO-’ M) on the HDL (0.2 mg/ml)- and thrombin (10 U/ml)-induced increase in cytosolic Ca2-l concentration of human umbilical vein endothelial cells. Data are mean & S.E.M. (n = 6; *P i 0.05, “P

<: 0.01).

and cell proliferation. Protein phosphorylation is an immediate consequence of the interaction between growth factors and the cell surface receptors. The receptors for growth factors are transmembrane glycoproteins that possess both intrinsic and ligand-stimulated protein tyrosine kinase activity. Other consequences of receptor activation include an immediate breakdown of inositol phospholipids, a transient rise in cytoplasmic free Ca2+ , alkalinization of intracellular pH and the stimulation of protein kinase C [24]. Epidemiologic studies have shown an inverse relationship between serum levels of high-density lipoprotein and the risk of coronary artery dis-

Fig. 7. Effect of IP, on release of Ca*+ from storage sites. Effect of pretreatment with antimycin A and oligomycin. Data are mean k S.E.M. (n = 6; “P < 0.01).

ease. HDL is reported to produce an anti-atherogenie effect via reverse cholesterol transport [25]. Badimon et al. found that atherosclerotic changes in cholesterol-fed rabbits disappeared after the administration of HDL [26]. However, the relationship between the anti-atherogenic effect of HDL and the HUVEC is unclear. According to the ‘response to injury’ hypothesis of Ross and Glomset, endothelial cell injury triggers the development of atherosclerosis [27]. HDL reportedly inhibits the oxidation of LDL by the endothelial cells [28]. Many investigators report that the stimulation or stabilization of PGI, production by HDL may be a general property of the endothelial cells [29,30]. There are, however, no reports on

$ 0.6 P $2 n$

0.6

ii{

o,4

8s is f” E 0.2 a,p .% 8 .c

0.0 control

HDL thrombln (mglml) (10unltdml)

Fig. 6. Effect of HDL (0.2 mg/ml) and thrombin (10 U/ml) on generation of IP:, by human vascular endothelial cells. Data are mean k S.E.M. (n = 6; *P < 0.05, l *P < 0.01).

pHofCa++-EGTAbuffer(3.0

xlO”M)

Fig. 8. Effect of alkalinization on release of ?a from storage sites already enhanced by IP, in saponin-treated HUVEC. Data are presented as mean f S.E.M. (n = 6; *P < 0.05, “P < 0.01).

80

T. Tamagaki et al. / Atherosclerosis

0

control

HDL

LDL

thmbln

(0.2 m~ml)(O.* mg/ml)(lO units/ml)

Fig. 9. Effect of a pretreatment with EGTA (10W3 M) on HDL (0.2 mg/ml), LDL (0.2 mg/ml) and thrombin (10 U/ml)induced promotion in human vascular endothelial cell proliferation. Data are mean + S.E.M. (n = 6; *P c 0.05, l *P < 0.01; N.S. means not statistically significant).

the effects of HDL on the relationship between IpHI, and PGI, on production or proliferation of endothelial cells. We therefore investigated the effects of HDL on the [PHI, and the proliferation of vascular endothelial cells with reference to prostacyclin. Our results suggest that [pH]i is mainly dependent on Na + /H + exchange and is partially de-

** I 1

lir

P e. 2 3 4 UE

g 100 E 2

3 ‘s % E 5 dp

O

control

LDL thlwombln HDL (O.2mgIml)(O.2m#nl)(lOuniWml)

Fig. 10. Effect of a pretreatment with EGTA (10 - 3 M) on HDL (0.2 mg/ml), LDL (0.02 mg/ml) and thrombin (10 U/ ml)-induced promotion in human vascular endothelial cell proliferation. Data are mean f S.E.M. (n = 6; l *P < 0.01).

123 (1996) 73-82

pendent on the extracellular Ca2+ of HUVEC, not only in the resting state, but also during HDL stimulation of those cells. LDL produced an acidification of [pHIi, which presumably depended mainly on the inhibition of Na+/H + exchange. Because the alkalinization of [PHI, is said to promote the proliferation in many types of cells, we also evaluated the effects of HDL and LDL on the proliferation of endothelial cells. HDL promoted the cell proliferation, while LDL inhibited the cell proliferation. When the cells were pretreated with EGTA, their proliferation was inhibited not only in the resting state but also during HDL stimulation. These results indicate that HDL may exert an anti-atherosclerotic action via the alkalinization of [pHIi and the promotion of endothelial cell proliferation, which leads to angiogenesis and the repair of injured endothelial cells. This mechanism is mainly dependent on an Nat/H + exchange and partly dependent on extracellular Ca’ + . LDL may stimulate atherosclerosis via the acidification of [pH]i and an inhibition of the proliferation of endothelial cells. This mechanism would presumably depend on the inhibition of Na +/H + exchange. The addition of HDL increased the generation of PGI, from cells. This enhancement was strongly inhibited by acetyl salicylic acid, a cyclooxygenase inhibitor, and was thus due to an increase in synthesis of PG12. And the addition of HDL also increased the generation of IP, and cytosolic Ca* + which is necessary for enhancement of PGI, generation [31]. This increased cytosolic Ca2+ was partly due to the influx of extracellular Ca2+ and perhaps in part from the mobilization of Ca2+ from intracellular stores. In our preliminary experiment, IP, enhanced the release of Ca2+ from storage sites [31], an action that is considered to be modulated by factors present in contractile cells. No previous studies were conducted in endothelial cells on the relationship between [pHIi and the release of Ca2+ from storage sites by IP,. In the present study, the IP, (10 - 5 M)-induced release of Ca’ + from storage sites was enhanced in conjunction with the alkalinization of [PHI,. HDL increased IP, and alkalized [pHIi. These findings support the possibility that HDL increases the release of Ca* +

T. Tamagaki et al. /Atherosclerosis 123 (1996) 73-82

from storage sites, not only via an increase in the generation of IP, but also by the alkalinization of [pH], which increases the release of Ca* + by IP,. Of course the concentration of IP, in Fig. 7 is non-physiologic. But the concentration of IP, in each part of cytosol is different. And so the differing concentrations of IP, used in our study can be explained by compartmentalization of IP,. These effects on the vascular endothelial cells may be important in the anti-atherosclerotic action of HDL. Lipoproteins have the risk of contamination and oxidation. But we used lipoproteins immediately (within 1 month), checked the same experiment with the other commercial source lipoproteins (such as Biomedical Technologies Inc., MA, U&4) and obtained a similar result. So we believe that the risk of contamination and oxidation of lipoproteins never exists. Thrombin induces the alkalinization of [PHI,, the proliferation of cells and an increase in the generation of PGI, and IP, by the endothelial cells to a greater extent than HDL. The thrombininduced increase in PGI, may protect against platelet aggregation or vasoconstriction [33]. The generation of PGI, enhanced by HDL may be similarly protective. The thrombin-induced alkalinization and increase of cytosolic Ca2+ and cell proliferation were inhibited in HUVEC pretreated with EGTA. Our data suggest that thrombin increases the intracellular Ca2+, not only via an influx from extracellular Ca*+ sites, but also via release from the intracellular Ca2+ stores. Thrombin may increase the release of Ca2+ from storage sites via an increase in the generation of IP, as well as by alkalinization of [PHI,, leading to an increased release of Ca2+ by IP,. Thrombin may produce an anti-atherosclerotic effect via the alkalinization of [pHIi and the promotion of endothelial cell proliferation, leading to angiogenesis and to the repair of injured endothelial cells. This action mainly depends on Nat/H+ exchange, but also on extracellular Ca2+ . While the changes induced by HDL were quantitatively smaller than those of thrombin, HDL may exert a qualitatively similar anti-atherosclerotic action. Although the alkalinization of [PHI, produced by growth factor or vasoactive agents has been

81

shown to promote cell proliferation, the relationship between cell proliferation and [pHIi is unclear. In our research, thrombin caused much more potent changes in the level of [PHI, than HDL and HDL-stimulated DNA synthesis than thrombin. And so at least the alkalinization of [pHIi is an initiation of cell proliferation and any other mechanism concerns the cell proliferations. Further studies are required on the mechanism of cell proliferations.

References

111Molski TFP, Naccache PH, Volpi M, Wolpert LM,

Sha’afi RI. Specific modulation of the intracellular pH of rabbit neutrophils by chemotactic factors. Biochem Biophys Res Commun 1980;94:508. [21Segal AW, Geisow M, Garcia R, Harper A, Miller R. The respiratory burst of phagocytic cells is associated with a rise in vacuolar pH. Nature 1981;290:406. [31 Simchowitz L. Intracellular pH modulates the generation of superoxide radicals by human neutrophils. J Clin Invest 1985;76:1079. 141 Grinstein S, Furuya W. Amiloride-sensitive Na + /H + exchange in human neutrophils: mechanism of activation by chemotactic factors. Biochem Biophys Res Commun 1984;122:755. 151Royal DS, Yvonne WB, George HD. Role of intracellular pH in the axolemma and myelin induced proliferation of Schwann cells. J Neurochem 1989;52:1576. 161Wounter HM. Effects of growth factors on intracellular pH regulation. Annu Rev Physiol 1986;48:363. [71 Philip MC, Johnson WJ, Rothblat GH. Mechanisms and consequencesof cellular cholesterol exchange and transfer. Biochim Biophys Acta 1987;906:223. 181Toyoda T, Sawada S, Niwa I, Maebo N, Tsuji H, Mikami K, Rin K, Nakagawa M. Effect of Ca antagonists on PGI, generation in cultured human vascular endothelial cells; relationship between intra- and extracellular Ca’+ and cyclic nucleotides. Blood Vessels 1985;16:245. 191Weibel ER, Palade GE. New cytoplasmic components in arterial endothelia. Cell Biol 1964;23:101. 1101Rink TJ, Tsien RY, Pozzan T. Cytoplasmic pH and free Mg* + in lymphocytes. J Cell Biol 1982;95:189. [I 11Thomas JA, Buchsbaum RN, Zimniak A, Racker E. Intracellular pH measurements in Ehrlich ascites tumour cells utilizing spectroscopic probes generated in situ. Biochemistry 1979;18:2210. Toyoda T, Nakagawa M, Maeda Y, Niwa I, Tsuji H, Rin 1121 K, Osamura K, Kitani T, Ijichi H. Effect of Ca antagonist on vascular prostacyclin generation. J Pharmacol Ther 1983:11:63.

a2

T. Tamagaki et al. /Atherosclerosis

[13] Burgess GM, McKinney JS, Fabiato A, Leslie BA, Putney JW Jr. Calcium pools in saponin-permeabilized guinea pig hepatocytes. J Biol Chem 1983;258:15336. [14] Kitazono T, Takeshige K, Edward JC, Minakami S. Involvement of calcium and protein kinase C in the activation of the Na + /H + exchange in cultured bovine aortic endothelial cells stimulated by extracellular ATP. Biochim Biophys Acta 1989;1013:152-158. [15] Ghigo D, Treves S, Turrini F, Pannocchia A, Pescarmona G, Bosia A. Role of Na +/H + exchange in thrombin and arachidonic acid induced CaZ+ influx in platelets. Biochim Biophys Acta 1988;940:141. [16] Gillies RJ, Zaguilan RM. Regulation of intracellular pH in BALB/c 3T3 cells. J Biol Chem 1991;266:1551. [17] Hallam TJ, Tashjian AH. Thyrotropin-releasing hormone activates Na + /H + exchange in rat pituitary cells. Biochem J 1987;242:411. [18] Levin EG, Marzec U, Underson J, Harker LA. Thrombin stimulates tissue plasminogen activator release from cultured human endothelial cells. J Clin Invest 1984;74:1988. [19] Meng HP, Michael JB, Pierce GN. Effect of platelet-activating factor (PAF) on sodium calcium exchange in cardiac sarcolemmal vesicles. Mol Cell Biochem 1990;92:45. [20] Gardner JP, Maher E, Ariv A. Calcium mobilization and Na+/H+ antiport activation by endothelin in human skin fibroblasts. FEBS Lett 1989;256:38. [21] Murakami M, Kudo I, Inoue K. Molecular nature of phospholipases A, involved in prostaglandin I, synthesis in human umbilical vein endothelial cells. J Biol Chem 1993;268:839. [22] Hara S, Kudo I, Chang HW, Matsuta K, Miyamoto T, Inoue K. Purification and characterization of extracellular phospholipase A, from human synovial fluid in rheumatoid arthritis. Biochem J 1989;105:395. [23] Kim DK, Kudo I, Inoue K. Purification and characterization of rabbit platelet cytosolic phospholipase A,.

123 (1996) 73-82

Biochim Biophys Acta 1991;1083:80. [24] Ulrich A, SchlessingerJ. Signal transduction by receptors with tyrosine kinase activity. Cell 1990;61:203. [25] Tall AR, Small DM. Body cholesterol removal: role of plasma high-density lipoproteins. Adv Lipid Res 1980;17:1. [26] Badimon JJ, Badimon L, Galvez A, Dische R, Fuster V. High density lipoprotein plasma fractions inhibit aortic fatty streaks in cholesterol-fed rabbits. Lab Invest 1989;60:455. [27] Ross R. The pathogenesis of atherosclerosis - an update. N Engl J Med 1986;31:488. [28] Ohta T, Takata K, Horiuchi S, Yoshimasa M, Matsuda I. Protective effect of lipoproteins containing apoprotein A-I on Cu’ + -catalyzed oxidation of human low density lipoprotein. FEBS Lett 1989;257:435. [29] Yui Y, Aoyama T, Morishima H, Takahashi M, Takatsu Y, Kawai C. Serum prostacyclin stabilizing factor is identical to apoprotein, Al. J Clin Invest 1988;82:803. [30] Fleisher NL, Tall AR, Witte LD, Miller RW, Cannon PJ. Stimulation of arterial endothelial cell prostacyclin synthesis by high density lipoproteins. 1982;257:6653. [31] Toyoda T. Role of Ca2+ calmodulin and cyclic nucleotides in the PGI, production in cultured human vascular endothelial cells with special reference to mechanism of enhanced PGI, production by Ca antagonists. J Kyoto Pref Univ Med 1986;95:205. [32] Yamamoto K, Toyoda T, Sawada S, Shirai K, Kobayashi K, Katoh K, Takada 0, Watanabe M, Uno M, Tsuji H, Nakagawa M. The regulatory mechanisms of inositol trisphosphate generation and Ca’ + release from the storage sites of human umbilical vein endothelial cells with references to the prostacyclin generation. J Thromb Hemost 1990;1:321. [33] Weksler BB, Ley CW, Jaffe EA. Stimulation of endothelial cell prostacyclin production by thrombin trypsin and the ionophore A23187. J Clin Invest 1978;62:923.