- Email: [email protected]
Transfer of [3H]cholesterol between lipid vesicles and rat arterial smooth muscle cells in vitro
Biochinuca et Biophywu Arta, 750 (1983) 434-439 Elsevier Biomedical Press
434
BBA 51335
TRANSFER OF I 3H]CHOLESTEROL BETWEEN LIPID VESICLES AND RAT ARTERIAL SMOOTH MUSCLE CELLS IN VITRO J. PETER
SLOTTE
and BO LUNDBERG
Department of Biochemistry and Pharmacy, kbo Akademi, Porthansgaten 3- 5, SF-20500 (Received
October
Turku (iho) {Finland)
Ist, 1982)
Key words: Cholesterol exchange; Yesicle - ceil transfer; Phosphaiidylcholine; Cholesterol esterification; (Rat smooth muxie cell)
Unesterified [ 3H]cholesterol is rapidly transferred between cholesterol-phosphatidylcholine vesicles and rat arterial smooth muscle cells in vitro. Exchange rate is influenced by the vesicle/cell ratio in a saturable way. The maximal transfer of cholesterol, which is 3.76 pg per mg cell protein during 4 h, is achieved with a vesicle/cell ratio of 3.4.10’. Bovine serum albumin enhances the exchange by a factor of 4.5 compared to a protein-free system. The activation energy for the process is + 38.5 kJ*mol - ’ with vesicles of 1: 1 mole ratio of cholesterol to phosphatidylcholine (C/P). A fraction of the inco~rat~ free [ ‘H~c~iesterol is esterified within 4 h with donor vesicles of over 1: 1 C/P. When cetls were incubated with vesicles of low C/P mole ratio (1: 2) a fraction of the incorporated free i3H]cholesteroi was esterified within 16 h. Our results are compatible with the aqueous diffusion mechanism of cholesterol exchange. Furthermore, we suggest that, in rat smooth muscle cells, the cell membrane cholesterol pool is not metabolically isolated from internal cholesterol pools, at least as judged by the ability of the ceils to esterify incorporated free cholesterol.
The surface transfer or exchange of lipids in general, and of cholesterol in particular, has recently been the subject of intense investigation. It has been shown that the exchange of free cholesterol between lipoproteins and cultured cells can play a significant role in the maintenance of the cellular sterol balance [I]. The mechanism of lipid transfer is a matter of debate by several authors 12-41. Phillips et al. (2f have postulated that the exchange process involves the desorption into and diffusion through the water interphase of the exchanging lipid molecules. This exchange mechanism would predict that the hydrophobicity of the lipids involved is one of the main determinants of the exchange rate. According to this it Abbreviations: Hepes, 4-(2-hydroxyethyi)-t-piperazineethanesulphonic acid; LDL, low-density lipoprotein. ~05-2760/83/~-~/$03.00
0 1983 Elsevier Science Publishers
would be easy to understand why phospholipids exchange much more slowly than cholesterol, since their monomeric water solubility (critical micellar concentration) is far less than that for cholesterol [3]. One must note that the difference in exchange rate is valid only in the absence of proteins. The existence of several phospholipid-specific exchange proteins has been shown by several investigators [5,6]. Although the picture of lipid exchange is growing clearer, it is still difficult to estimate the quantitative si~ificance of the process in vivo. In order to understand the contribution of lipid exchange to the complex lipid metabolism of the arterial wall, it is of great importance to characterize the effects of lipid exchange on the predominant cell type of the arterial wall, the smooth muscle cell. It is mainly unresolved whether smooth muscle cells are able to esterify membrane-derived free cholesterol or if the membrane cholesterol
435
pool is metabolically isolated from intracellular cholesterol pools. In the present study we have examined the exchange of free cholesterol between lipid vesicles of different C/P mole ratios, and cultured rat smooth muscle cells. We have also examined the effect of serum albumin on the rate of cholesterol exchange between vesicles and cells. Furthermore, we have followed the fate of the incorporated cholesterol molecules, that is the ability of the cells to esterify the exchanged free cholesterol. Experimental procedures Ceil culture Primary cultures of rat arterial smooth muscle cells from the thoracic aorta of adult SpragueDawley rats were used. The cell lines were started from explants of the intima-medial layer according to methods reported by Ross [7] and Bierman et al. [8]. The cells were grown in Dulbecco’s modified Eagle medium supplemented with 2 mM Lglutamine, 0.1 mM non-essential amino acid solution, 20 mM Hepes buffer, 0.08% (w/v) sodium bicarbonate, 10 pg/ml of gentamycin and 10% (v/v) newborn calf serum. The cells showed typical growth characteristics of cultured smooth muscle cells, as described by Ross [7]. The cells grew in multiple, overlapping layers with a doubling time of approximately 48 h. Lipid vesicles Vesicles containing different mole ratios of [3H]cholesterol (1.6. lo4 dpm/pg) to egg phosphatidylcholine were prepared by co-lyophilization from benzene stock solutions in glass tubes. After completed co-lyophilization, phosphate-buffered saline (pH 7.4) was added to give a final cholesterol concentration of 1 mg/ml. The suspension was sonicated with an MSE Ultrasonic Disintegrator (titanium probe) for 30 min at 20°C at maximal effect. The opalescent solution was centrifuged at 20000 X g for 30 min at 4°C in order to sediment titanium particles and larger lipid aggregates. When the lipid vesicles contained trace amounts of cholesteryl[‘4C]oleate, this was added from a benzene solution prior to co-lyophilization. Characterization of the lipid vesicles revealed a particle diameter of 25 nm and average particle weight of 1.7 . 10’ daltons (1 : 1 C/P vesicles) [9].
Incubation procedures All experiments were performed in 50 mm diameter petri dishes with confluent cells between the 7th and the 12th passage. About 2.9. lo5 cells were seeded in each dish, where after the cells grew to confluency in 4-5 days. A confluent petri dish contained about 1.5 . lo6 cells, corresponding to about 300 pg cell protein. Growth medium without serum supplement was used as the incubation medium, with final volume adjusted to 3.0 ml. Bovine serum albumin was added from a sterile 100 mg/ml stock solution made up in phosphatebuffered saline. The albumin preparation was purified from contaminating apolipoprotein A-I by the method reported by Fainaru and Deckelbaum [lo]. The purified albumin was delipidized according to the method of Rothblat et al. [ll]. The albumin preparation was over 98% pure by polyacrylamide gel electrophoresis. When cells were incubated at different temperatures, the appropriate dishes were equilibrated at respective temperatures for 30 min before adding the vesicle preparation. After completed incubation, cells were washed twice with cold phosphate-buffered saline and the cells were detached by gentle scraping. The cells were sedimented by low-speed centrifugation, and the cell pellet was disrupted in 500 ~1 phosphate-buffered saline by sonication. Aliquots of this suspension were taken for the different analytical procedures. Analytical procedures The cellular sterol mass was determined by the enzymatic method presented by Gamble et al. [ 121. Isotopic free and esterified cholesterol was extracted with chloroform/methanol (2 : 1, v/v) containing lipid carriers. The neutral lipids were separated by thin-layer chromatography on Kieselgel 60 (DC-Alufolien Kieselgel 60, Merck, Darmstadt, F.R.G.) with n-hexane/diethyl ether (90 : 10, v/v) and detected by spraying the thin-layer plates with 50% sulphuric acid and charring at 150°C for 5 min. The distribution of lipid radioactivity between free and esterified cholesterol positions were determined by cutting the respective positions into liquid scintillation vials. A toluene-based scintillation cocktail was added 30 min before counting. Cell protein was determined by the method of Lowry et al. [ 131 as modified by Markwell et al. 1141.
436
Reagents The egg phosphatidylcholine employed was a highly purified product prepared by a method developed in this laboratory [15]. The cholesterol used (Merck, Darmstadt, F.R.G.) was chromatographically pure by GLC. [3H]Cholesterol (58 Ci/mmol) was obtained from the Radiochemical Centre, Amersham, and the product was over 98% pure by TLC. Cholesteryl[‘4C]oleate was prepared according to a modified acid chloride method [ 161 and the radioactive [ “C]oleic acid (50 mCi/mmol) was purchased from the Radiochemical Centre, Amersham. Bovine serum albumin and all reagents for cholesterol determination were obtained from Sigma Chemicals, St. Louis, MO, U.S.A. Cell growth medium and supplements were purchased from Gibco Europe, U.K. Gentamycin was a product of Neofarma, Finland.
11 : / : -0.6
-0.4
-0,2
0 1 /
0.2
AwnIN
0.4
CONCE~~TRATION
0.6
0.8
(MG/ML)
Fig. 2. Effect of albumin concentration on the exchange rate. Cells were incubated at 37°C for 4 h with 33 pg vesicle [3H]cholesterol per ml medium (1 : 1 C/P) and different amounts of purified albumin. The vesicle-cholesterol/cellcholesterol mole ratio was approx. 12: 1. The radioactivity of the cells was counted after thorough washing.
Results The rate of [3H]cholesterol transfer from lipid vesicles into rat smooth muscle cells were determined from the amount of [ 3H]cholesterol accumulated in the cellular sterol pool. Measurements VESICLE / CELL
2.16
1.08
RATIO 3.24
4.32 x10'
I
of cellular sterol mass after incubation with lipid vesicles of 1 : 1 C/P mole ratio did not reveal any significant transfer of cholesterol mass. Fig. 1 demonstrates the effect of varying the vesicle/cell ratio on the exchange rate. During 4 h of incubation, a maximum of 3.76 pg of [3H]cholesterol was incorporated per mg cell protein, with a vesicle/cell ratio of 3.4. 10’ (values obtained from the double reciprocal insert). The vesicle/cell ratio of 3.4 . IO’ corresponds to a ves-5.50 0
OL-----;;60 50
150
E, =
+ 38.5 ~Jno~-l
200
p4 VESICLE CHOLESTEROLPER ML MEDIUM -7.5 ,
Fig. 1. Effect of vesicle/cell ratio (donor/acceptor) on the transfer of [‘HIcholesterol from lipid vesicles (1 : 1 C/P) to cultured smooth muscle cells. Each petri dish contained about 1.5. lo6 cells corresponding to 8.4 gg total cellular cholesterol. The cells were incubated at 37°C for 4 h with different amounts of vesicles in 3.0 ml medium. The amount of [7H]cholesterol transferred was determined by counting the cell radioactivity and the counts were correlated to the vesicle [‘HIcholesterol specific activity.
3.2
3.3
3.4 1 /
T
3.5
i i.6
X103 lo K-l)
Fig. 3. Effect of temperature on the transfer of [ ‘HIcholesterol between lipid vesicles (1 : 1 C/P) and cultured cells. Cells were incubated at different temperatures between 6 and 37’C with 33 pg vesicle [3H]cholesterol per ml medium. The vesiclecholesterol/cell-cholesterol mole ratio was approx. 12 : 1. The activation energy is obtained from the Arrhenius plot.
437
0
4
8
12
16
20
24
(H)
lNC”BATlON TIME
0
4
8 INCUBATION
12
16
20
24
CM)
TIME
Fig. 4 A. Effect of vesicle C/P mole ratio on the incorporation of [ 3H]cholesterol into cells. The incubation medium contained 33 pg vesicle cholesterol per ml. The vesicle-cholesterol/cell-cholesterol mole ratio was approx. 12: 1. 0, greater than 1 : 1 C/P; 0, 1 : 1 C/P; A 1 :2 C/P. B. Esterification of the incorporated [ 3H]cholesterol as a function of vesicle C/P mole ratio and time. Incubation conditions and symbols are identical to those in A.
icle-cholesterol/cell-cholesterol mole ratio of approx. 60 : 1. Half-maximal rate was achieved at a vesicle/cell ratio of 0.77 . lo’, which corresponds to a vesicle cholesterol concentration of 36 pg per ml medium. (Vesicle-cholesterol/cell-cholesterol mole ratio, approx. 13 : 1). Isotopic equilibration was achieved in 16 h, and a plot of In dpm versus time gives a straight line, indicating first-order kinetics (plot not shown). The transfer of [3H]cholesterol from lipid vesicles to cells is not due to fusion of vesicles with cell membranes, since cholesteryl[ “C]oleate, when used as a non-exchangeable marker, it not found in the cellular lipid pool even after prolonged incubations (48 h). Since serum albumin has been shown to induce sterol efflux from cultured cells [ 17,181 we examined whether the presence of albumin has any effect on the exchange rate. As shown in Fig. 2, the presence of albumin has a profound effect on the incorporation of [ 3H]cholesterol into cells. Maximal effect was achieved with an albumin concentration of about 20 mg/ml. The V,,, obtained, 7 pg of [3H]cholesterol incorporated per mg cell protein during 4 h, is about 4.5 times higher than the value obtained without albumin (compare to Fig. 1). The albumin concentration,
which gave half-maximal transfer was only 1.7 mg/ml. The rate of [ 3H]cholesterol transfer was strongly temperature-dependent, as shown in Fig. 3. In a protein-free system, with equimolar C/P vesicles, the half-time for the transfer was found to be 3.6 h at 37°C. An Arrhenius plot of In k versus l/T gives the activation energy for the process, which was found to be about + 38.5 kJ . mall’ with 1 : 1 C/P vesicles and cells. In Fig. 4, panel A, is shown the effect of different C/P mole ratios on the incorporation of vesicle [ 3HIcholesterol into cells. Considerably more cholesterol was transferred from vesicles with high cholesterol saturation (greater than 1 : 1 C/P) compared to vesicles of lower C/P ratio (1 : 1 and 1 : 2, respectively). This is true in spite of similar cholesterol concentration per ml medium in each case. Panel B in Fig. 4 shows that the incorporated [ ‘HIcholesterol was readily esterified within the cells. [ 3H]Cholesterol originating from cholesterol-rich vesicles (greater than 1 : 1 C/P) was found in the cholesteryl ester pool within 4 h of incubation. [ 3H]Cholesterol from 1 : 1 C/P vesicles was esterified within 6-8 h, while the [ 3H]cholesterol originating from 1 : 2 C/P vesicles was found in the ester pool after about 16 h.
438
Discussion The modes of cholesterol uptake by cells in culture are complex, and are not known in detail. In a lipoprotein-containing medium, sterols are transported into cells by the specific binding and internalization of low-density lipoproteins [ 191, by exchange of lipids between cell membranes and lipoproteins [20], and finally by pinocytosis of the medium. The receptor-mediated endocytosis is believed to be the dominating mechanism, although spontaneous lipid exchange can play an important role in the regulation of cellular sterol content ]1,211. It has been shown that lipid exchange between lipoproteins and biological membranes is bidirectional [22]. The net transfer of free cholesterol has been shown to be dependent on the mole ratio of cholesterol to phospholipids in the lipid pools examined [23]. In our model system, with lipid vesicles and smooth muscle cells, we have not found any measurable net transfer of free cholesterol between 1 : 1 C/P vesicles and cultured cells. Our results show that the exchange rate is dependent on the ratio of vesicles to cells (donors to acceptors) in a saturable way. Maximal transfer is achieved at a vesicle/cell ratio of 3.4. 10’. This ratio dependence can be explained on the basis of the exchange mechanism first proposed by Phillips et al. [2]. According to them, monomeric cholesterol desorbs from the donor membrane into the water interphase, after which the molecule diffuse until it collides with an acceptor structure, where it can be solubilized. When the vesicle/cell ratio is increased, and the amount of donor particles is increased as well, a larger fraction of [3H]cholesterol can be partitioned into the water interphase. This will eventually lead to a larger incorporation of [ 3H]cholesterol into the cellular lipid pool. At the saturation level of vesicles, the desorption of monomeric [ 3H]cholesterol is ratelimiting, and no further increase in exchange rate will be found. Rothblat and Phillips [24] have recently presented similar results for a reversed system; the efflux of sterol from cultured cells into lipid vesicles. Since serum albumin is a protein with high-affinity to amphiphilic ligands, and since albumin
has been shown to induce sterol efflux from cultured cells, we wanted to study the effect of albumin on the exchange process. We found a marked increase in [ 3H]cholesterol incorporation into cells in the presence of bovine serum albumin. The maximal rate was achieved at an albumin concentration of 20 mg/ml. Rottem et al. [25] have presented similar results of the albumin effect on lipid transfer. They showed that the transfer of cholesterol from Mjxoplasma gallisepticum membranes to lipid vesicles was enhanced by albumin in a saturable way, with maximal effect at a concentration of 20 mg/ml. In other studies, albumin was added to increase the stability of membranes and vesicles [26], but the initial rate of cholesterol exchange was not found to be affected by the presence of albumin. The authors did not, however, make any systematic study of the albumin effect. The effect of albumin on the exchange rate is not due to contaminating apolipoprotein A-I, although its presence in commercial albumin preparations has been demonstrated by Fainaru and Deckelbaum [lo]. We obtained identical effect of albumin on the exchange rate, with both unpurified and purified preparations. Bartholow and Geyer [ 17,181 have shown that albumin can bind dipalmitoylphosphatidylcholine, and that this complex can induce sterol efflux from cultured cells. We propose that the fraction of [3H]cholesterol, which is solubilized by the albumin-phosphatidylcholine complex, is readily available for exchange. The finding that the exchange process is temperature dependent, with an activation energy of + 38.5 kJ . mall ‘, is in good agreement with activation energies reported by Poznansky and Czekanski [27] for membranes with similar C/P mole ratios (1 : 1). The activation energy for the process can be thought of as the sum of the individual free energies required to transfer 1 mol of cholesterol from a donor membrane, through the water interphase, to an acceptor membrane pool. It is evident that rat arterial smooth muscle cells differ significantly from the human skin fibroblast cell type, at least in the ability to esterify exchanged free cholesterol. The fact that rat smooth muscle cells esterify free cholesterol which origins from the cell membrane indicates strongly
439
that the cell membrane cholesterol pool is not metabolically isolated from internal sterol pools. Poznansky and Czekanski [26] could not find any of the incorporated free cholesterol in esterified form in human skin fibroblasts. Although the exchanged free [3H]cholesterol is found in the rat smooth muscle cellular cholesterol ester pool quite rapidly (in 4 h with vesicles of greater than 1 : 1 C/P), we are not able to define whether the cholesterol is esterified by acyl-CoA : cholesterol acyltransferase, or by some other enzyme with similar properties. The fact that exchanged cholesterol is esterified even when there is no net transfer into cells (vesicles of 1 : 2 C/P) indicates that the membrane cholesterol pool is in a dynamic balance with internal sterol pools. It seems reasonable that the aqueous diffusion mechnism has a physiological significance in distributing the unesterified cholesterol between different lipid pools. Quite recently, it has been shown that receptor-negative human fibroblasts can incorporate free cholesterol from LDL by an exchange process that is independent of receptormediated uptake of LDL [28]. It has also been shown that cholesteryl ester accumulation in cultured cells can be mobilized by phospholipid dispersions [29]. In spite of the growing knowledge in the exchange process, it is far too early to speculate on the quantitative significance of the process for the sterol balance in vivo.
11 Rothblat, G.H., Arbogast, L.Y., Quellette, L. and Howard, B.V. (1976) In Vitro 12, 554-57 12 Gamble, W., Vaughn, M., Kruth, H.S. and Avigan, J. (1978) J. Lipid Res. 19, 1068-70 13 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R. (1951) J. Biol. Chem. 193, 265-275 14 Markwell, M.A.L., Hass, SM., Bieber, L.L. and Tolbert, N.E. (1978) Anal. Biochem. 87, 206-10 15 Lundberg, B. (1973) Acta Chem. Stand. 27, 3435-39 16 Lundberg, B. (1973) Acta Academiae Aboensis 33, 1-4 17 Bartholow, L.C. and Geyer, R.P. (1981) B&him. Biophys. Acta 665, 40-47 18 Bartholow, L.C. and Geyer, R.P. (1981) J. Biol. Chem. 257, 3126-30 19 Goldstein, J.L. and Brown, MS. (1974) J. Biol. Chem. 249, 5153-62 20 Angel, A., Yuen, R. and Nettleton, J.A. (1981) Can. J. Biochem. 59, 655-61 21 Rothblat, G.H., Arbogast, L.Y. and Ray, E.K. (1978) J. Lipid Res. 19, 350-58
Acknowledgements
22 Nichols, 2783-90
These studies have been financed in part by grants from the Nordiska Samfundets Stiftelse for Vetenskaplig Forskning utan Djurforsok, Stockholm (Sweden). One of us (J.P.S.) has received grants from the Tor, Joe, and Pentti Borgs fond at Abe Akademi, Turku, Finland. These grants are gratefully acknowledged. We thank Dr. E.-M. Suolinna for comments on the manuscript. References I Goldstein, J.L., Helgeson, J.A.S. and Brown, MS. (1979) J. Biol. Chem. 254, 540335409
2 Phillips, M.C., McLean, L.R., Stoudt, G.W. and Rothblat, G.H. (1980) Atherosclerosis 36, 409-422 3 McLean, L.R. and Phillips, M.C. (1981) Biochemistry 20, 2893-300 4 Backer, J. and Dawidowicz, E.A. (1981) Biochemistry 20, 3805- 10 5 Bloj, B. and Zilversmith, D.B. (1976) Biochemistry 15, 1277-83 6 Wirtz, K.W.A. and Zilversmith, D.B. (1969) B&him. Biophys. Acta 193, 105-16 7 Ross, R. (1971) J. Cell Biol. 50, 172-186 8 Bierman, E.L., Stein, 0. and Stein, Y. (1974) Circ. Res. 35, 136-150 9 Lundberg, B. (1977) Chem. Phys. Lipids 18, 212-20 10 Fainaru, M. and Deckelbaum, R.J. (1979) FEBS Lett. 97, 171-174
J.W. and
Pagano,
R.E. (1981)
Biochemistry
20,
23 Lange, Y. and D’Alessandro, J.S. (1977) Biochemistry 16, 4339-43 24 Rothblat, G.H. and Phillips, M.C. (1982) J. Biol. Chem. 257, 4775-82 25 Rottem, S., Shinar, D. and Bittman, R. (1981) Biochim. Biophys. Acta 649, 572-80 26 Poznansky, M.J., and Czekanski, S. (1982) Biochim. Biophys. Acta 685, 182-90 27 Poznansky, M.J. and Czekanski, S. (1979) B&hem. J. 177, 989-91 28 Shireman, R.B. and Remsen, J.F. (1982) B&him. Biophys. Acta 711, 281-289 29 Yau-Young. A., Rothblat. G.H. and Small, D. (1982) Biochim. Biophys. Acta 710, 181-82
434
BBA 51335
TRANSFER OF I 3H]CHOLESTEROL BETWEEN LIPID VESICLES AND RAT ARTERIAL SMOOTH MUSCLE CELLS IN VITRO J. PETER
SLOTTE
and BO LUNDBERG
Department of Biochemistry and Pharmacy, kbo Akademi, Porthansgaten 3- 5, SF-20500 (Received
October
Turku (iho) {Finland)
Ist, 1982)
Key words: Cholesterol exchange; Yesicle - ceil transfer; Phosphaiidylcholine; Cholesterol esterification; (Rat smooth muxie cell)
Unesterified [ 3H]cholesterol is rapidly transferred between cholesterol-phosphatidylcholine vesicles and rat arterial smooth muscle cells in vitro. Exchange rate is influenced by the vesicle/cell ratio in a saturable way. The maximal transfer of cholesterol, which is 3.76 pg per mg cell protein during 4 h, is achieved with a vesicle/cell ratio of 3.4.10’. Bovine serum albumin enhances the exchange by a factor of 4.5 compared to a protein-free system. The activation energy for the process is + 38.5 kJ*mol - ’ with vesicles of 1: 1 mole ratio of cholesterol to phosphatidylcholine (C/P). A fraction of the inco~rat~ free [ ‘H~c~iesterol is esterified within 4 h with donor vesicles of over 1: 1 C/P. When cetls were incubated with vesicles of low C/P mole ratio (1: 2) a fraction of the incorporated free i3H]cholesteroi was esterified within 16 h. Our results are compatible with the aqueous diffusion mechanism of cholesterol exchange. Furthermore, we suggest that, in rat smooth muscle cells, the cell membrane cholesterol pool is not metabolically isolated from internal cholesterol pools, at least as judged by the ability of the ceils to esterify incorporated free cholesterol.
The surface transfer or exchange of lipids in general, and of cholesterol in particular, has recently been the subject of intense investigation. It has been shown that the exchange of free cholesterol between lipoproteins and cultured cells can play a significant role in the maintenance of the cellular sterol balance [I]. The mechanism of lipid transfer is a matter of debate by several authors 12-41. Phillips et al. (2f have postulated that the exchange process involves the desorption into and diffusion through the water interphase of the exchanging lipid molecules. This exchange mechanism would predict that the hydrophobicity of the lipids involved is one of the main determinants of the exchange rate. According to this it Abbreviations: Hepes, 4-(2-hydroxyethyi)-t-piperazineethanesulphonic acid; LDL, low-density lipoprotein. ~05-2760/83/~-~/$03.00
0 1983 Elsevier Science Publishers
would be easy to understand why phospholipids exchange much more slowly than cholesterol, since their monomeric water solubility (critical micellar concentration) is far less than that for cholesterol [3]. One must note that the difference in exchange rate is valid only in the absence of proteins. The existence of several phospholipid-specific exchange proteins has been shown by several investigators [5,6]. Although the picture of lipid exchange is growing clearer, it is still difficult to estimate the quantitative si~ificance of the process in vivo. In order to understand the contribution of lipid exchange to the complex lipid metabolism of the arterial wall, it is of great importance to characterize the effects of lipid exchange on the predominant cell type of the arterial wall, the smooth muscle cell. It is mainly unresolved whether smooth muscle cells are able to esterify membrane-derived free cholesterol or if the membrane cholesterol
435
pool is metabolically isolated from intracellular cholesterol pools. In the present study we have examined the exchange of free cholesterol between lipid vesicles of different C/P mole ratios, and cultured rat smooth muscle cells. We have also examined the effect of serum albumin on the rate of cholesterol exchange between vesicles and cells. Furthermore, we have followed the fate of the incorporated cholesterol molecules, that is the ability of the cells to esterify the exchanged free cholesterol. Experimental procedures Ceil culture Primary cultures of rat arterial smooth muscle cells from the thoracic aorta of adult SpragueDawley rats were used. The cell lines were started from explants of the intima-medial layer according to methods reported by Ross [7] and Bierman et al. [8]. The cells were grown in Dulbecco’s modified Eagle medium supplemented with 2 mM Lglutamine, 0.1 mM non-essential amino acid solution, 20 mM Hepes buffer, 0.08% (w/v) sodium bicarbonate, 10 pg/ml of gentamycin and 10% (v/v) newborn calf serum. The cells showed typical growth characteristics of cultured smooth muscle cells, as described by Ross [7]. The cells grew in multiple, overlapping layers with a doubling time of approximately 48 h. Lipid vesicles Vesicles containing different mole ratios of [3H]cholesterol (1.6. lo4 dpm/pg) to egg phosphatidylcholine were prepared by co-lyophilization from benzene stock solutions in glass tubes. After completed co-lyophilization, phosphate-buffered saline (pH 7.4) was added to give a final cholesterol concentration of 1 mg/ml. The suspension was sonicated with an MSE Ultrasonic Disintegrator (titanium probe) for 30 min at 20°C at maximal effect. The opalescent solution was centrifuged at 20000 X g for 30 min at 4°C in order to sediment titanium particles and larger lipid aggregates. When the lipid vesicles contained trace amounts of cholesteryl[‘4C]oleate, this was added from a benzene solution prior to co-lyophilization. Characterization of the lipid vesicles revealed a particle diameter of 25 nm and average particle weight of 1.7 . 10’ daltons (1 : 1 C/P vesicles) [9].
Incubation procedures All experiments were performed in 50 mm diameter petri dishes with confluent cells between the 7th and the 12th passage. About 2.9. lo5 cells were seeded in each dish, where after the cells grew to confluency in 4-5 days. A confluent petri dish contained about 1.5 . lo6 cells, corresponding to about 300 pg cell protein. Growth medium without serum supplement was used as the incubation medium, with final volume adjusted to 3.0 ml. Bovine serum albumin was added from a sterile 100 mg/ml stock solution made up in phosphatebuffered saline. The albumin preparation was purified from contaminating apolipoprotein A-I by the method reported by Fainaru and Deckelbaum [lo]. The purified albumin was delipidized according to the method of Rothblat et al. [ll]. The albumin preparation was over 98% pure by polyacrylamide gel electrophoresis. When cells were incubated at different temperatures, the appropriate dishes were equilibrated at respective temperatures for 30 min before adding the vesicle preparation. After completed incubation, cells were washed twice with cold phosphate-buffered saline and the cells were detached by gentle scraping. The cells were sedimented by low-speed centrifugation, and the cell pellet was disrupted in 500 ~1 phosphate-buffered saline by sonication. Aliquots of this suspension were taken for the different analytical procedures. Analytical procedures The cellular sterol mass was determined by the enzymatic method presented by Gamble et al. [ 121. Isotopic free and esterified cholesterol was extracted with chloroform/methanol (2 : 1, v/v) containing lipid carriers. The neutral lipids were separated by thin-layer chromatography on Kieselgel 60 (DC-Alufolien Kieselgel 60, Merck, Darmstadt, F.R.G.) with n-hexane/diethyl ether (90 : 10, v/v) and detected by spraying the thin-layer plates with 50% sulphuric acid and charring at 150°C for 5 min. The distribution of lipid radioactivity between free and esterified cholesterol positions were determined by cutting the respective positions into liquid scintillation vials. A toluene-based scintillation cocktail was added 30 min before counting. Cell protein was determined by the method of Lowry et al. [ 131 as modified by Markwell et al. 1141.
436
Reagents The egg phosphatidylcholine employed was a highly purified product prepared by a method developed in this laboratory [15]. The cholesterol used (Merck, Darmstadt, F.R.G.) was chromatographically pure by GLC. [3H]Cholesterol (58 Ci/mmol) was obtained from the Radiochemical Centre, Amersham, and the product was over 98% pure by TLC. Cholesteryl[‘4C]oleate was prepared according to a modified acid chloride method [ 161 and the radioactive [ “C]oleic acid (50 mCi/mmol) was purchased from the Radiochemical Centre, Amersham. Bovine serum albumin and all reagents for cholesterol determination were obtained from Sigma Chemicals, St. Louis, MO, U.S.A. Cell growth medium and supplements were purchased from Gibco Europe, U.K. Gentamycin was a product of Neofarma, Finland.
11 : / : -0.6
-0.4
-0,2
0 1 /
0.2
AwnIN
0.4
CONCE~~TRATION
0.6
0.8
(MG/ML)
Fig. 2. Effect of albumin concentration on the exchange rate. Cells were incubated at 37°C for 4 h with 33 pg vesicle [3H]cholesterol per ml medium (1 : 1 C/P) and different amounts of purified albumin. The vesicle-cholesterol/cellcholesterol mole ratio was approx. 12: 1. The radioactivity of the cells was counted after thorough washing.
Results The rate of [3H]cholesterol transfer from lipid vesicles into rat smooth muscle cells were determined from the amount of [ 3H]cholesterol accumulated in the cellular sterol pool. Measurements VESICLE / CELL
2.16
1.08
RATIO 3.24
4.32 x10'
I
of cellular sterol mass after incubation with lipid vesicles of 1 : 1 C/P mole ratio did not reveal any significant transfer of cholesterol mass. Fig. 1 demonstrates the effect of varying the vesicle/cell ratio on the exchange rate. During 4 h of incubation, a maximum of 3.76 pg of [3H]cholesterol was incorporated per mg cell protein, with a vesicle/cell ratio of 3.4. 10’ (values obtained from the double reciprocal insert). The vesicle/cell ratio of 3.4 . IO’ corresponds to a ves-5.50 0
OL-----;;60 50
150
E, =
+ 38.5 ~Jno~-l
200
p4 VESICLE CHOLESTEROLPER ML MEDIUM -7.5 ,
Fig. 1. Effect of vesicle/cell ratio (donor/acceptor) on the transfer of [‘HIcholesterol from lipid vesicles (1 : 1 C/P) to cultured smooth muscle cells. Each petri dish contained about 1.5. lo6 cells corresponding to 8.4 gg total cellular cholesterol. The cells were incubated at 37°C for 4 h with different amounts of vesicles in 3.0 ml medium. The amount of [7H]cholesterol transferred was determined by counting the cell radioactivity and the counts were correlated to the vesicle [‘HIcholesterol specific activity.
3.2
3.3
3.4 1 /
T
3.5
i i.6
X103 lo K-l)
Fig. 3. Effect of temperature on the transfer of [ ‘HIcholesterol between lipid vesicles (1 : 1 C/P) and cultured cells. Cells were incubated at different temperatures between 6 and 37’C with 33 pg vesicle [3H]cholesterol per ml medium. The vesiclecholesterol/cell-cholesterol mole ratio was approx. 12 : 1. The activation energy is obtained from the Arrhenius plot.
437
0
4
8
12
16
20
24
(H)
lNC”BATlON TIME
0
4
8 INCUBATION
12
16
20
24
CM)
TIME
Fig. 4 A. Effect of vesicle C/P mole ratio on the incorporation of [ 3H]cholesterol into cells. The incubation medium contained 33 pg vesicle cholesterol per ml. The vesicle-cholesterol/cell-cholesterol mole ratio was approx. 12: 1. 0, greater than 1 : 1 C/P; 0, 1 : 1 C/P; A 1 :2 C/P. B. Esterification of the incorporated [ 3H]cholesterol as a function of vesicle C/P mole ratio and time. Incubation conditions and symbols are identical to those in A.
icle-cholesterol/cell-cholesterol mole ratio of approx. 60 : 1. Half-maximal rate was achieved at a vesicle/cell ratio of 0.77 . lo’, which corresponds to a vesicle cholesterol concentration of 36 pg per ml medium. (Vesicle-cholesterol/cell-cholesterol mole ratio, approx. 13 : 1). Isotopic equilibration was achieved in 16 h, and a plot of In dpm versus time gives a straight line, indicating first-order kinetics (plot not shown). The transfer of [3H]cholesterol from lipid vesicles to cells is not due to fusion of vesicles with cell membranes, since cholesteryl[ “C]oleate, when used as a non-exchangeable marker, it not found in the cellular lipid pool even after prolonged incubations (48 h). Since serum albumin has been shown to induce sterol efflux from cultured cells [ 17,181 we examined whether the presence of albumin has any effect on the exchange rate. As shown in Fig. 2, the presence of albumin has a profound effect on the incorporation of [ 3H]cholesterol into cells. Maximal effect was achieved with an albumin concentration of about 20 mg/ml. The V,,, obtained, 7 pg of [3H]cholesterol incorporated per mg cell protein during 4 h, is about 4.5 times higher than the value obtained without albumin (compare to Fig. 1). The albumin concentration,
which gave half-maximal transfer was only 1.7 mg/ml. The rate of [ 3H]cholesterol transfer was strongly temperature-dependent, as shown in Fig. 3. In a protein-free system, with equimolar C/P vesicles, the half-time for the transfer was found to be 3.6 h at 37°C. An Arrhenius plot of In k versus l/T gives the activation energy for the process, which was found to be about + 38.5 kJ . mall’ with 1 : 1 C/P vesicles and cells. In Fig. 4, panel A, is shown the effect of different C/P mole ratios on the incorporation of vesicle [ 3HIcholesterol into cells. Considerably more cholesterol was transferred from vesicles with high cholesterol saturation (greater than 1 : 1 C/P) compared to vesicles of lower C/P ratio (1 : 1 and 1 : 2, respectively). This is true in spite of similar cholesterol concentration per ml medium in each case. Panel B in Fig. 4 shows that the incorporated [ ‘HIcholesterol was readily esterified within the cells. [ 3H]Cholesterol originating from cholesterol-rich vesicles (greater than 1 : 1 C/P) was found in the cholesteryl ester pool within 4 h of incubation. [ 3H]Cholesterol from 1 : 1 C/P vesicles was esterified within 6-8 h, while the [ 3H]cholesterol originating from 1 : 2 C/P vesicles was found in the ester pool after about 16 h.
438
Discussion The modes of cholesterol uptake by cells in culture are complex, and are not known in detail. In a lipoprotein-containing medium, sterols are transported into cells by the specific binding and internalization of low-density lipoproteins [ 191, by exchange of lipids between cell membranes and lipoproteins [20], and finally by pinocytosis of the medium. The receptor-mediated endocytosis is believed to be the dominating mechanism, although spontaneous lipid exchange can play an important role in the regulation of cellular sterol content ]1,211. It has been shown that lipid exchange between lipoproteins and biological membranes is bidirectional [22]. The net transfer of free cholesterol has been shown to be dependent on the mole ratio of cholesterol to phospholipids in the lipid pools examined [23]. In our model system, with lipid vesicles and smooth muscle cells, we have not found any measurable net transfer of free cholesterol between 1 : 1 C/P vesicles and cultured cells. Our results show that the exchange rate is dependent on the ratio of vesicles to cells (donors to acceptors) in a saturable way. Maximal transfer is achieved at a vesicle/cell ratio of 3.4. 10’. This ratio dependence can be explained on the basis of the exchange mechanism first proposed by Phillips et al. [2]. According to them, monomeric cholesterol desorbs from the donor membrane into the water interphase, after which the molecule diffuse until it collides with an acceptor structure, where it can be solubilized. When the vesicle/cell ratio is increased, and the amount of donor particles is increased as well, a larger fraction of [3H]cholesterol can be partitioned into the water interphase. This will eventually lead to a larger incorporation of [ 3H]cholesterol into the cellular lipid pool. At the saturation level of vesicles, the desorption of monomeric [ 3H]cholesterol is ratelimiting, and no further increase in exchange rate will be found. Rothblat and Phillips [24] have recently presented similar results for a reversed system; the efflux of sterol from cultured cells into lipid vesicles. Since serum albumin is a protein with high-affinity to amphiphilic ligands, and since albumin
has been shown to induce sterol efflux from cultured cells, we wanted to study the effect of albumin on the exchange process. We found a marked increase in [ 3H]cholesterol incorporation into cells in the presence of bovine serum albumin. The maximal rate was achieved at an albumin concentration of 20 mg/ml. Rottem et al. [25] have presented similar results of the albumin effect on lipid transfer. They showed that the transfer of cholesterol from Mjxoplasma gallisepticum membranes to lipid vesicles was enhanced by albumin in a saturable way, with maximal effect at a concentration of 20 mg/ml. In other studies, albumin was added to increase the stability of membranes and vesicles [26], but the initial rate of cholesterol exchange was not found to be affected by the presence of albumin. The authors did not, however, make any systematic study of the albumin effect. The effect of albumin on the exchange rate is not due to contaminating apolipoprotein A-I, although its presence in commercial albumin preparations has been demonstrated by Fainaru and Deckelbaum [lo]. We obtained identical effect of albumin on the exchange rate, with both unpurified and purified preparations. Bartholow and Geyer [ 17,181 have shown that albumin can bind dipalmitoylphosphatidylcholine, and that this complex can induce sterol efflux from cultured cells. We propose that the fraction of [3H]cholesterol, which is solubilized by the albumin-phosphatidylcholine complex, is readily available for exchange. The finding that the exchange process is temperature dependent, with an activation energy of + 38.5 kJ . mall ‘, is in good agreement with activation energies reported by Poznansky and Czekanski [27] for membranes with similar C/P mole ratios (1 : 1). The activation energy for the process can be thought of as the sum of the individual free energies required to transfer 1 mol of cholesterol from a donor membrane, through the water interphase, to an acceptor membrane pool. It is evident that rat arterial smooth muscle cells differ significantly from the human skin fibroblast cell type, at least in the ability to esterify exchanged free cholesterol. The fact that rat smooth muscle cells esterify free cholesterol which origins from the cell membrane indicates strongly
439
that the cell membrane cholesterol pool is not metabolically isolated from internal sterol pools. Poznansky and Czekanski [26] could not find any of the incorporated free cholesterol in esterified form in human skin fibroblasts. Although the exchanged free [3H]cholesterol is found in the rat smooth muscle cellular cholesterol ester pool quite rapidly (in 4 h with vesicles of greater than 1 : 1 C/P), we are not able to define whether the cholesterol is esterified by acyl-CoA : cholesterol acyltransferase, or by some other enzyme with similar properties. The fact that exchanged cholesterol is esterified even when there is no net transfer into cells (vesicles of 1 : 2 C/P) indicates that the membrane cholesterol pool is in a dynamic balance with internal sterol pools. It seems reasonable that the aqueous diffusion mechnism has a physiological significance in distributing the unesterified cholesterol between different lipid pools. Quite recently, it has been shown that receptor-negative human fibroblasts can incorporate free cholesterol from LDL by an exchange process that is independent of receptormediated uptake of LDL [28]. It has also been shown that cholesteryl ester accumulation in cultured cells can be mobilized by phospholipid dispersions [29]. In spite of the growing knowledge in the exchange process, it is far too early to speculate on the quantitative significance of the process for the sterol balance in vivo.
11 Rothblat, G.H., Arbogast, L.Y., Quellette, L. and Howard, B.V. (1976) In Vitro 12, 554-57 12 Gamble, W., Vaughn, M., Kruth, H.S. and Avigan, J. (1978) J. Lipid Res. 19, 1068-70 13 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R. (1951) J. Biol. Chem. 193, 265-275 14 Markwell, M.A.L., Hass, SM., Bieber, L.L. and Tolbert, N.E. (1978) Anal. Biochem. 87, 206-10 15 Lundberg, B. (1973) Acta Chem. Stand. 27, 3435-39 16 Lundberg, B. (1973) Acta Academiae Aboensis 33, 1-4 17 Bartholow, L.C. and Geyer, R.P. (1981) B&him. Biophys. Acta 665, 40-47 18 Bartholow, L.C. and Geyer, R.P. (1981) J. Biol. Chem. 257, 3126-30 19 Goldstein, J.L. and Brown, MS. (1974) J. Biol. Chem. 249, 5153-62 20 Angel, A., Yuen, R. and Nettleton, J.A. (1981) Can. J. Biochem. 59, 655-61 21 Rothblat, G.H., Arbogast, L.Y. and Ray, E.K. (1978) J. Lipid Res. 19, 350-58
Acknowledgements
22 Nichols, 2783-90
These studies have been financed in part by grants from the Nordiska Samfundets Stiftelse for Vetenskaplig Forskning utan Djurforsok, Stockholm (Sweden). One of us (J.P.S.) has received grants from the Tor, Joe, and Pentti Borgs fond at Abe Akademi, Turku, Finland. These grants are gratefully acknowledged. We thank Dr. E.-M. Suolinna for comments on the manuscript. References I Goldstein, J.L., Helgeson, J.A.S. and Brown, MS. (1979) J. Biol. Chem. 254, 540335409
2 Phillips, M.C., McLean, L.R., Stoudt, G.W. and Rothblat, G.H. (1980) Atherosclerosis 36, 409-422 3 McLean, L.R. and Phillips, M.C. (1981) Biochemistry 20, 2893-300 4 Backer, J. and Dawidowicz, E.A. (1981) Biochemistry 20, 3805- 10 5 Bloj, B. and Zilversmith, D.B. (1976) Biochemistry 15, 1277-83 6 Wirtz, K.W.A. and Zilversmith, D.B. (1969) B&him. Biophys. Acta 193, 105-16 7 Ross, R. (1971) J. Cell Biol. 50, 172-186 8 Bierman, E.L., Stein, 0. and Stein, Y. (1974) Circ. Res. 35, 136-150 9 Lundberg, B. (1977) Chem. Phys. Lipids 18, 212-20 10 Fainaru, M. and Deckelbaum, R.J. (1979) FEBS Lett. 97, 171-174
J.W. and
Pagano,
R.E. (1981)
Biochemistry
20,
23 Lange, Y. and D’Alessandro, J.S. (1977) Biochemistry 16, 4339-43 24 Rothblat, G.H. and Phillips, M.C. (1982) J. Biol. Chem. 257, 4775-82 25 Rottem, S., Shinar, D. and Bittman, R. (1981) Biochim. Biophys. Acta 649, 572-80 26 Poznansky, M.J., and Czekanski, S. (1982) Biochim. Biophys. Acta 685, 182-90 27 Poznansky, M.J. and Czekanski, S. (1979) B&hem. J. 177, 989-91 28 Shireman, R.B. and Remsen, J.F. (1982) B&him. Biophys. Acta 711, 281-289 29 Yau-Young. A., Rothblat. G.H. and Small, D. (1982) Biochim. Biophys. Acta 710, 181-82