| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Atherosclerosis and Lipoproteins |
From the Laboratoire de Biochimie (N.F., D.G., N.M., J.-L.P.), Faculté des Sciences Pharmaceutiques, Châtenay-Malabry; Laboratoire de Biochimie (N.F., A.C., V.A., N.M., J.-L.P.), Hôpital Européen Georges Pompidou, AP-HP, Paris; U505 INSERM (D.P., J.C., A.-D.K.), Université Pierre et Marie Curie, Centre de Recherche des Cordeliers, Paris; and U430 INSERM (A.N.), Hôpital Broussais, AP-HP, Paris, France.
Correspondence to Dr N. Fournier, Laboratoire de Biochimie, Faculté des Sciences Pharmaceutiques, 5 rue JB Clément, 92296 Châtenay-Malabry, France. E-mail natalie_fournier{at}yahoo.fr
| Abstract |
|---|
|
|
|---|
Key Words: human apolipoprotein A-II transgenic mice Fu5AH rat hepatoma cells J774 mouse macrophages cholesterol efflux ATP binding cassette transporter 1
| Introduction |
|---|
|
|
|---|
ApoA-I is the major apolipoprotein component of HDL, and its role in RCT has been extensively studied. ApoA-II is the second most abundant HDL apolipoprotein, but its contribution to the function of HDL is controversial and poorly understood. We recently generated transgenic mice overexpressing human apolipoprotein (hapo) A-II, which display a marked hypertriglyceridemia and a great decrease in plasma HDL and apoA-I concentrations.12 HDLs from our transgenic mice are characterized by an enrichment in triglycerides, the predominance of smaller sized particles (7.8- versus 10-nm diameter), a low apoA-I/apoA-II ratio, and the presence of pre-ß HDLs containing hapo A-II.13
Our aim was to investigate the ability of plasma from hapo A-II transgenic (hAIItg) mice to promote cholesterol efflux from 2 cell types representative of SR-BImediated or ABCA1-mediated cholesterol efflux. Fu5AH rat hepatoma cells, which have a high expression level of SR-BI and lack ABCA1,7,10 were used to assess the contribution to cholesterol efflux of plasma PL-rich acceptors.1416 The ability of plasma with a high hapo A-II content to promote cellular cholesterol efflux was assessed by use of the J774 mouse macrophage cell system, which expresses SR-BI at very low levels.7 Exposure of J774 cells to cAMP upregulates the expression of ABCA1, which is closely correlated with increased cholesterol efflux to several unassociated apolipoproteins, such as apoA-I,17 apoE,18 and apoA-IV.9 Therefore, we compared the efflux capacity of purified hapo A-II with that of hapo A-I.
| Methods |
|---|
|
|
|---|
| Results |
|---|
|
|
|---|
and
), we compared its ability to promote cholesterol efflux with that of control mouse plasma. After 4-hour incubations with 2.5% diluted plasma from animals fed ad libitum, the fractional cholesterol efflux was 36% and 57% lower in plasma from
and
mice, respectively, than in plasma from control mice (Table 1). To determine whether the low plasma concentration or the different properties of HDL from transgenic mice were responsible for the decreased cholesterol efflux, we compared HDL from transgenic and control mice fed ad libitum at the same HDL-PL concentration. HDL from transgenic mice displayed increased protein and triglyceride contents and decreased cholesteryl ester, resulting in a lower ratio of HDLtotal lipid to protein (Table 2). Cholesterol efflux was greatest to control HDL and was inversely related to the level of hapo A-II expression in transgenic mice, displaying 25% and 50% decreases in HDL from
and
mice, respectively (Figure 1).
|
|
|
Plasma From hAIItg Mice Stimulates Cholesterol Efflux From J774 Cells
In a first attempt, we tested the same plasma pools used in Fu5AH cells for their capacity to promote cholesterol efflux from J774 cells. Surprisingly, we obtained the opposite effect, inasmuch as plasma pools from
and
mice exhibited a greater efflux capacity from cAMP-treated cells (0.72±0.02%, 1.18±0.05%, and 2.46±0.09% for control,
, and
mice, respectively). Therefore, we precisely documented this effect with plasma from individual mice and used the feeding-fasting transition as a means to modulate the plasma concentrations of hapo A-II, mouse apo A-I (mapo A-I), and HDL. Indeed, plasma hapo A-II from animals in the fasted state is decreased 2-fold, leading to concomitant increases in plasma HDL levels and in the ratio of apoA-I to A-II in HDL; conversely, hypertriglyceridemia is normalized.13 When cells were not pretreated with cAMP, the fractional efflux was comparable to plasma from all mouse groups (Table 3). Pretreatment of the cells with cAMP markedly increased the cholesterol efflux to plasma from fed transgenic animals, with 40% and 73% cAMP stimulations of efflux for
and
mice, respectively. Interestingly, plasma from fasted
and
mice, which have similar plasma hapo A-II concentrations, induced comparable cAMP stimulations of efflux. By contrast, the fractional efflux to control mouse plasma was not stimulated by cAMP. Linear regression analysis revealed a strong positive correlation between the percent cAMP stimulation of efflux and the plasma concentration of hapo A-II (see Figure I, which can be accessed online at http://atvb.ahajournals.org).
|
Hapo A-II and Hapo A-I Stimulate Cholesterol Efflux From J774 Cells
Because cholesterol efflux from J774 macrophages is enhanced in the presence of unassociated apolipoproteins9,17,18 and because plasma from hAIItg mice has elevated concentrations of hapo A-II, we compared the efflux capacities of lipid-free hapo A-II and hapo A-I. In 3 separate experiments, marked cAMP stimulations of efflux were promoted by both apolipoproteins to similar degrees (Table 4). These values were very close to those previously obtained for hapo A-I and hapo A-IV.9
|
Pre-ß Migrating HDLs Containing Hapo A-II Are Present in Plasma of Transgenic Mice
Because pre-ß HDL is an efficient acceptor of cellular cholesterol, we analyzed the proportions of
and pre-ß HDL in control and transgenic mice (Figure 2). In plasma from control mice either fed or fasted, the bulk of apoA-I was present in large-sized
HDL and a very small amount in pre-ß HDL, whereas mapo A-II was present only in
HDL (not shown). Plasma from fed or fasted
mice contained 2 main HDL subpopulations with
mobility; apoA-I was present in the larger
HDL and in pre-ß HDL (not shown), whereas hapo A-II was present in all
HDL subpopulations as well as in pre-ß HDL. Irrespective of nutritional state, plasma from
mice contained little apoA-I in
HDL (not shown), whereas hapo A-II was abundantly present in all
HDL subfractions, especially in the smaller ones. For the fed state, pre-ß HDL with hapo A-II was more abundant in
than in
mouse plasma. Of note, pre-ß HDL with hapo A-II was not detected when
and
mice were fasted overnight.
|
Absence of Atherosclerotic Lesions in Transgenic Mice
Aortic sections from chow-fed control mice (not shown) and hapo A-II high-expressing
mice stained with oil red O were totally devoid of lipid infiltrations (Figure II, which can be accessed online at http://atvb.ahajournals.org). "En face" specimens of the thoracic aorta from chow-fed control and high-expressing
mice were stained with oil red O and were also totally devoid of lipid infiltrations (data not shown).
| Discussion |
|---|
|
|
|---|
In the Fu5AH system, cholesterol efflux correlates best with HDL-PL.15,16 The markedly reduced cholesterol efflux capacity of plasma from hAIItg mice is consistent with the major decrease in HDL.12,13 Plasma from independently established mice overexpressing hapo A-II has also displayed lower HDL-PL concentrations and a reduced ability to efflux cholesterol from fibroblasts.19 The lower efflux capacity of HDL from hAIItg mice compared with HDL from control mice may be attributed to their high hapo A-II content. Indeed, cholesterol efflux from Fu5AH cells to human HDL3 was negatively correlated with their apoA-II/apoA-I+apoA-II ratio.20 Our results are also in accordance with the lower cholesterol efflux capacity from Fu5AH cells of HDL from hapo A-I/hapo A-II transgenic mice compared with HDL from hapo A-I transgenic mice.21 Furthermore, we have demonstrated by use of a large number of human serum samples that LpA-I is a better acceptor than LpA-I/A-II of cholesterol effluxing from Fu5AH cells.14
Hapo A-I and hapo A-II bind to rodent SR-BI,22 and this binding is essential for cholesterol efflux.23 However, although apoA-IIenriched HDLs bind to human SR-BI with higher affinity than apoA-Irich HDLs, they display a lower capacity to deliver cholesteryl esters to human adrenal cells.24 By analogy, the lower efflux capacity of hapo A-IIrich HDL from our hAIItg mice may indicate that hapo A-II is a negative modulator of SR-BImediated cholesterol efflux. This negative effect of hapo A-II may be related to the altered cholesterol distribution in plasma membrane domains induced by SR-BI.25
Apart from the greater apoA-II/apoA-I ratio, some of the other HDL modifications of our transgenic mice may contribute to their lower efflux capacities from Fu5AH cells. It is unlikely that the smaller particle size explains their reduced efflux effectiveness, inasmuch as smaller particles, at a given PL concentration, more efficiently remove cellular cholesterol.26 However, the triglyceride enrichment of HDL may affect cholesterol efflux because, at a given PL concentration, triglyceride-enriched HDLs isolated from hypertriglyceridemic subjects were less efficient in extracting cholesterol.27
The major finding of the present study was that cAMP treatment of J774 cells, which upregulates ABCA1,10,11 markedly increased cholesterol efflux to the plasma of hAIItg mice, in a manner directly proportional to the plasma concentration of hapo A-II; conversely, no variation in efflux to the plasma of control mice was observed. Exposure of macrophages to cAMP stimulated the release of cholesterol and PL to lipid-free apoE,18 apoA-I, 17 and apoA-IV.9 Our hapo A-II transgenic mice have small amounts of apoE and apoA-IV, which are comparable to those of control mice, whereas HDL and apoA-I are greatly decreased.12,13 On the other hand, plasma from hAIItg mice contains small-sized
HDLs rich in hapo A-II as well as pre-ß HDLs with hapo A-II.13 The greater cAMP stimulation of efflux to plasma from fed relative to fasted
mice may be linked to the appreciable amount of hapo A-IIcontaining pre-ß HDLs in the fed animals. Moreover, in the present study, free hapo A-II stimulated cholesterol efflux to the same extent as did free hapo A-I. Accordingly, similar efflux stimulations were reported for unassociated apoA-II and apoA-I from human fibroblasts that express ABCA1,28 whereas expression of ABCA1 in Hela cells markedly increased specific binding of apoA-I and apoA-II to a common binding site.29 Therefore, the greater efflux ability of plasma from hapo A-II transgenic mice may be attributed to the pre-ß HDLs containing hapo A-II and/or to the high total hapo A-II content.
Studies of cholesterol efflux from cultured cells ultimately aim to assess the efficacy of the first step of RCT as a predictor of atherogenic risk.2 Thus, it is of interest to compare the results on cholesterol efflux elicited by plasma from various types of mice overexpressing hapo A-II with the degree of aortic lesions developed by the corresponding mice.
Transgenic mice with a high hapo A-II expression did not develop lesions when they were fed a chow diet,30 whereas their plasma displayed a 40% lower efflux capacity from fibroblasts than did plasma from chow-fed control C57BL/6 mice.19 Conversely, plasma from hapo A-II transgenic and control mice fed the atherogenic diet had similar efflux capacities,19 although the former developed more aortic lesions than did the latter.30 Independently, established transgenic mice with a moderate expression of hapo A-II developed less aortic lesions on a cholesterol-rich diet than did control mice, although their plasma elicited less cholesterol efflux from Fu5AH cells.31 On the other hand, apoA-I knockout (AI-KO) mice have very low plasma HDL levels but are not at increased risk of aortic lesion development, even after administration of the atherogenic diet.32 Chiesa et al33 expressed hapo A-I, hapo A-II, and both hapo A-I and A-II in the AI-KO background, and they measured cholesterol efflux from Fu5AH cells. Efflux was equally low in plasma from AI-KO and AI-KO/hAIItg mice, whereas it increased appreciably in plasma from AI-KO/hAItg mice, with or without the concomitant expression of hapo A-II. In the present study, the hypertriglyceridemic high-expressing
mice did not develop atherosclerotic lesions, although their HDL and plasma displayed lower efflux capacities from Fu5AH cells than did HDL and plasma from control mice.
Taken together, the results of the above studies clearly show that variations in the ability of plasma and HDL to induce cholesterol efflux from Fu5AH cells is not predictive of aortic lesion development of the corresponding mice. To what extent SR-BIexpressing cells display appreciable cholesterol efflux in vivo has not been established at present. Conversely, it is well known that SR-BI mediates selective uptake of cholesteryl esters in steroidogenic tissues and the liver.34 Because SR-BI is equally efficient at mediating the import and export of cholesterol to and from cells to lipoproteins and other acceptors,7,8 the specific metabolic conditions in vivo favoring cholesterol influx or efflux remain to be established.
By contrast, expression of a nonfunctional ABCA1 protein leads to increased atherosclerotic risk in patients with Tangier disease,35 whereas the expression of a partially functional ABCA1 is correlated with HDL deficiency in humans.2,36 In this case, lack of cholesterol efflux is correlated with greater atherosclerotic risk, thus providing evidence that strengthens the link between HDL, RCT, and atherosclerosis.35 Plasma from our hAIItg mice markedly stimulated cholesterol efflux from J774 macrophages pretreated with cAMP, which upregulates ABCA1 expression,10,11 whereas the coronary arteries of hAIItg mice were totally devoid of lipid infiltrations.
In conclusion, our data demonstrate opposite effects of plasma from hapo A-II transgenic mice on cholesterol efflux from Fu5AH hepatoma cells and from J774 macrophages. Because macrophages are implicated in the early steps of atherogenesis, the increased ability of plasma with a high hapo A-II content to extract cholesterol from such cells indicates an antiatherogenic influence of hapo A-II.
| Acknowledgments |
|---|
Received November 27, 2001; accepted January 30, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. D. Wool, C. A. Reardon, and G. S. Getz Apolipoprotein A-I mimetic peptide helix number and helix linker influence potentially anti-atherogenic properties J. Lipid Res., June 1, 2008; 49(6): 1268 - 1283. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. Adorni, F. Zimetti, J. T. Billheimer, N. Wang, D. J. Rader, M. C. Phillips, and G. H. Rothblat The roles of different pathways in the release of cholesterol from macrophages J. Lipid Res., November 1, 2007; 48(11): 2453 - 2462. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Zimetti, G. K. Weibel, M. Duong, and G. H. Rothblat Measurement of cholesterol bidirectional flux between cells and lipoproteins J. Lipid Res., March 1, 2006; 47(3): 605 - 613. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Rotllan, V. Ribas, L. Calpe-Berdiel, J. M. Martin-Campos, F. Blanco-Vaca, and J. C. Escola-Gil Overexpression of Human Apolipoprotein A-II in Transgenic Mice Does Not Impair Macrophage-Specific Reverse Cholesterol Transport In Vivo Arterioscler. Thromb. Vasc. Biol., September 1, 2005; 25(9): e128 - e132. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. J. Beulens, A. Sierksma, A. van Tol, N. Fournier, T. van Gent, J-L. Paul, and H. F. J. Hendriks Moderate alcohol consumption increases cholesterol efflux mediated by ABCA1 J. Lipid Res., September 1, 2004; 45(9): 1716 - 1723. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Sauvaget, V. Chauffeton, S. Dugue-Pujol, A.-D. Kalopissis, I. Guillet-Deniau, F. Foufelle, J. Chambaz, A. Leturque, P. Cardot, and A. Ribeiro In Vitro Transcriptional Induction of the Human Apolipoprotein A-II Gene by Glucose Diabetes, March 1, 2004; 53(3): 672 - 678. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2002 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |