LIPOPROTEIN A-I CONTAINING PARTICLES

LIPOPROTEIN A-I CONTAINING PARTICLES

INTRODUCTION

Many epidemiological studies have indicated that the plasma level of high density lipoproteins (HDL) is inversely correlated with the risk for coronary artery disease1. It has been hypothesized that HDL exerts this protective effect by the "reverse" transport of excess cholesterol from peripheral tissues to the liver2.

Nevertheless, in spite of its role in anti-atherogenesis, the true mechanisms of HDL uptake of peripheral cholesterol and subsequent delivery to the liver are still under investigation.

Conventionally HDL are isolated by ultracentrifugation in the density range of 1.063 to 1. 1 g/ml. HDL represents a heterogenous population of particles which differ in size, in lipid and protein composition and overall in their metabolic functions.

Although nearly all the apoproteins that have been characterized thus far can be found in HDL in variable proportions, apo A-I and apo together comprise 85-90% of the total HDL protein.

Ultracentrifugation has been an invaluable tool for subfractionation of lipoprotein particles but it has been shown that this procedure alters the structure and composition of the particles3.

we have been using immunological procedures to isolate the different particles of the HDL fraction. Particles containing apo A-I and apo A-II (LpA-I:A-II) and particles that do not contain apo A-II (LpA-I).

Studies published by our group in collaboration with Ailhaud's group have shown that on long-term exposure to LpA-I and LpA-I:A-II particles ; only the LpA-I particles were able to promote cholesterol efflux from cholesterol preloaded, differentiated OB 1771 adipose cells4.

More recently5 it has been shown that particles containing apo A-IV are equally effective in producing cholesterol efflux.

Taking into account these results we decided to further study the different HDL particles ; to do so we prepared directly from whole plasma LpA-I and LpA-I:A-II particles and simultaneously we isolated from the same plasma sample, the particles containing apo A-IV (LpA-IV) and the particles containing apo A-I, A-IV and A-II (LpA-I:A-IV:A-II).

The present studies were undertaken with two major goals :

- to define the lipid and protein composition of the different particles ;

- to correlate their composition to their ability to promote the cholesterol efflux from adipocytes in culture.

In order to understand the intracellular IOOchanism of cholesterol efflux and to appreciate the different roles of apo A-I and apo A-II, we decided to test the PKC involvement in this phenomenon.

It was recently described6 that binding of HDL3 to 3H-phosphatidyl­ choline (PC) labelled platelets stimulated a transient biphasic increase in diacylglycerol (DAG). We have analysed this stimulatory effect using HDL3 and proteoliposomes containing apo A-I or A-II on incubation with the adipose cells 08 1771.

MATERIAL AND METHODS

The study was carried out on the plasma of five normolipidemic male subjects ; the blood was drawn into tubes containing EDTA and a mixture of protease inhibitors. The plasma was promptly separated by low-speed centrifugation at 4°C and immediately used for isolation of the parti­ cles. All manipulations were performed at 4°C.

The particles were prepared by immunoaffinity chromatography as outlined before1 ,2. On each fraction, we determined :the lipid composi­ tion by enzymatic methods ; the apoprotein composition by ELISA.

We measured the concentration of the two major phospholipids phosphatidylcholine and shingomyelin isolated by thin layer chromato­ graphy?. We estimated the proportion of the molecular species of fatty acids on the lipid extract purified by thin layer chromatography. The fatty acids I).Ydrolyzed and methylated were measured on gas liquid chromatography<>.

The activity of the lecithin cholesterol acyl transferase (LCAT) was measured by the proteoliposome method of Chen and Albers9. Cholesterol efflux was determined on differentiated 08 1771 cells preloaded with 3H-cholesteryl ester LDL prepared by the method of Craig et al10.

3H-diacylglycerol was measured after separation on thin layer chromatography. The apo A-I and apo A-II containing liposomes were prepared by the cholate dialysis procedure9, molar ratio DMPC to protein 150:1.

Quantitation of apo A containing lipoprotein particles

In order to quantify LpA-I and LpA-I:A-II, two tests have been developed. TO directly determine LpA-I:A-II, we have used an enzyme­ linked differential antibody bnmunosorbent assay11. TO directly determine LpA-I we have developed a simpler procedure using differential electro­ immunoassay12. By using a large excess of anti A-II, LpA-I:A-II particles are retained in the first peak and LpA-I migrates as a second peak. A monoclonal anti A-I labelled with peroxidase revealed the two peaks while a monoclonal anti A-II revealed only one peak.

RESULTS

Figure 1 shows the representation of the composition by weight percentage of the different particles.

The four types of particles have about the same proportion of protein, but they show significant differences in cholesterol and trigly­ ceride content. LpA-I and LpA-I:A-II particles contained more cholesterol (11% of the total mass) than LpA-IV and LpA-I:A-IV:A-II which contain 6%, conversely the particles containing apo A-IV have more triglycerides (12%).

Apolipoprotein analysis showed that LpA-I and LpA-IV contain a single apolipoprotein, 97 and 98.6% respectively, LpA-I:A-II contain 53% of apo A-I and 45% of apo A-II. The LpA-I:A-IV:A-II contain 65% of apo A-I, 18% of apo A-IV and 14% of apo A-II.

We determined the proportion of the different molecular species of fatty acids, the most striking results were found in the phospholipid fractions. The LpA-IV particles contain a high proportion of saturated fatty acids (76%) significatively different from LpA-I (58%) and LpA-I:A-II (47%).

Figure 2 shows the results of the determination of the activity of LCAT. Taking the value for the LpA-I particle as one hundred percent activity, we found that the LpA-IV particles have the most activity, followed by LpA-IV:A-I:A-II and the LpA-I : the LpA-I:A-II particles have very little LCAT activity.

The phospholipid analysis revealed differences among the particles in the type of phospholipid constituents. The phosphatidylcholine/sphin­ gomyelin ratios were 3.9, 5.3 for LpA-I, LpA-I:A-II respectively and around 1 for the particles containing apo A-IV.

Incubation of 3a-PC prelabelled adipose cells in the presence of HDL3 or liposome containing apo A-I results in 3a-DAG production with a maximum at 5 min. On the other hand, tetranitromethane modified HDL3 TNM-HDL3) which is not recognized by the HDL receptor, did not induce C breakdown and DAG release. Moreover despite their ability to be effective competitors for apo A-I binding sites, apo A-II containing liposomes were ineffective in stimulating DAG production by phospholipase activation. It seems likewise that only the binding of apo A-I to cell surface receptors is able to promote DAG generation (Figure 4).

The presence of phorbol esters in the incubation medium enhanced the cholesterol efflux efficiency in LDL-cholesterol loaded 0817 adipose cells (Figure 5). This data strongly suggests that PC-breakdown, DAG production and protein kinase C activation are involved in the cholesterol efflux.

DISCUSSICN

Our results show that factors other than the lipid composition of the particles studied are the major determinant of the ability to promote cholesterol efflux.

The compositional data of LpA-I and the two apo A-IV containing particles reveals that these lipoproteins are very different, but on incubation with adipocytes they are equally effective in facilitating the efflux of cholesterol. On the other hand LpA-I:A-II are very similar in composition to LpA-I but are unable to promote cholesterol efflux. These results confirm the antagonist role proposed for apo A-II.

This antagonist role of apo A-II is demonstrated in our studies on the production of DAG and correlates very well with the results obtained in epidemiological studies.

We have shown recently that the lower apo A-I levels for patients with significant coronary artery disease were reflecting, in fact, a decrease in LpA-I particles13. Other data obtained in octogenarians14 supports, also, the view that LpA-I might represent the "anti-athero­ genic" fraction of HDL.

Apo A-I in fenales and apo A-II in males were lower in octogenarians while apo A-I in males and apo A-II in females were similar in octogenarian and control subjects. However, LpA-I was significantly elevated in octogenarian males and females by comparison with younger control subjects.

Recently it has been observed that the level of LpA-I in children whose patients suffer from premature coronary heart disease (CHD) was lower than that of a cnntrol group without any familial history of CHD15.

The clinical interest in the quantification of LpA-I and LpA-I:A-II is illustrated by the effect of moderate alcohol consumption on HDL16. We have measured HDL cholesterol, apo A-I, apo A-II, LpA-I and LpA-I:A-II in plasma from three hundred and fifty male subjects matched for age and clinical data and divided into five groups according to their alcohol consumption. Results confirm that alcohol consumption increases LpA-I:A-II and decreases LpA-I. These opposite variations are dose dependent and the differences are highly significant. Our findings indicate that an increase in HDL cholesterol can reflect an increase in LpA-I :A-II and a decrease in LpA-I. r.bre011er, assuming that LpA-I is the "anti-atherogenic" subfraction, alcohol " <A::mld oot have any anti-athero­ genic effect through the increase in HDL.

REFERENCES

1. T. Gordon, W.P. Castelli, M.C. Hjortland, W.B. Kannel, T.R. Dawber, High density lipoprotein as protective factor against coronary heart disease, Am. J. Med. 62:707 (1977).

2. J. A. Glomset, The plasma lecithin-cholesterol acyl transferase reaction, J. Lipid Res. 9:155 (1968).

3. G. R. Castro, and C.F. Fielding, Evidence for the distribution of apolipoprotein E between lipoprotein classes in human normo­ cholesterolemic plasma and for the origin of unassociated apolipo­ protein E (LpE), J. Lipid Res. 25:58 (1984).

4. R. Barbaras, P. Puchois, J.C. Fruchart, and G. Ailhaud, Cholesterol efflux fran cultured adipose cells is mediated by LpA-I particles but not by LpA-I:A-II particles, Biochem. Biophys. Res. Oommun. 142:63 (1987).

5. A. Steinmetz, R. Barbaras, N. Ghalim, V. Clavey, J .C. Fruchart, and G. Ailhaud, Human apolipoprotein A-IV binds to apolipoprotein A-I/A-II receptor sites and promotes cholesterol efflux from  adipose cells, J. Biol. Chern. 265:7859 (1990).

6. C. Delbart, N. Theret, G. Ailhaud, J.C. Fruchart, Phosphatidylcholine breakdown during receptor binding of HDL3, 62nd Scientific Sessions of American Heart Association, New-Orleans, 13-16 novembre 1989, Circulation, 80/4, supplement II, II-487, abstract 1938 (1989).

7. A. Nouvelot, G. Sezille, P. Dewailly, J.C. Fruchart, Chranatographie monodimensionnelle des lipides polaires sur couche mince de gel de silice en gradient discontinu d'humidite, Revue Francaise des Cbrps Gras 7:251 (1977).

8. L. Hagenfelt, A gas chromatography method for the determination on individual free fatty acids in plasma, Clin. Chim. Acta 13:266 (1966).

9. C. H. Chen and J.J. Albers, Characterization of proteoliposomes containing apolipoprotein A-I :a new substrate for the measurement of lecithin cholesterol acyl transferase activity, J. Lipid Res. 23:680 (1982).

10. J. F. Craig, D.P. Via, B.C. Sharril, L.A. Labar, W.W. Mantulin, A.M. Gotto, and L. Smith, Incorporation of defined cholesteryl esters into lipoproteins using cholesterol ester-rich microemulsions, J. Biol. Chern. 257:330 (1982).

11. E. Koren, P. Puchois, P. Alaupovic, J. Fesmire, A. Kandoussi, J.C. Fruchart, Quantification of two different types of apolipoprotein A-I containing lipoprotein particles in plasma by enzyme linked differential antibody immunosorbent assay, Clin. Chern. 33:38 (1987).

12. H. J. Parra, H. Mezdour, N. Ghalim, J.M. Bard, J.C. Fruchart, Differential electroimmunoassay on ready-to-use plates for human LpA-I lipoprotein particles, Clin. Chern. in press (1990).

13. P. Puchois, A. Kandoussi, P. Fievet, J.L. Fourrier, M. Bertrand, E. Koren, J .C. Fruchart, Apolipoprotein A-I containing lipoproteins in coronary artery disease, Atherosclerosis 68:35 (1987).

14. G. Luc, J.M. Bard, s. Lussier-Cacan, H.J. Parra, J.C. Fruchart, J. Davignon, High density lipoprotein particles in octogenarians, Metal:x>lism in press (1990)•

15. P. Amouyel, H.J. Parra, D. Boute, J.M. Bard, D. Isorez, D. Barbier,

J.C. Fruchart, G. Zylberberg, Inlfuence des antecedents cardio­

vasculaires familiaux sur le profil lipoproteique d'une population issue de la metropole lilloise, XVeme Reunion de l'ADELF, Bordeaux, 12-15 septembre 1989.

16. N. Ghalim, P. Puchois, G. Zylberberg, P. Fievet, C. Demarquilly, J.C. Fruchart, Effect of alcohol intake on human apolipoproteins A-I containing lipoproteins subfractions, Arch. Int. Med. in press (1990).