Interaction between high-density lipoprotein subpopulations in apo B-free and abetalipoproteinemic plasma

Interaction between high-density lipoprotein subpopulations in apo B-free and abetalipoproteinemic plasma

Biochimica et Biophysica A cta, 1128(1¢992)244-249 © 1992 ElsevierScience Publishers B.V. All rights reserved tX)05-2760/92/$05.00 244 BBALIP 54026 ...

649KB Sizes 0 Downloads 62 Views

Biochimica et Biophysica A cta, 1128(1¢992)244-249 © 1992 ElsevierScience Publishers B.V. All rights reserved tX)05-2760/92/$05.00

244

BBALIP 54026

Interaction between high-density lipoprotein subpopulations in apo B-free and abetalipoproteinemic plasma M a r i a n C. C h e u n g

", A n i t r a C. W o l f " a n d D . R o g e r l l l i n g w o r t h b

"Department of Medicine, Unit'ersity of Washington, Seattle WA (USA) and t, Department of Medicine; Oregon Health Sciences Unit'ersity, Portland OR (USA)

(Received 19 September 1991) (Revised manuscript received26 June 1992) Key words: Abetalipoproteinemia:ttDL subpopulation:Immunoaffi~itychromatography;LCAT: Lipid; Metabolism Two populations of high-density lipoprotein (HI)L) particles exist in human plasma. Both contain apolipoprotein (apo) A-I, but only one contains apo A-Ih Lp(A! w AID and Lp(A! w / o AII), To study the extent of interaction between these particles, apo B-free plasma prepared by the selective removal of apo B-containing lipoproteins (LpB) from the plasma of three normolipidemic (NL) subjects and whole plasma from t~:o patients with abetalipoprotcinemia (ABL) were incubated at 37°C for 24 h. Apo B-free ptasJ~a samplcs were uscd to avoif~ lipid-exchange between HDL and LpB. Lp(AI w All) and Lp(AI w / o All) were isolated from each apo B-free plasma sample before and after incubation and ',heir protein and lipid contents quantified. Before incubation, ABL plasma had reduced levels of Lp(AI w AID and Lp(AI w / o All). (40% and 70% of normals, respectively). Compared to the HDL of apo B-free NL plasma, ABL HDL had higher relative contents of free cholesterol, phospholipid and total lipid, and contained more particles with apparent hydrated Stokes diameter in the 9.2-17.0 nm region. These differences were particularly pronounced in particles without apo A-II. Despite their differences, the total cholesterol contents of Lp(A1 w AID increased, while that of Lp(AI w / o All) decreased in all five plasma samples and the amount of apo A-! in Lp(AI w All) increased by 6-8 mg/dl in four during the incubation. These compositional changes were accompanied by a relative reduction of particles in the 7.0-8.2 nm Stokes diameter size region and an increase of particles in the 9.2-11.2 nm region. These data are consistent with intravascular modulation between HDL particles with and without apo A-II. The ob~rved increase in apo A-IZ-associatcd cholesterol and apo A-I, could involve either the transfer of cholesterol and apo A-! from particles without apo A-II to those with A-If, or the transfer of apo A-If from LIRAI w AID to LIRAI w / o AID. The exact mechanism and direction of the transfer remain to be determined.

Introduction Human high-density lipoproteins (HDL) are made up of two populations of particles. Botb contain apolipoprotein (apo) A-l, but only one contains apo A-il: Lp(AI w A l l ) and Lp(AI w / o A l l ) [1]. These apo-defined HDL particles are heterogeneous in size and in their lipid and protein composition [I-4]. They are present in both conventional H D L 2 and H D L 3 density subfractions [2,5]. In clinical studies, it has been found that compared to healthy, normolipidemic individuals, the levels and physicochemical properties of Lp(AI w All) and Lp(A! w / o AID are altered in patients with coronary artery diseases (CAD) or those at risk for CAD [6-8]. in intervention studies, Lp(A! w A l l ) and Lp(AI w / o All) are affected to different extents by

Correspondence to: M.C. Cheung, Department of Medicine,University of Washinglon,2121 N 35th St., Seattle, WA 98103, USA.

lipid-lowering agents [9-11] and by diets high in polyunsaturated fats [12]. These observations suggest that H D L particles with and without apo A-ll are metabolically distinct, in order to further define their metabolic interrelations, we have characterized the H D L in hepatoblastoma G2 (HepG2) conditioned medium [13] and have studied the role of the lower-density lipoproteins and lecithin-cholesterol acyltransferase (LCAT) in modulating their chemical composition and physical properties in vitro [14]. We found that HepG2-conditioned medium contained nascent Lp(AI w A l l ) and Lp(AI w / o All), suggesting that these particles may be individually synthesized and secreted. We also found that LCAT could alter the chemical composition and particle size of plasma Lp(A! w / o A l l ) in the absence of other plasma iipoproteins. However, both LCAT and the lower-density lipoproteins were necessary to induce significant changes in the composition and size of plasma Lp(A! w A l l ) in vitro. Not addressed in our previous studies is the extent of lipid and protein interaction between H D L particles

245 with and without apo A-I1. in normal plasma, the presence of lipid-exchange or net transfer between the lower-density lipoproteins and HDL makes it difficult to differentiate HDL lipid from non-HDL lipid. To circumvent this situation, apo B-free plasma generated by the removal of apo B-containing liporoteins (LpB) from the plasma of three normolipidcmic (NL) subjects was incubated in this study to assess the movement of lipid and protein among HDL particles. Since the removal of LpB from a NL plasma may disrupt the established equilibrium state of HDL in the plasma and affect subsequent observations, we also performed studies with plasma from two previously described patients with abetalipoproteinemia (ABL), a rare genetic disease in which LpB are absent from plasma [15,16]. Materials and Methods

Plasma and apo B-free plasma Blood samples from all but the female ABL subject who was pregnant, were obtained after a 12-14 h overnight fast and were collected into tubes containing EDTA (1 mg/ml). Upon separation of plasma at 4°C by low-speed centrifugation, preservatives were promptly added to final concentrations of 0.5 g / I sodium azide, 0.01 g / I chloramphenicol and 0.005 g / l gentamycin. Plaslna samples from the patients with ABL were shipped on wet ice to Seattle. The NL subjects were personnel of the Northwest Lipid Research Laboratories. Their plasma were adsorbed with dextran sulfate cellulose (Kanegafuchi Chemical Industry, Osaka, Japan) to remove all LpB [i7] prior to incubation and lipoprotein isolation. Apo B-free NL plasma prepared by this metho0 has been shown to retain 86% to 100% of plasma LCAT and cholesteryl ester transfer protein (CETP) activities [14]. Furthermore, the HDL particles isolated from plasma before and after dextran sulfate absorption were comparable

[14]. Incubation studies 5-10 ml aliquots of apo B-free NL plasma and ABL plasma were separately incubated at 37°C for 24 h in sterilized glass vials, Similar aliquots were kept at 4°C as controls. At the end of the incubation, Lp(Ai w All) and Lp(A! w / o A l l ) w e r e immediately isolated from each plasma aliquot for lipid, protein and size analyses. Isolation of apo A-i-containing lipoproleins APO A-l-containing lipoproteins with and without apo A-II were isolated from apo B-free NL plasma and ABL plasma by an established two-step immunoaffinity chromatography procedure [1,18]. Briefly, plasma aliquots were sequentially adsorbed with anti-A-lI-Sepharose CL 4B and anti-A-I-Sepharose CL 4B to remove Lp(AI w AID and Lp(A! w / o AID, respectively.

No:'.'~dsorbed proteins were washed with 0.01 M TrisHC1 buffer (pH 7.4), containing 0.15 M NaCI. 1 mM EDTA and 0.05% sodium azide. Lipoproteins bound to the immunoadsorbents were eluted with 3 M NaSCN in 0.02 M sodium phosphate (pH 7.0) and immediately filtered through a column packed with Sephadex G-25 (Pharmacia LKB Biotechnology) to remove the thiocyanate. Nonadsorbed proteins and adsorbed lipoproteins were concentrated by MicroConfilt concentrator (Biomolecular Dynamics, 8eaverton, OR) for composition and size analyses. All isolation and concentration processes were performed at 4°C.

Analytical procedures Total cholesterol (C), unesterified C (FC), phospholipid (PL) and triacylglycerol (TG) in the plasma and liooprotein fractions were measured by en~matic methods [19]. Cholesterol ester (CE) mass was calculated as the difference between C and FC with adjustment for the difference between the molecular weight of FC (387) and CE (650). Apo A-I and A-ll were quantitated by specific immunoassays [20,21]. HDL particle sizes were determined by non-denaturing gradient polyac~lamide gel electrophoresis (gPAGE) using precast 4-30% gels (Pharmacia LKB) [22] an~ proteins were visualized with 0.04% Coomassie G-250 dissolved in 3.5% perchioric acid. Gel scanning and integration were performed with the LKB 2400 laser densitometer and Gelscan xL software. Results

Apo B-free NL plasma and ABL plasma The apo A-l, A-I! and lipid contents of the apo B-free NL plasma and ABL plasma are shown in Table I. The lipid level and composition of these samples were similar to those of plasma HDL. However, compared to the apo B-free NL plasma, ABL plasma contained significantly more FC relative to CE. Also, the apo A-! and A-ll levels in the plasma from the

TABLE I Apo ,4-1. A-H and lipid lerelsof apo B-free NL and ABL plasma Subject I was a 27 year old female (Ref. 15) and subject 2 a 37 year old male (Ref. 16) ABL patient. Subjects 3-5 were normolipldemic individuals with plasma C and TG beow 170 mg/dl and 50 mg/dl, respectively. Subject

(mg/dl) A-! A-II I (ABL) 74 13.0 2 (ABL) 69 |8.5 3 t34 3l.I 4 122 30.4 5 145 32.1

C 38 42 49 53 56

TG 5 4 2 4 9

PL 64 67 84 103 104

FC/CE (molar ratio) control incubated 0.583 0.387 0.579 0.400 0.162 0.050 0.298 0.051 0.270 0.047

246 TABLE !! Apo A-! attd lipM composithm o[ Lp(A-! w .4-H) isolated from control and incubated alu~-B-freeplasma Subject

1 2

Control mg/dl (~) '' A-I FC

CE

TG

PL

FC/CE (mol)

31.0 (3,1.8) 37.0

3,3 (3.7) 6.4

!7, I [ 19.2) 26.[]

1.3 (1.5) 1.9

23.3 (26.2) 32.5

0.324

(3(1.4)

(5.3)

(21.4)

(1.6)

(26.71

0.413

Incubated mg/dl (%) A-! FC

CE

TG

PL

FC/CE (mot)

38.0 (34.5) 43.5

4,2 (3.8) 6.(J

25.4 (23.1]) 34.1

1.5 (1.4) 1.4

28.1 (25.5) 33.6

0.278

(31.8)

(4.4)

(25.11)

(l.[l)

(24.6)

0.296

3

93.5

3.6

47.9

1.8

51.7

0.126

I01.5

11.6

65.0

! .4

43.7

U.016

4

(40.7) 75.6 (34. ] )

(1.6) 6.2 (2.8)

(20.9) 48.5 (21.9)

(0.8) 2. I ((].9)

(22.5) 58.9 (26.6)

(].215

(41.7) 8 I. 7 (36,2]

(0.2) 0.8 (0,4)

(26,7) 62,6 (27,8)

(0.6) 1.8 (0.8)

(18.0) 48.4 (21.4)

0.021

0.210

0.037

5

76.8

5.0

40.11

3.8

53.9

(36.3)

(2.4)

(18.9)

( !.81

(25.5t

72.5

1.t

51].6

3,8

43.5

(35.6)

(0.51

(24.8)

( 1.91

(21.4)

;' % was calculated by assuming the sum of apo A-I, A-I! and lipid mass to be i(hF'/~. See Table I fi~r apo A-I! content. A B L patients w e r e low, below the 5th p e r c e n t i l e of a population control [21]. W h e n the plasma samples w e r e incubated at 37°C for 24 h, the C E c o n t e n t s increased in all the samples. B e t w e e n 67% and 79% of the F C in the apo B-free N L plasma samples was c o n v e r t e d to C E during the incubation. In contrast, only 22% a n d 24% of the FC in the A B L plasma samples was esterifled in the same time period. Thus, the F C / C E ratio difference b e t w e e n A B L and apo B-free plasma was f u r t h e r magnified upon incubation.

,:t,o A-L' A - H and lipid cornposition o f H D L subpopulations Tables II and Ill show the composition of Lp(AI w A l l ) and Lp(AI w / o A l l ) isolated from 4°(2 control and 37°C incubated samples of plasma. ( T h e m e a n recovery of apo A-l, A-t1 and lipid after c o l u m n chrom a t o g r a p h y and c o n c e n t r a t i o n was 86%. D a t a in Tables Ii and Ill have b e e n c o r r e c t e d for recovery.) T h e quantities of Lp(Al w A l l ) and Lp(AI w / o A l l ) in the

t h r e e apo B-free N L plasma samples w e r e c o m p a r a b l e to o t h e r N L plasma samples s t u d i e d [1,8,14,23]. T h e two A B L plasma samples had r e d u c e d levels o f Lp(AI w A l l ) and Lp(AI w / o A l l ) , 40% and 70% of normals, respectively. Thus, the r e d u c e d level of H D L in A B L was d u e p r e d o m i n a n t l y to a r e d u c t i o n in L p ( A I w A l l ) . In general, Lp(AI w / o A l l ) c o n t a i n e d relatively m o r e lipid than Lp(AI w A l l ) . C o m p a r e d to the H D L from N L plasma, A B L H D L had h i g h e r relative c o n t e n t s of F C and total !ipid. T h e s e d i f f e r e n c e s w e r e m o r e pron o u n c e d in partieles w i t h o u t a p o A-II which w e r e also particularly e n r i c h e d with phospholipid. U p o n 37°C incubation, in both n o r m a l a n d A B L H D L , the relative a m o u n t of C E increased, while those o f F C and P L d e c r e a s e d , a result indicative o f the L C A 1 reaction. H o w e v e r , w h e n t h e F C and C E contents iv, the control a n d i n c u b a t e d H D L w e r e comp a r e & differences b e t w e e n A B L a n d n o r m a l H D I . , and b e t w e e n H D L particles with a n d w i t h o u t apo A-II w e r e observed. In the t h r e e n o r m a l Lp(AI w A l l ) a n d

TABLE 111

Apo A-i and lilfid compositi¢nt of Lp(A-i w~ o A-~I) ,'.~olatedfrom cotttrol and incubated aim-B-free plasma Subject

Control mg/dl (~):'

Subject

A-I

1

FC

lncubaled mg/dl (r&) CE

TG

PL

43.0 (38.3) 32,0

9.5 (8.5) 7. I

21.8 (19.5) 17.6

1.1 ( 1.01 0.7

36.8 (32,8) 26.5

(38. I )

(8.5)

(21.0)

(0.8)

(31.6)

4

40.1 (48,4) 46.4

1.4 (1.7) 3.3

20.3 (24.5) 19.5

0.4 (0.5) i .0

20.6 (24.9) 25.9

(3.4)

(2(].3)

(1.0)

(27.0)

5

(48.3) 08.2

5.4 (3.5)

32.4 (21.3)

4.8 (3.2)

41.5 (27.2)

2 3

(44.8)

FC/CE (mol)

A-!

0.731

36.1] (39.4) 25,5

5.3 i5,81 4.0

211.3 122,3) 16,11

I. 2 (1.3) i .3

28.5 (31.21 17.5

(39.7)

(6.2)

(24.8)

(2.0)

127.2)

32,0 (57.7) 41).3

l) 6 (I. 1) (L6

13. I (23.6) 20.8

0.1 (0.2) !. 1

9.7 (17.5) 17.4

(50.2)

(I).7) I.I

(26.0) 36.6

(1.4)

(21.7)

(0.7)

(24.6)

4.8 (3.2)

33,8 (22.7)

0.676 0. I 16 0.284 11,280

FC

CE

TG

PL

FC/CE (tool)

72.5 (48.7)

" % was calculated by assuming Ihe sum of al~ A-I and lipid mass to be 1009~.

0.438 I).42I 0.()77 0.048 {1,0511

247

Lp (AI w All)

Lp(AI w / o All), FC contents decreased by an average of 83% and 73%, respectively. In contrast, the FC content of ABL Lp(Al w All) was essentially unchanged and that of Lp(AI w / o All) decreased by only 44% during the incubation. Consequently. the F C / C E ratio in the incubated ABL HDL was five to eight times higher than incubated normal HDL. Although there was no change in the FC content of ABL Lp(Al w All) upon incubation, the CE content in those particles increased by 8.2 mg/dl. Similarly, in the normal Lp(Al w AID, the increase in CE was more than the corresponding decrease in FC. As a result, the total C content of incubated LptN! w All) increased by 6.7-14.1 m g / d l in all fiw plasma samples studied. This increase was also accompanied by an increase of 6-8 mg/dl ape A-! in four of the five plasma samples. No consistent difference between the TEl contents of control and incubated HDL was detected. As expected, the PL content of the HDL particles decreased whenever a decrease in FC was observed. The PL lost from HDL was recovered in the ape A-l-free plasma (data not shown). In the two ABL Lp(Al w All), where changes in FC were minimal, PL increased rather than decreased, and resulted in an absolute increase of total lipid in those particles. In all other cases, the total lipid contents of the HDL particles decreased after incubation due mostly to the loss of phospholipid.

,:v

I.p(At w/o All)

,

J

L

,:

i

Particle size profile of HDL subpop,dations Non-denaturing gPAGE provided an estimate of the size of HDL particles. As depicted from the densitometric scans (Fig. l) control Lp~AI w All) contained a spectrum of particles mostly between 7.0-10.2 nm in apparent hydrated Stokes diameters. The size profiles ~f the two ABL Lp(A! w All) were broader than the ¢hree normal Lp(Ai w All), with relatively more particles larger than 9.2 r~m. Based on gel scanning and integration, incubation reduced the relative amount of ABL particles in the entire 7.0-9.2 nm range from an average of 48% to 31% with a compensatory increase of particles in the 9.2-11.2 nm size range (from 3651%). In normal Lp(A! w All), the shift in particle size was relatively limited, mostly from the 7.0-8.2 nm to the 9.2-11.0 nm range and averaged only 4% of the scan, much less than that observed in incubated ABL Lp(AI w All) where the total lipid content increased during incubation. The particle size profiles of control normal and ABL Lp(AI w / o All) were clearly different. In agreemcnt with previous studies [I,14], the three normal Lp(A! w / o All) contained two distinct size species with apparent mean Stokes diameter of IO.l nm for Lp(AI w / o All)= and 8.5 nm for Lp(AI w / o All) z (Fig. 1) Each of the two size species represented between 30% and 45% of total Lp(AI w / o All). In ABL plasma, Lp(AI v , / o AII)~ comprised nearly 70% of all

I

I

tT.o ~ . 2

i

I

9,s s.~

I 7.0

~7.o ~ . 2

9.2

o.z

7,0

Stokes Diameter (nm) Fig. i, Densit~metric .~ans of 4-30q;- non-denaluring gradient gel

electrophoresis of Lp(A! w All) (left panel) and Lp(AI w/o All) (right panel) isolated from l,h¢ plasma of subjects I-5 (lop-bottom) before ( ~ ) and after (. . . . . . ) incubation. 1 and 2 are ABL subjecls and 3-5 are NL subjects. The bottom scale marks the fimr size intervals we u.~ to describe particle size profiles of Ihesc Iwo HDL subl~)pulalions.

Lp(AI w / o All), while only 12-15% of the particles were in the size region of Lp(Al w / o Ail)2o Furthermore, the apparent mean S~okes diameter of the two ABL Lp(Ai w / o All)~ was considerably larger: 11.6 and l l.2 nm. Despite these differences, both normal and ABL Lp(AI w / o All) contained about 16% of particles smaller than 8.2 nm. After incubaiion, the mean particle size of Lp(A! w / o All) I in both ape B-free NL and ABL plasma samples reduced by 0.4-0.8 nm and the relative amount of particles in the 7.0-8.2 nm region decreased by 6-11%. The effect of incubation on the relative arr.ount of Lp(AI w / o AIl)~ was

248 more variable, ranging from minimal changes (2-3% in subjects one and five) to a near abolishment of that size species in subject three. Discussion

The purpose of this study was t.o determine the extent of lipid and protein interaction between the two major ape-specific HDL subpopulations. We have chosen the mass approach rather than the radioactive tracer approach to monitor the movement of lipid and proteins because: (1) the pre-incubation required tc introduce tracer compounds into the lipoproteins may alter tb: properties of the lipoproteins; and (2) the labeled molecules may not be in complete equilibrium with endogenous molecules. To avoid the masking of HDL lipid mass by the lipid of LpB, ape B-free plasma generated from the plasma of NL subjects was used. To control for potential artefacts in generating ape B-free plasma in vitro, we also studied plasma from two ABL subjects devoid of LpB. Similar results were obtained with both types of ape B-free plasma samples indicating that ape B-free NL plasma generated by dextran sulfate cellulose absorption behaved similarly to ABL plasma. In every case, the CE content of HDL particles with ape A-ll increased after incubation and the increase in CE was more than the corresponding decrease in FC. in a previous study, a significant increase in the CE level of Lp(AI w AID but not Lp(A! w / o AID was also observed when whole plasma was incubated [14]. Increase in the CE content of Lo(AI w All) was accompanied by the increase of 6 - 8 mg/dl of ape A-I in Lp(AI w All) in plasma incubations from four of the five subjects. These observations are consistent with the net transfer of ape A-! and CE from Lp(A! w / o A-ll) to Lp(A! w All). Alternatively, the transfer of an ape A-II molecule from Lp(AI w All) hav:,ng more than one ape A-ll per particle to Lp(A! w / o All) could also increase ape A-ll-associated A-I and CE. The former mechanism implies remodeling of existing Lp(A! w All), while the latter suggests format~op of new Lp(AI w All) in the vascular compartment, Our present data cannot differentiate between these two mechanisms. Since ape A-I is known to exchange between HDL particles [24-27] and ape A-! (but not aoo A-II) dissociates readily from HDL, particularly from Lp(AI w / o All) during ultracentrifugation [28,29], we hypothesize that the observed increase of ape A-I and CE in Lp(A! w All) is due to the transfer of ape A-I and CE from particles without ape A-ll to those with A-ll, but not the transfer of ape A-ll from Lp(AI w AII) to Lp(AI w / o All). If this is true, the present data along with the previous observation that most CETP activity in NL plasma is located in Lo(AI w / o All) [30] would also infer that, of the two major ape-

specific HDL subpopulations, particles without ape A-It are the preferred CE donors and those with ape A-ll are the preferred CE accepters. We have reported that in HepG2 conditioned medium, relatively less ape A-I was found in Lp(AI w All) than in Lp(Al w / o AID (40% vs. 60%) and the molar A - l / A - I I ratio of nascent Lp(A! w AID was about 1.2 [13]. However, in plasma 50-75% ape A-! are usually A-ll associated and the molar A - I / A - I I ratio of mature Lp(AI w AID is significantly higher, between 1.5 and 2.0 [1,2,31]. Although these differences between nascent HepG2 and mature plasma Lp(A! w All) and Lp(AI w / o All) could be related to the cell culture conditions a n d / o r to the faster catabolic rates of Lp(A! w / o AID [9,27], another explanation which is compatible with the present observations is that intravasular remodeling a n d / o r formation of Lp(AI w All) change both the amount and the observed A - I / A - l l ratio of nascent Lp(AI w All) towards those of plasma Lp(AI w AIi). We speculate that as one specie~ of Lp(AI w / o All) of a particular size is converted to another size species in the vascular compartment, some ape A-I and perhaps lipids are released. Since Lp(AI w All) appears to be an avid accepter of CE, the released ape A-I can be transferred to particles with ape A-II, cxpandil~g the surface of existing particles or creating new particles for additional lipids fro~', LpB or Lp(AI w / o AID. Release of ape A-i from HDL particles during incubation of plasma in vitro has been re.rotted [32,33]. The finding that the reduced level of HDL in ABL is due predominiantly to the reduction in Lp(AI w All) also lends support to this speculation. The transfer of apo A-I from Lp(A! w / o All) to Lp(A! w All) may be accelerated by the transfer of LpB lipid into those particles, gradually expands the pool of plasma Lp(AI w AID. Sir:ce there is no LpB in ABL plasma, the ABL Lp(AI w ,~,,, Tool was nat expanded to th ,~ same exte-.' ay in normal plasma. We have previously performed seven whole plasma incubation studies [14,23]. In every case, essentially all Lp(AI w / o All) I and Lp(AI w / o All) 2 were transformed to a single maior subpopulation several nanometers smaller than Lp(Ai w / o All) I, but containing the same number (4) of ape A-! molecules per particle as Lo(AI w / o All) I [23]. in the present study, a similar transformation was observed with Lp(AI w / o All)~, but in two ape B-free NL plasma samples, minimal transformation in Lp(AI w / o All) 2 was observed. This indicates that the transformations of Lp(Al w / o All) i and Lp(AI w / o All) 2 occur independently of each other. Also, while Lp(AI w / o All) 2 can transform in ape B-free plasma, this process is facilitated by LpB. The relative amount of Lo(A! w / o All) particles in the 7.0-8.2 nm region decreased by 6-11% in all five plasma samples after incubation. A compensatory

249

increase in the 9.2-11.2 nm region suggests a precursor-product relationship between Lp(A! w / o All) particles of these two size regions. Although ABL HDL have. been extens~:ely studied [34-37], to our knowledge this is the first report describing the characteristics of A B L Lp(AI w A l l ) and L p ( A I w / o A l l ) . A s in o t h e r A B I . cases, t h e t w o A B L

patients in this study had low plasma levels of apo A-I and A-II. The apo A-I was distributed fairly evenly between particles with and without apo A-II. Consistent with previous observations, the average ABL HDL particles were larger and contained more lipid than normal HDL. Based on gPAGE and densitometric scans, half of the ABL Lp(AI w All) and 70% of ABL L p ( A I w / o A l l ) w e r e l a r g e r t h a n 9.2 n m in a p p a r e n t

hydrated Stokes diameter. Although the gross C / T G / PL ratio of ABL and normal HDL was comparable, the F C / C E ratio was considerably higher in ABL than in normal HDL and is consistent with previous reports of decreased CE content in ABL plasma [34,36]. This may be due to reduced LCAT activity as observed in the present study as well as to the absence of LpB to accept the CE formed in the LCAT reaction. However the observation that 76% of the FC in the apo B-free plasma was converted to CE during incubation whereas only 23% ABL FC was esterified suggests that other differences such as the fa,ty acid composition of lecithin or the reduced lecithin/sphingomyelin ratio in HDL [37,38] may represent additional substrate abnormalities which may contribute to the high plasma FC content and low C esterifieation rate observed in ABL plasma. In conclusion, we have provided evidence for lipid and protein transfer between Lp(Ai w All) and Lp(Ai w / o All) and have characterized these apo-specific HDL subpopulations in the plasma of two patients with ABL. The mechanisms and direction of lipid and protein transfer between these two HDL subpopulations remain to be determined. Acknowledgements The authors wish to thank the Northwest Lipid Research Core Laboratories for performing lipid analyses. This'c~ork was supported by NIH grants HL-30086 (MCC, ACW) and HL 28399 (DRi). References I Cheung. M.C. and Albers, J.J. (1984) J. Biol. Chem. ~ q 1220112209. 2 Ch~ung. M.C., Segrest. J.P., Albers, J.J., Cone, J.T., Brouillette, C.G., Chun8, B.H., Kashyap, M., Glasscock, M.A. a~d AnanIharamaiah, G.M. {1987) J. Lipid Res. 28, 91.)-~29. 3 James, R.W., Hochstrasset, D., Tissot, J.-D., Funk, M., Appel, R., Bahia, F., Pellegrini, C., Muller, A.F. and Pometta, D. (1988) J. Lipid Res. 29, 1557-1571.

4 0 h t a . 'U,, |tattori, S.. Nishiyama, S. and Matsuda. ]. ( l q ~ ) J. Lipid Rcs. 2~). 721-728. 5 James, R.W., Proudfoot. A. and Pometta, D. (1989) Biochim. Biophys. Acta I(H)2. 292-3OI. 6 Puchois, P., Kandoussi, A., Fievet, P., Fourrier, J.L., Bertrand, M., Koren, E. and Fruchart, J.C. (1987) Atherosclerosis 68, 35-40. 7 0 h t a . T.. tlattoz:, S., Nishiyama S., Higashk A. and Matsuda. I. (1989) Metabolism 38, 843-849. 8 Cheung, M.C., Brown, B.G., Wolf, A.C. and Atbers, J.J. (lqgl) J. Lipid Res. 32, 383-394. 9 Attach, R.F., Shepherd, J. and Packard, C.J. (1983) Biochim. Bioph~s. Acta 75I, 175-1~. I0 Fruchart. J.C.. Bard, J.M., and Parra, HJ. (1990) J. Drug Dcv. 3, lfi3- i06. II Bard, J.M, Parra, H.J., Douste-Blazy, P. and Fruchart, J.C. (1990) Metabolism 39. 269-273. 12 Fumeron, F., Brigant, L., Parra, |l.J., Bard, J.M., Fruchart, J.C. and Apfelbaum, M. (lqgl) Am. J. Clin. Nutr. 53, 655-631}. 13 Cheung, M.C., Lure. K.D., Brouillette, C.G. and Bisgaier, C.L. (1989) J. Lipid Res. 30, 1429-1436. 14 Cheung, M.C, and Wolf, A.C. (t989) J. Lipid Res. 30, 499-5049. 15 IIlingworth. D.R., Connor, W.E., Buist, N.R.M,, Jhaveti, B,M., Lin, D.S. and McMurry, M.P. (1979) Metabolism 28, 1152-1160. 16 lllingworth, D.R., Connor, W,E. and Miller, R.G, (1980) Arch. Neur;)), (Chicago) 37, 659-662. 17 Yo[q)yama, S., Hayashi, R., Satani, M. and Yamamoto, A. (lgaq) Arlertosclerosis 5, 613-622. 18 Cheung, M.C. (1986) Methods EnzymoI, 129, 130-145. 19 Warnick, G.R. (1986)Methods En~mol. 1,39. 101-123. 20 AIbets, J.J., Wahl, P., Cabana, V.G., Hazzard, W.R. and Hoover, J.J. (1976) Metabolism 25. 633-644. 21 Cheung, M.C. and Alhers, J.J. (1977) J. Clin. Invest. 60, 43-50. 22 Blanche, PJ., Gong, E.L, Forte, T.M. and Nichols. A.V. (1981) Biochim. Biophys. Acta 665, 408-419. 23 Cheung, M.C., Wolh A.C. and Brunzeil0 J,D. (1990) in Disorders of HDL. Prodeedings of the International Symposium, Stockholm (Carlson, LA,, ed.), pp. 89-97, Smith,Gordon, London. 24 Shepherd. J., Golto, A.M., Taunton, O.D., Caslake, ~l.J. and Farish, E. (1977) Biochim. Biophys_ Acta 489, 486-501. *,2.5 Grow, T.E. and Fried, M. (IO78) J. Biol. Chem. 253, 8034-8041. 26 Cheung, M.C., Wolf, A.C., Foster, D.M. and Knopp, R.H. (1988) Circulation 78, 576A. 27 Rader, D,J. Castro, G., Zech, L.A., Kindt, M.R., Fruchart, J.C. and Brewer, H.BJr. (1991) J. Lipid Res. 3Z 1849-1859. 28 Kunitake, S.T. and Kane, J.P. (1982) J. Lipid Res. Z3, 936-940. 29 Cheung, M.C. and Wolf, A.C. (1988)J. Lipid Res. 29, 15-25. 30 Cheung, M.C., Wolf, A.C., Lure, K.D.. Tollef~n, J.H. and AIb e t , J.J. (1986) J. Lipid Res. 23, 747-753. 31 Cheung. M.C. (1986) in Proceedings of the workshop on lipoprorein heterogeneity (Lippel, K., ed.), pp. 341-349, USDHEW Pub, No. (NIH) 8"-2646. 32 Clay, MA., Newnham, H.H. and Barter, PJ. (1991) Arteriosclerosis rhromb. ! i, 415-422. 33 Musliner, T.A., Long, M.D., Forte, T.M., Nichols, A.V., Gong. E.L,, Blanche, PJ. and Krauss, R.M. (199I) J. Lipid Res. 32, 917-933. 34 Jones, J,W. and Ways, P. (1967) J. Ctin. Invest. 46, ! 151-1161. 35 Scanu, A.M.. Aggerbeck, L,P., Krushi, A,W., Lira, C.T. a~d Kayden. H.J. (1974)J. Clin. Invest. 53, 441)-453. 36 K0stner, G., Holasek, A., Bohlmann, G. and Thilde, H. (1974) Clip. Sc. Moi. Med. 46, 457-468. 37 Deckelbaum. R.J., Eisenberg, S., Oschry, Y., Cooper, M. and Blum, C. 11982) J. Lipid Res. 23, 1274-1282. 38 Banerji, B,P., Subbaiah, P.V., Gregg, R.E. and Bagdad, J.D. (t989) J. Lipid Res. 30, 1907-1916,