Biochimica et Biophysics Acta, 1165 (1992) 61-67 0 1992 Elsevier Science Publishers B.V. All rights reserved
BBALIP
61 0005-2760/92/$05.00
54030
Identification
and characterization of apolipoprotein(AII-E2-AII) complex in human plasma lipoprotein
Minoru Tozuka a, Hiroya Hidaka a, Mami Miyachi ‘, Ken-ichi Furihata ‘, Tsutomu Katsuyama b and Masamitsu Kanai b ’ Central Clinical Laboratories, Shinshu University Hospital, Asahi, Matsumoto (Japan) and ’ Department of Laboratory Medicine, Shinshu Unicersity School of Medicine, Asahi, Matsumoto (Japan) (Received
Key words:
Apolipoprotein(AII-E2-AI11
15
June 1992)
complex; Identification; ApoE subunit; Immunoblotting; (Human)
Enzyme-linked
immunosorbent
assay;
A new apolipoprotein complex designated as the apo(AII-EZAII) complex was identified in the lipoprotein fractions of human plasma with apoE phenotypes containing apoE (E4/E2, E3/E2, and E2/E2). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by an immunoblotting assay using anti-apoE or anti-apoAI1 antibodies, established that the apo(AII-E2-AII) complex, with a molecular weight of 58000, was identical to the complex consisting of apoE and apoAI1, and that it also dissociated following reduction with &mercaptoethanol. This new complex was also demonstrated to be distinct from the apo(E-AII) complex and apoE monomer by isoelectric focusing, in the samples that were not treated with P-mercaptoethanol. In apoE phenotype E3/E2, the apo(AII-E2-AII) complex was primarily included in the high-density lipoprotein (HDL, 1.063 < d < 1.21 g/ml) fraction, but was also observed in a small quantity in the very-low-density lipoprotein (VLDL, d < 1.006 g/ml) fraction. For further characterization, the apo(AII-E2-AID complex was isolated by preparative SDS-PAGE, and no contamination of apo(E-AII) complex and apoE monomer was detected by immunoblotting assay using an anti-apoE antibody. It was confirmed by an enzyme-linked immunosorbent assay (ELBA) system that a molecular ratio between apoAI1 monomer and apoE in the isolated apo(AII-E2-AII) complex was approx. 2, when the apo(E-AII) complex was used as a standard with the ratio of 1: 1. It indicates that the apo(AII-EZAII) complex is formed from two molecules of apoAI1 monomer and one molecule of apoE. The apoE component of the apo(AII-EZAII) complex was identified with apoE by isoelectric focusing of the isolated complex. The new complex, therefore, was designated as the apo(AII-E2-AII) complex to distinguish it from the apo(E-AII) complex.
Introduction Apolipoprotein E (apoE) was first demonstrated in normal human VLDL as a minor constituent [1,2]. A portion of apoE formed the complex with an apoAI1 monomer which was designated as the apo(E-AII) complex, both in the VLDL of type III hyperlipoproteinemic plasma and the HDL-I [3]. This HDL-I was a subfraction of the d = 1.063 to 1.21 ultracentrifugal fraction isolated by preparative block electrophoresis
Correspondence to: M. Tozuka, Central Clinical Laboratories, Shinshu University Hospital, 3-l-l Asahi, Matsumoto 390, Japan. Abbreviations: apoE, apolipoprotein E; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; VLDL, very-low-density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline (pH 7.4); DTNB, 5,5’-dithio-bis(2-nitrobenzoic acid).
and included a predominant apo(E-AI11 complex compared with apoE. The ability of the HDL-I to displace the ‘*‘I-LDL from the LDL receptor was enhanced after its reduction and alkylation [4]. It was assumed that the apo(E-AI11 complex was an inactive form of apoE [4]. ApoE exists as three major isoforms (E4, E3, and E21, of which biosynthesis was controlled by three independent alleles at a single genetic locus [5,6]. As a result, three homozygous (E4/E4, E3/E3, and E2/E2), and three heterozygous (E4/E3, E4/E2, and E3/E2) phenotypes are represented. ApoE and apoE have structural differences from apoE3, by a single substitution of arginine for cysteine, and of cysteine for arginine at residues 158 and 112, respectively [7]. It has been established that apoE plays an important role in the metabolism of cholesterol and triacylglycerol through the LDL receptor and the specific receptor for apoE [8-161.
62
Through the studies on the distribution of apoE in plasma lipoproteins and its binding with other apoproteins, a new apoprotein with a molecular weight of 58000 was identified in the subject with apoE phenotype E4/E2, E3/E2, or E2/E2. Although this apoprotein was formed by apoE and apoAI1, it was apparently distinguished from the apo(E-AII) complex described previously [3]. The present study was undertaken to characterize this apoprotein and to explore its distribution in plasma lipoproteins. materials and Methods Isolation of lipoproteins. VLDL (d < 1.006 g/ml>, LDL (d 1.006-3.063 g/ml>, HDL, (d 1.063-1.125 g/ml>, HDL, (d 1.125-1.21 g/ml), and whole lipoprotein fractions Cd < 1.21 g/ml) were isolated from the plasma of the subjects with various apoE phenotypes by ultracentrifugation as described previously [17]. Lipoproteins were washed by re-centrifugation at the appropriate densities. Immz~~ob~utti~g assay. Lipoproteins or gel-filtration fractions were loaded on 8-16% gradient polyacrylamide gels (containing SDS) and electrophoret~cally separated by the method of Laemmli [IS]. The separated proteins were electrophoretically transferred onto nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany) [19], which were then incubated with a blocking buffer (50 mM Tris-HCl (pH 8.0) containing 5% (w/v) skirn milk) for 1 h at room temperature, followed by washing 3 times with PBS containing 0.1% (v/v) Tween 20 (washing buffer). The membranes were then incubated with a blocking buffer containing antiapoE (goat) or anti-apoAII (rabbits antibody (Daiich Chemical, Tokyo, Japan) for 1 h. After washing 3 times with the washing buffer, the membranes were incubated with horseradish-peroxidase-conjugated anti-goat IgG or anti-rabbit IgG (MBL, Nagoya, Japan). The bands containing apoE or apoAI1 were visualized using 3,3’-diaminobenzidine tetrahydrochloride (Daido, Kumamoto, Japan) and hydrogen peroxide (Wako Pure Chemicals, Osaka, Japan). Determination of apoE isoforms. ApoE isoforms were determined by the method described previously [ZO]. Briefly, 10 ~1 of plasma or lipoprotein fractions were incubated with an equal volume of a neuraminidase (Nakarai Chemical, Kyoto, Japan) solution (5 U/I) for 12-16 h at 37°C foilowed by a treatment with 5% (v/v) P-mercaptoethanol for 10 min at 80°C. The mixture was then delipidated with chloroform/methanol/ether (4: 2 : 1, v/v) followed by a washing with ether. The dried precipitate was dissolved with 20 mM Tris-HCI (pH 8.0) containing 8 M urea, and was applied to isoelectric focusing, which was followed by a transfer of the separated proteins onto a nitrocelluiose membrane.
ApoE isoforms were visualized by an anti-apoE antibody as described above. DTNB treatment. Blood was obtained from two fasted individuals with apoE phenotypes E3/E3 and E3/E2, and was immediately treated with 5,5’-dithiobis(2nitrobenzoic acid) (DTNB) (Boehringer-Mannheim GmbH, Mannheim, Germany) by drawing 2 ml of blood directly into 2 ml of the DTNB reagent (20 mg/ml of DTNB, 2 mg/mi EDTA, 0.07 M phosphate buffer, pH 7.4) [3]. Following separation of the plasma fraction from 3 ml of the treated blood, an additional 75 r;tl of the DTNB reagent was added. The whole lipoprotein fraction was then isolated from I ml of the plasma fraction as described above. An additionai 30 ~1 of the DTNB reagent was added, and the whole lipoprotein fraction was dialyzed against 0.07 M phosphate buffer containing 2 mg/ml DTNB. Immediately following each preparation, the blood, the plasma, or the whole lipoprotein fraction was supplied to an immunoblotting assay for the apolipoprotein complexes containing apoE, Gel ~liration. The whole lipoprotein fraction isolated from the subject with apoE phenotype E3/E2 was deIipidated as described above. Apolipoproteins were dissolved with SO mM Tris-HCl (pH 8.0) containing 6.8 M urea, and applied to the Sephadex G-100 column (1.5 x 100 cm), previously equilibrated with the same buffer. The column fractions were determined by immunoblotting assay as described above. Isolation of apo(E-AH) plex. The whole lipoprotein
and apo(AII-E2-AU)
com-
fraction of apoE phenotype E3/E2 was dissolved in 0.1 M Tris-HCl (pH 6.8) containing 2% SDS, and Ioaded onto preparative SDS-PAGE (8- 16% gradient polyac~lamide gel). After electrophoresis, a slice of gel was removed for immunoblotting assay using an anti-apoE antibody, and the remainder was frozen until the bands containing apoE were visualized by the assay. Apo(AII-E2-AI11 complex, apo(E-AII) complex, and apoE monomer were cut out in comparison with the immunoblot pattern, and were eluted by an incubation with PBS containing 1% SDS. The eluates were dialyzed against 5 mM NH,HCO, (pH 8.0) and lyophilized. Purification ofantibodies. Anti-apoE and anti-apoAII IgG were purified by affinity chromatographies as described previously 1201. Determinations E2-AU) compkx
of apoAI1 /apoE
ratio of the apo(AIl-
~n~rne-linked immunosorbent assay (ELISA) was used to confirm the apoAII/apoE ratio of the apo(AII-E2-AII) complex. Briefly, polystyrene plates (Nunc, Roskilde, Denmark) were coated with an isolated apo(E-AH) complex and apo(AII-E2-AII) complex, which had no cross-contamination by an incubation at 4°C overnight with 50 ~1 per well of complexes diluted from 50- to 6400-fold with 0.1 M sodium carbonate buffer (pH 9.6). After washing twice with the
63
washing buffer (PBS containing 0.1% Tween 201, nonspecific binding sites were blocked with 200 ~1 of 1% skim milk in PBS for 2 h at room temperature. We then washed the wells three times and added 100 ~1 of purified anti-apoE or anti-apoAII IgG diluted 250-fold with PBS, followed by incubation for 2 h at room temperature. After three washings, we added 100 ~1 of horseradish-peroxidase-conjugated anti-goat or antirabbit IgG diluted lOOO-fold in PBS and incubated the mixture for 2 h at room temperature. We washed the plates three times and then added 100 ~1 of 24 mM citrate-phosfate buffer (pH 5.0) containing 400 mg ophenylenediamine dihydrochloride (Nakarai Chemical, Kyoto, Japan) and 400 ~1 .hydrogen peroxide per 1. After a 20 min incubation at room temperature, the reaction was stopped by adding 100 ~1 of 0.4 mol/l sulfuric acid, and the absorbance was measured at 492/650 nm with Behring ELISA Processor II (Behringwerke AG, Marburg, Germany). The results for the apoAI1 monomer and the apoE of the apo(E-AID complex, which consists of one molecule of apoAI1 monomer and apoE, were used as the standards to determine the apoAI1 monomer and the apoE concentrations of the apo(AII-EZAII) complex, respectively. The concentrations of the apoAI1 monomer and the apoE of the apo(E-AII) complex at four different dilutions, which showed an appropriate absorbance, were arbitrarily defined as 1, 2, 4, and 8 M. Results
Whole lipoprotein fractions from various apoE phenotypes were determined by an immunoblotting assay using anti-apoE and anti-apoAI1 antiserum (Fig. 1). An apoprotein with a molecular weight of approx. 58000 was observed in the apoE phenotypes E4/E2, E3/E2, and E2/E2. This apoprotein included both apoE and apoAI1, and was referred to as the apo(AII-E2-AII)
complex which differed from the apo(E-AII) complex detected in the remaining five phenotypes, but not for E4/E4. After treatment of the whole lipoprotein fractions with &mercaptoethanol, the apo(AII-EZAII) complex band as well as the apo(E-AII) complex disappeared, and the apoE monomer apparently increased (Fig. 2). This suggested that the apo(AII-E2-AII) complex was formed with disulfide linkages between apoE and apoAI1. Isoelectric focusing followed by immunoblotting assay using anti-apoE antibody was carried out for the serum samples with various apoE phenotypes with or without a treatment of P-mercaptoethanol (Fig. 31. ApoE phenotypes were clearly demonstrated by a treatment with &mercaptoethanol, whereas at least two extra bands in the phenotype E4/E3 and E3/E3, and several extra bands in the phenotype E4/E2, E3/E2, and E2/E2, were observed without the treatment. No extra band was detected in the phenotype E4/E4. This suggested that these extra bands were identical to the apo(E-AI11 complex and the apo(AIIE2-AI0 complex, respectively. To confirm if the apo(AII-EZAII) complex exists in lipoproteins in vivo or if it is formed during the isolation procedures, the blood sample with apoE phenotype E3/E2 was immediately treated with DTNB, which protect free sulfhydryl groups from a formation of disulfide linkage. The apo(AII-E2-AII) complex was detected regardless of the procedures of preparation of the blood, the plasma, and the lipoprotein samples (Fig. 4). To investigate the distribution of the apo(AII-E2AI11 complex in lipoproteins, VLDL, LDL, HDL,, and HDL, were isolated from the plasma with apoE phenotype E3/E2, and were applied to SDS-PAGE, followed by an immunoblotting assay for apoE (Fig. 5). On the immunoblot pattern, the apo(AII-E2-AII) complex was mainly included in the HDL fraction, but it
A MUM e&i 77.0 ) 66.2 * 45.0 *
17.2 * __ .-
ab-6de
.“m .-
f
All -
(dimer)
abcdef
Fig. 1. Immunoblotting for apoE and apoAI1 in the plasma with various apoE phenotypes. Whole lipoprotein fractions (d < 1.21 g/ml) from the subjects with apoE phenotypes of (a) E4/E4, (b) E4/E3, (c) E3/E3, (d) E4/E2, (e) E3/E2, and (f) E2/E2 were applied to S-16% gradient SDS-polyacrylamide gels, electrophoresed and immunoblotted using an anti-apoE or anti-apoAII(B) antibody followed by the color development as described in Materials and Methods. Positions of molecular-mass standards (kDa) are indicated.
64
C
Ail-Et&AllE-All E-+-+-+-+
a
73’ 7
ab
7
Fig. 2. The effect of treatment with ~-mercaptoethanol on the immunoblot pattern for apoE. Whole lipoprotein fractions (d < 1.21 g/ml) from the subjects with apoE phenotypes of (a) E3/E3, tb) E4/E2, (cl E3/E2, and (d) E2/E2 were incubated with h2.5mM Tris-HCI (pi-I 6.8) containing 2% SDS in the absence t-1 or presence (+ ) of 5% P-mercaptoethanol, and applied to 8- 16% gradient SDS-polyacrylamide gel. After electrophoresis, proteins were transferred onto the nitroceilulose membrane which was then incubated with an anti-apoE antibody fohowed by the color development as described in Materials and Methods.
was also observed in the VLDL fraction as a minor constituent. The whole lipoprotein fractions from the subjects with apoE phenotypes of E4/E2, E3/E2, and E2/E2 were supplied to an immunoblotting assay for apoE followed by a densitometric assay at 485 nm. The various proportions of apoE included in the apo(AIIEZAII) complex were obtained in each phenotype, and no significant difference was observed among the three phenotypes (Table I). The result of a gel-filtration suggested that the linkage between apoE and apoAI1 monomer in the
ab
ab
Fig. 4. The effect of DTNB treatment on the immuaobl(~t patterns. Fresh blood (AI, plasma (BI, and whole hpoprotein fraction (0, obtained from two fasted individuals with apoE phenotypes of E3/E3 (a) and E3/E2 (b), were immediately treated with DTNB as described in Materials and Methods, and were subsequently supplied to SDS-PAGE followed by the immunoblotting assay.
apo(AII-EZ-AII) complex was not dissociated by 6.8 M urea (Fig. 6). The apo(AII-E2-AII) complex, apo(E-AI11 complex, and apoE monomer were isolated by a preparative SDS-PAGE from the whole lipoprotein fraction of the subject with apoE phenotype E3/E2. No contamination among each other was detected by an immunochemical procedure (Fig. 7). A molecular ratio between the apoE and the apoAI1 monomer in the apo(AII-E2-AI11 complex was determined by the ELISA system using isolated apo(E-AH) complex as a standard. The concentrations of the apoE and the apoAI1 monomer in three different dilutions of apo(AII-EZAII) complex were 1.0, 2.3, and 4.8 M for the apoE, and 1.7, 3.6, and 7.9 M for the apoAI1 monomer, respectively. The apoAI1 monomer/ apoE ratios in these diiutions were 1.70, 1.57, and 1.63,
A
w.,
-‘a
_~”
bcdef
-a’bc
.!yp
+ AH-EP- All
def
Fig. 3. Isoelectric focusing for apoE in various apoE phenotypes. Plasma with apoE phenotypes of (a) te) E3/E2, and tff E2/E2 were incubated with neuraminidase (2.5 U/I) followed by treatment with After isoelectric focusing, proteins were transferred onto nitrocellulose membranes which were then bands containing apoE were visualized as described in Materials and
E4/E4, (b) E4/E3, (c) E3/E3, (d) E4/E2, (A) or without (B) 5% P-mercaptoethanot. incubated with an anti-apoE antibody. The Metbods.
65
8
12341234 Fig. 5. Determination of the distribution of the dpo(AII-E2-AII) complex in plasma. (1) VLDL, (2) LDL, (3) HDL,. and (4) HDL, were isolated from the plasma with apoE phenotypes of E3/E3 (A) and E3/E2 (B). and applied to an immunoblotting assay using anti-apoE antibody.
41
40
42
b
C
Q
Fig, 7. Immunoblotting assay of the isolated apo(AII-E2-AID complex, apo(E-AID complex, and apoE. The apo(AII-E2-AI11 complex (b), the apo(E-AID complex (cl, and apoE Cd) isolated from the subject with apoE phenotype of E3/E2 by preparative SDS-PAGE, were applied to an immunoblotting assay for apoE. Whole lipoprotein fraction (a) supplied to this purification was also immunoblotted as a control.
43
44
45
46
50 FRACTION
47
48
60 NUMBER
Fig. 6. Sephadex G-100 column elution profile of the urea soluble apolipoproteins of the whole lipoprotein fractions from the subject with an apoE phenotype of E3/E2. Whole lipoprotein fractions were delipidated with chloroform/methanol/ether (4: 2: 1, v/v) followed by a washing with ether. The dried proteins were solubilized in 50 mM Tris-HCI (pH 8.0) containing 6.8 M urea and ~hromatographed on a Sephadex G-100 column (1.5 X 100 cm). The inset shows an imm~noblotting assay for apoE of gel-filtration fractions, and the number of each lane corresponds to the fraction number.
66 TABLE
I
Discussion
Whole lipoprolein fractions isolated from the subjects with apoE phenotypes of E4/E2, E3/E2, and E2/E2 were supplied to the immunoblotting of apoE. The proportions of apoE included in the apo(AII-E2-AII) complex were determined by a densimetric assay at a wavelength of 485 nm. Pheno-
N “
E
2 7 2
39.7 * 14.0 36.3 * 10.2 36.1 rir 10.0
type
(mean or SD.,
EJ,/E?
E.t/Ed? E2/EZ
E-AH
” Number of subjects; h high-molecular-weight
AII-E2-AI1
Es ”
32.5 + 6.2 25.6+7.8 33.7+X7
3.5 * 0.1 14,7&S.l II.21 1.6
%) 24.5 & 8.3 23.4+5.1 19.2k2.9
bands.
respectively. Taking the molecular weight of the apo(AII-E2-AH) complex into consideration, it was confirmed that the apo(AII-E2-AH) complex was formed with two molecules of apoAI1 monomer and one molecule of apoE. The apo(AIl-~Z-AII~ complex isolated from the subject with apoE phenotype E3/E2 was also applied to isoelectric focusing followed by an immunoblotting assay for apoE (Fig. 8). It was clarified that the apoE subunit of the apo(AII-E2-AI11 complex was comprised of apoE2, and that apoE also formed the apo(E2-AI11 complex and the apoE monomer.
E3L E2’
-k-+-+-
a
b
c
d
Fig. 8. Isoelectric focusing of the isolated apo(AII-EZAII) complex. The isolated apo(AII-EZ-AII) complex (b), the apo(E-AII) complex Cc), and the apoE (d) from the subject with an apoE phenotype of E3/E2 were treated with neuraminidase followed by a treatment with (+ ) or without (- ) P-mercaptoethanol, and were applied ta isoelectric focusing (pH 4-6). The separated proteins were transferred onto the nitrocellulose membrane, and the apoE bands were visualized by an immunoblotting assay as described in Materials and Methods. Whole lipoprotein fraction (al was also applied to isoelectric focusing followed by immunoblotting as a control.
A new apolipoprotein with a molecular weight of 58000 was identified in the subjects with apoE phenotypes E4/E2, E3/E2, and E2/E2 by SDS-PAGE and isoelectric focusing followed by immunoblotting. This apolipoprotein containing both apoE and apoAI1 differs from the apo(E-AI11 complex previously reported (31. The new apoIipoprotein reduced its disulfide linkage by treatment with fi-mercaptoethanol, and was not dissociated by sodium dodecyl sulfate or urea. The molecular ratio of the apoAI1 monomer to the apoE for the new apolipoprotein, which had been postulated to be two because of its molecular weight, was determined by the ELISA method using the isolated apo(EAII) complex as a standard of the ratio 1 : 1. It was confirmed that the new apolipoprotein was a disolfidelinked complex consisting of one molecule of the apoE and two molecules of the apoAl1 monomer. This apolipoprotein was represented as the apo(All-E2-AI11 complex in order to distinguish it from the apo(E-AIIf complex. The treatment with DTNB showed that the apo~AII-E2-AII) complex originalIy existed in lipoproteins in vivo, not formed during isolation procedures in vitro. The apo(AII-EZAlI) complex was one of the predominant styles for existence of apoE in the l-IDL fraction. The apo(AII-EZAII> complex exists in the plasma with three apoE phenotypes, E4/E2, E3/E2, and E2/E2. By contrast, the apo(E-AII) complex is found in the plasma with five of the six common apoE phenotypes except for E4/E4. This means that only the apoE isoform, which contains two cysteine residues [213, is able to form both the apo(AII-E2-AII) complex and the apo(E-AI11 complex with apoAI1 monomers. Actually, the apoE components were identified with apoE for the apo(AII-EZAII) complex and both apoE and apoE for the apo(E-AI11 complex by isoelectric focusing of the isolated complexes from the subject with the apoE phenotype E3/E2. Both apoE and apoE were also identified in the purified apoE monomers. These results suggest that apoE takes three forms of the apo(AII-E2-AII) complex, the apo(E2-AII) complex, and the apoE monomer in Iipoprotein particles. It was aIso demonstrated that a predominant apo(AII-EZ-AII~ complex compared to an apo(E-AII~ compfex was included in phenotype E4JE2 and E2/E2 on SDS-PAGE, followed by immunoblotting assay. This means that larger quantities of apoE exists as the apo(AII-EZAII) complex rather than as the apo(E-AH) complex. The apoE species with a size smaller than apoE and the apo(E-AII) complex doublets were frequently observed on the immunoblot patterns (Figs. 1,2,5). It probably caused by the differences of the glycation and(or) the sialylation of apoE molecules. The high-
67 molecular-weight bands detected by the immunoblotting could be complexes with disulfide bond between apoE and other plasma protein(s), or the products formed by self-association af the apoE because of an absence of apoAI1. It is an obvious fact that the number of cysteine residue is concerned in the formation of these bands, since the several bands are observed in the phenotypes containing the apoE but a single and no band in the phenotype E3/E3 and E4/E4, respectively. We previously reported the assay procedure for the apo(E-AII) complex by the ELISA system [20] and obtained abnormally high values for phenotypes E4/E2, E3/E2, and E2/E2. It was clarified that these high values were caused by the apo(AII-EZAII) complex which is different from the apo(E-AII) complex in the number of detected apoAI1 monomer to one molecule of apoE. It was reported that four apoAI1 isoproteins existed, designated apoAII-1, apoAII-2, apoA_II-3, and apoAII4, with apparent isoelectric points of 5.16, 4,89, 4.58, and 4.31, respectively [22]. However, a major isoform of apoAI1 is apoAII-2. It is unclear if all of the apoAII isoproteins are able to form the apo(E-AII) complex and the apo(AII-E2-AI11 complex. There also remains to be elucidated a possibility of interconversion among the apo(AII-E-AH) complex, the apo(E-AII) complex and apoE in vivo, as well as a physiological or a pathological role of the apo(AII-E2-AII) complex. References 1 Shore, V.G. and Shore, B (1973) ~i~hemist~ 12,502-507. 2 Shelburne, F.A. and Quarfordt, S.H. (19’74)J. Biol. Chem. 249, 1428-1433.
3 Weisgraber, K.H. and Mahley, R.W. (1978) J. Biol. Chem. 253, 6281-6288. 4 Innerarity, T.L., Mahley, R.W., Weisgraher, K.H. and Bersot, T.P. (1978) J. Biol. Chem. 253, 6289-6295. 5 Zannis, V.I. and Breslow, J.L. (1980) J. Biol. them. 255, 17591762. 6 Zannis, V.I. and Breslow, J.L. (1981) Biochemistry 20, 1033-1041. 7 Weisgraber, K.H., Rail, S.C.Jr. and Mahley, R.W. (1981) J. Biol. Chem. 256, 9077-9083. 8 Havel, R.J., Chao, Y-S., Windler, E.E., Kotite, L. and Gun, L.S.S. (1980) Proc. Natl. Acad. Sci. USA ?7,4349-4353. 9 Mahley, R.W. (1982) Med. Clin. North Am. 66, 375-402. 10 Mahley, R.W. and Angelin, B. (1984) Adv. Intern. Med. 29, 385-411. 11 Innerarity, T.L., Pitas, R.E. and Mahley, R.W. (1986) Methods Enzymol. 129, 542-565. 12 Hui, D.Y.. Brecht, W.J., Hall, EA., Friedman, G., Innerarity, T.L. and Mahley, R.W. (1986) J. Biol. Chem. 261, 42.56-4267. 13 Mahley, R.W. (1988) Science 240, 622-630. 14 Beisiegel, U., Weber, W., Havinga, J.R., Ihrke, G., Hui, D.Y., Wernette-Hammond, M.E., Turck, C.W., Knnerarity, T.L. and Mahley, R.W. (1988) Arteriosclerosis 8, 288-297. 15 Beisiegel, U.,‘Weber, W., lhrke, G., Herz, J. and Stanley, K.K. (1989) Nature 341, 162-164. 16 Kowal, R.C., Herz, J., Goldstein, J.L., Esser, V. and Brown, MS. (1989) Proc. Natl. Acad. Sci. USA 86,5810-5814. 17 Havel, R.J., Eder, H.A. and Bragdon, J.H. (19%) J. Clin. Invest. 34, 1345-1353. 18 Laemmli, U.K. (1970) Nature 227, 680-685. 19 Burnette, W.N. (1981) Anal. Biochem. 112, 195-203. 20 Tozuka, M., Yoshida, Y., Tanigami, J., Miyachi, M., Katsuyama, T. and Kanai, M. (1991) Clin. Chem. 37, 164.5-3648. 21 Rail, S.C.Jr., Weisgraber, K.H. and Mahley, R.W. (1982) J. Biol. Chem. 257, 4171-4178. 22 Schmitz, G., Ilsemann, K., Melnik, B. and Assmann, G. (1983) J. Lipid Res. 24, 1021-1029.
BBALIP
61 0005-2760/92/$05.00
54030
Identification
and characterization of apolipoprotein(AII-E2-AII) complex in human plasma lipoprotein
Minoru Tozuka a, Hiroya Hidaka a, Mami Miyachi ‘, Ken-ichi Furihata ‘, Tsutomu Katsuyama b and Masamitsu Kanai b ’ Central Clinical Laboratories, Shinshu University Hospital, Asahi, Matsumoto (Japan) and ’ Department of Laboratory Medicine, Shinshu Unicersity School of Medicine, Asahi, Matsumoto (Japan) (Received
Key words:
Apolipoprotein(AII-E2-AI11
15
June 1992)
complex; Identification; ApoE subunit; Immunoblotting; (Human)
Enzyme-linked
immunosorbent
assay;
A new apolipoprotein complex designated as the apo(AII-EZAII) complex was identified in the lipoprotein fractions of human plasma with apoE phenotypes containing apoE (E4/E2, E3/E2, and E2/E2). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by an immunoblotting assay using anti-apoE or anti-apoAI1 antibodies, established that the apo(AII-E2-AII) complex, with a molecular weight of 58000, was identical to the complex consisting of apoE and apoAI1, and that it also dissociated following reduction with &mercaptoethanol. This new complex was also demonstrated to be distinct from the apo(E-AII) complex and apoE monomer by isoelectric focusing, in the samples that were not treated with P-mercaptoethanol. In apoE phenotype E3/E2, the apo(AII-E2-AII) complex was primarily included in the high-density lipoprotein (HDL, 1.063 < d < 1.21 g/ml) fraction, but was also observed in a small quantity in the very-low-density lipoprotein (VLDL, d < 1.006 g/ml) fraction. For further characterization, the apo(AII-E2-AID complex was isolated by preparative SDS-PAGE, and no contamination of apo(E-AII) complex and apoE monomer was detected by immunoblotting assay using an anti-apoE antibody. It was confirmed by an enzyme-linked immunosorbent assay (ELBA) system that a molecular ratio between apoAI1 monomer and apoE in the isolated apo(AII-E2-AII) complex was approx. 2, when the apo(E-AII) complex was used as a standard with the ratio of 1: 1. It indicates that the apo(AII-EZAII) complex is formed from two molecules of apoAI1 monomer and one molecule of apoE. The apoE component of the apo(AII-EZAII) complex was identified with apoE by isoelectric focusing of the isolated complex. The new complex, therefore, was designated as the apo(AII-E2-AII) complex to distinguish it from the apo(E-AII) complex.
Introduction Apolipoprotein E (apoE) was first demonstrated in normal human VLDL as a minor constituent [1,2]. A portion of apoE formed the complex with an apoAI1 monomer which was designated as the apo(E-AII) complex, both in the VLDL of type III hyperlipoproteinemic plasma and the HDL-I [3]. This HDL-I was a subfraction of the d = 1.063 to 1.21 ultracentrifugal fraction isolated by preparative block electrophoresis
Correspondence to: M. Tozuka, Central Clinical Laboratories, Shinshu University Hospital, 3-l-l Asahi, Matsumoto 390, Japan. Abbreviations: apoE, apolipoprotein E; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; VLDL, very-low-density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline (pH 7.4); DTNB, 5,5’-dithio-bis(2-nitrobenzoic acid).
and included a predominant apo(E-AI11 complex compared with apoE. The ability of the HDL-I to displace the ‘*‘I-LDL from the LDL receptor was enhanced after its reduction and alkylation [4]. It was assumed that the apo(E-AI11 complex was an inactive form of apoE [4]. ApoE exists as three major isoforms (E4, E3, and E21, of which biosynthesis was controlled by three independent alleles at a single genetic locus [5,6]. As a result, three homozygous (E4/E4, E3/E3, and E2/E2), and three heterozygous (E4/E3, E4/E2, and E3/E2) phenotypes are represented. ApoE and apoE have structural differences from apoE3, by a single substitution of arginine for cysteine, and of cysteine for arginine at residues 158 and 112, respectively [7]. It has been established that apoE plays an important role in the metabolism of cholesterol and triacylglycerol through the LDL receptor and the specific receptor for apoE [8-161.
62
Through the studies on the distribution of apoE in plasma lipoproteins and its binding with other apoproteins, a new apoprotein with a molecular weight of 58000 was identified in the subject with apoE phenotype E4/E2, E3/E2, or E2/E2. Although this apoprotein was formed by apoE and apoAI1, it was apparently distinguished from the apo(E-AII) complex described previously [3]. The present study was undertaken to characterize this apoprotein and to explore its distribution in plasma lipoproteins. materials and Methods Isolation of lipoproteins. VLDL (d < 1.006 g/ml>, LDL (d 1.006-3.063 g/ml>, HDL, (d 1.063-1.125 g/ml>, HDL, (d 1.125-1.21 g/ml), and whole lipoprotein fractions Cd < 1.21 g/ml) were isolated from the plasma of the subjects with various apoE phenotypes by ultracentrifugation as described previously [17]. Lipoproteins were washed by re-centrifugation at the appropriate densities. Immz~~ob~utti~g assay. Lipoproteins or gel-filtration fractions were loaded on 8-16% gradient polyacrylamide gels (containing SDS) and electrophoret~cally separated by the method of Laemmli [IS]. The separated proteins were electrophoretically transferred onto nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany) [19], which were then incubated with a blocking buffer (50 mM Tris-HCl (pH 8.0) containing 5% (w/v) skirn milk) for 1 h at room temperature, followed by washing 3 times with PBS containing 0.1% (v/v) Tween 20 (washing buffer). The membranes were then incubated with a blocking buffer containing antiapoE (goat) or anti-apoAII (rabbits antibody (Daiich Chemical, Tokyo, Japan) for 1 h. After washing 3 times with the washing buffer, the membranes were incubated with horseradish-peroxidase-conjugated anti-goat IgG or anti-rabbit IgG (MBL, Nagoya, Japan). The bands containing apoE or apoAI1 were visualized using 3,3’-diaminobenzidine tetrahydrochloride (Daido, Kumamoto, Japan) and hydrogen peroxide (Wako Pure Chemicals, Osaka, Japan). Determination of apoE isoforms. ApoE isoforms were determined by the method described previously [ZO]. Briefly, 10 ~1 of plasma or lipoprotein fractions were incubated with an equal volume of a neuraminidase (Nakarai Chemical, Kyoto, Japan) solution (5 U/I) for 12-16 h at 37°C foilowed by a treatment with 5% (v/v) P-mercaptoethanol for 10 min at 80°C. The mixture was then delipidated with chloroform/methanol/ether (4: 2 : 1, v/v) followed by a washing with ether. The dried precipitate was dissolved with 20 mM Tris-HCI (pH 8.0) containing 8 M urea, and was applied to isoelectric focusing, which was followed by a transfer of the separated proteins onto a nitrocelluiose membrane.
ApoE isoforms were visualized by an anti-apoE antibody as described above. DTNB treatment. Blood was obtained from two fasted individuals with apoE phenotypes E3/E3 and E3/E2, and was immediately treated with 5,5’-dithiobis(2nitrobenzoic acid) (DTNB) (Boehringer-Mannheim GmbH, Mannheim, Germany) by drawing 2 ml of blood directly into 2 ml of the DTNB reagent (20 mg/ml of DTNB, 2 mg/mi EDTA, 0.07 M phosphate buffer, pH 7.4) [3]. Following separation of the plasma fraction from 3 ml of the treated blood, an additional 75 r;tl of the DTNB reagent was added. The whole lipoprotein fraction was then isolated from I ml of the plasma fraction as described above. An additionai 30 ~1 of the DTNB reagent was added, and the whole lipoprotein fraction was dialyzed against 0.07 M phosphate buffer containing 2 mg/ml DTNB. Immediately following each preparation, the blood, the plasma, or the whole lipoprotein fraction was supplied to an immunoblotting assay for the apolipoprotein complexes containing apoE, Gel ~liration. The whole lipoprotein fraction isolated from the subject with apoE phenotype E3/E2 was deIipidated as described above. Apolipoproteins were dissolved with SO mM Tris-HCl (pH 8.0) containing 6.8 M urea, and applied to the Sephadex G-100 column (1.5 x 100 cm), previously equilibrated with the same buffer. The column fractions were determined by immunoblotting assay as described above. Isolation of apo(E-AH) plex. The whole lipoprotein
and apo(AII-E2-AU)
com-
fraction of apoE phenotype E3/E2 was dissolved in 0.1 M Tris-HCl (pH 6.8) containing 2% SDS, and Ioaded onto preparative SDS-PAGE (8- 16% gradient polyac~lamide gel). After electrophoresis, a slice of gel was removed for immunoblotting assay using an anti-apoE antibody, and the remainder was frozen until the bands containing apoE were visualized by the assay. Apo(AII-E2-AI11 complex, apo(E-AII) complex, and apoE monomer were cut out in comparison with the immunoblot pattern, and were eluted by an incubation with PBS containing 1% SDS. The eluates were dialyzed against 5 mM NH,HCO, (pH 8.0) and lyophilized. Purification ofantibodies. Anti-apoE and anti-apoAII IgG were purified by affinity chromatographies as described previously 1201. Determinations E2-AU) compkx
of apoAI1 /apoE
ratio of the apo(AIl-
~n~rne-linked immunosorbent assay (ELISA) was used to confirm the apoAII/apoE ratio of the apo(AII-E2-AII) complex. Briefly, polystyrene plates (Nunc, Roskilde, Denmark) were coated with an isolated apo(E-AH) complex and apo(AII-E2-AII) complex, which had no cross-contamination by an incubation at 4°C overnight with 50 ~1 per well of complexes diluted from 50- to 6400-fold with 0.1 M sodium carbonate buffer (pH 9.6). After washing twice with the
63
washing buffer (PBS containing 0.1% Tween 201, nonspecific binding sites were blocked with 200 ~1 of 1% skim milk in PBS for 2 h at room temperature. We then washed the wells three times and added 100 ~1 of purified anti-apoE or anti-apoAII IgG diluted 250-fold with PBS, followed by incubation for 2 h at room temperature. After three washings, we added 100 ~1 of horseradish-peroxidase-conjugated anti-goat or antirabbit IgG diluted lOOO-fold in PBS and incubated the mixture for 2 h at room temperature. We washed the plates three times and then added 100 ~1 of 24 mM citrate-phosfate buffer (pH 5.0) containing 400 mg ophenylenediamine dihydrochloride (Nakarai Chemical, Kyoto, Japan) and 400 ~1 .hydrogen peroxide per 1. After a 20 min incubation at room temperature, the reaction was stopped by adding 100 ~1 of 0.4 mol/l sulfuric acid, and the absorbance was measured at 492/650 nm with Behring ELISA Processor II (Behringwerke AG, Marburg, Germany). The results for the apoAI1 monomer and the apoE of the apo(E-AID complex, which consists of one molecule of apoAI1 monomer and apoE, were used as the standards to determine the apoAI1 monomer and the apoE concentrations of the apo(AII-EZAII) complex, respectively. The concentrations of the apoAI1 monomer and the apoE of the apo(E-AII) complex at four different dilutions, which showed an appropriate absorbance, were arbitrarily defined as 1, 2, 4, and 8 M. Results
Whole lipoprotein fractions from various apoE phenotypes were determined by an immunoblotting assay using anti-apoE and anti-apoAI1 antiserum (Fig. 1). An apoprotein with a molecular weight of approx. 58000 was observed in the apoE phenotypes E4/E2, E3/E2, and E2/E2. This apoprotein included both apoE and apoAI1, and was referred to as the apo(AII-E2-AII)
complex which differed from the apo(E-AII) complex detected in the remaining five phenotypes, but not for E4/E4. After treatment of the whole lipoprotein fractions with &mercaptoethanol, the apo(AII-EZAII) complex band as well as the apo(E-AII) complex disappeared, and the apoE monomer apparently increased (Fig. 2). This suggested that the apo(AII-E2-AII) complex was formed with disulfide linkages between apoE and apoAI1. Isoelectric focusing followed by immunoblotting assay using anti-apoE antibody was carried out for the serum samples with various apoE phenotypes with or without a treatment of P-mercaptoethanol (Fig. 31. ApoE phenotypes were clearly demonstrated by a treatment with &mercaptoethanol, whereas at least two extra bands in the phenotype E4/E3 and E3/E3, and several extra bands in the phenotype E4/E2, E3/E2, and E2/E2, were observed without the treatment. No extra band was detected in the phenotype E4/E4. This suggested that these extra bands were identical to the apo(E-AI11 complex and the apo(AIIE2-AI0 complex, respectively. To confirm if the apo(AII-EZAII) complex exists in lipoproteins in vivo or if it is formed during the isolation procedures, the blood sample with apoE phenotype E3/E2 was immediately treated with DTNB, which protect free sulfhydryl groups from a formation of disulfide linkage. The apo(AII-E2-AII) complex was detected regardless of the procedures of preparation of the blood, the plasma, and the lipoprotein samples (Fig. 4). To investigate the distribution of the apo(AII-E2AI11 complex in lipoproteins, VLDL, LDL, HDL,, and HDL, were isolated from the plasma with apoE phenotype E3/E2, and were applied to SDS-PAGE, followed by an immunoblotting assay for apoE (Fig. 5). On the immunoblot pattern, the apo(AII-E2-AII) complex was mainly included in the HDL fraction, but it
A MUM e&i 77.0 ) 66.2 * 45.0 *
17.2 * __ .-
ab-6de
.“m .-
f
All -
(dimer)
abcdef
Fig. 1. Immunoblotting for apoE and apoAI1 in the plasma with various apoE phenotypes. Whole lipoprotein fractions (d < 1.21 g/ml) from the subjects with apoE phenotypes of (a) E4/E4, (b) E4/E3, (c) E3/E3, (d) E4/E2, (e) E3/E2, and (f) E2/E2 were applied to S-16% gradient SDS-polyacrylamide gels, electrophoresed and immunoblotted using an anti-apoE or anti-apoAII(B) antibody followed by the color development as described in Materials and Methods. Positions of molecular-mass standards (kDa) are indicated.
64
C
Ail-Et&AllE-All E-+-+-+-+
a
73’ 7
ab
7
Fig. 2. The effect of treatment with ~-mercaptoethanol on the immunoblot pattern for apoE. Whole lipoprotein fractions (d < 1.21 g/ml) from the subjects with apoE phenotypes of (a) E3/E3, tb) E4/E2, (cl E3/E2, and (d) E2/E2 were incubated with h2.5mM Tris-HCI (pi-I 6.8) containing 2% SDS in the absence t-1 or presence (+ ) of 5% P-mercaptoethanol, and applied to 8- 16% gradient SDS-polyacrylamide gel. After electrophoresis, proteins were transferred onto the nitroceilulose membrane which was then incubated with an anti-apoE antibody fohowed by the color development as described in Materials and Methods.
was also observed in the VLDL fraction as a minor constituent. The whole lipoprotein fractions from the subjects with apoE phenotypes of E4/E2, E3/E2, and E2/E2 were supplied to an immunoblotting assay for apoE followed by a densitometric assay at 485 nm. The various proportions of apoE included in the apo(AIIEZAII) complex were obtained in each phenotype, and no significant difference was observed among the three phenotypes (Table I). The result of a gel-filtration suggested that the linkage between apoE and apoAI1 monomer in the
ab
ab
Fig. 4. The effect of DTNB treatment on the immuaobl(~t patterns. Fresh blood (AI, plasma (BI, and whole hpoprotein fraction (0, obtained from two fasted individuals with apoE phenotypes of E3/E3 (a) and E3/E2 (b), were immediately treated with DTNB as described in Materials and Methods, and were subsequently supplied to SDS-PAGE followed by the immunoblotting assay.
apo(AII-EZ-AII) complex was not dissociated by 6.8 M urea (Fig. 6). The apo(AII-E2-AII) complex, apo(E-AI11 complex, and apoE monomer were isolated by a preparative SDS-PAGE from the whole lipoprotein fraction of the subject with apoE phenotype E3/E2. No contamination among each other was detected by an immunochemical procedure (Fig. 7). A molecular ratio between the apoE and the apoAI1 monomer in the apo(AII-E2-AI11 complex was determined by the ELISA system using isolated apo(E-AH) complex as a standard. The concentrations of the apoE and the apoAI1 monomer in three different dilutions of apo(AII-EZAII) complex were 1.0, 2.3, and 4.8 M for the apoE, and 1.7, 3.6, and 7.9 M for the apoAI1 monomer, respectively. The apoAI1 monomer/ apoE ratios in these diiutions were 1.70, 1.57, and 1.63,
A
w.,
-‘a
_~”
bcdef
-a’bc
.!yp
+ AH-EP- All
def
Fig. 3. Isoelectric focusing for apoE in various apoE phenotypes. Plasma with apoE phenotypes of (a) te) E3/E2, and tff E2/E2 were incubated with neuraminidase (2.5 U/I) followed by treatment with After isoelectric focusing, proteins were transferred onto nitrocellulose membranes which were then bands containing apoE were visualized as described in Materials and
E4/E4, (b) E4/E3, (c) E3/E3, (d) E4/E2, (A) or without (B) 5% P-mercaptoethanot. incubated with an anti-apoE antibody. The Metbods.
65
8
12341234 Fig. 5. Determination of the distribution of the dpo(AII-E2-AII) complex in plasma. (1) VLDL, (2) LDL, (3) HDL,. and (4) HDL, were isolated from the plasma with apoE phenotypes of E3/E3 (A) and E3/E2 (B). and applied to an immunoblotting assay using anti-apoE antibody.
41
40
42
b
C
Q
Fig, 7. Immunoblotting assay of the isolated apo(AII-E2-AID complex, apo(E-AID complex, and apoE. The apo(AII-E2-AI11 complex (b), the apo(E-AID complex (cl, and apoE Cd) isolated from the subject with apoE phenotype of E3/E2 by preparative SDS-PAGE, were applied to an immunoblotting assay for apoE. Whole lipoprotein fraction (a) supplied to this purification was also immunoblotted as a control.
43
44
45
46
50 FRACTION
47
48
60 NUMBER
Fig. 6. Sephadex G-100 column elution profile of the urea soluble apolipoproteins of the whole lipoprotein fractions from the subject with an apoE phenotype of E3/E2. Whole lipoprotein fractions were delipidated with chloroform/methanol/ether (4: 2: 1, v/v) followed by a washing with ether. The dried proteins were solubilized in 50 mM Tris-HCI (pH 8.0) containing 6.8 M urea and ~hromatographed on a Sephadex G-100 column (1.5 X 100 cm). The inset shows an imm~noblotting assay for apoE of gel-filtration fractions, and the number of each lane corresponds to the fraction number.
66 TABLE
I
Discussion
Whole lipoprolein fractions isolated from the subjects with apoE phenotypes of E4/E2, E3/E2, and E2/E2 were supplied to the immunoblotting of apoE. The proportions of apoE included in the apo(AII-E2-AII) complex were determined by a densimetric assay at a wavelength of 485 nm. Pheno-
N “
E
2 7 2
39.7 * 14.0 36.3 * 10.2 36.1 rir 10.0
type
(mean or SD.,
EJ,/E?
E.t/Ed? E2/EZ
E-AH
” Number of subjects; h high-molecular-weight
AII-E2-AI1
Es ”
32.5 + 6.2 25.6+7.8 33.7+X7
3.5 * 0.1 14,7&S.l II.21 1.6
%) 24.5 & 8.3 23.4+5.1 19.2k2.9
bands.
respectively. Taking the molecular weight of the apo(AII-E2-AH) complex into consideration, it was confirmed that the apo(AII-E2-AH) complex was formed with two molecules of apoAI1 monomer and one molecule of apoE. The apo(AIl-~Z-AII~ complex isolated from the subject with apoE phenotype E3/E2 was also applied to isoelectric focusing followed by an immunoblotting assay for apoE (Fig. 8). It was clarified that the apoE subunit of the apo(AII-E2-AI11 complex was comprised of apoE2, and that apoE also formed the apo(E2-AI11 complex and the apoE monomer.
E3L E2’
-k-+-+-
a
b
c
d
Fig. 8. Isoelectric focusing of the isolated apo(AII-EZAII) complex. The isolated apo(AII-EZ-AII) complex (b), the apo(E-AII) complex Cc), and the apoE (d) from the subject with an apoE phenotype of E3/E2 were treated with neuraminidase followed by a treatment with (+ ) or without (- ) P-mercaptoethanol, and were applied ta isoelectric focusing (pH 4-6). The separated proteins were transferred onto the nitrocellulose membrane, and the apoE bands were visualized by an immunoblotting assay as described in Materials and Methods. Whole lipoprotein fraction (al was also applied to isoelectric focusing followed by immunoblotting as a control.
A new apolipoprotein with a molecular weight of 58000 was identified in the subjects with apoE phenotypes E4/E2, E3/E2, and E2/E2 by SDS-PAGE and isoelectric focusing followed by immunoblotting. This apolipoprotein containing both apoE and apoAI1 differs from the apo(E-AI11 complex previously reported (31. The new apoIipoprotein reduced its disulfide linkage by treatment with fi-mercaptoethanol, and was not dissociated by sodium dodecyl sulfate or urea. The molecular ratio of the apoAI1 monomer to the apoE for the new apolipoprotein, which had been postulated to be two because of its molecular weight, was determined by the ELISA method using the isolated apo(EAII) complex as a standard of the ratio 1 : 1. It was confirmed that the new apolipoprotein was a disolfidelinked complex consisting of one molecule of the apoE and two molecules of the apoAl1 monomer. This apolipoprotein was represented as the apo(All-E2-AI11 complex in order to distinguish it from the apo(E-AIIf complex. The treatment with DTNB showed that the apo~AII-E2-AII) complex originalIy existed in lipoproteins in vivo, not formed during isolation procedures in vitro. The apo(AII-EZAlI) complex was one of the predominant styles for existence of apoE in the l-IDL fraction. The apo(AII-EZAII> complex exists in the plasma with three apoE phenotypes, E4/E2, E3/E2, and E2/E2. By contrast, the apo(E-AII) complex is found in the plasma with five of the six common apoE phenotypes except for E4/E4. This means that only the apoE isoform, which contains two cysteine residues [213, is able to form both the apo(AII-E2-AII) complex and the apo(E-AI11 complex with apoAI1 monomers. Actually, the apoE components were identified with apoE for the apo(AII-EZAII) complex and both apoE and apoE for the apo(E-AI11 complex by isoelectric focusing of the isolated complexes from the subject with the apoE phenotype E3/E2. Both apoE and apoE were also identified in the purified apoE monomers. These results suggest that apoE takes three forms of the apo(AII-E2-AII) complex, the apo(E2-AII) complex, and the apoE monomer in Iipoprotein particles. It was aIso demonstrated that a predominant apo(AII-EZ-AII~ complex compared to an apo(E-AII~ compfex was included in phenotype E4JE2 and E2/E2 on SDS-PAGE, followed by immunoblotting assay. This means that larger quantities of apoE exists as the apo(AII-EZAII) complex rather than as the apo(E-AH) complex. The apoE species with a size smaller than apoE and the apo(E-AII) complex doublets were frequently observed on the immunoblot patterns (Figs. 1,2,5). It probably caused by the differences of the glycation and(or) the sialylation of apoE molecules. The high-
67 molecular-weight bands detected by the immunoblotting could be complexes with disulfide bond between apoE and other plasma protein(s), or the products formed by self-association af the apoE because of an absence of apoAI1. It is an obvious fact that the number of cysteine residue is concerned in the formation of these bands, since the several bands are observed in the phenotypes containing the apoE but a single and no band in the phenotype E3/E3 and E4/E4, respectively. We previously reported the assay procedure for the apo(E-AII) complex by the ELISA system [20] and obtained abnormally high values for phenotypes E4/E2, E3/E2, and E2/E2. It was clarified that these high values were caused by the apo(AII-EZAII) complex which is different from the apo(E-AII) complex in the number of detected apoAI1 monomer to one molecule of apoE. It was reported that four apoAI1 isoproteins existed, designated apoAII-1, apoAII-2, apoA_II-3, and apoAII4, with apparent isoelectric points of 5.16, 4,89, 4.58, and 4.31, respectively [22]. However, a major isoform of apoAI1 is apoAII-2. It is unclear if all of the apoAII isoproteins are able to form the apo(E-AII) complex and the apo(AII-E2-AI11 complex. There also remains to be elucidated a possibility of interconversion among the apo(AII-E-AH) complex, the apo(E-AII) complex and apoE in vivo, as well as a physiological or a pathological role of the apo(AII-E2-AII) complex. References 1 Shore, V.G. and Shore, B (1973) ~i~hemist~ 12,502-507. 2 Shelburne, F.A. and Quarfordt, S.H. (19’74)J. Biol. Chem. 249, 1428-1433.
3 Weisgraber, K.H. and Mahley, R.W. (1978) J. Biol. Chem. 253, 6281-6288. 4 Innerarity, T.L., Mahley, R.W., Weisgraher, K.H. and Bersot, T.P. (1978) J. Biol. Chem. 253, 6289-6295. 5 Zannis, V.I. and Breslow, J.L. (1980) J. Biol. them. 255, 17591762. 6 Zannis, V.I. and Breslow, J.L. (1981) Biochemistry 20, 1033-1041. 7 Weisgraber, K.H., Rail, S.C.Jr. and Mahley, R.W. (1981) J. Biol. Chem. 256, 9077-9083. 8 Havel, R.J., Chao, Y-S., Windler, E.E., Kotite, L. and Gun, L.S.S. (1980) Proc. Natl. Acad. Sci. USA ?7,4349-4353. 9 Mahley, R.W. (1982) Med. Clin. North Am. 66, 375-402. 10 Mahley, R.W. and Angelin, B. (1984) Adv. Intern. Med. 29, 385-411. 11 Innerarity, T.L., Pitas, R.E. and Mahley, R.W. (1986) Methods Enzymol. 129, 542-565. 12 Hui, D.Y.. Brecht, W.J., Hall, EA., Friedman, G., Innerarity, T.L. and Mahley, R.W. (1986) J. Biol. Chem. 261, 42.56-4267. 13 Mahley, R.W. (1988) Science 240, 622-630. 14 Beisiegel, U., Weber, W., Havinga, J.R., Ihrke, G., Hui, D.Y., Wernette-Hammond, M.E., Turck, C.W., Knnerarity, T.L. and Mahley, R.W. (1988) Arteriosclerosis 8, 288-297. 15 Beisiegel, U.,‘Weber, W., lhrke, G., Herz, J. and Stanley, K.K. (1989) Nature 341, 162-164. 16 Kowal, R.C., Herz, J., Goldstein, J.L., Esser, V. and Brown, MS. (1989) Proc. Natl. Acad. Sci. USA 86,5810-5814. 17 Havel, R.J., Eder, H.A. and Bragdon, J.H. (19%) J. Clin. Invest. 34, 1345-1353. 18 Laemmli, U.K. (1970) Nature 227, 680-685. 19 Burnette, W.N. (1981) Anal. Biochem. 112, 195-203. 20 Tozuka, M., Yoshida, Y., Tanigami, J., Miyachi, M., Katsuyama, T. and Kanai, M. (1991) Clin. Chem. 37, 164.5-3648. 21 Rail, S.C.Jr., Weisgraber, K.H. and Mahley, R.W. (1982) J. Biol. Chem. 257, 4171-4178. 22 Schmitz, G., Ilsemann, K., Melnik, B. and Assmann, G. (1983) J. Lipid Res. 24, 1021-1029.
