- Email: [email protected]
phospholipid microemulsion catalyzed by insect hemolymph lipid transfer particle
289
Biochimica et Biophysics Acta, 1043 (1990) 289-294 Elsevier
BBALIP
53365
Lipid transfer between human plasma low-density lipoprotein and a triolein/phospholipid microemulsion catalyzed by insect hemolymph lipid transfer particle Seiichi Ando ‘, * , Robert 0. Ryan 2 and Shinji Yokoyama Departments
’
of ’ Medicine and ’ Biochemistry, Lipid and Lipoprotein Research Group, University of Alberta, Edmonton (Canada) (Received
Key words:
Lipid transfer
particle;
14 September
1989)
LDL; Lipid microemulsion;
Insect; (M. sextn)
Lipid transfer between human plasma low-density lipoprotein (LDL) and an LDL-size microemulsion of triolein and phosphatidylcholine stabilized with human apolipoprotein A-I was catalyzed by the lipid transfer particle from hemolymph of the tobacco homworm (Manducu sexta). Net transfer of phospholipid and triacylglycerol from the emulsion to LDL was observed and the apparent initial rates of transfer were dependent on the amount of catalyst. Net transfer of phospholipid mass was twice as much as that of triacylglycerol with respect to both the initial rate and the final equilibrium state. The final amount of net transfer of both lipids was dependent upon the initial ratio of LDL : microemulsion present in the incubation mixture up to 1: 1 on the basis of phospholipid. The microemulsion lipid composition was maximally altered from an initial weight ratio of 1.09 f 0.08 (phospholipid/triolein) to 0.90 f 0.03 by this reaction. Further increase of LDL in the incubation caused neither further net transfer nor further change in the lipid composition of the microemulsion. The catalyst neither affected spontaneous transfer of free cholesterol between the emulsion and LDL nor enhanced cholesteryl ester transfer in this reaction system. As a result of the facilitated reaction, LDL gained a significant amount of phospholipid and triacylglycerol causing up to an 8% increase in core lipids and 14% in phospholipid. Some free cholesterol is recovered in the emulsions via spontaneous exchange. Transfer or exchange of apolipoproteins during the course of facilitated lipid transfer did not occur. Introduction In the circulatory fluid of animals, most lipids are complexed with protein in the form of lipoproteins. These particles function as major carriers of lipid and other hydrophobic compounds. The basic structure of lipoproteins consists of a spherical lipid microemulsion that is stabilized and biologically activated by apolipoproteins [l]. The major features of lipid transport by lipoproteins include their secretion into the circulatory system by the cells which synthesize or adsorb lipids, their uptake by cells through specific or non-specific cell-surface recognition proteins, and non-specific lipid exchange between cells and lipoproteins. Some enzymes
* Present address: kaido, Hakodate, Abbreviations: A-I.
Faculty of Fishery Hokkaido, Japan.
LDL, low-density
Science,
lipoprotein;
University
of Hok-
apo A-I, apolipoprotein
Correspondence: S. Yokoyama, Department of Medicine, Lipid and Lipoprotein Research Group, University of Alberta, Edmonton, Alberta, Canada T6G 2S2. 0005-2760/90/$03.50
0 1990 Elsevier Science Publishers
B.V. (Biomedical
such as lipoprotein lipase and 1ecithin:cholesterol acyltransferase are also involved in these transport systems. Lipid exchange between plasma lipoproteins represents an important reaction in the metabolism of lipoproteinassociated lipid. Whereas free cholesterol is spontaneously exchanged between lipoproteins, other lipids such as phospholipids, esterified cholesterol, triacylglycerol and diacylglycerol are exchanged or transferred between lipoprotein particles in the presence of catalysts. Mammalian blood plasma contains catalytic activities for transfer of cholesteryl ester and triacylglycerol, as well as phospholipids, between lipoproteins [2-51. The protein with cholesteryl ester transfer activity was isolated from plasma and amino acid sequence was deduced from cloned c-DNAs for human [6] or rabbit, respectively [7]. There also appears to be a separate phospholipid transfer protein in human plasma [29]. The cholesteryl ester transfer protein catalyzes a redistribution of cholesteryl ester and triacylglycerol between lipoprotein particles without altering in the basic lipid pool size of each lipoprotein [8-lo]. Although the reactions catalyzed by lipid transfer proteins are crucial for establishing the lipid composition of lipoprotein subDivision)
290 classes and may directly affect specific functions of lipoproteins including atherogenesis and antiatherogenesis, the mechanism of the reactions remain unknown. On the other hand, it has also been shown that insect hemolymph contains a lipid transfer catalyst [ll]. The active material identified in this system is a very-highdensity lipoprotein consisting of three apoprotein components and 14% lipid. It catalyzes exchange or net transfer of diacylglycerol between lipophorins, the major insect lipoprotein [12,13], and also between human low-density lipoprotein (LDL) and lipophorin (Ryan and Yokoyama, unpublished data), resulting in significant alteration of the density of these lipoproteins. Thus, insect lipid transfer particle-catalyzed lipid redistribution results in establishment of an entirely new thermodynamic equilibrium of lipid distribution among lipoproteins. At present it is not known if the reaction mechanism of insect lipid transfer particle is distinct from catalysts found in mammalian plasma. By comparison of the catalytic activity and lipid specificity of respective catalysts using a well-defined donor/acceptor system, direct comparisons between catalysts may be possible. Such a comparison in detail may provide the basis for understanding the mechanism of facilitated lipid transfer in both systems. We have used an LDL-sized phospholipid/triolein microemulsion together with human LDL to study the lipid transfer reaction catalyzed by human plasma lipid transfer protein [lo]. When the emulsion surface was stabilized by apolipoproteins facilitated lipid transfer between the emulsion and human LDL occurred [lo]. Each lipid component in the reaction mixture was redistributed in simple equilibrium between these two types of lipid particles without change in the pool size or the lipid mass of the respective particles [lo]. The major net changes observed in this system were, therefore, an increase in free and esterified cholesterol together with a decrease in triacylglycerol in the microemulsion [lo]. In the present report, we applied this reaction system to the lipid transfer reaction catalyzed by insect lipid transfer particle. In contrast to the results obtained with human lipid transfer protein, redistribution and net change in lipid mass is demonstrated for triacylglycerol and phospholipid in the emulsions, indicating that insect lipid transfer particle catalyzes transfer of lipids to establish a different equilibrium from that reached with human lipid transfer protein in this microemulsion-LDL system. Materials
and Methods
Prepurution of LDL, upolipoprotein A-I, lipid microemulsion und lipid transfer particle LDL was isolated from fresh human plasma between a density of 1.006 and 1.063 g/ml, by sequential ultra-
centrifugal flotation. Solution densities were adjusted by addition of solid NaBr. The preparation was dialyzed against 10 mM sodium phosphate (pH 7.4) containing 0.15 M NaCl. Apolipoprotein (apo) A-I was isolated from human plasma high-density lipoprotein and solubilized in the same buffer used for preparation of LDL according to procedures previously described [14]. The lipid microemulsion was prepared from egg phosphatidylcholine (Avanti Polar Lipids, Alabama) and triolein (Sigma, more than 99% grade) in the same buffer according to procedures described elsewhere [15,16], by co-sonication of the lipid mixture at an initial weight ratio of 1 : 1, ultracentrifugation at density 1.006 g/ml and Sepharose CL-4B gel permeation chromatography. In the fractions used for experiments, the final weight ratio of phospholipid/triolein was 1.09 f 0.08, Lipid transfer particle was isolated from hemolymph plasma of the 7-day-old fifth instar larval tobacco hornworms (M. sexta) according to the methods described by Ryan et al. [ll]. The preparation was dialyzed against the above stated buffer immediately before use. Lipid transfer reaction between the microemulsion and LDL Lipid transfer reactions between LDL and the apo A-I-stabilized lipid microemulsion were performed according to procedures described elsewhere [lo]. Briefly, the lipid microemulsion was prestabilized with apo A-I by incubating a mixture of the emulsion and apo A-I (3 : 1 phospholipid/protein weight ratio) at room temperature for 30 min. More than 95% of apolipoprotein binding sites on the emulsion surface are occupied by apo A-I under these conditions [15]. At 37” C, the lipid transfer particle was added to the stabilized emulsion and then LDL was introduced to initiate the reaction. In the final incubation mixture (120 to 300 ~1) the phospholipid concentration in the emulsion was 0.27 to 0.75 g/l and LDL phospholipid was 0.04 to 1.06 g/l. The lipid transfer particle protein concentration ranged from 1.7 to 13 mg/l. After given periods of incubation, the reaction was terminated by adding 2.5 to 3.8 ~01s. of ice-chilled Dextran sulfatecellulose (Kanegafuchi, Osaka, Japan) [ 17,181 suspended in the same buffer (0.7 g/ml). The samples were then vigorously mixed intermittently while kept chilled in an ice bath for 30 min, permitting complete adsorption of LDL. Dextran sulfate-cellulose was precipitated and microemulsion lipids left in the supernatant were determined. Cholesteryl ester transfer from LDL to the microemulsion was also measured using LDL containing [1,2-3H]cholesteryl oleate (Amersham) with the final specific radioactivity of 619 cpm/nmol cholesteryl ester prepared according to Nishikawa et al. [19]. The same assay system was employed and the radioactivity in the supernatant was used as a measure of cholesteryl ester
291
transfer. In some cases, the incubation mixture was chilled to 0°C and then applied to a mini column of dextran sulfate-cellulose (0.8 ml) equilibrated in assay buffer. The emulsion was eluted under starting conditions while the bound LDL fraction was eluted with the buffer containing 0.5 M NaCl [19].
pL ,
100
( pQ o-o
60
TG
i ‘controlo_
I I
1
f
I I
I
F
‘Control’
----Q-o-
I
I
I
1
‘LTP o.syg
LTP o.sJlg
Analytical procedures Free and esterified cholesterol, choline-contai~ng phospholipids and t~acylgly~erol were analyzed using enzymatic assay methods specific for each lipid [20-221 and protein content was measured according to Lowry et al. [23]. Proteins were also analyzed by polyacrylamide gel electrophoresis employing an acrylamide density gradient from 4 to 10% in the presence of 0.1% sodium dodecylsulfate [24]. Proteins were stained with Coomassie brilliant blue R-250.
I 1
O-
0 i
Results
Following incubation with human LDL, changes in the lipid content of apo A-I-stabilized lipid microemulsions as a function of time in the presence and absence of insect lipid transfer particle were determined. Microemulsions are retained in the supematant while LDL precipitates upon addition of dextran sulfate-cellulose to the incubation mixture. In the absence of transfer particle, there was no significant change in the phospholipid or triolein content of the microemulsion throughout the incubation period as previously shown [lo]. By contrast, insect lipid transfer particle caused a decrease of both lipids in the emulsion (Fig. 1). The initial rate of decrease was enhanced as the amount of the transfer catalyst was increased in the incubation mixture, but the reaction reached the same end-point within 30 min, regardless of the amount of catalyst. A faster transfer rate and a larger final net decrease was observed for phospholipid vs. triacylglycerol on a weight basis (Fig. 1). The apparent initial rate of the reaction was calculated by least squares linear regression analysis of the data from the first 33% of the total change for each lipid and plotted against the amount of transfer particle (Fig. 2). There was a clear dependency of transfer rate upon catalyst concentration for both phospholipid and triacylglycerol, though the rate for phospholipid was twice as fast as that for triacylglycerol. The final equilibrium state showed no dependency upon the amount of catalyst for either triacylglycerol or phospholipid (Fig. 1). Changes in free and esterified cholesterol in the supernatant are shown as a function of time in the Fig. 3. There was no significant effect of the transfer particle on spontaneous transfer of free cholesterol from LDL to the ~croemulsion. Furthermore, significant transfer of esterified cholesterol was not catalyzed by the trans-
60
0
-- 20
40
60
INCUBATION TIME, mln Fig. 1. Lipid transfer between the apo A-I-stabilized lipid microemulsion and LDL. Each incubation mixture, 186 PI, contained LDL and microemulsion (77 pg and 81 pg phospholipid, respectively). After incubation with various amounts of insect lipid transfer particle (LTP), 700 gl of dextran sulfate-cellulose suspension, containing 500 mg gel, was added and the lipid concentration in the supernatant determined. Details of the method are described in the text. The amount of LTP is expressed as its protein content.
fer particle in this system. The observed lack of facilitated cholesteryl ester transfer was confirmed by experiments using LDL containing radiolabeled cholesteryl where no significant difference was observed be-
6-
, PL
0
I
1
2
4
AMOUNT
OF LTP,
Apparent initial rates of triacylglycerol (TG). Apparent termined from the data shown regression of the first 33% of
I
I
I
0
2
4
TO
yg protein
transfer of phospholipid (PL) and initial rates of transfer were dein Fig. 1, by least-squares linear the ma~mum decrease of lipid.
292 30 20 IO 0 20 IO 0 20 IO 0
TABLE
I
Changes
in the lipid composition
lipid transfer
0 20
0
20
40
0 60
INCUBATION
0
20
TIME,
40
60
control LTP
PL control
min
Fig. 3. Transfer reaction of free and esterified cholesterol (FC and EC). Data represent the same experiments shown in Fig. 1. Concentration of esterified cholesterol is expressed as weight concentration of cholesterol moiety.
LTP
EC control LTP
tween the reactions with and without transfer particle (data not shown). Table I shows the change in lipid composition of LDL and the lipid emulsion following incubation and separation of the two particles on a mini column of
0
20
40
60
INCUBATION
0
TIME,
20
40
60
min
Fig. 4. The effect of LDL concentration on transfer of phosphohpid (PL) and triacylglycerol (TG) between the microemulsion and LDL. Lipid microemulsions stabilized with apo A-I (179 pg as as phospholipid) were incubated with various amounts of LDL in the presence of 0.8 ng of insect lipid transfer particle (LTP) in the final incubation volume of 240 pl; 2, 23 pg as LDL phospholipid; 3.46 ng; 4, 91 pg; 5, 182 gg; 1, 182 pg as LDL phospholipid without LTP. Transfer was expressed as percent decrease of lipids in the supernatant after LDL was precipitated with dextran sulfate-cellulose.
followrng
Emulsion
LDL
Total
Omin 60 min Omin 60 min
57 +7 53 * 50 +7 38 *
15+1 21*3 14*3 25+2
72+7 74 64+8 63
0 min 60 min Omin 60 min
60 54 55 41
62*0.3 60+2 62kl 71*1**
122&l 114*4 117k2 112*2
TG
IO 0
and microemulsion
Unit, pg/each fraction. Each incubation mixture, 156 ~1, contained LDL and microemulsion stabilized with apo A-I (77 ng and 81 ng as phospholipid, respectively) with 1 pg of insect lipid transfer particle (LTP) as protein. After a 60 min incubation, emulsion and LDL were separated on a column of dextran sulfate-cellulose (0.8 ml) as described in the text. Recovery of the lipids were 71&4% for emulsion and 80+ 1% for LDL. The values represent meanf S.D. for three experiments except for those with * value is the average of two determinations. * * Significant difference (P < 0.01) both from control and 0 time incubation.
20 IO
of LDL
reaction
FC control LTP
kl +1 *2 *1**
**
Omin 60 min Omin 60 min
0 3.7 + 0.7 0 2.4i1.2
62+2 60*2 61+2 57i2
62+2 64*3 61k-2 59*4
Omin 60 min Omin 60 min
0 3.1 +o 0 3.0*0
22+1 16fl 22f2 17*1
22*1 19*1 22f2 2Okl
dextran sulfate-cellulose. The results demonstrate that the transfer particle catalyzed a net decrease of phospholipid and triacylglycerol in the emulsion with a corresponding increase of these lipids in LDL. There was no significant difference between the control and transfer particle-containing incubations with respect to free and esterified cholesterol. Apolipoproteins were analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate for each of these fractions. There was no transfer of apolipoprotein between these two particles during the incubation with or without transfer particle (data not shown). Even after significant transfer or lipid mass had occurred from emulsion to LDL, no apo-A-I was detected in the LDL fraction. The final net transfer of phospholipid and triacylglycerol at the apparent end-point of the reaction depended on the LDL: emulsion ratio in the incubation mixture. Fig. 4 shows the results obtained when the reaction was conducted with a constant concentration of emulsion with different concentrations of LDL, expressed as relative decrease in the microemulsion phos-
293
af
501 0
I I.0
I 2.0
I 3.0
LDL/EMULSlON RATKf IN PHOSPHOLlPD
Fig. 5. Final net transfer of phosphohpid (open symbols) and triacylglycerol (closed symbols) as a function of LDL/emuision ratio in the incubation mixture. Data represent percent decrease of lipids in the supernatant after LDL was precipitated with Dextran sulfate-cellulose at 60 min incubation time. Concentration of apo A-I-stabilized emulsion as phospholipid: circles, 179 ng/240 pl in the presence of 0.8 ng insect lipid transfer particle (LTP); triangles, 81 pg/300 ~1 with 1 ng LTP; diamonds, 81 pg/186 ~1 with 1 ng LTP; squares, 38 ng/112 pl with 1 pg LTP.
pholipid and t~acyl~ycerol. The lipid transferred increased as the concentration of LDL increased relative to the microemulsion until the LDL: emulsion ratio exceeded 1: 1 on a phospholipid weight basis. Further increases in LDL concentration did not result in further net transfer of either lipid or further change in the phospholipid/triacylglycerol ratio in the microemulsion. This is shown in Fig. 5 as the relative decrease in phospholipid and triacylglycerol as a function of the LDL: emulsion ratio. The phospholipid/triacylglycerol ratio in the emulsion, on a weight basis, was maximally altered from an original value of 1.09 f 0.08 to 0.90 + 0.03 following facilitated lipid transfer. Discussion
The lipid transfer particle isolated from hemolymph of the tobacco homworm was studied with respect to lipid transfer between human LDL and an apo A-Istabilized phosphatidylcholine/triacylglycerol microemu&ion, because this reaction system was well-characterized for human plasma lipid transfer protein reaction [lo]. Both phospholipid and t~acyl~ycerol were transferred between these particles resulting in a net decrease of these lipids in the microemulsion to differing extents, thereby causing alterations of microemulsion lipid composition. Although some cholesterol was spontaneously transferred from LDL, the emulsion contained relatively less surface component after this reaction. Consequently, there was a reciprocal increase of phospholipid and triacylglycerol in LDL. Transfer particle did not affect spontaneous transfer of free cholesterol between the particles and did not enhance transfer of cholesteryl
ester between the particles si~ficantly. Althou~ the ~croemulsion was stabilized with apo A-I, there was no transfer of apo A-I from the emulsion to LDL, in spite of a substantial increase in LDL lipid mass catalyzed by the lipid transfer particle. This phenomenon contradicts the reaction catalyzed by human plasma lipid transfer protein under identical conditions. Human plasma lipid transfer protein catalyzed significant net transfer of esterified cholesterol from LDL to the emulsion [lo] with reciprocal net transfer of triacylglycerol from microemulsion to LDL [lo]. There was a slight suppressive effect on the rate of spontaneous transfer of free cholesterol from LDL to the microemulsion [lo]. At the endpoint of the reaction, however, each lipid reached a homogeneous distribution in equilibrium between the two particles and, therefore, the lipid composition of both particles ultimately became the same [lo]. Diacylglycerol is one of the major lipids carried by the insect hemolymph lipoprotein, lipophorin and this makes a structural model of lipophorin controversial [25,26]. Insect lipid transfer particle transfers diacylglycerol between lipophorins [12,13] or .between lipophorin and human LDL (Ryan and Yokoyama, unpublished data). The transfer is bidirectional but the reaction can lead to a final equilibrium in which the distribution of lipid is significantly different from that of the original particles [11,12]. Thus, it appears that major differences exist between the reactions catalyzed by human lipid transfer protein and insect lipid transfer particle with respect to their substrate specificity and the final equilibrium stage of the reaction. Human transfer protein prefers hydrophobic core lipids (triacylglycerol and cholesteryl ester) as substrates and facilitates their homogeneous redistribution among different types of lipid particle. It also seems to catalyze transfer of phospholipids, but it is not clear if this reaction results in net transfer of phospholipids between lipoproteins [29]. Insect lipid transfer particle, on the other hand, seems to prefer glycerolipids, especially diacylglycerolipids including at least diacylglycerol and diacylglycerophospholipids. While human transfer protein does not cause massive changes in the structure of the lipid particles, the insect transfer particle does induce alterations of lipid particle structure. It has not been shown whether the vectorial transfer of lipids demonstrated here is a simple unidirectional transfer reaction, or a result of bidirectional transfer. Further investigation is currently being undertaken to clarify these mechanistic aspects of the reaction. The insect lipid transfer particle is not a single polypeptide, but exists as a very-high-density lipoprotein composed of three apoproteins and a significant amount of lipid [13]. This particle provides an extra lipid pool with which the lipids of potential donor or acceptor
294 particles are exchangeable [13]. If the exchange equilibrium of the lipids is not the same between the transfer particle and various substrate carrier lipid particles, facilitated transfer reaction will lead to establishment of a new lipid distribution equilibrium. On the other hand, mammalian lipid transfer protein is a single polypeptide [6,7] capable of binding small amounts of lipid [27]. Postulated mechanisms have suggested that human transfer protein facilitates lipid exchange when donor and acceptor particles collide [28], or may act as a lipid carrier between the lipid particles [27]. Conceivably, either of these mechanisms could account for the observed redistribution of lipids among substrate carrier lipid particles. Insect lipid transfer particle provides us with a new tool to introduce specific modifications into human lipoproteins. Even though the mass of lipids is substantially increased by this reaction, LDL did not gain additional apolipoprotein. Thus, we are now capable of loading additional di- and triacylglycerol or phospholipid by selecting the proper experimental conditions. This technique will potentially be useful to investigate the effect of these alterations on atherogenicity or antiathoregenicity of plasma lipoproteins. Acknowledgments This work was supported in part by research grants from the Medical Research Council of Canada (SY) and from Alberta Heart and Stroke Foundation (ROR). ROR and SY are both Medical Research Scholars of the Alberta Heritage foundation for Medical Research. References 1 Shen, B.W., Scanu, A.M. and Kbdy, F.J. (1977) Proc. Natl. Acad. Sci. USA. 74, 837-841. 2 Pattnaik, N.M., Montes, A., Hughes, L.B. and Zilversmit, D.B. (1978) Biochim. Biophys. Acta 530, 428-438. 3 Pajaram, C.V., White, G.H. and Barter, P.J. (1980) B&him. Biophys. Acta 617, 383-392. 4 Ihm, J., Elssworth, J.L., Chataing, B. and Harmony, J.A.K. (1982) J. Biol. Chem. 257, 4818-4827.
5 Albers, J.J., Tollefson, J.H., Chen, C-H. and Steinmetz, A. (1984) Arteriosclerosis 4. 49-58. 6 Drayna, D., Jarnagin, AS., Mclean, J., Henzel, W., Kohr, W.. Fielding, C. and Lawn. R. (1987) Nature 327, 632-634. 7 Nagashima, M., McLean, J.W. and Lawn, R.M. (1988) J. Lipid Res. 29, 1643-1649. 8 Ihm, J. and Harmony, J.A.K. (1980) Biochem. Biophys. Res. Commun. 93, 1114-1120. 9 Barter, P.J. and Jones, M.E. (1980) J. Lipid Res. 21, 2388249. 10 Nishikawa, O., Yokoyama, S., Okabe, H. and Yamamoto. A. (1988) J. B&hem. 103, 188-194. 11 Ryan, O.R., Prasad, S.U., Henriksen, E.J., Wells, M.A. and Law, J.H. (1986) J. Biol. Chem. 261, 563-568. 12 Ryan, R.O., Wells, M.A. and Law, J.H. (1986) Biochem. Biophys. Res. Commun. 136, 260-265. K.R., Wells, M.A. and Law, J.H. 13 Ryan, R.O., Senthilathipan, (1988) J. Biol. Chem. 263, 14140-14145. 14 Yokoyama, S., Tajima, S., and Yamamoto, A. (1982) J. Biochem. 91, 1267-1272. A. (1983) J. Biol. Chem. 15 Tajima, S., Yokoyama, S. and Yamamoto, 258, 10073-10082. S., Kawai, Y., Tajima, S. and Yamamoto, A. (1985) J. 16 Yokoyama, Biol. Chem. 260, 16375-16382. 17 Yokoyama, S., Hayashi, R., Kikkawa, T., Tani, N., Takada, S., Hatanaka, K. and Yamamoto, A. (1984) Arteriosclerosis 4, 276281. 18 Yokoyama, S., Hayashi, R., Satani, M. and Yamamoto, A. (1985) Arteriosclerosis 5, 613-622. O., Yokoyama, S. and Yamamoto, A. (1986) J. Bio19 Nishikawa, them. 99, 295-301. 20 Heider, J.G. and Boyett, R.L. (1978) J. Lipid Res. 19, 514-518. 21 Bucolo, G. and David, H. (1973) Clin. Chem. 19, 476-482. 22 Takayama, M., Itoh, S., Nagasaki, T. and Tanimizu, I. (1977) Clin. Chim. Acta 79, 93-98. N.J., Farr, A.L. and Randall, R.J. 23 Lowry, O.H., Rosebrough, (1951) J. Biol. Chem. 193, 265-275. 24 Laemmli, U.K. (1970) Nature 227, 680-685. Insect Physiology Biochem25 Chino, H. (1985) In: Comprehensive istry and Pharmacology (Kerkut, G.A. and Gilbert, L.I., eds.), pp. 115-136, Pergamon Press, Oxford. J.P., Law, J.H. and Wells, M.A. (1988) Annu. Rev. 26 Shapiro, Entomol. 33, 297-318. 27 Swenson, T.L.. Brocia, R.W. and Tall, A.R. (1988) J. Biol. Chem. 263. 5150-5175. B. and Harmony, 28 Ihm, J., Quin, D.M., Busch, S.J., Chataing, J.A.K. (1982) J. Lipid Res. 23, 1328-1341. 29 Tollefson, J.H., Ravnik, S. and Albers, J.J. (1988) J. Lipid Res. 29, 1988.
Biochimica et Biophysics Acta, 1043 (1990) 289-294 Elsevier
BBALIP
53365
Lipid transfer between human plasma low-density lipoprotein and a triolein/phospholipid microemulsion catalyzed by insect hemolymph lipid transfer particle Seiichi Ando ‘, * , Robert 0. Ryan 2 and Shinji Yokoyama Departments
’
of ’ Medicine and ’ Biochemistry, Lipid and Lipoprotein Research Group, University of Alberta, Edmonton (Canada) (Received
Key words:
Lipid transfer
particle;
14 September
1989)
LDL; Lipid microemulsion;
Insect; (M. sextn)
Lipid transfer between human plasma low-density lipoprotein (LDL) and an LDL-size microemulsion of triolein and phosphatidylcholine stabilized with human apolipoprotein A-I was catalyzed by the lipid transfer particle from hemolymph of the tobacco homworm (Manducu sexta). Net transfer of phospholipid and triacylglycerol from the emulsion to LDL was observed and the apparent initial rates of transfer were dependent on the amount of catalyst. Net transfer of phospholipid mass was twice as much as that of triacylglycerol with respect to both the initial rate and the final equilibrium state. The final amount of net transfer of both lipids was dependent upon the initial ratio of LDL : microemulsion present in the incubation mixture up to 1: 1 on the basis of phospholipid. The microemulsion lipid composition was maximally altered from an initial weight ratio of 1.09 f 0.08 (phospholipid/triolein) to 0.90 f 0.03 by this reaction. Further increase of LDL in the incubation caused neither further net transfer nor further change in the lipid composition of the microemulsion. The catalyst neither affected spontaneous transfer of free cholesterol between the emulsion and LDL nor enhanced cholesteryl ester transfer in this reaction system. As a result of the facilitated reaction, LDL gained a significant amount of phospholipid and triacylglycerol causing up to an 8% increase in core lipids and 14% in phospholipid. Some free cholesterol is recovered in the emulsions via spontaneous exchange. Transfer or exchange of apolipoproteins during the course of facilitated lipid transfer did not occur. Introduction In the circulatory fluid of animals, most lipids are complexed with protein in the form of lipoproteins. These particles function as major carriers of lipid and other hydrophobic compounds. The basic structure of lipoproteins consists of a spherical lipid microemulsion that is stabilized and biologically activated by apolipoproteins [l]. The major features of lipid transport by lipoproteins include their secretion into the circulatory system by the cells which synthesize or adsorb lipids, their uptake by cells through specific or non-specific cell-surface recognition proteins, and non-specific lipid exchange between cells and lipoproteins. Some enzymes
* Present address: kaido, Hakodate, Abbreviations: A-I.
Faculty of Fishery Hokkaido, Japan.
LDL, low-density
Science,
lipoprotein;
University
of Hok-
apo A-I, apolipoprotein
Correspondence: S. Yokoyama, Department of Medicine, Lipid and Lipoprotein Research Group, University of Alberta, Edmonton, Alberta, Canada T6G 2S2. 0005-2760/90/$03.50
0 1990 Elsevier Science Publishers
B.V. (Biomedical
such as lipoprotein lipase and 1ecithin:cholesterol acyltransferase are also involved in these transport systems. Lipid exchange between plasma lipoproteins represents an important reaction in the metabolism of lipoproteinassociated lipid. Whereas free cholesterol is spontaneously exchanged between lipoproteins, other lipids such as phospholipids, esterified cholesterol, triacylglycerol and diacylglycerol are exchanged or transferred between lipoprotein particles in the presence of catalysts. Mammalian blood plasma contains catalytic activities for transfer of cholesteryl ester and triacylglycerol, as well as phospholipids, between lipoproteins [2-51. The protein with cholesteryl ester transfer activity was isolated from plasma and amino acid sequence was deduced from cloned c-DNAs for human [6] or rabbit, respectively [7]. There also appears to be a separate phospholipid transfer protein in human plasma [29]. The cholesteryl ester transfer protein catalyzes a redistribution of cholesteryl ester and triacylglycerol between lipoprotein particles without altering in the basic lipid pool size of each lipoprotein [8-lo]. Although the reactions catalyzed by lipid transfer proteins are crucial for establishing the lipid composition of lipoprotein subDivision)
290 classes and may directly affect specific functions of lipoproteins including atherogenesis and antiatherogenesis, the mechanism of the reactions remain unknown. On the other hand, it has also been shown that insect hemolymph contains a lipid transfer catalyst [ll]. The active material identified in this system is a very-highdensity lipoprotein consisting of three apoprotein components and 14% lipid. It catalyzes exchange or net transfer of diacylglycerol between lipophorins, the major insect lipoprotein [12,13], and also between human low-density lipoprotein (LDL) and lipophorin (Ryan and Yokoyama, unpublished data), resulting in significant alteration of the density of these lipoproteins. Thus, insect lipid transfer particle-catalyzed lipid redistribution results in establishment of an entirely new thermodynamic equilibrium of lipid distribution among lipoproteins. At present it is not known if the reaction mechanism of insect lipid transfer particle is distinct from catalysts found in mammalian plasma. By comparison of the catalytic activity and lipid specificity of respective catalysts using a well-defined donor/acceptor system, direct comparisons between catalysts may be possible. Such a comparison in detail may provide the basis for understanding the mechanism of facilitated lipid transfer in both systems. We have used an LDL-sized phospholipid/triolein microemulsion together with human LDL to study the lipid transfer reaction catalyzed by human plasma lipid transfer protein [lo]. When the emulsion surface was stabilized by apolipoproteins facilitated lipid transfer between the emulsion and human LDL occurred [lo]. Each lipid component in the reaction mixture was redistributed in simple equilibrium between these two types of lipid particles without change in the pool size or the lipid mass of the respective particles [lo]. The major net changes observed in this system were, therefore, an increase in free and esterified cholesterol together with a decrease in triacylglycerol in the microemulsion [lo]. In the present report, we applied this reaction system to the lipid transfer reaction catalyzed by insect lipid transfer particle. In contrast to the results obtained with human lipid transfer protein, redistribution and net change in lipid mass is demonstrated for triacylglycerol and phospholipid in the emulsions, indicating that insect lipid transfer particle catalyzes transfer of lipids to establish a different equilibrium from that reached with human lipid transfer protein in this microemulsion-LDL system. Materials
and Methods
Prepurution of LDL, upolipoprotein A-I, lipid microemulsion und lipid transfer particle LDL was isolated from fresh human plasma between a density of 1.006 and 1.063 g/ml, by sequential ultra-
centrifugal flotation. Solution densities were adjusted by addition of solid NaBr. The preparation was dialyzed against 10 mM sodium phosphate (pH 7.4) containing 0.15 M NaCl. Apolipoprotein (apo) A-I was isolated from human plasma high-density lipoprotein and solubilized in the same buffer used for preparation of LDL according to procedures previously described [14]. The lipid microemulsion was prepared from egg phosphatidylcholine (Avanti Polar Lipids, Alabama) and triolein (Sigma, more than 99% grade) in the same buffer according to procedures described elsewhere [15,16], by co-sonication of the lipid mixture at an initial weight ratio of 1 : 1, ultracentrifugation at density 1.006 g/ml and Sepharose CL-4B gel permeation chromatography. In the fractions used for experiments, the final weight ratio of phospholipid/triolein was 1.09 f 0.08, Lipid transfer particle was isolated from hemolymph plasma of the 7-day-old fifth instar larval tobacco hornworms (M. sexta) according to the methods described by Ryan et al. [ll]. The preparation was dialyzed against the above stated buffer immediately before use. Lipid transfer reaction between the microemulsion and LDL Lipid transfer reactions between LDL and the apo A-I-stabilized lipid microemulsion were performed according to procedures described elsewhere [lo]. Briefly, the lipid microemulsion was prestabilized with apo A-I by incubating a mixture of the emulsion and apo A-I (3 : 1 phospholipid/protein weight ratio) at room temperature for 30 min. More than 95% of apolipoprotein binding sites on the emulsion surface are occupied by apo A-I under these conditions [15]. At 37” C, the lipid transfer particle was added to the stabilized emulsion and then LDL was introduced to initiate the reaction. In the final incubation mixture (120 to 300 ~1) the phospholipid concentration in the emulsion was 0.27 to 0.75 g/l and LDL phospholipid was 0.04 to 1.06 g/l. The lipid transfer particle protein concentration ranged from 1.7 to 13 mg/l. After given periods of incubation, the reaction was terminated by adding 2.5 to 3.8 ~01s. of ice-chilled Dextran sulfatecellulose (Kanegafuchi, Osaka, Japan) [ 17,181 suspended in the same buffer (0.7 g/ml). The samples were then vigorously mixed intermittently while kept chilled in an ice bath for 30 min, permitting complete adsorption of LDL. Dextran sulfate-cellulose was precipitated and microemulsion lipids left in the supernatant were determined. Cholesteryl ester transfer from LDL to the microemulsion was also measured using LDL containing [1,2-3H]cholesteryl oleate (Amersham) with the final specific radioactivity of 619 cpm/nmol cholesteryl ester prepared according to Nishikawa et al. [19]. The same assay system was employed and the radioactivity in the supernatant was used as a measure of cholesteryl ester
291
transfer. In some cases, the incubation mixture was chilled to 0°C and then applied to a mini column of dextran sulfate-cellulose (0.8 ml) equilibrated in assay buffer. The emulsion was eluted under starting conditions while the bound LDL fraction was eluted with the buffer containing 0.5 M NaCl [19].
pL ,
100
( pQ o-o
60
TG
i ‘controlo_
I I
1
f
I I
I
F
‘Control’
----Q-o-
I
I
I
1
‘LTP o.syg
LTP o.sJlg
Analytical procedures Free and esterified cholesterol, choline-contai~ng phospholipids and t~acylgly~erol were analyzed using enzymatic assay methods specific for each lipid [20-221 and protein content was measured according to Lowry et al. [23]. Proteins were also analyzed by polyacrylamide gel electrophoresis employing an acrylamide density gradient from 4 to 10% in the presence of 0.1% sodium dodecylsulfate [24]. Proteins were stained with Coomassie brilliant blue R-250.
I 1
O-
0 i
Results
Following incubation with human LDL, changes in the lipid content of apo A-I-stabilized lipid microemulsions as a function of time in the presence and absence of insect lipid transfer particle were determined. Microemulsions are retained in the supematant while LDL precipitates upon addition of dextran sulfate-cellulose to the incubation mixture. In the absence of transfer particle, there was no significant change in the phospholipid or triolein content of the microemulsion throughout the incubation period as previously shown [lo]. By contrast, insect lipid transfer particle caused a decrease of both lipids in the emulsion (Fig. 1). The initial rate of decrease was enhanced as the amount of the transfer catalyst was increased in the incubation mixture, but the reaction reached the same end-point within 30 min, regardless of the amount of catalyst. A faster transfer rate and a larger final net decrease was observed for phospholipid vs. triacylglycerol on a weight basis (Fig. 1). The apparent initial rate of the reaction was calculated by least squares linear regression analysis of the data from the first 33% of the total change for each lipid and plotted against the amount of transfer particle (Fig. 2). There was a clear dependency of transfer rate upon catalyst concentration for both phospholipid and triacylglycerol, though the rate for phospholipid was twice as fast as that for triacylglycerol. The final equilibrium state showed no dependency upon the amount of catalyst for either triacylglycerol or phospholipid (Fig. 1). Changes in free and esterified cholesterol in the supernatant are shown as a function of time in the Fig. 3. There was no significant effect of the transfer particle on spontaneous transfer of free cholesterol from LDL to the ~croemulsion. Furthermore, significant transfer of esterified cholesterol was not catalyzed by the trans-
60
0
-- 20
40
60
INCUBATION TIME, mln Fig. 1. Lipid transfer between the apo A-I-stabilized lipid microemulsion and LDL. Each incubation mixture, 186 PI, contained LDL and microemulsion (77 pg and 81 pg phospholipid, respectively). After incubation with various amounts of insect lipid transfer particle (LTP), 700 gl of dextran sulfate-cellulose suspension, containing 500 mg gel, was added and the lipid concentration in the supernatant determined. Details of the method are described in the text. The amount of LTP is expressed as its protein content.
fer particle in this system. The observed lack of facilitated cholesteryl ester transfer was confirmed by experiments using LDL containing radiolabeled cholesteryl where no significant difference was observed be-
6-
, PL
0
I
1
2
4
AMOUNT
OF LTP,
Apparent initial rates of triacylglycerol (TG). Apparent termined from the data shown regression of the first 33% of
I
I
I
0
2
4
TO
yg protein
transfer of phospholipid (PL) and initial rates of transfer were dein Fig. 1, by least-squares linear the ma~mum decrease of lipid.
292 30 20 IO 0 20 IO 0 20 IO 0
TABLE
I
Changes
in the lipid composition
lipid transfer
0 20
0
20
40
0 60
INCUBATION
0
20
TIME,
40
60
control LTP
PL control
min
Fig. 3. Transfer reaction of free and esterified cholesterol (FC and EC). Data represent the same experiments shown in Fig. 1. Concentration of esterified cholesterol is expressed as weight concentration of cholesterol moiety.
LTP
EC control LTP
tween the reactions with and without transfer particle (data not shown). Table I shows the change in lipid composition of LDL and the lipid emulsion following incubation and separation of the two particles on a mini column of
0
20
40
60
INCUBATION
0
TIME,
20
40
60
min
Fig. 4. The effect of LDL concentration on transfer of phosphohpid (PL) and triacylglycerol (TG) between the microemulsion and LDL. Lipid microemulsions stabilized with apo A-I (179 pg as as phospholipid) were incubated with various amounts of LDL in the presence of 0.8 ng of insect lipid transfer particle (LTP) in the final incubation volume of 240 pl; 2, 23 pg as LDL phospholipid; 3.46 ng; 4, 91 pg; 5, 182 gg; 1, 182 pg as LDL phospholipid without LTP. Transfer was expressed as percent decrease of lipids in the supernatant after LDL was precipitated with dextran sulfate-cellulose.
followrng
Emulsion
LDL
Total
Omin 60 min Omin 60 min
57 +7 53 * 50 +7 38 *
15+1 21*3 14*3 25+2
72+7 74 64+8 63
0 min 60 min Omin 60 min
60 54 55 41
62*0.3 60+2 62kl 71*1**
122&l 114*4 117k2 112*2
TG
IO 0
and microemulsion
Unit, pg/each fraction. Each incubation mixture, 156 ~1, contained LDL and microemulsion stabilized with apo A-I (77 ng and 81 ng as phospholipid, respectively) with 1 pg of insect lipid transfer particle (LTP) as protein. After a 60 min incubation, emulsion and LDL were separated on a column of dextran sulfate-cellulose (0.8 ml) as described in the text. Recovery of the lipids were 71&4% for emulsion and 80+ 1% for LDL. The values represent meanf S.D. for three experiments except for those with * value is the average of two determinations. * * Significant difference (P < 0.01) both from control and 0 time incubation.
20 IO
of LDL
reaction
FC control LTP
kl +1 *2 *1**
**
Omin 60 min Omin 60 min
0 3.7 + 0.7 0 2.4i1.2
62+2 60*2 61+2 57i2
62+2 64*3 61k-2 59*4
Omin 60 min Omin 60 min
0 3.1 +o 0 3.0*0
22+1 16fl 22f2 17*1
22*1 19*1 22f2 2Okl
dextran sulfate-cellulose. The results demonstrate that the transfer particle catalyzed a net decrease of phospholipid and triacylglycerol in the emulsion with a corresponding increase of these lipids in LDL. There was no significant difference between the control and transfer particle-containing incubations with respect to free and esterified cholesterol. Apolipoproteins were analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate for each of these fractions. There was no transfer of apolipoprotein between these two particles during the incubation with or without transfer particle (data not shown). Even after significant transfer or lipid mass had occurred from emulsion to LDL, no apo-A-I was detected in the LDL fraction. The final net transfer of phospholipid and triacylglycerol at the apparent end-point of the reaction depended on the LDL: emulsion ratio in the incubation mixture. Fig. 4 shows the results obtained when the reaction was conducted with a constant concentration of emulsion with different concentrations of LDL, expressed as relative decrease in the microemulsion phos-
293
af
501 0
I I.0
I 2.0
I 3.0
LDL/EMULSlON RATKf IN PHOSPHOLlPD
Fig. 5. Final net transfer of phosphohpid (open symbols) and triacylglycerol (closed symbols) as a function of LDL/emuision ratio in the incubation mixture. Data represent percent decrease of lipids in the supernatant after LDL was precipitated with Dextran sulfate-cellulose at 60 min incubation time. Concentration of apo A-I-stabilized emulsion as phospholipid: circles, 179 ng/240 pl in the presence of 0.8 ng insect lipid transfer particle (LTP); triangles, 81 pg/300 ~1 with 1 ng LTP; diamonds, 81 pg/186 ~1 with 1 ng LTP; squares, 38 ng/112 pl with 1 pg LTP.
pholipid and t~acyl~ycerol. The lipid transferred increased as the concentration of LDL increased relative to the microemulsion until the LDL: emulsion ratio exceeded 1: 1 on a phospholipid weight basis. Further increases in LDL concentration did not result in further net transfer of either lipid or further change in the phospholipid/triacylglycerol ratio in the microemulsion. This is shown in Fig. 5 as the relative decrease in phospholipid and triacylglycerol as a function of the LDL: emulsion ratio. The phospholipid/triacylglycerol ratio in the emulsion, on a weight basis, was maximally altered from an original value of 1.09 f 0.08 to 0.90 + 0.03 following facilitated lipid transfer. Discussion
The lipid transfer particle isolated from hemolymph of the tobacco homworm was studied with respect to lipid transfer between human LDL and an apo A-Istabilized phosphatidylcholine/triacylglycerol microemu&ion, because this reaction system was well-characterized for human plasma lipid transfer protein reaction [lo]. Both phospholipid and t~acyl~ycerol were transferred between these particles resulting in a net decrease of these lipids in the microemulsion to differing extents, thereby causing alterations of microemulsion lipid composition. Although some cholesterol was spontaneously transferred from LDL, the emulsion contained relatively less surface component after this reaction. Consequently, there was a reciprocal increase of phospholipid and triacylglycerol in LDL. Transfer particle did not affect spontaneous transfer of free cholesterol between the particles and did not enhance transfer of cholesteryl
ester between the particles si~ficantly. Althou~ the ~croemulsion was stabilized with apo A-I, there was no transfer of apo A-I from the emulsion to LDL, in spite of a substantial increase in LDL lipid mass catalyzed by the lipid transfer particle. This phenomenon contradicts the reaction catalyzed by human plasma lipid transfer protein under identical conditions. Human plasma lipid transfer protein catalyzed significant net transfer of esterified cholesterol from LDL to the emulsion [lo] with reciprocal net transfer of triacylglycerol from microemulsion to LDL [lo]. There was a slight suppressive effect on the rate of spontaneous transfer of free cholesterol from LDL to the microemulsion [lo]. At the endpoint of the reaction, however, each lipid reached a homogeneous distribution in equilibrium between the two particles and, therefore, the lipid composition of both particles ultimately became the same [lo]. Diacylglycerol is one of the major lipids carried by the insect hemolymph lipoprotein, lipophorin and this makes a structural model of lipophorin controversial [25,26]. Insect lipid transfer particle transfers diacylglycerol between lipophorins [12,13] or .between lipophorin and human LDL (Ryan and Yokoyama, unpublished data). The transfer is bidirectional but the reaction can lead to a final equilibrium in which the distribution of lipid is significantly different from that of the original particles [11,12]. Thus, it appears that major differences exist between the reactions catalyzed by human lipid transfer protein and insect lipid transfer particle with respect to their substrate specificity and the final equilibrium stage of the reaction. Human transfer protein prefers hydrophobic core lipids (triacylglycerol and cholesteryl ester) as substrates and facilitates their homogeneous redistribution among different types of lipid particle. It also seems to catalyze transfer of phospholipids, but it is not clear if this reaction results in net transfer of phospholipids between lipoproteins [29]. Insect lipid transfer particle, on the other hand, seems to prefer glycerolipids, especially diacylglycerolipids including at least diacylglycerol and diacylglycerophospholipids. While human transfer protein does not cause massive changes in the structure of the lipid particles, the insect transfer particle does induce alterations of lipid particle structure. It has not been shown whether the vectorial transfer of lipids demonstrated here is a simple unidirectional transfer reaction, or a result of bidirectional transfer. Further investigation is currently being undertaken to clarify these mechanistic aspects of the reaction. The insect lipid transfer particle is not a single polypeptide, but exists as a very-high-density lipoprotein composed of three apoproteins and a significant amount of lipid [13]. This particle provides an extra lipid pool with which the lipids of potential donor or acceptor
294 particles are exchangeable [13]. If the exchange equilibrium of the lipids is not the same between the transfer particle and various substrate carrier lipid particles, facilitated transfer reaction will lead to establishment of a new lipid distribution equilibrium. On the other hand, mammalian lipid transfer protein is a single polypeptide [6,7] capable of binding small amounts of lipid [27]. Postulated mechanisms have suggested that human transfer protein facilitates lipid exchange when donor and acceptor particles collide [28], or may act as a lipid carrier between the lipid particles [27]. Conceivably, either of these mechanisms could account for the observed redistribution of lipids among substrate carrier lipid particles. Insect lipid transfer particle provides us with a new tool to introduce specific modifications into human lipoproteins. Even though the mass of lipids is substantially increased by this reaction, LDL did not gain additional apolipoprotein. Thus, we are now capable of loading additional di- and triacylglycerol or phospholipid by selecting the proper experimental conditions. This technique will potentially be useful to investigate the effect of these alterations on atherogenicity or antiathoregenicity of plasma lipoproteins. Acknowledgments This work was supported in part by research grants from the Medical Research Council of Canada (SY) and from Alberta Heart and Stroke Foundation (ROR). ROR and SY are both Medical Research Scholars of the Alberta Heritage foundation for Medical Research. References 1 Shen, B.W., Scanu, A.M. and Kbdy, F.J. (1977) Proc. Natl. Acad. Sci. USA. 74, 837-841. 2 Pattnaik, N.M., Montes, A., Hughes, L.B. and Zilversmit, D.B. (1978) Biochim. Biophys. Acta 530, 428-438. 3 Pajaram, C.V., White, G.H. and Barter, P.J. (1980) B&him. Biophys. Acta 617, 383-392. 4 Ihm, J., Elssworth, J.L., Chataing, B. and Harmony, J.A.K. (1982) J. Biol. Chem. 257, 4818-4827.
5 Albers, J.J., Tollefson, J.H., Chen, C-H. and Steinmetz, A. (1984) Arteriosclerosis 4. 49-58. 6 Drayna, D., Jarnagin, AS., Mclean, J., Henzel, W., Kohr, W.. Fielding, C. and Lawn. R. (1987) Nature 327, 632-634. 7 Nagashima, M., McLean, J.W. and Lawn, R.M. (1988) J. Lipid Res. 29, 1643-1649. 8 Ihm, J. and Harmony, J.A.K. (1980) Biochem. Biophys. Res. Commun. 93, 1114-1120. 9 Barter, P.J. and Jones, M.E. (1980) J. Lipid Res. 21, 2388249. 10 Nishikawa, O., Yokoyama, S., Okabe, H. and Yamamoto. A. (1988) J. B&hem. 103, 188-194. 11 Ryan, O.R., Prasad, S.U., Henriksen, E.J., Wells, M.A. and Law, J.H. (1986) J. Biol. Chem. 261, 563-568. 12 Ryan, R.O., Wells, M.A. and Law, J.H. (1986) Biochem. Biophys. Res. Commun. 136, 260-265. K.R., Wells, M.A. and Law, J.H. 13 Ryan, R.O., Senthilathipan, (1988) J. Biol. Chem. 263, 14140-14145. 14 Yokoyama, S., Tajima, S., and Yamamoto, A. (1982) J. Biochem. 91, 1267-1272. A. (1983) J. Biol. Chem. 15 Tajima, S., Yokoyama, S. and Yamamoto, 258, 10073-10082. S., Kawai, Y., Tajima, S. and Yamamoto, A. (1985) J. 16 Yokoyama, Biol. Chem. 260, 16375-16382. 17 Yokoyama, S., Hayashi, R., Kikkawa, T., Tani, N., Takada, S., Hatanaka, K. and Yamamoto, A. (1984) Arteriosclerosis 4, 276281. 18 Yokoyama, S., Hayashi, R., Satani, M. and Yamamoto, A. (1985) Arteriosclerosis 5, 613-622. O., Yokoyama, S. and Yamamoto, A. (1986) J. Bio19 Nishikawa, them. 99, 295-301. 20 Heider, J.G. and Boyett, R.L. (1978) J. Lipid Res. 19, 514-518. 21 Bucolo, G. and David, H. (1973) Clin. Chem. 19, 476-482. 22 Takayama, M., Itoh, S., Nagasaki, T. and Tanimizu, I. (1977) Clin. Chim. Acta 79, 93-98. N.J., Farr, A.L. and Randall, R.J. 23 Lowry, O.H., Rosebrough, (1951) J. Biol. Chem. 193, 265-275. 24 Laemmli, U.K. (1970) Nature 227, 680-685. Insect Physiology Biochem25 Chino, H. (1985) In: Comprehensive istry and Pharmacology (Kerkut, G.A. and Gilbert, L.I., eds.), pp. 115-136, Pergamon Press, Oxford. J.P., Law, J.H. and Wells, M.A. (1988) Annu. Rev. 26 Shapiro, Entomol. 33, 297-318. 27 Swenson, T.L.. Brocia, R.W. and Tall, A.R. (1988) J. Biol. Chem. 263. 5150-5175. B. and Harmony, 28 Ihm, J., Quin, D.M., Busch, S.J., Chataing, J.A.K. (1982) J. Lipid Res. 23, 1328-1341. 29 Tollefson, J.H., Ravnik, S. and Albers, J.J. (1988) J. Lipid Res. 29, 1988.