Are fatty acid-binding proteins involved in fatty acid transfer?

Are fatty acid-binding proteins involved in fatty acid transfer?

Biochimica et Biophysica Acta, 1002 (1989) 8-13 Elsevier 8 BBA 53047 Are fatty acid-binding proteins involved in fatty acid transfer? Roger A. Peet...

894KB Sizes 0 Downloads 21 Views

Biochimica et Biophysica Acta, 1002 (1989) 8-13 Elsevier

8

BBA 53047

Are fatty acid-binding proteins involved in fatty acid transfer? Roger A. Peeters t, Jacques H. Veerkamp and Rudy A. Demel 2 s Department of Biochemistry, University of Nijmegen and z Department of Biochemistry, Universityof Utrecht (The Netherlands)

(Received 30 June 1988) (Revised manuscript received 18 November 1988)

Key words: Fatty acid binding protein; Fatty acid transfer; Fatty acid oxidation; Monolayer; (Human): (Rat)

The pesdble function of fatty ccid-binding protein (FABP) to act as a fatty acid carrier protein was investigated in model systems with ~ to three mpeets. (I) does FABP release fatty acids from membranes? (2) does it facilitate fatty ~ transport in an aqueous environment? (3) ave FABP.bouml fatty acids released for use by mitoehondria? FABI~ ~ blml elek acid from lipmmaes and m i t o e ~ a l membranes with a ratio of I mot per tool protein. Oleic acid was withdrawn from negative, neutral or chollesterol-contalning monolayers by FABP with rates up to 10~/min. Only about ~$ of FABP penetrated into the mono]ayer. Spontaneous transfer of oleic acid between mitoehondria and v ~ or lipmma~ e e e t m ~ so rapidly that an effect of FABP was not detectable. When the mitoehondria were separated from the vesicles in an equilibrium d~ysis ceil, a stimulating effect of FABP on fatty acid transfer could be demonserate$. 10jected FABP increased also mun~der of oleic acid between two separate monolayers. FABP-bound fatty acid was well oxidized by rat liver mitochondria. The results indicate that the FABP-fatty acid complex may function as an intermediate in the transfer of fatty acids between membranes. No functional differences were detected between heart and liver FABPs in this respect.

InUmktetiee Fatty acids play an important role in cell metabolism. They are not only used for mitochondrial and peromsomal p-oxidation, but also for anabolic processes, such as triacylglycerol and phospholipid synthesis. A family of fatty acid-binding proteins (FABPs)is thought to be involved in the intracellular transport of these hydrophobic compounds [1-4]. FABPs are present in several mammalian cell types [5,6], but are characterized best in liver and heart of man [7-9], pig [7,8], rat [2,7,8,10] and cattle [11]. There are also data about the primary structure of some FABPs [12-14], but little evidence is available about their function as fatty acid carriers. Ockner and co-workers [2,4] proposed that FABPs may markedly enhance translational diffusion of fatty acid. FABPs may, however, also facilitate the desorption of fatty acid from the cytosolic side of the plasma

Abbreviations: FABP(s), fatty acid-binding protein(s); PC, phosphatidylcholine; PS, phosphatidyiserine; PE, phosphatidylethanolamine. C ~ : J.H. Veerkamp, Department of Biochemistry, University of Nijmegen, p.o. Box 9101, 6500 HB Nijmegen, The Netherlands.

membrane [2,15]. Experimental work [2,16,17] and calculations [18] showed that almost all cellular nonesterified fatty acids are bound to FABP. Studies on the effect of sp~ific proteins on fatty acid transport are complicated by the relatively high solubility and the rapid spontaneous transfer of fatty acids in in vitro systems [15,19-21]. These conditions do not impede studies on binding proteins of phospholipids [22] or tocopherol [23]. Uptake of fatty acids from vesicles or liposomes was demonstrated for rat and bovine liver and human heart FABP and albumin either by fluorescence [24,25], radiochemical procedures [7,26] or t3C-NMR [27]. In contrast to these FABPs, bovine heart FABP could only donate fatty acid to vesicles [24,25]. Catala and Avanzati [28] reported that rat liver FABP could significantly enhance oleic acid transfer from microsomes or mitochondria to unUamellar egg phosphatidylcholine (PC) liposomes. Their observation of a low spontaneous transfer did not appear probable to use because of the high exchange rates of fatty acids between vesicles [20,21], and we could not establish their results in preliminary experiments. Others [29] excluded collision transfer between liposomes and microsomes by using an equilibrium dialysis cell and demonstrated fatty acid transfer by rat liver FABP by an increased acyl-CoA production.

000~2"/60/89,~03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

9 In this study, we describe the capacity of different FABPs to release fatty acid from mitochondrial membranes and several phospholipid monolayers. The rapid spontaneous transfer of fatty acid between mitochondria and vesicles or liposomes in one suspension made it impossible to establish an effect of FABP. We could, however, demonstrate an enhancement of fatty acid transport by FABP between those membranes in an equilibrium dialysis cell and between separate phospholipid monolayers. The release of FABP-bound fatty acids to mitochondria was investigated by assay of fatty acid oxidation.

Mater|als and Methods

Mitochondria (100-250 pg protein), loaded with [14C]oleic acid, were incubated with 7-30 pg FABP for 15 min at 37°C in 150 #1 buffer AoThe mixture was then centrifuged for 10 min at 1 0 0 0 0 × g and the amount of FABP-bound fatty acid was calculated from the radioactivity in the supernatant. Mitochondria (5 mg protein), loaded with [14C]oleic acid, were incubated with small unilamellar vesicles or liposomes (25-200 nmol lipid phosphorus) in 500 ~1 buffer A for 5 or 10 min at 37 ° C. A volume of 300 ~1 0.2 M sodium acetate (pH 5.0) was added and the mixture was centrifuged for 5 min 10000 × g at 4°C. The amount of [14C]oleic acid transferred was determined in the supernatant.

Transfer of oleic ~cid between separate compartments. Materials. [1-14C]Oleic acid and [1-14C]palmitic acid were obtained from Amersham International, Little Chalford, U.K.; egg PC, PE and lyso-PC and bovine PS from Lipid Products, South Nutfield, U.K.; cholesterol from Merck, Darmstadt, F.R.G. All other reagents were of analytical gra,:ie. Loading of [14C]oleic acid on mitochondria. Porcine liver or heart tissue (50 g) was homogenized (10~; w / v ) in ice-cold buffer A (0.25 M sucrose/2 mM EDTA/10 mM Tris-HCl (pH 7.4)). The homogenate was centrifuged for 5 rain at 600 × g. The supernatant was then centrifuged for 10 min at 10000 x g, and the pellet was resuspended in 50 ml of the same buffer and again centrifuged for 10 min at 10000 × g. This pellet was resuspended in 35 ml buffer A and incubated for 10 rain at 37°C with a solution of 50 pmol [1-~4C]oleic acid (430 dpm/nmol) in 2 ml ethylene glycol. The suspension was centrifuged for 15 min at 12000 ×g. The pellet was washed twice with 30 ml buffer A, resuspended in 45 ml buffer A and stored in portions of 1.5 ml at - 2 0 ° C . Preparation of small unilamellar vesicles. A suspension of unilamellar vesicles was prepared by sonicating (55 W) of 8.0 pmol lipid (PC/PS, 10 : 1) for five 3-mincycles at 0 °C in 5 ml buffer A with the microtip probe of a Branson sonifier. Vesicles were stored at 4 ° C under nitrogen. Preparation of large unilamellar vesicles. 8.0 #mol lipid (PC/PS, 10:1) was rotary-evaporized and 5 ml diethyl ether (washed with water) and 1 ml buffer A were added to the lipid film. After 20 min sonication in a waterbath at 0°C, vesicles were formed by rotaryevaporation of diethyi ether for 30 min at room temperature. Vesicles were stored at 4 ° C under nitrogen. Preparation of multilamellar liposomes. Lipcsomes were prepared by first dissolving a mixture of 8.0 pmol P C / P S (10: 1) in 2 ml chloroform, evaporating the organic solvent and resuspending the lipid residue in 5 ml buffer A under gentie agitation. Liposomes were stored at 4 ° C under nitrogen.

Transfer assays with oleic acid-loaded mitocho,,ldria.

The two compartments (1.0 ml each) of perspex equilibrium dialysis cells were separated by a polycarbonate membrane (Nucleopore, membrane diameter 13 mm, pore diameter 0.2 pm). Before use, the membranes were put into water for 10 min and the chamber walls were silanized with 0.01% acrylsilan in ethanol. Cells were incubated at 20 °C under rotation. After 2 and 4 h, 200 pl aliquots were taken from compartment 2 to determine the concentration of oleic acid. Percentages of oleic acid transferred were calculated from these values and corrected for the amount of oleic acid bound to FABP, assuming maximal binding. Oieic acid withdrawal from monolayers. Monolayers were formed at the air/water interface in a Teflon trough (5.9 × 5.4 × 0.5 cm wide). A 1.5 x 1.5 cm extended corner with a hole of 0.8 cm was made for injection of the protein underneath the monolayer. The trough was filled with 19 ml 10 mM Tris-HC! (pH 6.9). The subphase was stirred with a magnetic bar. Phospholipid monolayer films, containing 0.8 nmol [114C]oleic acid, were spread from a chloroform/methanol (9 : 1) solution with a capillary pipette untill the desired interracial pressure (20 m N / m ) was achieved. The phospholipid composition is given in the tables and figures. The withdrawal of oleic acid from the monolayer was measured with a gas flow detector (Berthold LB 6280) determining the surface radioactivity. The oleic acid concentration of the subphase was determined from the radioactivity in 1 ml samples. The off-rate was determined from the slope of the exponential part of the oleic acid withdrawal curve (5-15 min) by measuring the surface radioactivity and the concentrations of oleic acid present in the subphase at the beginning and the end 30 min) of the experiments.

Transfer of oleic acid between two separate monolayers. The surface area of the trough was divided in two equal compartments by a Teflon bar. On one side a P C / P S / [1-14C]oleic acid monolayer ( 8 : 1 : 1 , 8.0 nmol) was spread and on the other side a PC/PS monolayer (8 : 2, 8.0 nmol). 4 pg FABP was injected into the subphase of 19 ml 10 mM Tris-HCl (pH 6.9). After the experiments,

10 both monolayers were collected and the amount of [l-t4C}oleic acid in each monolayer was determined. Other procedures. FABPs were isolated from pig heart and from human, pig and rat liver as described previously [7,81. The protein concentration of FABPs was determined from amino acid analysis. Mitochondrial protein was measured by the procedure of Lowry et al. [30] using bovine serum albumin as standard. The phospborus content of the vesicles or liposomes was determined by the method of Bartlett [31]. Fatty acid oxidation experiments were performed as described by Veerkamp et al. [32]. FABP was labeled with [t4C]formaldehyde by the method of Okada and Spire [33]. The product was checked after gel electrophoresis by autoradiography. The fatty acid.binding capacity of pure, delipidated FABP preparations was assayed as described previously [7,8].

TABLE I Migration of [1. i ¢C]oleic acid from phospholipid monolayers to FA BP

Results

monolayer did not exceed the sum of the effects of FABP and vesicles. The penetration of FABP into the monolayers was investigated with [t4C]formaldehyde-labeled pig liver FABP for a PC/P$/oleic acid (8 : 1 : 1) and a PC/lysoPC/oleic acid (8:1:1) monolayer. After 15 min, only about 5~$ of the injected FABP was found in the monolayers.

Transfer of fatty acid from mitochondria to FABP When [l-t4C]oleic acid-loaded pig heart or liver mitochondria were incubated with FABPs, pig liver and heart FABP were able to bind 56 + 18 (n = 29) and 17 :!: $ (n - 11) nmol oleic acid/rag FABP, respectively. These values indicate a binding stoichiometry of 0.78 and 0.26 reel fatty acid/reel protein. The differences between liver and heart FABPs are due to loss of fatty acid-binding activity during the isolation of heart FABP [7]. Comparable data were obtained with human and rat liver FABPs [8]. The data indicate the same binding stoichiometry of I reel fatty acid per reel FABP, as was observed with the Lipidex assay [8].

Withdrawal of fatty acid from phospholipid monolayers We prepared phospholipid monolayers containing [l-14C]oleic acid at a surface pressure of 20 mN/m. After stabilization, FABP was injected under the monolayer, which resulted in a decrease of oleic acid in the monolayer. Table ! shows the effects of pig liver and heart FABPs on negative, neutral and cholesterol-conraining monolayers. The off-rates obtained with all three types of monolayers are in the same range for the same FABP preparation. Liver FABP caused a higher off-rate, in agreement with its higher binding activity. .A maximal rate of oleic acid withdrawal was obtamed when the amount of FABP injected under the monolayer was increased (Fig. 1). Nearly all fatty acid was bound to FABP at the end of the experiments, when higher concentrations of FABP were used. A comparable hyperbolic curve was obtained when increasing amounts of large unilamellar vesicles or albumin were brought under the monolayer (data not shown). A transport function of FABP could not be demonstrated by injecting both FABP and vesicles under a monolayer. The decrease of oleic acid in the

The monolayer contained 0.3-0.5 nmoi [1-14C]oleic acid and 80 pg FABP (of pig origin) was injected in the subphase of 19 mi 10 mM Tris-HCI (pH 6.9). The results of individual experiments are given. The binding capacity of liver and heart FABP was 50 and 16 pmol/Fg protein, respectively. Monolayer

FABP

Off-rate [1- t4 C]oleic acid 7o/min

pmoi/min

PC/PS/oleic acid (8:1 : 1)

liver heart

6.8; 8.8 3.4

22.1; 21.1 10.5

PC/PE/oleic acid (8: I : 1)

liver heart

8.1; 10.9 3.9; 4.5

41.3; 28.6 20.6; 14.2

PC/chol/oleic acid (8:1 : 1)

liver heart

6.7 3.0

25.8 11.3

Transfer of fatty acids from mitochondria to small unilameilar vesicles and multilamellar liposomes and the reverse Spontaneous transfer of oleic acid from loaded mitochondria may occur by diffusion through the water phase, but associative interaction of vesicles and mitochondria may also contribute. When vesicles or multilamellar liposomes were incubated with oleic acidloaded mitochondria from pig liver, an equilibrium was achieved within 5 rain, since the same values were M

c 15

~. ~ lo .. o

/

O

O

s

0 0

'

,tO

'

' 80 FABP (,g)

Fig. I. Withdrawal of [l-14C]oleic acid from a phospholipid monolayer by pig liver FABP. Different amounts of pig liver FABP were injected under PC/PS/oleic acid (8: I : I ) monolayers, containing 0.3-0.5 nmol oleic acid. Values were corrected for the control (no FABP present).

|1 TABLE ll

TABLE I11

Spontaneous transfer of [l-s4C]oleic acid from mitochondria to model membranes

Transfer of [l.J'tC]oleic acid from mitochondria to vesicles in separate compartments

The system contained 5 mg mitochondrial protein (pig liver), to which 71 nmol [l-14C]oleic acid was bound, in 0.5 ml buffer A (pH 7.4). The model membranes were composed of P C / P S in a ratio of 10:1. Values given were obtained after 5 min incubation at 37 o C, but the percentages a r e similar after 10 rain.

Compartment I contained pig heart mitochondria (0.5 mg protein) to which 32 nmol [14C]oleic acid was bound and 100/Lg FABP; compartment 2 contained vesicles (200 nmol lipid phosphorus, P C / P S 10: 1). Both compartments contained 0.9 ml sucrose-Tris-EDTA buffer (pH 7.4) and were separated by a 0.2 /~m polycarbonate membrane. Values of oleic acid transfer to vesicles were obtained from the oleic acid concentration in compartment 2, corrected for the amount bound to FABP (assuming maximal binding). The proportion of FABP present in compartment 2 was 15 and 22~, respectively, after 2 and 4 h.

Acceptor concentration (nmol lipid phosphorus)

Fatty acid transfer (%)

25 50 100 200

vesicles

liposomes

13.9 26.6 44.4

0.9 1.2 1.8 3.4

Addition

Binding capacity (pmol/Fg)

-

obtained after 10 min incubation. The relative amount of fatty acid transferred was proportional to the concentration of vesicles or multilamellar liposomes (Table !1). However, with the latter model membranes, it was relatively much lower, indicating limitation to the outermost layer, possibly due to the slow rate of transbilayer movement [21]. A change of temperature from 37 to 20 ° C did also not result in an observable slower transfer. Similar results were obtained with oleic acid-loaded mitochondria from pig heart and with liposomes of other phospholipid compositions. Due to this rapid spontaneous transfer, the addition of FABP or albun~n had no discernible effect on oleic acid transport in all these systems, if oleic acid concentrations in the supernatants were corrected for fatty acid bound to the protein. A rapid spontaneous transport from vesicles to mitochondria was also noticed in oxidation experiments with intact rat muscle mitochondria and [1-t4C]oleic acid-loaded vesicles. The oxidation rate of 2.5 nmol/min per mg protein did not increase when FABP or albumin were added.

Transfer of fatty acid in an equilibrium dialysis cell In this system, we excluded collision-mediated transfer of oleic acid by a polycal~onate membrane. Small @ 100

Heart FABP Liver FABP

Oleic acid transfer (%) 2h

4h

-

1.6

16 50

2.2 4.1

2.1 6.1 10.4

unilamellar vesicles (200 nmol lipid phosphorus) and oleic acid-loaded mitochondria (500/~g) were present in the separate compartments. The spontaneous transfer of oleic acid to the vesicle-containing compartment amounted only to a few percent. Recovery of oleic acid in the water phase was always more than 90%, and less than 2% was bound to the polycarbonate membrane. The addition of pig heart or liver FABP to the mitochondria-containing compartment resulted in an increase of fatty acid transfer to the vesicles (Table III). The passage of [t4C]formaldehyde-labeled FABP over the polycarbonate membrane amounted to 22% after 4 h. This implicates that up to 1.1 and 3.1% of total fatty acid was bound to heart and liver FABP after 4 h in compartment 2. The liver FABP preparation, which had a higher binding activity than heart FABP, showed a proportionally higher transfer activity. The fatty acid in compartment 2, which is not bound to FABP, will be present in the vesicles, since the partition coefficients of long-chain fatty acids added to a suspension of vesicles B

A

100.~

v

1

.~. =w

U

U

0

'0

"C3 m

50

.-

50

a3 ~3

T

]I

o

!

~o

i

6o

l

'~

90

------, Time(rain)

0

o

i

" "

30

i

e0

90

~

Time(rain)

Fig. 2. Transfer of [l-14C]oleic acid between two separate monolayers (PC/PS 8 : 1). Monolayer 1 contained at the start of the experiments 168 nmol [l-14C]oleic acid. (A) Exchange of [1-14C]oleic acid between two separate monolayers by pig heart FABP (addition of 4 Fg pig heart FABP ( 1'). (B) Spontaneous exchange of [l-14C]oleic acid between two separate monolayers (without FABP).

12 TABLE IV Oxidation of albumin- and FABP.bound palmitic acid by rat liver mitockondria Concentrations were 10 ~M [1-t4Clpalmitic acid and 1.25 mg mitochondrial protein in 0.5 ml medium. Incubation took place for 5 rain at 37 o C. Protein

Molar ratio (palmitic acid/ protein)

Oxidation rate (nm¢l/min per mg protein)

Albumin

5: ! 3:1 I :i

2,52 2.80 1.65

Rat liver FABP

t :!

3.00

rf

~r

in water are of the order of 106-10 ~ in favour of the vesicles [20,M].

Transfer of oleic acid between separate monolayers In a system of two separate monolayers, oleic acid transport could also take place by diffusion or by facilitated transport by FABP. Because fatty acid is readily bound from the monolayer to FABP (Fig. 1), only a small amount of pig heart FABP was injected. At the beginning, the first monolayer showed a release of fatty acid and after a while, fatty acid appeared in the second monolayer (Fig. 2A). After 95 rain, an equilibrium was reached at which each of both monolayers contained 155 of the original amount of oleic acid. Without the injection of pig heart FABP, the release of fatty acid was much slower and no equilibrium was achieved (Fig. 2B). Only 8.25 of oleic acid was transferred to the second monolayer and 21.1~ was still present in the first monolayer after 95 rain.

Fatty acid oxidation experiments FABP cannot only withdraw fatty acids from model and natural membranes or transport fatty acids to model membranes, it also donates fatty acids to membranes of metabolically active systems. Fatty acids bound to rat liver FABP are equally well oxidized by rat liver mitochondna as albumin-bound fatty acids (Table IV). Comparable results were previously obtained by us with rat heart mitochondria and rat heart FABP [35]. Discussion The use of phospholipid monolayers is a new technique in studying fatty acid transport by FABP. Monolayers are often used in studies of phospholipid and sterol transfer [36,37]. FABP cannot only bind fatty acids from vesicles or llposomes, as observed previously [7,15,2&-27], but also from mitochondrial membranes and different phospholipid monolayers. The binding data are in good accordance with those obtained with

the Lipidex assay [6-8]. The binding stoichiometry of 1 mol fatty acid per reel protein was also found by others with liposome assays [7,26] and fluorescence [27,38] or NMR studies [39]. Some authors [40-43] reported, however, a 2 : 1 ratio. Solvation of fatty acid from membranes appears to be a rate-limiting step in fatty acid transfer from vesicles to vesicles and to albumin [15,20]. The maximal fatty acid withdrawal rate at increasing amounts of FABP or liposomes under the monolayer is in agreement with those observations. We observed a rapid spontaneous fatty acid transfer between mitochondria and unilamellar vesicles in accordance with the results of others on donor and accepter vesicles [15,19,20]. Only Catala and Avanzati [28] found a slight spontaneous fatty acid transfer between microsomes and vesicles. Transfer of fatty acids between unUamellar vesicles occurs through the water phase [20,21], but the rapid spontaneous fatty acid transfer in our system of vesicles and mitochondria must be due mainly to collision or association, because transfer was dramatically reduced in an equilibrium dialysis cell with separate compartments. Others observed also a marked interaction of phospholipid vesicles with mitochondria [44] and microsomes [29]. With the equilibrium dialysis cell, we could show that pig heart or liver FABP increased the fatty acid transfer between the two membrane systems. McCormack and Brecher [29] demonstrafed fatty acid transfer by rat liver FABP in such a cell from liposomes to microsomes by an increased acyl-CoA production. Since the effects of the unphysiological polycarbonate membrane on transfer are, however, unknown, we also used a system of two separate monolayers. In this system, FABP and fatty acid can freely move in the aqueous phase. Fatty acid transfer between the separate monolayers was also markedly enhanced by the presence of pig heart FABP in the water phase. These results establish the role of the FABP-fatty acid complex in tile water phase as an intermediate of fatty acid transfer between membranes. We did not observe differences in fatty acid binding or transfer between liver and heart FABPs in any system or study. Some authors [24,25] observed different transport properties of bovine liver and heart FABP and related this to a difference in function. Bovine heart FABP could only donate 16-anthroyloxypalmitic acid to vesicles and bovine liver FABP could only bind this fatty acid from vesicles [24,25]. In this study, we demonstrated that pig heart FABP did not only donate fatty acid to membranes, but also bound fatty acid from membranes, like the liver FABPs. Human heart FABP also bound fatty acid from iiposomes [7]. Liver FABP showed a similar fatty acid donor function in fatty acid oxidation experiments, like heart FABP [35,45]. Brecher and co-workers [29,43] also described the production of acyl-CoA by liver microsomes and heart sarcoplasmic

13

reticulum from fatty acid bound to rat liver or heart FABP. In conclusion, both liver and heart FABP are able to ~nction as fatty acid-acceptor, -carrier and -donor proteins. More investigations are necessary to evaluate the suggested role of FABP in cellular fatty acid transport and trafficing. Molecular biological techniques may assist in this respect. Acknowledgements The investigations were supported by the Nederlands Foundation for Chemical Research (SON) with financial aid from the Nederlands Organization for Scientific Research (NWO). References 1 0 c k n e r , R.K., Manning, J.A., Poppenhausen, R.B. and Ho, W.K.U. (1972) Science 177, 56-58. 2 0 c k n e r , R.K., Manning, J.A. and Kane, J.P. (1982) J. Biol. Chem. 257, 7872-7878. 3 Ketterer, B., Tipping, E., Hackney, J.A. and Beale, D. (1976) Biochem. J. 155, 511-521. 4 Burnett, D.A., Lysenko, N., Manning, J.A. and Ockner, R.K. (1979) Gastroenterology 77, 241-249. 5 Giatz, J.F. and Veerkamp, J.H. (1985) Int. J. Biochem. 17, 13-22. 6 Bass, N.M. (1985) Chem. Phys. Lipids 38, 95-114. 7 Paulussen, R.A., Van der Logt, C.P. and Veerkamp, J.H. (1988) Arch. Biochem. Biophys. 264, 533-545. 8 Peeters, R.A., In het Groen, M.A., De Moel, M.P., Van Moerketk, H.T.B. and Veerkamp, J.H. (1988) Int. J. Biochem. (in press). 9 Takikawa, H. and Kaplowitz, N. (1986) Arch. Biochem. Biophys. 251, 385-392. 10 Takahashi, K., Odani, S. and Ono, T. (1983) Eur. J. Biochem. 136, 589-601. 11 Jagschies, G., Reers, M., Unterherg, C. and Spener, F. (1985) Eur. J. Bioehem. 152, 537-545. 12 Gordon, J. and Lowe, J. (1985) Chem. Phys. Lipids 38, 137-158. 13 Chan, L., Wei, C., Li, W., Yang, C. and Smith, L. (1985) J. Biol. Chem. 260, 2629-2632. 14 Sacchetini, J.C., Said, B., Schulz, H. and Gordon, J.l. (1986) J. Biol. Chem. 261, 8218-8223. 15 Daniels, C., Noy, N. and Zakim, D. (1985) Biochemistry 24, 3286-3291. 16 Glatz, J.F., Janssen, A.M., Baerwaldt, C.F. and Veerkamp, J.H. (1985) Biochim. Biophys. Acta 837, 57-66.

17 Pignon, J.P., Bailey, N.C., Baraona, E. and Lieber, S.C. (1987) Hepatology 7, 865-871. 18 Cooper, R., Noy, N. and Zakim, D. (1987) Biochemistry 26, 5890-5896. 19 Sengupta, P., Sackmann, E., Kuhnle, W. and Scholz, H.P. (1976) Biochim. Biophys. Acta 436, 869-878. 20 Doody, M.C., Pownall, H.J., Kao, Y.J. and Smith, L.C. (1980) Biochemistry 19, 108-116. 21 Storch, J. and Kleinfeld, A. (1986) Biochemistry 25, 1717-1726. 22 Zilversmit, D.B. (1983) Methods Enzymol. 98, 565-573. 23 Mowri, H., Nakagawa, Y., Inoue, K. and Nojima, S. (1981) Eur. J. Biochem. 117, 537-542. 24 Riidel, H., Unterberg, Ch. and Spener, F. (1985) Fette. Seifen. Anstrichmittel 87, 561-567. 25 Reefs, M., EIbracht, R., Riidel, H. and Spener, F. (1984) Chem. Phys. Lipids 36, 15-28. 26 Brecher, P., Sauoaf, R., Sugarman, J., Eisenberg, D. and LaRosa, K. (1984) J. Biol. Chem. 259, 13395-13400. 27 Hamilton, J.A. and Cistola, D.C. (1986) Proc. Natl. Acad. Sci. USA 83, 82-86. 28 Catala, A. and Avanzati, B. (1983) Lipids 18, 803-807. 29 McCormack, M. and Brecher, P. (1987) Biochem. J. 244, 717-723. 30 Lowry, O.H., Roseburgh, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 31 Bartlett, J. (1959) J. Biol. Chem. 234, 466-468. 32 Veerkamp, J.H., Van Moerkerk, H.T, Glatz, J., Zuurveld, J., Jacobs, A. and Wagenmakers, A.J. (1986) Biochem. Med. Metab. Biol. 35, 248-259. 33 Okada, Y. and Spiro, R. (1980) J. Biol. Chem. 255, 8865-8872. 34 Noy, N. and Zakim, D. (1985) Biochemistry 24, 3521-3525. 35 Glatz, J.F., Paulussen, R.A. and Veerkamp, J,H. (1985) Chem. Phys. Lipids 38, 115-129. 36 Demel, R.A., Kalsbeek, R., Wirtz, K.W. and Van Deenen, LL.M. (1977) Biochim. Biophys. Acta 466, 1•-22. 37 Theunissen, J.J., Jackson, R.L., Kempen, H.J. and Demel, R.A. (1986) Biochim. Biophys. Acta 860, 66-74. 38 Wilkinson, T. and Wilton, D. (1987) Biochem. J. 247, 485-488. 39 Cistola, D.P., Walsh, M.T., Corey, R.P., Hamilton, J.A. and Brecher, P. (1988) Biochemistry 27, 711-716. 40 Lowe, J., Sacchetini, J., Laposata, M., McQuillian, J. and Gordon, J. (1987) J. Biol. Chem. 262, 5931-5937. 41 Bordewick, U., Schulenberg, H., Schafer, P. and Spener, F. (1986) Fette. Seifen, Anstrichmittei 88, 588-592. 42 Offner, G., Troxler, R. and Brecher, P. (1986) J. Biol. Chem. 261, 5584-5589. 43 Burrier, R.E., Manson, C.R. and Brecher, P. (1987) Biochim. Biophys. Acta 919, 221-230. 44 DiCorleto, P.E. and Zilversmit, D.B. (1977) Biochemistry 16, 2145-2150. 45 Fournier, N.C. and Rahim, M. (1985) Biochemistry 24, 2387-2396.