THE JOURNAL BIOLOGICAL OF CHEMISTRY 0 1985 by The American Society of Biological Chemists, Inc.

Vol. 260, No. 8, Issue of April 25, pp. 4733-4739,1986 Printed in U. A. S.

Sterol Carrier Protein and Fatty Acid-binding Protein 2

SEPARATE AND DISTINCT PHYSIOLOGICAL FUNCTIONS*
(Received for publication, July 27, 1984)

Terence J. ScallenSQ, Billie J. Noland*, Kathleen L. GaveyS, Nathan M. Bassn, Robert K. Ocknern(( , Ronald Chanderbhan**, and George V. Vahouny**$$
From the $Department of Biochemistry and Department of Medicine, Schoolof Medicine, University of New Mexico, Albuquerque, New Mexico 87131, the TDepartment of Medicine, Schoolof Medicine, University of California, San Francisco, of California 94143,and the **Department of Biochemistry, The George Washington University School Medicine and Health Sciences, Washington, D. C. 20037

Sterol carrier protein 2 (SCP-2) participates in the of translocating cholesterol from the outer to the inner mitomicrosomal conversion of lanosterol to cholesterol, in chondrial membrane (7). Furthermore, the stoichiometry of the conversion of cholesterol to cholesterol ester, and binding of cholesterol by SCP-2 is 1:l ( 6 ) , and pretreatment in intracellular cholesterol transfers. The stoichiome- of adrenal cytosol with anti-SCP-2 IgG abolishes all of the try of binding between cholesterol and SCP-2 is 1:l. stimulatory activity of this cytosol on adrenal mitochondrial However, reports have appeared attributing sterol carpregnenolone production (8). rier protein activity to protein preparation identical However, reports have appeared attributing sterol carrier a to hepaticfatty acid-binding protein(FABP). protein activity to a protein preparation (9, 10) identical to Therefore, the present investigation was conducted hepatic FABP (11).There exists, therefore, confusion as to to compare homogeneous preparations of FABP and as the true nature of hepatic sterol carrier protein. Is it SCP-2 SCP-2 withrespect to their capacities to participate carrier proteins in reactions involving sterols fatty or FABP, or do both proteins participate as sterol carriers? or acids. The results show that SCP-2 and FABP have Therefore, the present investigation was conducted to comseparateanddistinct physiological functions, with pare homogenous preparations of FABPandSCP-2 with SCP-2 participatingin reactions involving sterols and respect to their capacities to participate as carrier (binding) FABP participating in reactions involving fatty acid proteinsin reactions involving sterols or fatty acids. The FABP have separate and distinct is no results show that SCP-2 and binding and/or transport. Furthermore, there overlap in substrate specificities, i.e. FABP does not physiological functions, with SCP-2 participating in reactions possess sterol carrierprotein activity and SCP-2 does involving sterols and FABP participating in reactions involvnot specifically bind or transport fatty acid. ing fatty acids. There is no overlap in substrate specificities, As long as only small quantities of organic solvent i.e. FABP does not possess sterol carrier protein activity and ( . volume %) were used for substrate addition, the SCP-2 does not specifically bind or transport fatty acid. The 16 sterol A-reductase liver microsomal assay for SCP-2 results of the present study also explain why sterol carrier correlated well with the physiologically relevant as- protein activity was previously attributed to FABP (9). says employed in the reconstituted adrenal system. The Preliminary accounts of these experiments have appeared sterol carrierprotein activitypreviously attributed to (12, 13). rat hepatic FABP is explained by the presence of significant quantities of propylene glycol (15volume %) EXPERIMENTALPROCEDURES or Tween 80 in theassay procedure.
Materials--Rat liver SCP-2 was purified to homogeneity as previously reported (5). The yield, however, was improved to approximately 30% by the following modifications. (i) During the heat step, the 303,000 X g liver supernatant (Sm) was heated more slowly (18-20 min) to reach a final temperature of 50 C. Heating a t 50 C was then conducted for 1 min. (ii) A 2-liter Amicon ultrafiltration cell, using a YM-5 membrane, was used in place of the hollow-fiber concentrator to concentrate the SCP-2 fraction obtained from the blue DextranSepharose 4B affinity column. (iii) After concentration and before gel filtration on Sephadex G-75 superfine, the sample was centrifuged a t 5000 rpm for 10 min and the precipitate discarded. Using the procedure as previously described (5) and modified as described above, 6 to 7 mg of homogeneous SCP-2 was obtained from the livers of 100 male Sprague-Dawley rats. Hepatic FABP was purified to homogeneity as described (11). Polyacrylamide Gel Electrophoresis-Polyacrylamide gel electrophoresis was performed as described by Gabriel (14), and gels were stained with Coomassie Brilliant Blue G-250 according to theprocedure of Blakesley and Boezi (15). Stacking gels were not used, and gels were subjected to pre-electrophoresis a t 2 mA/gel for 3 h. Assay of SCP-2-SCP-2 activity of rat liver cytosol (303,000 X g supernatant, S3m) and of the homogeneous protein was determined by the stimulation of the conversion of 7-dehydrocholesterol to cholesterol using rat liver microsomes (51, except that the final assay volume was reducedto 0.65 ml.

Rat liver cytosol contains two sterol carrier proteins (1). SCP-1 ( A d r 47,000) participates in the microsomal conversion of squalene to lanosterol (2-4),while SCP-2 (Adr 13,500) participates in the microsomal conversion of lanosterol to cholesterol (5). In addition, SCP-2 is required for the transport of cholesterol from cytoplasmic lipid inclusion droplets to mitochondria in the adrenal cortex (6),and it also is capable

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. B Reciuient of United States Public Health Service Grants AM10628 and HL-16796. I Reciuient of United States Public Health Service Grants AMt 13328 and AM-26743. $$ Recipient of United States Public Health Service Grant AM32309. The abbreviations used are: SCP, sterol carrier protein; FABP, fatty acid-binding protein; ACTH, corticotropin.

4733

4734

Comparison of SCP-2 and FABP; Separate Physiological Roles

was produced in rabbits as described (22). Enzyme-linked Immunosorbent Assay with Anti-SCP-2 IgG-Proteins to be tested (SCP-2, FABP, or bovine serum albumin) were diluted, as indicated, in sodium carbonate buffer (0.06 M, pH 9.6), and 200-pl aliquots were added to theappropriate wells of disposable polyvinyl chloride microtitration plates, 12.7 X 9.5 cm, 96 flat bottom wells (Dynatech Labs, Inc.). The plates were incubated in a moisture chamber for 3 h a t 37 "C and then overnight at 4 'C. Protein solutions were aspirated from the wells, the plates were washed three times (3 min each) with PBS/T (0.01 M sodium phosphate, pH 7.2,0.15 M sodium chloride, and 0.05% Tween 20), and then 200 pl of a solution of 0.3% gelatin (Type IV, Sigma) in PBS was added to each well and incubation was conducted for 30min at 37 "C. The plates were washed again three times as described above. Anti-SCP-2 IgG and preimmune (control) IgG were adjusted to equivalent protein concentrations (1.5 mg/ml), diluted 1/256 using PBS/T, and 200-4 aliquots were added to appropriate wells. After incubation at 37 "C for 30 min, the solutions were aspirated and the wells were washed as described above. Goat anti-rabbit IgG, (200 pl), peroxidase conjugated (Cappell, -1 mg/ml and then diluted 1/800), was added to appropriate wells. The plates were incubated for 30 min at 37 "C, solutions aspirated, and washing conducted as described above. The substrate solution for the peroxidase reaction (200 pl), containing 0.02% (w/v) 2,2'-azinodi-(3ethylbenzthiazoline) sulfonic acid, (Sigma) in 0.12 M citric acid (adjusted to pH 4 with 5 M KH2P04) was added at timed intervals and, after 15 min, the reaction was stopped with 50 pl of37 m sodium M cyanide. The absorbance at 410 nm was determined spectrophotometrically (Dynatech, MR 580 Microelisa Auto Reader). Adrenal Preparations-Adult male Wistar rats (250-300 g; Charles River Breeding Laboratories) were "quiescent killed" as described earlier (6).Adrenal mitochondria were prepared as described (8) using 250 m sucrose containing 1 m EDTA, 20 M M mM nicotinamide, and 15 m Tris buffer (pH 7.4) (23). The washed mitochondrial pellets M were resuspended in pH 7.5 buffer (24) consisting of 36 m Tris, pH M 7.5,10.4 m sodium phosphate, pH 7.5, 13 m succinate, 5.2 m M M M magnesium chloride, and 0.25 M sucrose. Rat adrenal cytoplasmic lipid inclusion droplets were isolated as M described (6) using a 50 m sodium phosphate buffer, pH 7.4, containing 5 mMMgC1, and 154 m sodium chloride. [1-"CIArachM idonate was incorporated into the lipid droplets by injection of 5 pCi in 5 pl of acetone into the suspension of lipid droplets using a Hamilton syringe. The suspension was incubated, and thelipid droplets were reisolated and washed by ultracentrifugation as described for the incorporation of [4-14C]cholesterol into the lipid droplets (6). Under these conditions, 0.4pCiof labeled aracbidonate was incorporated. Incubations of Adrenal Preparations-For adrenal mitochondria, incubations were carried out at 37 "C for 30 min using a final volume of 2.0 ml. This included 1.0 ml of mitochondrial suspension, pH 7.5, containing 1-2 mg of protein, and 1.0 mlof indicated additions. Protein was estimated by the method of Lowry et al. (25), and pregnenolone was determined by radioimmunoassay (26) (Radioassay Systems Laboratory, Inc.). For incubations involving adrenal cytoplasmic lipid inclusion droplets, the lipid droplets resuspended in phosphate buffer to a volume of 1 ml (6) were incubated with SCP-2 or FABP in 1 mlof buffer, and incubations were conducted at 37 "C for 30 min in air. Droplets were reisolated by ultracentrifugation at 4 "C (50,000rpm for 60min), and the cholesterol (Fig. 2 A ) or the arachidonate (Table 111) was measured in the soluble subnatant fraction as previously described (27).
RESULTS AND DISCUSSION

Briefly, the standard assay procedure is conducted as follows. Incubations contained in a final volume of 0.65 ml: Buffer P (15 m M M potassium phosphate (pH 6.80) containing 0.1 m EDTA), NADPH (1.8 mM), buffer-washed rat liver microsomes (2 mgof protein) (5), or SCP-2 or the protein to be evaluated, and 7-dehydrocholesterol added in the above (140 nmol in 10 pl of 2:l dioxane/l,2-propanediol), sequence. The assay incubation mixture contained the above components in the volumes indicated Buffer P (300 p l ) , NADPH in Buffer P (50 p l ) , microsomes in Buffer P (200 p l ) , protein sample to be evaluated (100 pl), and 7-dehydrocholesterol added as described above. Incubations were conducted in an atmosphere of nitrogen a t 37 "C for 2 h in a Dubnoff shaker. The disappearance of 7-dehydrocholesterol was linear over the 2-h incubation period. Control incubations, minus any SCP-2 or S303 or protein addition (microsome blank), or minus SCP-2 or S303 or protein addition and microsomes (substrate blank) were always conducted. No evidence was observed for nonenzymatic oxidation of the 7-dehydrocholesterol substrate. Furthermore, the disappearance of absorbance measured in the ultraviolet at 294 nm corresponded precisely with assays conducted using [14C]7"dehydrocholesterolsubstrate in which [14C]cholesterol formation was measured by chromatography on silver nitrate-impregnated silicic acid columns (5). Saponification, extraction, and ultraviolet spectroscopy were conducted as described (5). The value obtained for cholesterol formation for the microsomal blanks was subtracted from the cholesterol formed in incubations containing both microsomes and the protein to be evaluated (5). For comparison, the assay conditions previously described in reports (9, 16) which attributed sterol carrier protein activity to FABP were evaluated with respect to theconcentration of propylene glycol and preincubation procedures as follows, combining the components listed Buffer P (213 pl) and propylene glycol (87 pl) mixed together, NADPH in Buffer P (50 pl), the protein sample to be evaluated (100 p l ) , and 7-dehydrocholesterol. This incubation mixture was preincubated at 37 "C for 15 min. Then microsomes (200 p l ) were added, and incubation was conducted for 2 h at 37 "C. Fatty Acid-bindingAssay-Fatty acid binding to hepatic FABP or SCP-2 was analyzed by a modification of the assay described by Glatz and Veerkamp (17). The assay relies on the temperature-dependent binding of hydrophobic ligands to Lipidex 1000 (Packard Instrument Co.), a substituted hydroxyalkoxypropyl derivative of Sephadex G-25 (17, 18).Thus, free, but not protein-bound fatty acids, partition into Lipidex at 0-4 "C. The assay, as carried out in these studies, was conducted as follows. The protein to be evaluated (10-14pg)was incubated with [14C]oleatein 1.5-ml polypropylene vials in 10 m M potassium phosphate buffer, pH 7.4. Radiolabeled oleic acid, dissolved in 5 pI of propylene glycol, wasadded to obtain a concentration range of 0.5 to 20 p ~The final assay volume was 0.455 ml. . After incubation of protein and oleate for 10 min at 37 "C, the vials were cooled, and 50 p1 of ice-cold Lipidex 1000 suspension (l:l, v/v), in buffer, was added to remove unbound fatty acids. After mixing, samples were incubated at 0 "C for 10 min, centrifuged at 20,000 X g at 4 "C for 2 min, and the supernatantswere assayed for radioactivity in 10 ml of Liquifluor (New England Nuclear) in toluene containing 10% BioSolv BBS, (Beckman Instruments). Protein-free blanks were included for each concentration of oleic acid analyzed. Fatty acid binding to protein was analyzed by means of computerized nonlinear least square curve fitting. Models with one or two saturable and/or nonsaturable binding processes were considered. Molecular weights of 14,200 for FABP and 13,500 for SCP-2 were used to calculate the number of binding sites. Acyl-CoA CholesterolAcyl Transferase-Rat liver microsomes were prepared from 50 Sprague-Dawley male rats (Simonson), 55-65 days old (250-350 g) as described (19) using Buffer P. The microsomes were washed three times and frozen in liquid nitrogen and stored at -80 "C in aliquots until needed. Cholesterol and palmitoyl-CoA, both labeled and unlabeled, were prepared as described (19), except that the concentration of palmitoyl-CoA was determined by the method of Seubert (20) using a Hewlett-Packard 8450A UV/VIS spectrophotometer. Acyl-CoA cholesterol acyl transferase assays were conducted at 37 "C in a Dubnoff shaker, 1 cycle/s. Reactions were stopped and the amount of cholesterol ester determined as described (19). Anti-SCP-2 IgG-Anti-SCP-2 IgG was prepared in rabbits using homogeneous rat liver SCP-2 by the technique of Van Vunakis et al. (21). Anti-SCP-2 IgG was characterized by immunoelectrophoresis (8). A single immunoprecipitation arc was observed in the same location (basic PI) for both purified SCP-2 and rat liver cytosol. Antiserum to FABP-Antiserum to homogeneous rat liver FABP

Characterization of SCP-2and FABP-A comparison ofthe electrophoretic mobility purified SCP-2 and rat liver of FABP at pH 4.3 wasperformed (Fig. 1. As can be seen, SCP-2 ) migrates more rapidly toward negative electrode the than does FABP. This is consistent with the PI for SCP-2 which is -8.6 (5), while the PI for FABP is 6.9 (11). It is also apparent (Fig. 1) that both SCP-2 and rat liver FABP are essentially homogeneous. The amino acid compositions of homogeneous SCP-2 (5), FABP (11), the protein preparation reported have SCP and to activity (9) are shown in Table I. As can be seen, there are marked differences in the amino acid composition of SCP-2

Comparison of SCP-2 and FABP; Separate Physiological Roles

4735

A
SCP2

TABLE I Amino acid compositions of SCP-2, FABP,and the protein preparation of Dempsey
SCP-2" Dernpsey proteinb
mol %

FABP'

Asp(n) Thrd Ser Glu(n) Pro GlY Alad 'hCys Vald Met Ile Leud Tyr Phe L s Y His '4%

Trp
a

10.8 3.6 6.0 11.1 3.4 11.5 9.1 1.1 4.8 3.2 4.9 10.0 0.0 5.6 14.0 0.4 0.0 0.6

8.9 9.4 3.6 13.6 1.6 10.2 1.6 1.0 9.5 5.4 6.7 5.1 2.3 4.1 13.4 1.6 1.7 0.0

8.7 9.4 4.8 13.1 1.6 9.6 1.7 1.7 9.1 5.3 6.6 4.8 2.5 4.7 13.0 1.5 1.8 0.0

FABP

See Ref. 5. See Ref. 9. e See Ref. 11. Indicates amino acid in which a large difference in composition exists between SCP-2 and FABP or Dempsey protein.

I A
FIG. 1. Densitometry tracing of a polyacrylamide gel electrophoresis of purified SCP-2and FABP. SCP-2 (10 pg) and FABP (10 pg) were subjected to electrophoresis in a 15% polyacrylamide gel at pH 4.3 for 2 h, 3 mA/gel.Following staining with Coomassie Brilliant Blue (G-250), densitometry tracings were made using a Gilford transport attachment (2410-5) at 600 nm.
Proteins

TABLE I1 Enzyme-linked immunosorbent assay utilizing anti-SCP-2 IgG Anti-SCP-2 IgG was thefirst antibody. Peroxidase-conjugated anti-rabbit IgG was the second antibody. Due to the high coating concentrations of proteins used, in order to allow maximum sensitivity, the absorbance values observed for SCP-2 were well beyond the linear region of the concentration curve.
Protein dilutions
1/25

1/100
AllO,,

1/400

SCP-2 (162 pg/ml) FABP (124 pglml) Bovine serum albumin (23 mg/ml) and eitherof the other proteins, particularly in the quantities

0.692 0.000 0.000

0.666 0.000 0.000

0.493 0.000 0.000

of threonine, alanine, valine, and leucine. In addition while tyrosineispresentinFABPandtheproteinpreparation of cholesterol mass from the cytoplasmic lipid inclusiondropreported to have SCP activity (9), it is not found in SCP-2. lets (Fig. 2 A ) . The correlation coefficient between SCP-2 and the protein Pregnenolone Production by Adrenal Mitochondria-The with reported SCP activity (9) is 0.736. Clearly, SCP-2 and synthesis of pregnenolone from endogenous cholesterol was this protein are separate and distinct proteins. measured as a function of either SCP-2 or FABP concentraOn the other hand, the amino compositions of purified tion. This assay procedure mirrors the final step in adrenal acid of hepatic FABP and the protein preparation Dempsey et al. steroidogenesis, i.e. the productionof the first adrenal steroid (9) (Table I) are virtually identical, with correlation coeffi- hormone from endogenous mitochondrial cholesterol. With a cient between the two protein preparations being 0.997. SCP-2, but not with FABP, there was a concentration-deHomogeneous SCP-2 and FABP were compared for im- pendent formation of pregnenolone by the adrenal mitochonmunological cross-reactivity against anti-SCP-2 IgG (Table dria (Fig. 2B). 11). There was no immunological cross-reactivity between Arachidonate Release from Adrenal Cytoplasmic Lipid IncluSCP-2andFABP, using anextremelysensitive enzyme- sion Droplets-The release of [l-'4C]arachidonate from cytolinked immunosorbent assay procedure. Also, it was found plasmic lipid inclusion droplets was measured for FABP and that anti-rat liver FABP antiserum did not react with pure SCP-2 (Table 111). With FABP, but not with SCP-2, there rat liver SCP-2 on radialimmunodiffusion. was a substantial increase in the release of [l-'4C]arachidonFABP and SCP-2 were then evaluated by means of the ate from the cytoplasmic lipid inclusion droplets. following six assay procedures. Oleic Acid Binding-The binding of oleic acid was measured Cholesterol Release from Adrenal Cytoplasmic Lipid Inclu- for both proteins (Fig. 3). With FABP, but not with SCP-2, sion Droplets-The release of cholesterol from adrenal cyto- there was specific and saturable binding of oleic acid (Kd = plasmiclipidinclusion droplets is an essential step in the 1.5 f 0.22 FM) involving a single binding site/molecule and a process of steroidogenesis in the adrenal. Therefore, we ex- minor nonsaturable process with a slope of 0.025. Binding of amined the capacity SCP-2 or FABP to of release cholesterol the fatty acid to SCP-2 fitted best by a linear nonsaturable was massfromthese cytoplasmiclipidinclusion droplets as a process (slope = 0.12), possibly explained by nonspecific asfunction of SCP-2 or FABP concentration. With SCP-2, but sociation of fatty acid aggregates with SCP-2. not with FABP, there was a concentration-dependent release Cholesterol Esterification-Rat liver acyl-CoA cholesterol

4736 Comparison

of SCP-2 and FABP; Separate Physiological Roles

0.4

1.2

2.0

2.8

PROTEIN 0 1 ) 11

IV
40 80 0

- - " / F40p - a " " .

l
2.8

a 4

1.2

2D

PROTEIN (vMI

FIG. 2. A comparison of the effect of SCP-2 and FABP on cholesterol release from adrenal lipid droplets and pregnenoloneproduction by adrenalmitochondria. A, adrenal lipid droplets (1ml), f SCP-2 or FABP,were incubated in a final volume of2.0ml at 37 'C for 30 min. The droplets were reisolated by ultracentrifugation, and the cholesterol mass of the subnatant fraction was assayed by gas-liquid chromatography. B , a 1.0-ml suspension of adrenal mitochondria (2.65 mg of protein), f protein (SCP-2 Free OleicAcid (AM 1 or FABP), was made to a final volume of 2 ml and incubated for 30 min at 37 "C. Pregnenolone production was determined by radioimFIG. 3. Binding of radioactive oleic acid to purified FABP munoassay. Each point represents the mean of duplicate determina- and SCP-2. ["CIOleic acid was incubated with 10.8 pg of FABP or tions. 13.6pgof SCP-2 at 37 "C for 10 min. Unbound fatty acid was separated from bound fatty acid by Lipidex 1000 as described under TABLE I11 "Experimental Procedures." Binding data were analyzed by means of . , . FABP MSCP, Release of [I-"C]arachidonate from prelabeled adrenal lipid inclusion nonlinear least squares curve fitting. " 2. Each pointrepresents the mean of duplicate determinations. droplets Adrenal cytoplasmic lipid inclusion droplets (1ml), containing 8.83 x 106 dpm of [I-"C]arachidonic acid, were incubated with or without SCP-2 or FABP made to a final incubation volume of 2 ml with Tris buffer (6). The incubation was conducted at 37 'C for 30 min. The droplets were reisolated by ultracentrifugation at 50,000 rpm for 60 min, and [1-"Clarachidonate was measured in the infranatant by liquid scintillation spectrometry.
Additions Release of 11-"Clarachidonate
dpm

None SCP-2 (0.74 p M ) FABP (0.70 u M )

14,978 f 690 15,852 f 122 32,734 f 241

acyl transferase activity was evaluated in the presence of SCP-2 or FABP. With [4-"C]cholesterol as substrate, SCP-2 produced a marked enhancement of acyl-CoA cholesterol acyl transferase activity (Fig. 4A), when compared with microsomes with no protein added or when FABP was added. A similar experiment (data not shown), using unlabeled cholesterol and labeled palmitoyl-CoA, also showed marked enhancement of cholesterol palmitate formation by SCP-2, but FABP (1.06 PM)was without effect. The results of studies in which acyl-CoA cholesterol acyl transferase activity was measured in the presence of FABP, but at much higher concentrations, are shown in Fig. 4B. Using concentrations of FABP which were 10- to 30-fold higher than in Fig. 4A, FABP produced a modest enhancement of acyl-CoA cholesterol acyl transferase activity using labeled palmitoyl-CoA as substrate. The effective concentration of FABP in this instanceis similar to that shown previously byBurnett etal. (28) for other reactions involving fatty acyl-CoA compounds as substrates. Effect of Assay Conditions on Sterol A'-Reductase-The question remained as to why sterol carrier protein activity has been attributed to a preparation of hepatic FABP (9). This question was investigated by comparing the assay conditions employed in the in vitro assay (9, 16),which contains

TIME (mi")

FABP (pM1

FIG. 4. A comparison of the effect of SCP-2 and FABP on cholesterol ester formation rat liver microsomes. A, cholesby terol ester formation with [4-1'C]cholesterol as substrate in the presence of SCP-2, FABP, or fatty acid-free albumin. Preincubations (30 min at 37 "C) containedmicrosomes (4.5 mg), ATP (0.5 mM), [4-"C] cholesterol (500 nmol/mg microsomes), and either no additional protein (o"-o), SCP-2 ( M pg/ml), FABP (", 15 , c. 15 pglml) orfatty acid-free albumin (A-A, 15 pg/ml). Thetotal volume was 9.0 ml. At the end of preincubation, an aliquot (1.0 ml) ) was removed for assay. Unlabeled palmitoyl-CoA (15 p ~ was then added, and additional aliquots (1.0 ml) were removed at the times indicated. Each point represents the mean of duplicate determinations. B, ["Clpalmitoyl-CoA (15nmol) was added either to phosphate buffer (15 mM, pH 6.8) or to varying amounts of FABP in buffer (800 p ) After prewarming (5 min, 37 "C), rat liver microsomes (0.5mg in l. 200 plof buffer) were added. The final incubation volume was 1.0 ml. Assay of acyl-CoA cholesterol acyl transferase activity was conducted for 3 min at 37 'C. Each point represents the mean of duplicate determinations.

approximately 15 volume % organic solvent (propylene glycol) used for substrate addition, with the procedure described (5) which uses small quantities (1.6 volume%) of organic solvent for substrate introduction.

Comparison of SCP-2 and FABP; Separate Physiological Roles

The results of studies in which the liver microsomal conversion of 7-dehydrocholesterol to cholesterol (sterol A-reductase) was measured with varying amounts of SCP-2 or FABP are shown in Fig. 5. In the experiment shown in Fig. 5A, the assay procedure (5), which uses a small quantity of organic solvent (di0xane:propylene glycol, 2:l; 1.6 volume % in the incubation) for substrate addition, showed marked activity for purified SCP-2 in stimulating the conversion of 7-dehydrocholesterol to cholesterol by rat liver microsomes. When the assay conditions of Dempsey, (9, 16) are employed (preincubation for 15 min at 37 C with the protein addition and with approximately 15 volume % of organic solvent (propylene glycol) present), the activity of purified SCP-2 was 12.0 markedly inhibited. When a similar experiment was conducted with FABP (Fig. 5B), sterol carrier protein activity no was observed as long as only small amounts of organic solvent (1.6 volume %) were used for substrate addition ( 5 ) (see Experimental Procedures). However, when approximately 15 volume % organic solvent (propylene glycol) was present in the incubations as in (9, 16, 29), small amounts of SCP activity were detected with FABP, but only at levels of protein far greater than for SCP-2. A similar phenomenon was observed with bovine serum albumin (Table IV), i.e. no SCP activity was observed when

4737

TABLE IV The effect of adding propylene glycol in thepresence of bovine serum

albumin on the microsomal conversion 7-dehydrocholesterolto of cholesterol Cholesterol formed Propylene glycol added to assay preincubation No Preincubation

nmol/2 h

0 0 0 10 10 0 16 20 0 28 50 3.0 34 87 13.5 55 125 200 34.5 17 Amount of propylene glycol similar to Dempsey et al. (9,16). The protein addition in each flask wasbovine serum albumin (4 mg/ assay). Assayswere conducted as described under Experimental Procedures.

small amounts of organic solvent (1.6 volume % in the incubations) were used for substrate introduction; however, SCPlike activity was observed when bovine serum albumin (4 mg) was preincubated with increasing amounts of propylene glycol prior to the addition of microsomes. With no preincubation, the effect of the propylene glycol was somewhat less. I I I I 1 The reconstituted adrenal assays employed in the present a investigation, i.e. cholesterol mass release from adrenal cyto,*----a I plasmic lipid inclusion droplets or pregnenolone production 1.6% Org. Sol. from endogneous cholesterol present in adrenal mitochondria, mirror precisely the physiologically important process of steroidogenesis in the adrenal and other endocrine tissues. These I / I assays do not require the use of organic solvents for substrate addition. The substrate cholesterol is presented in physiological donors, i.e. cytoplasmic lipid inclusion droplets or mito15% Org. Sol. chondria. Both of these organelles are known to be involved, 2 I 0 in uiuo, in the biosynthesis of pregnenolone from cholesterol. Since it is shown in the present investigation that FABP is not capable of releasing cholesterol mass from adrenal lipid droplets (Fig. 2A) or stimulating the utilization of endogenous mitochondrial cholesterol for pregnenolone production (Fig. 2B), it is clear that FABP is not involved as a sterol carrier protein in a physiological sense. Consistent with this finding is the observation of Mishkin et al. (30) that FABP does not I 1 I I I bind sterol. It is also clear that the physiologically relevant assays employed in the reconstituted adrenal system (Fig. 2) correlate well with the rat liver microsomal sterol A7-reductase assay (Fig. 5 ) ,in which small quantities (1.6 volume % in the incubations) of organic solvent are used for substrate addition. However, it is apparent that significant difficulties occur when larger quantities of organic solvent (propylene glycol) are used in the assay procedure (Fig. 5 and Table IV) and as described (9, 16). Problems similar to those observed with propylene glycol (Fig. 5 and Table IV) were also observedwhenTween 80 1.6% Org. Sol. 0 I (0.05%)was present during the assay (Table V), i.e. bovine -serum albumin shows SCP-like activity in the presence of 20 40 60 80 1 0 0 detergent. Therefore, it is notdesirable to use detergent as a substrate vehicle as described (31) for studies utilizing a FABP (pg) FIG. 5. The conversion of 7-dehydrocholesterol to choles- preparation of hepatic FABP. Previously (5),it was shown that the amino acid compositerol by rat liver microsomes is compared using two assay procedures. The assay procedure which uses a small amount (1.6 tion of a protein described as a nonspecific phospholipid volume %) of organic solvent (Org. Sol.) for substrate introduction is exchange protein (CM,) (32) was virtually identical to SCPcompared with the assay (9, 16) which contains approximately 15 volume % organic solvent (propylene glycol). See Experimental 2. The correlation coefficient between SCP-2and CM, is Procedures for additional details. Each point represents the mean of equal to 0.992 (5). We have shown that SCP-2 has in vitro duplicate determinations. A, the two assay procedures are compared phospholipid exchange activity as assayed by using a microfor SCP-2; B, the two assay procedures are compared for FABP. somejliposome exchange system. But since a liposome is an

4738

Comparison of SCP-2 and FABP; Separate Physiological Roles

TABLE V acids. Furthermore, there is no overlap in substrate specificity, The effect of Tween 80 detergent in the presence of bovine serum i.e. FABP does not possess sterol carrier protein activity and albumin on themicrosomal conversion of 7-dehydrocholesterol to SCP-2 does not specifically bind or transport fatty acid. cholesterol Thus, to avoid further confusion in nomenclature, we sugThe standard assay was conducted as described under Experimen- gest that, in the context of the present studies, the term tal Procedures. The protein addition in each flask was bovine serum albumin (7.5mg/assay). For Tween 80 experiments, 0.05% Tween 80 sterol carrier protein be applied only to the M, 13,500 was added in acetone tothe incubation tubes, the acetone was peptide, currently referred to asSCP-2, CM, (32), nonspecific evaporated under a stream of nitrogen, and theassay was conducted lipid transfer protein (36), or nonspecific phospholipid transwith or without preincubation. For preincubation experiments, Tween fer protein (32, 37), and that the term fatty acid-binding 80 (0.05%), bovine serum albumin (7.5 mg), NADPH (1.4mM), and protein be used to describe the M, 14,200 protein, currently 7-dehydrocholesterol (140 nmol) were preincubated for 15 min at referred to as FABP (11, 38), Z-protein (39, 40), and SCP (9, 37 C. The incubation was begun with the addition of microsomes (2 16). Furthermore, it is logical to subdivide the term sterol mg).
Cholesterol formed Tween 80 (0.05%) Standard assay

No preincubation
nmoll2 h

Preincubation

carrier protein into SCP-1 (for the protein required for the microsomal conversion of squalene to lanosterol) and SCP-2 (for the protein required for the conversion of lanosterol to cholesterol, for the conversion of cholesterol to cholesterol ester, and for intracellular cholesterol transfers).
Acknowledgment-We acknowledge the expert technical assistance of Joan A. Manning in the fatty acid-binding studies.
REFERENCES

0.4

20.0

artificial acceptor/donor, does this finding have physiological meaning? This question was examined using a physiological donor, i.e. adrenal cytoplasmic lipid inclusion droplets. The lipid inclusion droplets were incubated with SCP-2, and the mass of lipids released from the lipid droplets was measured. The results (1) show that cholesterol was the only lipid released from the lipid droplet in the presence of SCP-2. Phospholipid, cholesterol ester, triglyceride, and free fatty acid were not released from the cytoplasmic lipid droplets by SCP-2. These findings (1) strongly support the conclusion that SCP-2 does not play a role in the physiological transfer of lipids other than sterols between organelles. A recent report (33) on the distribution and diurnal variations of FABP in the adrenal gland is difficult to reconcile in light of the present studies with homogeneous preparations of SCP-2 and FABP., These studies (33) suggest that FABP can attain levels of 10% of thetotal protein of adrenal homogenates and that 35% of this is associated with mitochondria. Furthermore, data is presented to suggest that mitochondrial FABP is increased under in vivo conditions known to stimulate cholesterol movement to the mitochondrial inner membrane (i.e. ACTH and aminoglutethimide). The presentstudies clearly demonstrate the specificity of SCP-2 in sequestrationof cholesterol from lipid droplets and stimulation of mitochondrial pregnenolone production. We have also shown that SCP-2 can facilitate translocation of cholesterol from mitochondrial outer to inner membranes using mitochondria which are unable to further utilize this cholesterol (blocked with aminoglutethimide) (7). A possible function for FABP in the adrenal cortex can, however, be anticipated. ACTH can activate adrenal sterol ester hydrolase via the CAMPcascade (34,35),and this results in hydrolysis of the high levels of endogenous esterified cholesterol of adrenal lipid inclusion droplets (27). Mediation of cholesterol release from these droplets and transfer to mitochondria by SCP-2 has been demonstrated in model systems (6,8). However, disposition of the fattyacid moiety, resulting from hydrolysis of esterified cholesterol, may be a function of FABP. Delivery of this lipid to mitochondria could account for the high levelsof FABP observed in adrenal mitochondria (32). The reason for increased levels of FABP inmitochondria from aminoglutethimide-treated animals, however, is unclear. In summary, the results of the present investigation show that SCP-2 and FABP have separate and distinct physiological functions, with SCP-2 participating in reactions involving sterols and FABP participating in reactions involving fatty

1 Scallen, T. J., and Vahouny, G. V. (1985) in Comprehensive . Biochemistry: Sterols and Bile Acids, press in 2. Srikantaiah, M. V., Hansbury, E., Loughran, E. D., and Scallen, T. J.(1976)J. B i d . Chem. 251,5496-5504 J. 3 Gavey, K. L., and Scallen, T. J. (1978) Biol. Chem. 253,5476. 5483 4 Ferguson, J. B., and Bloch, K. (1977)J. Biol. Chem. 252, 5381. 5385 5 Noland, B. J., Arebalo, R. E., Hansbury, E., and Scallen, T. J. . (1980) Biol. Chem. 255,4282-4289 J. 6. Chanderbhan, R., Noland, B. J., Scallen, T. J., and Vahouny, G. V. (1982) Biol. Chem. 257,8928-8934 J. 7 Vahouny, G. V., Dennis, P., Chanderbhan, R., Fiskum, G., No. land, B. J., and Scallen, T. J. (1984)Biochem. Biophys. Res. Commun. 122,509-515 8. Vahouny, G. V., Chanderbhan, R., Noland, B. J., Irwin, D., Dennis, P., Lambeth, J. D., and Scallen, T. J. (1983)J. Biol. Chem. 258, 11731-11737 9. Dempsey, M. E., McCoy, K. E., Baker, H. N., DimitriadouVafiadou, A. D., Lorsbach, T., and Howard, J. B. (1981) Biol. J. Chem. 256,1867-1873 10. McGuire, D. M., Olson, C. D., Towle, H. C., and Dempsey, M. E. (1984) Biol. Chem. 259,5368-5371 J. 11. Ockner, R. K., Manning, J. A., and Kane, J. P. (1982)J. Biol. Chem. 257,7872-7878 12. Vahouny, G. V., Chanderbhan, R., Noland, B. J., Bass, N. M., Ockner, R. K., and Scallen, T. J. (1984)Fed. Proc. 43, 1900 13. Scallen, T. J., Noland, B. J., Gavey, K. L., Bass, N. M., Ockner, R. K., and Vahouny, G. V. (1984)Fed. Proc. 43,1901 14. Gabriel, 0. (1971)Methods Enzymol. 22, 565-578 15. Blakesley, R. W.,and Boezi, J. A. (1977) Anal. Biochem. 82,580582 16. Song, M. H., K. and Dempsey, M. E. (1981)Arch.Biochem. Biophys. 211, 523-529 17. Glatz, J. F. C., and Veerkamp, J. H. (1983) Anal. Biochem. 132, 89-95 Anal. 18. Dahlberg, E., Snochowski, M., and Gustafsson, J. A. (1980) Biochem. 106,380-388 19. Gavey, K. L., Noland, B. J., and Scallen, T. J. (1981)J . Biol. Chem. 256,2993-2999 20. Seubert, W . (1960)Biol. Prep. 7,580-583 21. Van Vunakis, H., Kaplan, J., Lehrer, N., and Levine, L. (1966) Immunochemistry 3,393-402 M., Manning, J. A., Ockner, R. K., Gordon, J. I., 22. Bass, N. Seetharam, S., and Alpers, D. H. (1985)J. Bwl. Chem. 260, 1432-1436 23. Bergon, L., Gallant, S., and Brownie, A. C. (1974)Endocrinology 94,336-345 24. Koritz, S. B., and Kumar, A. M. (1970) Biol. Chem. 245, 152J. 159

Comparison of SCP-2 and FABP; Separate

25. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 26. Johnson, L. R., Ruhmann-Wennhold, A., and Nelson, D. H. (1973) Ann. N. Y. A c d . Sci. 212, 307-318 27. Vahouny, G. V., Chanderbhan, R., Hinds, R., Hodges, V. A., and Treadwell, C. R. (1978) J. Lipid Res. 1 9 , 570-577 28. Burnett, D.A., Lysenko, N., Manning, J., and Ockner, R. K. (1979) Gastroenterology 77, 241-249 29. Daum, H. A. (1979) Ph.D thesis, University of Minnesota 30. Mishkin, S., Stein, L., Gatmaitan, Z., and Arias, I. M. (1972) Biochem. Biophys. Res. Commun. 47,997-1003 31. Ishibashi, T., and Bloch, K. (1981) J . Biol. Chen. 256, 1296212967 32.Bloj,B., Hughes, M. E., Wilson, D.B., and Zilversmit, D. B. (1978) FEBS Lett. 96,87-89 33. Conneely, 0. M., Headon, D. R., Olson, C. D., Ungar, F., and

Physiological Roles

4739

34.
35.

36. 37. 38. 39. 40.

Dempsey, M. E. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 2970-2974 Beckett, G. J., and Boyd, G. S. (1977) Eur. J. Biochem. 72, 223233 Nashshineh, S., Treadwell, C. R., Gallo, L. L., and Vahouny, G . V. (1978) J. Lipid Res. 19,561-569 Trzaskos, J. M., and Gaylor, J. L. (1983) Eiochim. Biophys. Acta 751,52-65 Poorthius, B. J. H. M., Glazta, J. F. C., Akeroyd, R., and Wirtz, K. W. A. (1981) Biochim. Biophys. Acta 665,256-261 Ockner, R. K., and Manning, J. A. (1974) J. Clin. Inuest. 5 4 , 326-338 Levi, A. J., Gatmaitan, Z., and Arias, I. M. (1969) J. Clin. Invest. 48,2156-2167 Mishkin, S., and Turcotte, R. (1974) Biochem. Biophys. Res. Commun. 57,918-926.