purification and some properties of the glycolipid transfer protein

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

Vol. 260, No. 20, Issue of September 15, 11231-11239,1985 Printed in U. S. A .

Purification and Some Properties of the Glycolipid Transfer Protein from Pig Brain*

(Received for publication, January 28, 1985)

Akira Abe and Terukatsu SasakiS From the Department of Biochemistry, Sapporo Medical College, Sapporo 060, Japan

A glycolipid-specific lipid transfer protein has been purified to apparent homogeneity from pig brain postmitochondrial supernatant. The purified protein was obtained after about 6,000-fold purification at a yield of 19%. Evidence for the homogeneity of the purified protein includes the following: (i) a single band in acidic gel electrophoresis, in sodium dodecyl sulfate- gel electrophoresis, (ii) a single band in analytical gel isoelectric focusing, (iii) exact correspondence between the glycolipid transfer activity and stained protein absorbance in the acidic gel electrophoresis, and (iv) coincidence between the transfer activity and protein absorption at 280 nm in gel filtration through Ultrogel AcA 54. The protein has an isoelectric point of about 8.3 and a molecular weight of 22,000, as measured by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. A molecular weight of 15,000 was calculated from AcA 54 gel filtration. The amino acid composition has been determined. The protein binds [3H]galactosylceramide but not [3H]phosphatidylcholine. Under the conditions used, 1 mol of the transfer protein bound about 0.13 mol of [3H]galactosylceramide. The glycolipid transfer protein-[3H]galactosylceramide complex was isolated by a Sephadex G-75 chromatography. An incubation of the complex with liposomes resulted in the transfer of [3H]galactosylceramide from the complex to the acceptor liposomes. The result indicates that the complex functions as an intermediate in the glycolipid transfer reaction. The protein facilitates the transfer of [3H]galactosylceramide from donor liposomes to acceptor liposomes lacking in glycolipid as well as to acceptor liposomes containing galactosylceramide.

Proteins that facilitate the transfer of phospholipids between membranes in vitro have been found in animal cells, plant cells, yeast, and a Gram-negative bacterium (for reviews, see Refs. 1-3). The transfer of phospholipids between intracellular membranes is presumably mediated by the activity of phospholipid transfer proteins. Phospholipid transfer proteins different in their phospholipid specificities have been purified from bovine liver (4, 5), heart (6), and brain (7), rat liver (8-10) and hepatoma (ll), yeast (12), and a bacterium (13).

The mechanism of intracellular translocation of glycosphingolipids in animal cells remains to be elucidated. I t is possible that glycosphingolipids are transferred by a mechanism different from that found in phospholipids. Still, it is

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in aceerdance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom all correspondence should be addressed.

interesting to find proteins which facilitate the transfer of glycosphingolipids between membranes because such proteins may possibly be involved in the biosynthesis or degradation of glycosphingolipids. Metz and Radin (14) found in bovine spleen a protein which accelerates the transfer of glucosylceramide between erythrocytes and liposomes. The protein was partially purified (15). Bloj and Zilversmit (16) showed that a nonspecific lipid transfer protein purified from bovine liver (5), which is identical to sterol carrier protein I1 (17, 18), accelerates the transfer of globoside and GMll ganglioside from liposomes to erythrocyte ghosts.

A sensitive and specific determination of the glycolipid transfer activity is possible by a modified procedure of the phospholipid transfer assay which we previously described (19). By the use of this assay, we studied the glycolipid transfer activity present in the postmicrosomal supernatant of rat brain and liver (20, 21). We obtained a small amount of purified glycolipid transfer protein from pig brain (22). The purified protein facilitates the transfer of various glycosphingolipids and glyceroglycolipids between membranes (23). However, the transfer of phosphatidylcholine, phosphatidylinositol, phosphatidylethanolamine, cholesterol, or cholesteryloleate was not accelerated by the protein (23).

In this paper, we wish to present an improved purification procedure of pig brain glycolipid transfer protein. The molecular weight and the isoelectric point of the protein were estimated. Amino acid composition of the protein was determined. It was found that the protein binds [3H]galactosylceramide. The protein-galactosylceramide complex functions as an intermediate in the glycolipid transfer reaction.

EXPERIMENTAL PROCEDURES*

Materials-Galactosylceramide containing nonhydroxy fatty acids was purchased from Sigma. The glycolipid was tritium-labeled (525 pCi/pmol) by the galactose oxidase/sodium b~ro[~H]hydride method according to the procedure of Radin (24). [3H]Phosphatidylcholine (200 pCi/pmol) was prepared from cells of a human lymphoblastoid cell line, CCRF-CEM, grown in RPMI 1640 medium containing [9,10- 3H]palmitic acid, which had been bound to fatty acid-free bovine serum albumin. Phosphatidylcholine was prepared from rat liver by the method described previously (19). [‘4C]Cholesteryloleate (52.5

‘The abbreviations and trivial names used are: GMl, I13-a-N- acetylneuraminosylgangliotetraosylceramide; ConA, concanavalin A; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; G M 2 , II3-cu-N-acetylneuraminosylgangliotriaosylceramide; Temed, N,N,N’,N’-tetramethylethylenediamine.

Portions of this paper (including part of “Experimental Proce- dures” and Figs. 1-4) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biolog- ical Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 85M-0260, cite the authors, and include a check or money order for $3.60 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

11231

11232 Glycolipid Transfer Protein

pCi/pmol), [9,10-3H]palmitic acid (15.2 mCi/pmol), and sodium b~ro[~H]hydride (8.8 mCi/pmol) were obtained from New England Nuclear. Commercial sources of chemicals were as follows: DL-dithiothreitol, phenylmethylsulfonylfluoride, and dicetyl phosphate, Sigma; concanavalin A, Pharmacia; Ampholines and Ultrogel AcA 54, LKB; acrylamide (for disc electrophoresis), E. Merck; SDS (specially pure), BDH Chemicals Ltd.; ammonium sulfate (for biochemical research), Nakarai Chemicals Ltd. Protein molecular weight standards used for SDS-polyacrylamide gel electrophoresis and Ultrogel AcA 54 gel filtration included the following: bovine serum albumin, 68,000 (Sigma); ovalbumin, 43,000 (Sigma); chymotrypsinogen A, 25,000 (Boehringer); soybean trypsin inhibitor, 21,500 (Worthington); myoglobin, 17,500 (Sigma); and cytochrome c, 12,500 (Boehringer).

Polyacrylamide Gel Electrophoresis-Polyacrylamide gel electrophoresis (in the absence of SDS) was performed according to the procedures of Reisfeld et al. (25) on 15% acrylamide gels at pH 4.3, except (i) pre-electrophoresis of separating gel was conducted in 0.06 N KOH-acetic acid, pH 4.3, at 4 mA/tube for 2 h, (ii) stacking gel was prepared from a buffered monomer solution to which glycerol and 2-mercaptoethanol were added at a final concentration of 5% and 0.6 mM, respectively, (iii) stacking gel was polymerized by the addition of ammonium persulfate, and (iv) the electrophoresis was carried out for 2 h with 2 mA/tube.

When the glycolipid transfer activity was recovered from a polyacrylamide gel, two gels were loaded with the purified glycolipid transfer protein (6.2 pg of protein/gel) in Buffer A (10 mM sodium phosphate, pH 6.0, 1 mM dithiothreitol, and 0.02% NaN3) (70 pl) containing NaCl (90 mM) and glycerol (10%). Electrophoresis was carried out in a cold room. One gel was stained, as described below, and the other gel was sliced into 2-mm segments. Each gel slice was fragmented by a glass bar and soaked overnight in 0.5 ml of Buffer B (0.15 M NaCl, 10 mM sodium phosphate, pH 7.4, 1 mM dithiothreitol, and 0.02% NaN3) at 4 "C. The samples were then centrifuged to remove the gel particles, and the galactosylceramide transfer activity was measured for each sample using 100 pl of supernatant.

Polyacrylamide gel electrophoresis, performed in the presence of SDS, was carried out according to the methods of either Laemmli and Favre (26) or Weber and Osborn (27). At the end of the electro- phoretic run, gels were stained with 0.25% (w/v) Coomassie Brilliant Blue G-250 in ethanollacetic acid/water (929, v/v). The gels were destained in ethanol/acetic acid/water first in a ratio of 25:8:65 (v/v) and then in a ratio of 1015:175 (v/v). The stained gel was scanned at 570 nm with a OZUMOR densitometer (model 02-802) (Asuka Mfg. Co., Tokyo).

Polyacrylamide Gel Isoelectric Focusing-Isoelectric focusing was carried out at 2 "C in cylindrical 5% acrylamide gels (5 mm in diameter and 8 cm in length) with 10 mM H3P0, and 1 M NaOH in the anode and cathode compartments, respectively. The ratio of acrylamide and bisacrylamide was 20 in the monomer solution used to prepare the gel. The total concentration of Ampholines was 2% and it was comprised of the pH ranges 6-8, 8-9.5, and 3.5-10 in the ratio of 2:2:1. The gel solution was photopolymerized under daylight fluorescent lamps with a mixture of catalysts consisting of Temed, ammonium persulfate, and riboflavin at final concentrations of 0.056% (v/v), 0.01% (w/v), and 0.0005% (w/v), respectively. Samples containing the purified glycolipid transfer protein (3.93 pg of protein/

gel) in 2% Ampholines and 30% glycerol were applied on top of the gel facing the anode. The samples were overlayed with 70 pl of 2% Ampholines in 15% glycerol. Voltage was applied at 140 V for 14 h and then at 280 V for 1 h. After electrofocusing, a gel was fixed in 15% (w/v) trichloroacetic acid. Ampholines in the gel were eluted in ethanol/acetic acid/water (25:8:65, v/v). The gel was stained in 0.1% (w/v) Coomassie Brilliant Blue R-250 in ethanol/acetic acid/water (9:29, v/v), and destained as described above on the gels of polyacrylamide gel electrophoresis. pH gradients were measured by section- ing a gel into 5-mm slices, soaking the slices in 1 ml of water at 25 "C for 2 h, chilling the samples to 0 'C, and reading the pH with a glass electrode. Cytochrome c and myoglobin were run as standards.

Determination of ~H]Galactosylceramide Binding to the Glycolipid Transfer Protein-Liposomes for the binding experiment were prepared by sonication from a lipid mixture containing phosphatidylcholine, dicetyl phosphate, cholesterol, and [6-3H]galactosylceramide (10' dpm/nmol) in molar ratio of 86:5:45:5. Electron microscopic exami- nation indicates that the liposomes were heterogeneous in size ranging from 25 to 150 nm in diameter. The liposomes (86 nmol as phosphatidylcholine) were incubated in the presence or in the absence of the glycolipid transfer protein (15.3 pg of protein) in 0.2 ml of Buffer B at 25 "C for 1 h. After the incubation, 31 p1 of 75% glycerol and 2 pl of 2-mercaptoethanol were added. Portions (70 pl/gel) of the sample were subjected to polyacrylamide gel electrophoresis at pH 4.3 by the method described above. After electrophoresis, a gel was stained with Coomassie Blue to locate protein. To determine the distribution of 3H radioactivity, gels were sectioned into 2-mm slices with a slicer. The slices were dissolved in 0.5 ml of 30% hydrogen peroxide at 50 "C for 20 h. The samples were mixed with 5 ml of toluene scintillation medium containing 33% Triton X-100 (28) and radioactivity was determined.

In experiments of [3H]phosphatidylcholine binding to the glycolipid transfer protein, liposomes were prepared by sonication from a lipid mixture containing [3H]phosphatidylcholine (9.3 X 10' dpm/ nmol), dicetyl phosphate, and cholesterol in a molar ratio of 865:45. The liposomes were incubated in the presence or in the absence of the glycolipid transfer protein (15.3 pg of protein/0.2 ml of incubation mixture), the incubation mixture was subjected to gel electrophoresis, and the gels were processed as described above.

Isolation of the Glycolipid Transfer Protein-~H]Galactosylceramide Complex and Transfer of rH]Galactosylceramide from the Complex to Acceptor Liposomes-Liposomes for these experiments were prepared by sonication from a lipid mixture containing phosphatidylcholine, dicetyl phosphate, and cholesterol in molar ratio of 86:5:45. Where indicated, either [6-3H]galactosylceramide (1.4 X lo5 dpm/nmol) or galactosylceramide was added as a constituent of the liposomes at the same molar concentration as dicetyl phosphate. The liposomes were incubated either with the glycolipid transfer protein or with the glycolipid transfer pr~tein-[~H]galactosylceramide complex as described in the legends to Figs. 10 and 12. Each incubated mixture was subjected to gel filtration on a column (1.2 X 45 cm) of Sephadex G- 75 (fine) in a cold room. A new Sephadex column was used in each gel filtration since a significant portion of the applied materials, both liposomes and the glycolipid transfer protein, was adsorbed by the column; the adsorbed material obscured the second analysis by the use of the same column. The column was eluted with 10 mM sodium

TABLE I Purification of glycolipid transfer protein from pig brain

Step

Postmitochondrial supernatantb (NH,),SO, precipitation 1st phosphocellulose Cm-cellulose Sephadex G-75 Phenyl-Sepharose 2nd phosphocellulose

Volume

ml 17,150

670 1,613

155 64

100 25

Protein

mg 82,375 28,609 4,242

389 64.6 4.53 2.6

Activity

nmollmin 545.3' 786.8 339.4 367.6 200.9 155.4 105.6

Recovery

% 100 144 62 67 37 28 19

Specific activity

nmollmin f mg

0.00662 0.0275 0.080 0.945 3.11

34.3 40.6

Purification factor

-fold (1) 4.2

12.1 143 470

5,181 6,133

(I Galactosylceramide transfer activity. * In order to obtain figures in this row, the postmitochondrial supernatant was centrifuged at 100,000 X g for 60

min and the protein concentration and the specific activity of the membrane-free supernatant were determined. This figure is exceptionally small probably due to a partial inactivation of the activity during the storage of a

sample for the assay. In most preparative runs, the recovery of the transfer activity in Step 2 was 70-90% of the activity in Step 1.

Glycolipid Transfer Protein 11233

ABSORBANCE AT 570 nm 1 2

bb “““””““ i 5 E Y C

t! “ ””””””” 4 1 0

@- ~

20 10 0 TRANSFER ACTIVITY (‘I. I 60 min )

FIG. 5. Recovery of the activity of purified glycolipid transfer protein from a gel after polyacrylamide gel electrophoresis. Electrophoresis and elution of protein from a gel were performed as described under “Experimental Procedures.” Purified glycolipid transfer protein (6.2 pg of protein/gel) was applied to two 15% acrylamide gels in a buffer a t pH 4.3. Electrophoresis was carried out for 2 h with 2 mA/tube in a cold room. Galactosylceramide transfer activity (0) was assayed using the protein eluted from 2-mm segments. The assay was performed as described under “Experimental Procedures” except the mixture was incubated for 60 min. One gel was stained with Coomassie Blue and the stained gel was scanned with a densitometer a t a full scale absorbance equal to 2.5 (-).

GEL SECTIONS 5 10 15

I n

FIG. 6. Determination of isoelectric point of the glycolipid transfer protein. Polyacrylamide gel isoelectric focusing was performed as described under “Experimental Procedures.” Purified glycolipid transfer protein (3.93 pg of protein/gel) was applied to three gels. A stained gel was scanned with a densitometer at a full scale absorbance equal to 0.5 (”-). A gel was used to determine the pH gradient (0).

phosphate, pH 7.4, containing 0.15 M NaCl and 0.02% NaN3 a t a flow rate of 9 ml/h. One-ml fractions were collected. The glycolipid transfer activity in the fractions of the Sephadex chromatography was determined by a fluorimetric assay (29).

Amino Acid Analysis-Protein was extensively dialyzed against distilled water, lyophilized, and then hydrolyzed in 6 N HCI in uacuo a t 110 “C for 24 h. The amino acid composition was determined on a JEOL amino acid analyzer (model JLC-6AH).

RESULTS

Purification of Glycolipid Transfer Protein from Pig Brain- Table I summarizes the results of the purification procedure for pig brain glycolipid transfer protein. A seven-step procedure was employed and the transfer activity was assayed by the galactosylceramide transfer throughout the purification.

FRACTION NUMBER FIG. 7. Gel filtration of the glycolipid transfer protein on a

Ultrogel AcA 54 column. The purified protein (740 pg in 1.2 ml of Buffer B) was applied to a column (1.2 X 95 cm) and eluted with Buffer B. The flow rate was 6 ml/h and 1.14-ml fractions were collected. Five-pl aliquots were assayed for galactosylceramide transfer activity (X). 0, absorbance at 280 nm. Molecular weight standards were subjected to gel filtration on the same column. Void volume (V,) and total volume were determined with blue dextran and p-nitro- phenol. Molecular weight standards are bovine serum albumin (a), ovalbumin (b) , myoglobin (c), and cytochrome c (d ) .

The postmitochondrial supernatant was fractionated with ammonium sulfate. The protein precipitated by 35-70% ammonium sulfate was applied to a phosphocellulose column, and the column was eluted by a stepwise gradient. A Cm- cellulose column was utilized in Step 4. This step produced a purification factor of about 12-fold. The fifth step employed a Sephadex G-75 gel filtration column, which produced an additional 3.3-fold purification. In the sixth step phenyl- Sepharose CL-4B column was utilized. Addition of 50% eth- ylene glycol to the elution buffer was required to elute the transfer activity from the column. This step produced an additional 11-fold purification. The final step was a chromatography on a phosphocellulose column, from which purified glycolipid transfer protein was eluted at about 98 mM NaCl. The resulting purified protein was about 6000-fold purified as compared with the membrane-free supernatant. The yield from 44 brains (3950 g) was 2.6 mg with a specific activity of 40.6 nmol/min/mg of protein under the conditions of the [3H] galactosylceramide transfer assay. The overall recovery of the activity amounted to 19%.

Purity and Estimation of Molecular Weight and Isoelectric Point-When each fraction in the second phosphocellulose chromatography (Step 7) was analyzed by SDS-polyacrylamide gel electrophoresis using the Laemmli gel system (26), a single band was observed in all the fractions (fractions 84-93 in Fig. 4) under the peak of the glycolipid transfer activity. The purified protein moved as a single band on polyacrylamide gel electrophoresis at pH 4.3 (Fig. 5). Fig. 5 also shows the recovery of the glycolipid transfer activity from a gel after the gel electrophoresis in acidic buffer. A precise correspondence was observed between the absorbance of the single protein band and the transfer activity. An analytical gel isoelectric focusing experiment also showed a single band with a PI equal to about 8.3 (Fig. 6).

The purified protein was subjected to molecular sieve chromatography on a column of Ultrogel AcA 54. The results of this procedure showed coincident profiles of protein and the glycolipid transfer activity (Fig. 7). When the same column was calibrated with several proteins of known molecular weight, the elution behavior of the glycolipid transfer protein


M W X I O - ~ a b

TOP - 7 *

68.0 - 43.0 -

17.5 - 12.5-

1

- FIG. 8. SDS-polyacrylamide gel electrophoresis of the gly-

colipid transfer protein. Electrophoresis on 12.5r;) acrylamide gels in t he presence of 0 . 1 '';, SI>S was performed by the met hod of Wel)er and Oshorn (27). The ratio of acrylamide and hisacrylamide was 37.5 in the monomer solution used to prepare the gel. Electrophoresis was performed at mA/tuhe for 0.7 h and then at 6.7 mA/tuhe for 3.5 h. l'nrified glycolipid transfer protein (13 .4 pg) was applied to the gel shown in h. Molecular weight standards and the glycolipid transfer protein were applied to the gel shown in a. Arrnul indicates the transfer protein. Molecular weight standards are hovine serum alhu- min (68.000). ovalhumin (43,000). chymotrypsinogen A (25,000). mv- oglohin (17,500), and cytochrome c (12,500).

indicates a molecular weight of 15,000 (Fig. 7). Furthermore, electrophoresis on 12.5% acrylamide gels in the presence of SDS according to the method of Weber and Osborn (27) revealed one single band (Fig. 8). Calculation of the molecular weight by this technique gives a value of 22,000 (Fig. 8).

Amino Acid Annlysis-The amino acid composition (Table 11) indicates that the glycolipid transfer protein is rich in glutamate (including glutamine), leucine, alanine, lysine, and aspartate (including asparagine). No hexosamine was detected by the amino acid analysis. The amino acid composition differs from the compositions of phosphatidylcholine-transfer proteins from bovine liver (4, 30) and from rat liver (10) and of nonspecific lipid t.ransfer proteins from bovine liver (Fi), and from rat liver (9, 17, 18, 31). The percentage of amino acids with nonpolar side chains, which include proline, alanine, valine, methionine, isoleucine, leucine, and phenylala- nine, is 46.3% in the glycolipid transfer protein. In this calculation, it is assumed that the transfer protein does not contain half-cystine. Tryptophan is not included in the calculation because the number of tryptophan residue in the prot,ein is not yet determined. The percentage obtained on the glycolipid transfer protein is larger than the values calculated for other lipid transfer proteins, which have been

TARIX I I Amino mid composition r~fpuri f i~d p!\co/ipid f m n s / v r prorr>in

The data is the average v>~lue of two analyses o f onr protein preparation.

~ ~ ~~ ~ - ~~~~~~

Amino arid mol ";

Aspartic acid" 8.79 Threonine 5.12 Serine Glutamic acid"

"X6 1 1.67

Proline KO9 Glycine 4 . 5 4 Alanine ! J . . 5 0 Half-cystine plus cysteine 1 .:< 1 Valine 6.21 Methionine 2.22 Isoleucine 5.95 Leucine I 1 . 3 ) Tyrosine :!.90 J'henylalanine 4 .x 1 Histidine 1 .:io Lysine 9.39 Arginine 2 . w Tr?rptophan + h

_____ ~~ ~~

a Amide content has not heen determined. ~ ~~~

Tr-yptophan was detected by fluorescence sprctroscopy. An exri- tation at 280 nm gave an emission maximum of :M2 nm. The ('V absorption spectrum of the protein exhibits a shoulder nt 290 nm. suggesting the presence of tryptophan.

counted as proteins high in the proportion of the amino acids with nonpolar side chains.

~H](;alactosylceramide Binding to thr (&olipid Tmnsfrr Protein-Binding of ["Hjgalactosvlceramide to the purified glycolipid transfer protein was examined bv gel electrophoresis a t p H 4.3. A complex between the glycolipid transfer protein and ["H]galactosylceramide was found when an incubation mixture containing the transfer protein and liposomes was subjected to the electrophoresis (Fig. 9.4); the liposomes were prepared from a lipid mixture containing phosphatidylcholine, dicetyl phosphate, cholesterol. and [li- "H]galactosvlceramide in molar ratio of 86,:5:45:5. About 30 pmol of ["H]galactosvlceramide was located at the position of the transfer protein, whose quantitv was ahout 2:30 pmol i f the molecular weight ofthe protein is assumed to be 2 0 , 0 0 0 . Therefore, it was estimated that ahout 13'; of the transfer protein contained bound [. 'Hjgalactosvlceramide under the conditions used in this experiment. A tailing of [ 'Hjgalacto- sylceramide in the gel from the protein band towards anode prohablv implys dissociation of the "H lipid from the transfer protein during the electrophoresis. Addition of N-ethvlmal- eimide to the incubation mixture at 4.2 mM resulted in a 35'F reduction of the bound ['H]galactosvlceramide (Fig. 9 A ) . T h e rate of galactosylceramide transfer facilitated hv the transfer protein was also reduced to ahout one-half by the addition of N-ethylmaleimide to the assay mixture at 4.2 mM (data not shown). The inhibition of the transfer activitv by N-ethvl- maleimide had previously been reported (22, 29).

When liposomes prepared from a lipid mixture containing ["H]phosphatidylcholine, dicetyl phosphate, and cholesterol were used in the incubation with the glycolipid transfer protein, no complex between ["Hjphosphatidvlcholine and the transfer protein was found hy the gel electrophoresis (Fig. 9H); the appearance of "H radioactivitv near the tracking dve was not dependent on the addition of the transfer protein to the incubation mixture (Fig. 9B), and this "H peak probably represents [:'H]phosphatidvlcholine monomer. In the experiment shown in Fig. 9N, the specific radioactivity of the liposomal ["Hjphosphatidvlcholine was adjusted to the same value as that of the liposomal [,"H]galactosvlceramide used in

Glycolipid Transfer Protein

1500

z = 500 cr W a Y, z 0 0 c

B

10 20 S L I C E NUMBER

FIG. 9. Analysis of 'H-labeled lipid binding to the glycolipid transfer protein by polyacrylamide gel electrophoresis at pH 4.3. The experiments were performed as descrihed under "Experi- mental I'rocedures." Liposomes containing either [:'HIgalactosvlcer- amide or ["Hlphosphatidvlrholine were incubated a t 25 "C for 1 h in the presence or in the ahsence of the glycolipid transfer protein. Port inns of t he incuhat ion mixture, containing 4.6 pg of protein when the protein had heen added to the incuhation mixture, were subjected to electrophoresis at pH 4.3 on 1 5 5 acrvlamide gels. The procedures for staining, slicing, and analyzing for radioactivity are descrihed under "l3xperimentaI I'rocedures." A. ['HH]gRlactosylceramide hinding t o the glycolipid transfer protein. H , experiments on ["Hlphosphati- dvlcholine hinding to the glycolipid transfer protein. 0, electrophoresis of incuhation mixtures containing the glycolipid transfer protein; .. electrophoresis of a mixture incubated in the presence of the glvcolipid t ransler protein and 4.2 mM N-ethvlrnaleimide; 0, electro- phnresis o f incuhation mixtures without the glvcolipid transfer protein. In each experiment, A and H , a gel, to which the incuhation mixture containing the transfer protein had been applied, was stained with Coomassic Hlue. The stained gels were shown in A and H .

the experiments shown in Fig. 9A. The results shown in Fig. 9 are in good agreement with the lipid specificity of the transfer reaction accelerated by the glycolipid transfer protein (22, 28).

Isolation of the Gl-vcolipid Transfer Protein-/RH]Galacto.~~vl- crramide Complex-The complex between [:'H]galactosylcer-

0 X

X

X

20 30 40 50 60 FRACTION NUMBER

FIG. 10. Isolation of the glycolipid transfer protein-[WIgaI- actosylceramidecamplex by Sephadex C-75 chromatography. The glvcolipid transfer protein f I:{!) o f protc,in) and 1 ' H I g a l ; ~ ~ t n - svlceramide-containing liposomes ( 7 t 5 nmol as phosphatidyl(~h~)line) were incuhated at 20 "(: for I h in 1 . 3 ml n f Fiuffer I < . 'I'hv incuhnted mixture was suhjected to Sephadex G-75 chromatography as descrihed under "Experimental Procedures." 0, 'H radioactivity; 0, nhwrhnnce at 280 nm; X, the glvcolipid trnnsfer activity. I t was not pnssihle t o obtain accurate values o f llhsorhance at 2x0 nm in the frnctinns at the void volume due to the light scattering hy liposomes.

amide and the glycolipid transfer protein was isolated hv a Sephadex (2-75 chromatography of the incuhated mixture of the protein and liposomes containing ["HJgalactosvlceramide (Fig. IO). The liposomes were eluted at the void volume (fractions 18-22 in Fig. 10) and did not contain protein (Fig. 11). A small amount of radioactivity corresponding to 0.46 nmol of ["H]galactosylceramide was eluted at the position of the protein, which was located both hv the ahsorption at 280 nm and by gel electrophoresis (Figs. 10 and 11 1. The optical densitv indicates that about 4 nmol of the glvcolipid transfer protein is present. in this peak if we assume the molecular weight of this protein to he 20,000. This estimation was made hased on the data that the optical densitv at 280 nm of a solution (100 pg of protein/ml) of the glycolipid transfer protein is 0.128 cm. Therefore, about 12r; of the glycolipid transfer protein contained hound ["Hjgalactosvlcernmide. It was found that the appearance of the small radioactive peak at around fraction 84 was totally dependent on the presence of the transfer protein in the applied material. The glycolipid transfer activity was found in the fractions around number 34 but not in the fractions at the void volume (Fig. IO). T h e material with absorption at 280 nm at around fraction 57 in Fig. 10 represents oxidized dithiothreitol present in the applied incubat.ion mixture. In the Sephadex G-75 chromatography shown in Fig. IO, about 7 nmol of the glycolipid transfer protein was applied to the column but onlv 4 nmol of the protein was recovered from the column. SDS-gel electrophoresis shown in Fig. 11 indicates that all the protein eluted from the column is present in the fractions around number 34 where "H radioactivity, material with absorption at 280 nm, and the glycolipid transfer activity were coeluted (Fig. IO). Low recovery of the transfer protein from the column is due to the adsorption of the protein to the column.

Participation of the (;l.vcolipid Transfrr I 'rotrin-~H]Galnr- tosylcwamide Complex as Intermrdiotc in (;lwolipid Tmnsfrr Reaction-An incubation of the glycolipid transfer protein- ["H]galactosvlceramide complex with liposomes resulted in

11236 Glycol ipid Transfer Protein

68.0 - 43.0 -

21.5-

12.5-

FIG. 1 1 . SDS-polyacrylamide gel electrophoresis of the Sephadex C-75 fractions obtained in the isolation of the glycolipid transfer pr~tein-[~H]gaIactosylceramide complex. Each sample ( G O p11 applicd to lnnes numhered from 18 t o 4 0 contained a 45-pl portion ofthe numheretl fraction ohtainetl in the rhromatogrnphy shown in Fig. 1 0 . Earh sample also contained SIX, 2-mercaptoethanol, and hromphennl hlue at final ronrentratir)ns of I";,, 1.5%, and O.OOR"i, respectivelv. The electrophoresis was performed hv the method o f I.nemmli and F'nvre ( 2 6 ) on a 15% acrylamide slah (2 m m ) gel. T h e following molecular weight standards were npplied t o the lane indicated hv S: hovine serum albumin (68.000). ovalhumin (43,000). soyhean tr.ypsin inhihitor (21.50IJ1, and cvtorhrome r (12,500). The major protein hand with a molecular weight of 22,000 represents the glyolipid transfer protein. The minor bands. which are not proteins, at the area around molecular weights o f . ~ : ~ , O ( ~ ~ ) - ~ ~ ~ , ~ J ~ ~ l l are an nrtifart produced by the addition of 2-merraptoethanol to the samples.

the transfer of [:'H]galactosylceramide from the complex to the acceptor liposomes (Fig. 12). When the complex was incuhated with liposomes lacking in glycolipids, about 46% of the ["H]galactosvlceramide originally present as a complex wit,h the transfer protein was transferred to the liposomes, which had heen prepared from phosphatidylcholine, cholesterol, and dicetyl phosphate (Fig. 12.4). About 54% of the ["HI galactosylceramide remained hound to the transfer protein after incuhation of the complex with the liposomes for 1 h a t 20 "C (Fig. 12A). When the glycolipid transfer protein-["HI galactosylceramide complex was incuhated with liposomes containing galactosylceramide, almost all the [:'H]galactosylceramide originally present as the complex was transferred to the liposomes during the incuhation (Fig. 12H). In this experiment, the incuhation mixture contained liposomal galactosylceramide in ahout 300-fold excess of the ["H]galactosylceramide added as the complex. Therefore, the result in Fig. 12R suggests an extensive mixing of the protein-hound ["Hlgal- actosylceramide with the liposomal galactosylceramide. The results in Fig. 12 indicate that the glycolipid transfer protein- glycolipid complex functions as the intermediate in the glycolipid t,ransfer reaction. The use of different. acceptor liposomes with or without galactosylceramide resulted in a marked difference in the amount of the ["HJgalact~osvlcer- amide eventually transferred from the complex to the acceptor liposomes at. an equilihrium point.

GI-vcolipid Transfer Pmtrin-fa~ilitntrd Transfrr o f (' la 1 nrto- sylcernmide from Donor Liposnrnes to Arrrptor 1,iposomr.s either Containing or Lucking in (~alacto.s~lcrrarnidr-'The glycolipid transfer protein facilitated the transfer of ["Hlgalac- tosylceramide from donor liposomes to acceptor liposomes lacking in glycolipids as well as to acceptor liposomes containing galactosylceramide at the same molar rat io as in the donor liposomes. The extents of ["H]gnlactosvlcerami~le transfer from the donor to acceptor liposomes approached the same equilibrium point for hoth glycolipid-containing and glycolipid-deficient acceptor liposomes. The results indicate that the transfer reaction to hot h acceptor liposomes proceeds in the same way up to the point where [. 'H]galactosvlceramide evenly distrihutes on the surfaces of the donor and acceptor liposomes.

DISCUSSION

In a purification of glycolipid transfer protein, wve used pig hrain as a starting material hecause it had been found that the specific activity of the glycolipid transfer reactinn facilitated by the memhrane-free supernatant of rat hrain was about 10 times higher than the specific activity of the reaction facilitated hv rat liver supernatant (21 1. Metz and Radin (1.5) reported partial purification and properties of a crrehroside transfer protein from hovine spleen which is similar t o h u t not identical with the glycolipid transfer protein from pig


A

n U

- 50 60

FRACTION NUMBER FIG. 12. Transfer of [SH]galactosylceramide from the gly-

colipid transfer pr~tein-[~H]galactosylceramide complex to acceptor liposomes. In A , the glycolipid transfer protein-f3H]galactosylceramide complex (a portion of fractions 33-36 in Fig. 10, 16,000 dpm) and liposomes lacking in galactosylceramide (715 nmol as phosphatidylcholine) were incubated at 20 "C for 1 h in 2.0 ml of Buffer B. In B, the glycolipid transfer pr~tein-[~H]galactosylceramide complex (a portion of fractions 33-36 in Fig. 10, 19,600 dpm) and galactosylceramide-containing liposomes (715 nmol as phosphatidylcholine) were incubated a t 20 "C for 1 h in 2.1 ml of Buffer B. The incubated mixtures were subjected to Sephadex G-75 chromatography as described under "Experimental Procedures." In A, the glycolipid transfer activity was found in fraction 33 but not in fraction 19.

brain. Similarities between the pig brain protein and the bovine spleen protein include the following properties: (i) the molecular weights are similar, for the pig brain protein, 22,000 (SDS-gel electrophoresis) and 15,000 (gel filtration), and for the bovine spleen protein, 20,300 (gel filtration); (ii) the isoelectric points are similar, for the pig brain protein, about 8.3, and for the bovine spleen protein, about 8.8 and 9.1; and (iii) both are active in facilitating the transfer of galactosylceramide, glucosylceramide, and lactosylceramide between membranes, but are not active in facilitating the transfer of phosphatidylcholine or cholesterol (15, 22, 23). Metz and Radin (15) obtained partially purified cerebroside transfer protein from bovine spleen at a yield of 2%. Their purest preparation still contained several protein components when analyzed by SDS-gel electrophoresis (15). They found that

the activity of cerebroside transfer protein was unstable during the last two steps of their purification procedure. We had no problem with the stability of the activity of pig brain transfer protein during the purification procedure described in this paper. Moreover, the purified transfer protein retained it's original activity for at least 4 months when stored at 0 "C in Buffer A containing 0.1 M NaCl at a protein concentration of 0.1 mg/ml. The purity of the glycolipid transfer protein prepared by the procedure described in this paper has been assessed by gel electrophoresis in acidic buffer, by gel filtration on Ultrogel AcA 54, by gel isoelectric focusing, and by SDS- gel electrophoresis (Figs. 5-8). All the results indicate that the protein has been purified to homogeneity.

The molecular weight of pig brain transfer protein estimated by SDS-gel electrophoresis was significantly larger than the molecular weight estimated by molecular sieve chromatography (Figs. 7 and 8). At present, we don't know the reason for the discrepancy of the estimated molecular weights of the protein. One possibility is the presence of covalently bound carbohydrate in the protein. However, no hexosamine was found by the amino acid analysis. The glycolipid transfer protein contains amino acids with nonpolar side chains at a high proportion (Table 11), which suggests the presence of a hydrophobic binding site for glycolipids in the protein. The presence of a highly hydrophobic region in the transfer protein may account for the effectiveness of phenyl-Sepharose chromatography in the purification of the transfer protein.

Conzelmann et at. (32) reported that an activator protein of lysosomal @-hexosaminidase A has an activity to facilitate the transfer of GW2 and other glycosphingolipids from donor to acceptor liposomes. The glycolipid transfer protein we purified is most likely different from the activator protein. The pig brain protein has a PI of 8.3 whereas the activator protein has a PI of 4.8. The activator protein accelerates the transfer of GMZ much faster than that of aSialO-GM2, whereas the rate of lactosylceramide and globotriaosylceramide transfer facilitated by the pig brain protein is faster than the rate of G M ~ transfer (23). It seems important to determine the intracellular localization of the glycolipid transfer protein, i.e. whether the glycolipid transfer protein is a cytoplasmic protein, a lysosomal protein, or a protein in the Golgi complex.

The glycolipid transfer protein does not facilitate the transfer of phosphatidylcholine, phosphatidylinositol, cholesterol, or dimannosyldiacylglycerol between membranes (22, 23). Therefore, galactosylceramide is the only one constituent of the liposomes used in this paper which can be transferred by the glycolipid transfer protein. Nevertheless, the transfer protein facilitates the transfer of [3H]galacto~ylceramide from donor liposomes to acceptor liposomes lacking in glycolipid. The results are consistent with the capability of the glycolipid transfer protein to facilitate a net mass transfer of glycolipid molecules from donor to acceptor membranes. The net mass transfer of galactosylceramide facilitated by the glycolipid transfer protein was shown by the use of monomolecular lipid film spread at the air-water interface (33).

A complex formation of the glycolipid transfer protein with pyrene-labeled galactosylceramide was indicated by fluorescence measurements of a mixture of the protein and liposomes containing the fluorescent glycolipid analogue (29). The results shown in Figs. 9 and 10 directly proved that the glycolipid transfer protein binds [3H]galactosylceramide. The lack of [3H]pho~phatidylcholine binding to the transfer protein (Fig. 9) is in good agreement with the specificity of the transfer reaction accelerated by the glycolipid transfer protein (23). Under the conditions used in Figs. 9 and 10, 1 mol of the transfer protein bound 0.12-0.13 mol of [3H]galactosylcer-


amide. Although it may be possible to find [3H]galactosylceramide-containing lipid film which will load a larger portion of the transfer protein with [3H]galactosylceramide (33), we take it significant to find a large portion of the transfer protein presumably in a lipid-free state. The results presented in Fig. 12 indicate that the transfer protein-glycolipid complex functions as the intermediate in the glycolipid transfer reaction. Most likely, the intermediate is the transfer protein containing 1 mol of bound glycolipid/mol of the protein as is the case with phosphatidylcholine-specific transfer protein purified from bovine liver (4, 34). Then, in order to perform a net mass transfer of a glycolipid molecule from the donor to acceptor liposomes, the transfer protein carrying a glycolipid monomer must unload the hound glycolipid onto an acceptor liposomes. Thus the glycolipid transfer protein without a bound lipid probably represents a state of the protein after net mass transfer of a glycolipid molecule to the acceptor membrane.

Wong et al. (38) showed an association of partially purified glycolipid transfer protein from bovine brain with phosphatidylcholine/[3H]glucosylceramide vesicles. We did not find any such binding of the purified glycolipid transfer protein to liposomes (Figs. 10 and 11). The significance of the results reported by Wong et al. (38) in considering the transfer mechanism by the glycolipid transfer protein is questionable in the light of the fact that the purity of the glycolipid transfer protein used in their experiments was too low to study the specific interaction of the transfer protein with glucosylceramide. They found that 30-40% of the protein present in the partially purified glycolipid transfer protein coeluted with the vesicles. The value is too large because the purification procedure and data reported by Wong et al. (38) suggest to us that the purity of their preparation is at most 10%. It is not possible to find the glycolipid transfer pr~tein-[~H]glycolipid complex by the use of a transfer protein preparation with low purity, which was the case in the experiments reported by Wong et al. (38). The glycolipid transfer protein is a basic protein and tends to form nonspecific complexes. Therefore, it is important to study the complex formation between the glycolipid transfer protein and lipids by the use of a pure protein.

Acknowledgments-We wish to thank Dr. Keiko Yamada for help

in the preparation of lipids. We are indebted to R. Ohtani for typing the manuscript.

REFERENCES 1. Wirtz, K. W. A. (1982) in Lipid-protein Interactions (Jost, P. C., and

2. Kader, J.-C., Douady, D., anfhazliak, P. (1982) in Phospholiptds (Haw-

3. Zilversmit, D. B., and Hughes, M. E. (1976) Methods Membr. Btol. 7 , 211-

4. Kamp, H. H., Wirtz, K. W. A,, and Van Deenen, L. L. M. (1973) Biochim.

5. Crain, R. C., and Zilversmit, D. B. (1980) Biochemistry 19 , 1433-1439 6. DiCorleto, P. E. Warach, J. B., and Zilversmit, D. B. (1979) J. Biol. Chem.

7. Helmkamp, G. M., Jr., Harvey, M. S., Wirtz, K. W. A., and Van Deenen,

8. Bloj B. and Zilversmit, D. B. (1977) J. Biol. Chem. 252 1613-1619 9. Pookhdis, B. J,. H. M., Glatz, J. F. C., Akeroyd, R., and Wirtz, K. W. A.

10. Poorthuis, B. J. H. I$, Van der Krift, T. P., Teerlink, T., Akeroyd, R., (1981) Biochtm. Blo hys Acta 6 6 5 , 256-261

Hostetler, K. Y., and Wirtz, K. W. A. (1980) Biochim. Biophys. Acta 600,376-386

11. Dyatlovitskaya, E. V., Timofeeva, N. G., Yakimenko, E. F., Barsukov, L. I.? "uzya, G. I., and Bergelson, L. D. (1982) Eur. J . Biochem. 123,311-

Griffith, 0. H., eds) Vol. 1, 151-231, Wiley-Interscience, New York

thorne, J. N., and Ansell, G. B., eds) pp. 279-311, Elsevier, Amsterdam

259

Biophys. Acta 3 1 8 , 313-325

264 , 7795-7802

L. L. M. (1974) J. Bid. Chem. 249,6382-6389

12. 13. 14. 15. 16. 17.

18.

Daum, G., and Paltauf F. (1984) Biochim. Biophys. Acta 7 9 4 , 385-391 Tai S:P. and Kaplan' S. (1984) J. Biol. Chem. 259 , 12178-12183 Me& R. j . , and Radin' N. S. (1980) J. Biol. Chem. 255,4463-4467 Metz: R. J., and Radin: N. S. (1982) J . Bid. Chem. 2 5 7 , 12901-12907 Blo', B., and Zilversmit, D. B. (1981) J. Biol. Chem. 256,5988-5991 Nofland, B. J., Arebalo, R. E., Hansbury, E., and Scallen, T. J. (1980) J.

Trzaskos, J. M., and Gaylor, J. L. (1983) Biochim. Biophys. Acta 7 5 1 , 52-

315

Biol. Chem. 255,4282-4289

Gr, 19. Saiaki, T., and Sakagami T. (1978) Biochim. Biophys. Acta 512,461-471 20. Yamada, K., and Sasaki, $. (1982) Biochim. Bio hys Acta 6 8 7 , 195-203 21. Yamada, K., and Sasaki, T. (1982) J. Biochem. g2,457-464 22. Abe, A., Yamada, K., and Sasaki, T. (1982) Biochem. Biophys. Res. Com-

23. Yamada, K., Abe, A., and Sasaki, T. (1985) J. Biol. Chem. 260,4615-4621 24. Radin, N. S. (1972) Methods Enzymol. 28,300-306 25. Reisfeld, R. A., Lewis, U. J., and Williams, D. E. (1962) Nature (Lord.)

27. Weber, K., and Osborn, M.'(1969) J. Biol. Chem. 244,'4406-4412 26. Laemmli U. K. and Favre M. (1973) J. Mol. Biol. 80 575-599

28. Patterson, M. S., and Greene, R. C. (1965) Anal. Chem. 37,854-857 29. Abe, A., Yamada, K., Sakagami, T., and Sasaki, T. (1984) Biochim. Biophys.

30. Akeroyd, R., Moonen P. Westerman J. Puyk, W. C., and Wirtz, K. W.

31. Bloj, B., Hughes, M. E., Wilson, D. B., and Zilversmit, D. B. (1978) FEES

32. Conzelmann, E., Burg J., Stephan, G., and Sandhoff, K. (1982) Eur. J .

33. Sasaki, T., and Demel, R. A. (1985) Biochemistry 2 4 , 1079-1083 34. Demel, R. A., Wirtz, K. W. A., Kamp, H. H., Geurts Van Kessel, W. S. M., 35. Hellings, J. A., Kamp H. H., Wirtz K. W. A,, and Van Deenen, L. L. M.

36. Bensadoun, A., and Weinstein, D. (1976) Anal. Biochem. 70 , 241-250 37. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J.

38. Wong, M.,, Brown, R. E., Barenholz, Y., and Thompson, T. E. (1984)

mun. 104 , 1386-1393

195,281-283

Acta 778,239-244

A. (1981) Eur. J . Bioc&m. 114,38&-3bl

Lett. 96,87-89

Biochem. 123,455-264

and Van Deenen, L. L. M. (1973) Nature (Lord.) 2 4 6 , 102-105

(1974) Eur. J . Biochem. 47,601-805

Biol. Chem. 193,265-275

Biochemrstry 23,6498-6505


S a t w a t i o n i n O r d e r t o p r e c l p i t d t e a l l p r o t e i n . The s o l u t i o n was a l l o w e d t o s t a n d o v e r n i g h t % 4: Cm-cellulose Chromatoqraphl - A m m n i u n s u l f a t e 160.3 g l l 0 0 m l ) was added t o 90%

and t h e n c e n t r i f u g e d a t 33,000 x g f o r ID mi". The P e l l e t was d i s s o l v e d i n 90 m l O f B u f f e r A . A f t e r d l a l y s i s a q a i n r t 4 x 1 l i t e r O f t h e same b u f f e r . t h e s o l u t i o n nas C e n t r i f u o e d a t 72.400 x g f o r 3 0 - m i n t i rermve a smal l m o u n t o f i n s o l u b l e m a t e r i a l . The supernatan t ;as a p p l i i d t o a column (2 .7 x 45 cm) of Cm-cellulose (Uhatman C H I Z l p r e e q u i l i b r a t e d w i t h B u f f e r A. The column #as washed w i t h 1 . 4 l i t e r o f t h e b u f f e r and e l u t e d w i t h a 1 . 6 - l i t e r l i n e a r g r a d i e n t o f NaCl

p i d t r a n s f e r a c t i v i t y YIP e l u t e d a t a b o u t 88 mi4 NaCl. b u t 50W transfer a c t i v i t y a p p e a l e d f rom 0 t o 0.5 H i n t h e same buf fe r a t a f l o w r a t e Of 37 ml Per h. The major peak o f g l y c o l i -

b e f o r e t h e main peak [F ig . I ) .

I 1

" J O 30

FRACTION NUMBER 60 w 120

Fig . I . E l u t l o n P d t t e r n o f g l y c o l l p i d t r a n s f e r p r o t e i n f r o m d Cm-cellulose column. P r o t e i n

t o t h e CM52 column and e l u t e d as described above. Seventeen-ml f ldCtlOnS were c o l l e c t e d . The (3,547 mg I " 151 ml o f Bu f fe r A) f rom the 1s t phosphoCel lu loSe chromatography S tep w a s a p p l i e d

" .I

30 60 w 1m FRACTION NUMBER

F i g . 2. E l u t i o n p a t t e r n o f glycolipid t r a n s f e r p r o t e l n f r o m d Sephadex 6-75 column. P r o t e l n

G-75 co lumn descr ibed above. E lut ion * I S w i t h t h e same b u f f e r . Four-ml f r a c t i o n s were c o l l e c - (311 mq i n 12 .7 m l O f B u f f e r B ) f rom the Cm-cellulose chromatography Step vas a p p l i e d t o t h e

t e d . f i f t e e n - y l d l i q u o t l were ar rayed f o r g a l a c t o r y l c e r a m i d e t r a n s f e r d c t i v l t y (H). M, absorbance a t 280 nm. H o r i z o n t a l b a r i n d i c a t e s f r a c t i o n s p o o l e d f o r t h e n e x t S t e p .

1.5 -

E R 0

I- 1-0 - W u 4 m a

2 0.5 - 0

4

- - a 0.2 2

E 2 E - > c-

0.1 E 4

a W U In z 4

c a

0 0 20 40 60 1M 140 1 6 0

FRACTION NUMBER

F l q . 3 . E l u t i o n p a t t e r n O f g l y c o l l p l d t r a n s f e r p l ' o t e l n f r o m a Phenyl-Sephdrose column. Pro- t e i n ( 6 4 . 6 mg i n 70 m l O f B u f f e r 0 from t h e Sephadex G-75 chromatography Step war a p p l l e d t o the Phenyl-Sepharose ~ O l u m n . The column w a s e l u t e d as described above. Arrows a. and b I n d i c a - t e t h e p o i n t s a t w h r c h e l u t i o n b u f f e r s were changed t o 0 .1 M T r l s - H C I , pH 6.8 , 1 mM d i t h l a t h - r e i t o l and 5 0 ( v i " ) e thy lene g l yco l , 0 .1 M Tr is -HCI , pH 6 .8 , 1 d J d l t h i o t h r e l t o l , respective-

152 where 4.;-ml f r d c t l o n l were m l l e c t e d . T W O - ~ I d l i q ~ o t s O f f r d t t l o n s 135-154 and 20-ul l y . The f l a w r a t e was 48 m l per h. and 8 . 6 - m l f r a c t i o n s were co l l ec ted excep t f l dC t lOns 120-

a l l q m t l o f o t h e r f r a c t l o n s were assdyed f o r g a l a c t o s y l c e r a m i d e t r a n s f e r a c t i v i t y I-). -, absorbance a t 280 nm. H o r l m n t d l b a r i n d i c a t e s f r a c t i o n s p o o l e d f o r t h e n e x t s t e p .

30 60 90 120 FRACTION NUMBER

t e i n ( 4 . 5 3 mg i n 100 ml O f B u f f e r A ) f r D n the Phenyl -Sepharore chromatography s tep was a p p l i e d F lg . 4. E l u t i o n p a t t e m O f g 1 y C O l i p i d t r a n s f e r p r o t e i n f r o m a phosphoce l lu lose column. Pro-

c o l l e c t e d i n t u b e s Ma. 1 - 3 5 and 36-130. r e s p e c t i v e l y . The shape o f t h e g r a d i e n t (x-x) w a s t o t h e P I 1 column and e l u t e d a6 described above. 5.2-ml f r a c t i o n s and 2 . 3 - 6 f r a c t i o n s were

estab l i shed f rom canduct lv i t y measurements . Two-"1 a l i q u o t s o f fractions 84-95 and 20-VI a l i - W O t l o f O t h e r f r a c t i o n s were assayed f o r g a l a c t o s y l c e r a m i d e t r a n s f e r activity ( H I . -,absorbance a t 280 nm. H o r l m n t s l b a r i n d i c a t e s f T a c t l O n S p o o l e d .

a g a i n s t 4 x 500 ml o f B u f f e r TM I r i s - H C I , pH 6.8, and 1 mt4 d i t h i o t h r e i t o l ) . The d i a l y s a t e was a p p l l e d t o 2 column ( 2 x 19 cm) O f Phenyl-Sepharore CL-40 (Phamacia) whlch had been e q u ~ l i b r d t e d w t h B u f f e r C . The column war washed w i t h 220 ml O f the Same buf fe r and 775 rnl of 0 .1 M Tris-HC1. pH 6.8. 1 mt4 d l t h l D t h r e l t D 1 . The g l y c o l i p i d t r a n s f e r activity was e l u t e d w l t h 250 ml o f 50X I v i u ) e t h y l e n e g l y c o l . 0 . 1 M Tvis-HCl, pH 6.8, and I d i t h i o t h r e i - t o 1 (Fig. 3 ) .

61 Phenyl-Sepharore Chromato r a h The combined a c t i v e f r a c t i o n s wele d i a l y z e d

purification and some properties of the glycolipid transfer protein

Documents