A novel cholesterol transfer protein in cardiac sarcolemma. Purification and initial characterization

Molecular and Cellular Biochemistry 100: 51-59, 1991. © 1991Kluwer Academic Publishers. Printed in the Netherlands. Original Article

A novel cholesterol transfer protein in cardiac sarcolemma. Purification and initial characterization

J. Santiago-Garc/a and J. Mas-Oliva Instituto de Fisiologia Celular, Universidad Nacional A u t 6 n o m a de MOxico, Mdxico

Key words: cholesterol transfer protein, plasma membrane, cardiac muscle cell

Summary In contrast to several sterol carrier proteins isolated from soluble cytosolic fractions, a cholesterol transfer protein (CHTP) with an apparent molecular weight of 73,000 was isolated from a cardiac sarcolemmal fraction by detergent solubilization, column chromatography, and preparative electrophoresis using nondissociating polyacrylamide gels. This protein must be reconstituted into an artificial membrane in order to mediate cholesterol transfer activity. For the expression of its full activity, CHTP must also be present in the membrane in a multimeric form, since the monomer was shown not to be active. We believe this novel protein might represent an important molecule in the regulation of the homeostasis of cholesterol in cardiac sarcolemma.

Introduction Several kinds of proteins that accelerate the exchange of different types of lipids between membranes, have been purified and characterized [1, 2]. Some of these cytosolic proteins have recently been used to study the distribution and transmembrane migration of phospholipids, fatty acids, and cholesterol [3, 4]. Studies by Brown and Goldstein [5, 6] have demonstrated that cholesterol homeostasis is mainly regulated by two proteins: the cell surface receptor for the low-density lipoprotein that specifically binds this cholesterol-carrying lipoprotein, and the 3-hydroxy-3-methylglutaryl-coenzyme A reductase, the rate-controlling enzyme in the internal synthesis of cholesterol [5, 6]. However, although both the endogenous cholesterol synthesized in the endoplasmic reticulum [7] and the exogenous cholesterol are rapidly transported throughout the liver cell and integrated in all intracellular membranes [8], the regulatory mechanisms for the

transport of cholesterol, and specifically the way cholesterol is integrated and withdrawn from plasma membranes are not well understood. Rat liver cytosol contains several proteins called sterol carrier proteins (SCP). Sterol carrier protein-1 apparently participates in the microsomal conversion of squalene to lanosterol, and SCP-2 in the conversion of lanosterol to cholesterol [9, 10]. A cholesterol-binding protein, also isolated from rat liver cytosol, has been reported to enhance the rate of cholesterol exchange between rat liver microsomes and mitochondria [11]. Because cytosols from several extrahepatic tissues including heart, have been shown to be immunoreactive to a cholesterol-binding protein antiserum [12], it has been suggested that this protein or a similar class of proteins might be involved in the intracellular transport of cholesterol in different tissues. On the other hand, an important group of blood plasma proteins involved in the transfer of neutral lipids and phospholipids between plasma lipoproteins, have been isolated [13-17]. In light of the

52 mechanism of transfer of cholesterol between the erythrocyte and plasma, it has been suggested that aqueous diffusion might govern the movement of cholesterol between compartments [18]. However, lipoproteins may also participate in the exchange process [19]. The present study describes the isolation and initial characterization of a protein from cardiac muscle sarcolemma that binds cholesterol, and facilitates the transfer of cholesterol from a serum/ buffer system into artificial membranes. These characteristics most probably reflect the properties of the cholesterol transfer protein in the sarcolemma.

Tris-HC1 pH7.4, I mM EDTA. The flow rate employed was 10 ml/h, and 0.5 ml fractions were collected at 4 ° C. After chromatography of the sample and collection of the first two peaks (peaks A and B), the column was washed with the same buffer until the optical density of the effluent was negligible. Further elution was achieved washing the column with 75 ml of a buffer containing 10 mM Tris-HC1 pH 7.4, 0.1% Triton X-100, 1 mM EDTA, and step increments in the ionic strength employing 0.1, 0.3, and 1.0 M KC1. Tubes containing the eluted proteins were pooled, concentrated at 4° C with a YM5 Amicon membrane, and tested for their cholesterol-binding activity in non-dissociating 5.0% polyacrylamide gels.

Materials and methods

Chemicals Cholesterol, phosphatidylserine, phosphatidylcholine type II-S and, cholesteryl hemisuccinate-agarose were obtained from Sigma Chem. Co. (St. Louis, MO, USA). [7-3H]cholesterol (specific activity 12 Ci/mmol) from ICN Radiochemicals (Irvine, CA, USA). All reagents used were of the highest quality.

Membrane preparation, solubilization and column chromatography The sarcolemmal preparation was isolated from New Zealand White rabbit hearts as previously described by us [20, 21]. Membranes were solubilized for 30 min at 4°C with gentle stirring, employing a medium containing 10mM Tris-HCl pH 7.4, 1 mM EDTA, 10 mM KC1 and 0.1% Triton X-100 (final concentrations) and centrifugated at 60 000 × g for 30 min at 4 ° C. The membrane protein to Triton X-100 ratio was maintained at 1.0/ 0.2. The supernatant was recovered and immediately applied to a cholesteryl hemisuccinate-agarose column (CHS-agarose column 1.0 × 5.0cm). The sample containing 10-15 ml was applied to the preequilibrated CHS-agarose column and eluted with the elution buffer (EB) containing 10 mM

Non-dissociating polyacrylamide gel electrophoresis and [3H]cholesterol binding Five percent Weber and Osborn gels [22] in the absence of SDS were run at constant 200 mA. A protein concentration of 100/~g was measured by the Bensadoun and Weinstein method [23] was usually employed. After electrophoresis, unfixed and unstained gels were incubated for one hour at 37°C in a medium containing 100 mM NaHPO4/ NaH2PO4 pH7.1), 10% DMSO, 100/zM cholesterol, and 10/zCi [7-3H]cholesterol/ml. After this procedure was carried out, the gels were washed five times with the same buffer without [3H]cholesterol, followed by slicing. Control gels were silver stained with the method described by Oakley et al. [24]. Following the same procedure, we used unfixed and unstained sliced gels for the specific extraction and isolation of the fractions named CI, CII, CIII, and CIV. Minced gels were placed in 20ml of a phosphate buffer (25mM NaHPO4/ NaH2PO4 pH 7.1), at 4°C for 24 h filtered through Millipore membranes (0.45/~m), and concentrated with an Amicon microconcentration unit employing a YM5 membrane. In a typical experiment we employed two gels (14.0 × 14.0 × 0.15 cm) loaded with 1.6 mg of protein from peak C obtained from the CHS-agarose column. From this procedure between 70 to 100/zg of CI fraction were obtained.

53

SDS-polyacrylamide gel electrophoresis

ried out at 37° C as previously described by us [20, 21]. At the end of the incorporation incubation, the liposomes were placed on Millipore filters (0.45/zm) and thoroughly washed with a 50mM Tris-HC1 buffer, pH7.4. The filters were transferred to scintillation vials, and 5 ml of tritosol added before counting in a Packard scintillation counter.

Fractions CI, CII, CIII and CIV obtained from the non-dissociating gel system were run in 7.5% Laemmli gels [25] at 10°C using 30 mA. The protein load was measured by the Bensadoun and Weinstein method [23]. After electrophoresis, the gel was fixed and stained with the method described by Oakley et al. [24].

Antibody characterization [3H]cholesterol incorporation into reconstituted liposomes

Sheep IgGs anti-rabbit CHTP were obtained following a classical inoculation scheme using the CI fraction isolated from native gels and complete Freund's adjuvant. IgGs were isolated using a protein A-Sepharose column [26]. The initial characterization of the antibody was performed using the Ouchterlony technique [27].

The different fractions obtained from the non-dissociating gel system were concentrated with a YM5 Amicon membrane. Using a lipid to protein ratio of 1,000/1, fractions CI, CII, CIII, CIV and cytochrome oxidase were incorporated into preformed unilamellar phosphatidylcholine/phosphatidylserine (9/1) liposomes, suspended in a 50mM TrisHCI buffer, pH 7.4. The mixture was sonicated for 1 min in a Brandsonic 8852 water bath sonicator. The incorporated protein was measured following the Bensadoun and Weinstein method [23]. The [3H]cholesterol incorporation procedure was car-

peak A

E c

Western Blots Proteins separated either with native or SDS polyacrylamide gels were transferred to nitrocellulose sheets as conventionally described [28], using the

peak B II

peak C

I

I

I

1,5

0 aO 1:7

1,0 o c-

,IIMKc,

o

c0 0.5

I0

20

.'30

40

50

60

70

8'0

FraGtion number Fig. 1. Continuous elution profile of the Triton X-100 solubilized sarcolemma on a CHS-agarose column (EB, elution buffer; S, 60,000 x g supernatant). For details see Materials and Methods section.

54

Fig. 2. Non-dissociatingpolyacrylamidegel electrophoresisof peak C obtained from the CHS-agarosecolumn. Gels incubated in a

[3H]cholesterolmedium(O). Gelspreincubatedwithunlabelledcholesterolbeforeincubationwith [3H]-cholesterol(Ik). For detailssee Materials and Methods section. 2117-250 LKB Nova Blot electrophoretic transfer kit. Nitrocellulose sheets with the different transferred proteins were first incubated overnight at 4°C with sheep IgGs anti-rabbit CHTP previously isolated with a protein A-Sepharose column. The reaction was visualized with the Vectastain-ABC kit (Vector Laboratories, Inc.).

Results

The procedure developed in this study for the isolation of the cholesterol-transfer protein consisted in the elution of a CHS-agarose column with step changes in ionic strength. As shown with other systems, CHS-agarose can be used in the isolation of cholesterol oxidase [29] and serum lipoproteins [30]. For the serum lipoproteins, specific desorp-

tion was achieved with detergents that apparently decrease the affinity of bound proteins to the adsorbent. Therefore, since we used the properties of the CHS-agarose for the selective separation of proteins in a detergent medium, our CHS-agarose column might be considered a weak affinity column. Elution of the CHS-agarose column resulted in several protein peaks (Fig. 1). Peak C, which eluted with a 100mM KC1 buffer, was concentrated and subjected to a further purification step employing preparative non-dissociating polyacrylamide gel electrophoresis. Among the different bands obtained, four fractions called CI, CII, CIII, and CIV were recovered from these preparative gels (Fig. 2). The CI fraction was able to bind [3H]cholesterol while embedded in the acrylamide gel (Fig. 2). When the gels were previously incubated with non

55 labeled cholesterol, [3H]cholesterol binding to the CI fraction was completely abolished (Fig. 2). The high background observed when the non labeled cholesterol was employed, could be due to the absorption of DMSO to the gel, that in turn could have facilitated the unspecific incorporation of [3H]cholesterol during the second incubation period. Although the non-dissociating gels were not successfully calibrated, the sequence of apparent molecular weights of CI, CII, CIII, and CIV suggested that these fractions might correspond to multiples of each other (Fig. 2). Therefore, the possibility of having isolated an active multimeric protein was initially considered. This possibility was primarily tested using SDSpolyacrilamide gel electrophoresis of the isolated CI, CII, CIII and CIV fractions. Despite the differences in the apparent molecular weights observed for the different fractions when the non-dissociating preparative gels were employed, interestingly, all four fractions exhibited the same main band under dissociation conditions using the SDS system (Fig. 3). The main band corresponded to a molecular weight of 73,000 daltons when silver stained (Fig. 3). This set of results together with the observation of [3H]cholesterol binding to the CI fraction, support the idea that under the conditions employed in this study, the 73,000 dalton protein isolated from cardiac sarcolemma might correspond to a protein with cholesterol binding activity only when associated in a multimeric form. To further test this hypothesis, fractions CI, CII, CIII, and CIV recovered from the preparative gels, were concentrated and incorporated by sonication into preformed unilamellar liposomes. In contrast to fractions CII, CIII, and CIV, which behaved like the control liposomes, the membranes containing the CI fraction showed a better capacity to incorporate cholesterol into the liposomal membranes (Fig. 4). The incorporated CI fraction transferred 95-115 pmoles of cholesterol/pmole of protein/h. The addition of albumin to the liposome incubation mixtures, substantially interfered with the incorporation of [3H]cholesterol, since the level of incorporation of the labelled lipid dropped be-

Fig. 3. Sodium dodecyl sulfate polycrylamide gel electrophoresis of fractions CI, CII, CIII and CIV, obtained from the nondissociating gel system, a) molecular weight standards; b) sarcolemmal fraction; c) 60,000 g supernatant; d) peak C; e) fraction CI; f) fraction CII; g) fraction CIII; h) fraction CIV. For details see Materials and Methods section.

low the control values of those samples incubated without albumin. Moreover, the reconstitution of cytochrome oxidase into the liposomal membranes, studied as a control measurement employing another integral membrane protein, did not increase the incorporation of [3H]cholesterol into the artificial membranes (Fig. 4). This result suggests that the incorporation of labelled cholesterol to the membrane, is not due to an unspecific effect given by intrinsec proteins in general, but due to the presence of CHTP. Moreover, these results support the conclusion from our early experiments showing the cholesterol binding properties of the CI complex and the requisite to have the 73,000 dalton protein in a multimeric form for the observation of its full cholesterol transfer activity. Based on these properties, the protein was named cholesterol transfer protein or CHTP. The initial characterization of the anti-CI (CHTP) antibodies was performed using the Ouch-

56

I00

0

~

- I00

.o 60

1:7

.O_o -

-

60

tl) + -

_5

:~

tO 0_. ~) 0 I_. t~.

40

~_~ 0

20-

-

-

40

0

"-~ "t~ - 20

-i0 a

c

"-"

20-

~

0

A

CO

CI

c'n" cTrr CI~Z

Fig. 4. [3H]cholesterol incorporation into preformed liposomes reconstituted with fractions CI, CII, CIII and CIV. Co, cytochrome oxidase, A, bovine serum albumin; C, control liposomes without incorporated protein. For details see Materials and Methods section.

Fig. 5. Double immunodiffusion in Ouchterlony plates. Immunoreaction was carried out at 20° C for 36 h. Gel stained with amido black. Center well, 3/zg of CHTP obtained from native gels. Outer wells filled with antiserum dilutions: a) 1 : 64; b) 1 : 32;c) 1: 16; d) 1: 8; e) 1: 4; f) 1: 2.

terlony method. Figure 5 shows the precipitation obtained for the antibodies contained in the sheep serum, reacting against CHTP. The single clear cut precipitation bands obtained for each one of the serum dilutions, demonstrate a high concentration of antibody in the serum, and a good antigen/antibody reactivity. In order to further characterize the apparent multimeric nature of CHTP, anti-CI sheep antibodies isolated with a protein-A Sepharose column were used. The antibodies were tested against the CI, CII, CIII and CIV fractions previously separated in a non-dissociating polyacrylamide gel and transferred to nitrocellulose. This set of experiments showed an important crossreactivity between the antibody and the CI, CII and CIV fractions (Fig. 6). Interestingly, the CI fraction when analyzed using polyacrilamide gel electrophoresis under dissociating conditions, showed the 73,000 dalton band that cross reacted with the anti-CHTP antibody (Fig. 7). However, a second band that runs below the 73,000 dalton band, was also apparent. This protein that could correspond to an im-

57

Fig. 6. Western blot of proteins contained in peak C. Proteins of peak C were separated in non-dissociating poliacrylamide gels, transferred to nitrocellulose paper and reacted with an IgG fraction anti-CI (1/500). Lane A, silver stained gel; Lane B western blot.

munologically related protein, or to a degradation product of the 73,000 dalton protein, will have to be further studied. Since there is a possibility that CHTP might correspond to a glycosilated protein, we will also explore if the two bands might represent the same protein with different carbohydrate content. The above results suggest that the 73,000 dalton protein apparently corresponds to the monomer that conforms the CI, CII and CIV complexes. The finding that the CIII fraction did not present cross reactivity with the antibody, could be explained in terms of a limited protein concentration that does not reaches the minimum required for visualization of the reaction. Nevertheless, higher amounts of antibody and longer development times did not modify the lack of reactivity under the conditions employed for this set of experiments. The possibility that CIII may be a protein not related to the other subunits is currently under investigation.

Fig. 7. Western blot of CHTP. The CI fraction obtained from non-dissociating polyacrylamide gels was run in 7.5% SDSpolyacrylamide gels and transferred to nitrocellulose. The nitrocellulose sheets were incubated with a sheep IgG anti-CI fraction (1/2000). Lane A, molecular weight markers; Lane B, CHTP.

Discussion

One of the interesting questions in cell biology concerning the homeostasis of cholesterol corresponds to the understanding of the control mechanisms that govern the nonhomogeneous distribution of cholesterol among the different membranes of the cell. Nowadays it is well known that cholesterol is found in relatively high concentrations in mammalian plasma cell membranes, much lower in endoplasmic reticulum, and lower yet in mitochondrial membranes. However, the factors controlling the establishment and maintenance of this cholesterol distribution are poorly understood and important to investigate. Since we have proposed that changes in the sarcolemmal cholesterol concentration might alter the equilibrium of cytoplasmic calcium through changes in the sarcolemmal calcium pump [20, 21], we believe that alterations in the expression or regu-

58 lation of CHTP in the membrane might be of major biological significance in the normal physiology of the cardiac muscle cell. One of the explanations put forward to explain the important differences in cholesterol concentration between intracellular membranes and the plasma membrane, has been through a passive equilibration of cholesterol between cell membranes and plasma lipoproteins. This proposal is based on the diffusion of molecules and transient collisional contacts [31, 32]. Nevertheless, specific sites of association that could facilitate the transfer of cholesterol through stable complexes have been proposed but much less studied [31]. We believe that CHTP, the novel protein described in this study might be representative of an entirely new class of membrane proteins involved in the regulation of the homeostasis of cholesterol in cardiac sarcolemma and probably the plasma membrane of other cell types. The understanding of the basic structural and kinetic properties of cardiac CHTP will unable us to continue the exploration of the mechanisms that regulate this phenomenon.

4.

5. 6.

7. 8.

9.

10.

11.

12.

13.

Acknowledgements

14.

We would like to thank Dr. Armando G6mezPuyou for his continuous advice, Dr. Antonio Pefia for his generous gift of yeast cytochrome oxidase, and Mrs. M. Elena Guti6rrez for the secretarial work. This work was in part supported by a grant from CONACyT (PCEXCNA050747). J. MasOliva; recipient of a fellowship from the John Simon Guggenheim Memorial Foundation.

15.

16.

17.

18.

References 1. Zilversmit DB, Hughes ME: Phospholipid exchange between membranes. In: ED Korn (ed.) Methods in Membrane Biology. Vol 7 Plenum Press, New York, 1976, pp 211-255 2. Bloj B, Zilversmit DB: Rat liver proteins capable of transferring phosphatidylethanolamine. J Biol Chem 252: 16131619, 1977 3. Rothman JE, Dawidowicz EA: Asymmetric exchange of

19.

20.

21.

vesicle phospholipids catalized by the phosphatidylcholine exchange protein. Measurement of inside-outside transitions. Biochemistry 14: 2809-2816, 1975 Lenard J, Rothman JE: Transbilayer distribution and movement of cholesterol and phospholipid in membrane of influenza virus. Proc Natl Acad Sci USA 73: 391-395, 1976 Brown MS, Goldstein JL: A receptor-mediated pathway for cholesterol homeostasis. Science 232: 34-47, 1986 Goldstein JL, Brown MS: The low-density lipoprotein pathway and its relation to atherosclerosis. Ann Rev Biochem 46: 897-930, 1977 O'Malley BW, Means AR: Female steroid hormones and target cell nuclei. Science 183: 610-620, 1974 Colbeau A, Nachbaur J, Vignais PM: Enzymic characterization and lipid composition of rat liver subcellular membranes. Biochim Biophys Acta 249: 462-492, 1971 Srickantaiah MV, Hansberg E, Loughan ED, Scallen TJ: Purification and properties of sterol carrier proteinl. J Biol Chem 251: 5496-5504, 1976 Noland BJ, Arebalo RE, Hansbury EH, Scallen TJ: Purification and properties of sterol carrier protein2. J Biol Chem 255: 4282-4289, 1980 Bell FP: Cholesterol exchange between microsomal, mitochondrial and erythrocyte membranes and its enhancement by cytosol. Biochim Biophys Acta 398: 18-27, 1975 Erickson SK, Meyer DJ, Gould RG: Purification and characterization of a new cholesterol binding protein from rat liver cytosol. J Biol Chem 252: 1817-1826, 1977 Morton RE, Zilversmit DB: Purification and characterization of lipid transfer protein(s) from human lipoproteindeficient plasma. J Lipid Res 23: 1058-1067, 1982 Barter PJ, Hopkins GJ, Calvert DG: Transfer and exchanges of sterified cholesterol between plasma lipoproteins. Biochem J 208: 1-7, 1982 Hesler CB, Swenson TL, Tall AR: Purification and characterization of a human plasma cholesteryl ester transfer protein. J Biol Chem 262: 2275-2282, 1987 Jarnagin AS, Kohr W, Fielding C: Isolation and specificity of a Mr 74,000 cholesteryl ester transfer protein from human plasma. Proc Natl Acad Sci USA 84: 1854-1857, 1987 Tollefson JH, Ravnik S, Albers JJ: Isolation and characterization of a phospholipid transfer protein (LTP-II) from human plasma. J Lipid Res 29: 1593-1602, 1988 Lange Y, Molinaro AL, Chauncey TR, Steck TL: On the mechanism of transfer of cholesterol between human erythrocyte and plasma. J Biol Chem 258: 6920-6926, 1983 Lange Y, Matthies JGH: Transfer of cholesterol from its site of synthesis to the plasma membrane. J Biol Chem 259: 14624-14630, 1984 Ortega A, Mas-Oliva J: Cholesterol effect on enzyme activity of the sarcolemmal (Ca2+,Mg2+)-ATPase from cardiac muscle. Biochim Biophys Acta 773: 231-236, 1984 Ortega A, Mas-Oliva J: Direct regulatory effect of cholesterol on the calmodulin stimulated calcium pump of cardiac sarcolemma. Biochem Biophys Res Commun 139: 868-874, 1986

59 22. Weber K, Osborn M: The realibility of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J Biol Chem 244: 4406--4412, 1969 23. Bensadoun A, Weinstein D: Assay of proteins in the presence of interfering materials. Anal Biochem 70: 241-251, 1976 24. Oakley RB, Kirsch RD, Morris RN: A simplified ultrasensitive silver stain for detecting proteins in polyacrilamide gels. Anal Biochem 105: 361-363, 1980 25. Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature Lond 227: 680--685, 1970 26. EyPL, ProwseSJ, JenkinCR: IsolationofpureIgGl, IgGza and IgGzbimmunoglobulins from mouse serum using Protein A-Sepharose. Immunochemistry 15: 429-436, 1978 27. Ouchterlony 13: Diffusion-in-gel. Methods for immunological analysis. In: P Kallos (ed.) Progress in Allergy, Vol V, Basel: Karger, 1958, pp 1-78 28. Towbin H, Staehlin T, Gondon J: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose

29.

30. 31.

32.

sheets. Procedure and some applications. Proc Natl Acad Sci USA 76: 4350-4354, 1979 Fukuyama M, Miyake Y: Purification and some properties of cholesterol oxidase from Schizophyllum commune with covalently bound flavin. J Biochem 85: 1183-1193, 1979 Wichman A: Affinity chromatography of human plasma low-density lipoproteins. Biochem J 181: 691-698, 1979 Phillips MC, Johnson WJ, Rothblat GH: Mechanism and consequences of cellular cholesterol exchange and transfer. Biochem Biophys Acta 906: 223-276, 1987 Wetterau JR, Zilversmit DB: Quantitationoflipidtransfer activity. In: D Glick (ed.) Methods of Biochemical Analysis, Vol 30, John Wiley & Sons, New York, 1984, pp 199226

Address for offprints: J. Mas-Oliva, Departamento de Bioenergdtica, Instituto de Fisiologia Celular, Universidad Nacional Aut6noma de M6xico. Apartado Postal 70-600, 04510 M6xico, D.F. M6xico