Cholesterol uptake capacity of Acholeplasma laidlawii is affected by the composition and content of membrane glycolipids

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 248, No. 1, July, pp. 282-288, 1986

Cholesterol Uptake Capacity of Acholeplasma laidlawii Is Affected by the Composition and Content of Membrane Glycolipids HAVA The Hebrew

EFRATI,

YOHANAN

WAX,*

AND

University-Hadassah Medical School, Jerusalem; The Hebrew University of Jerusalem, Received

September

3,1985,

and in revised

form

SHLOMO

ROTTEM’

and *Department Israel January

of Statistics,

30, 1986

The composition of the cell membrane of 20 Acholeplasma laidlawii strains grown under identical conditions was studied and correlated with the capacity of these strains to incorporate cholesterol. Membranes of these strains had similar sodium dodecyl sulfate-polyacrylamide gel electrophoresis patterns and contained the same lipid species, but the relative amounts of the major polar lipids varied. Statistical analyses revealed that the glycolipids, monoglucosyldiglyceride, and an unidentified glycolipid (glycolipidX) succeeded in explaining 90% (R2 = 0.90) of the cholesterol uptake variations. The regression coefficients for both glycolipids were negative (P < O.OOl), indicating that the capacity of A. laidlawii strains for cholesterol incorporation is inversely proportional to the relative amounts of these glycolipids. Accordingly, an increased capacity for cholesterol incorporation was detected upon aging of A. laidlawii cells. The aged cells contained significantly smaller amounts of both monoglucosyldiglyceride and glycolipidX, and a higher amount of diglucosuldiglyceride. The change in cholesterol incorporation as a response to glycolipid composition and content can be explained by the low solubility of cholesterol in glycolipids as well as by the induction by the sterol molecule of a nonlamellar phase state that will destabilize a membrane structure containing monoglucosyldiglyceride and glycolipid-X. o 1986 Academic press, lnc. Cholesterol is an important component of biological systems. A problem of biological significance concerns the mechanism by which cells control the amount of cholesterol incorporated into their plasma membrane. Mycoplasmas offer several unique advantages for investigating this problem. Unlike other prokaryotes, these organisms require cholesterol for growth (1,2) and their plasma membrane interacts directly with the exogenous cholesterol donors (2). Moreover, mycoplasma lipid composition can be manipulated in a controlled manner (4), facilitating the study of the factors influencing cholesterol incorporation. Different mycoplasmas vary widely in their cholesterol content when grown 1 Author dressed.

to whom

0003-9861186 Copyright All rights

correspondence

$3.00

0 1986 by Academic Press. Inc. of reproduction in any form reserved.

should

with the same supply of exogenous cholesterol. The cholesterol content of membranes of Mgcoplasma, Ureaplasma, and Spiroplasma species is much higher than that found in membranes of the Acholeplasma species, which were found to have a reduced capacity for cholesterol incorporation (5). Elucidating the factors responsible for the restricted cholesterol incorporation by Acholeplasma species may lead to a better understanding of the mechanisms controlling cholesterol incorporation. In previous studies it was shown that cholesterol incorporation into mycoplasmas is not influenced by variations in membrane protein content, whereas digesting membrane phospholipids markedly decreased the amounts of cholesterol incorporated (5,6). These observations led to the suggestion that the amount of choles-

be ad-

282

GLYCOLIPIDS

AFFECT

CHOLESTEROL

terol incorporated into mycoplasma membranes depends on the amount of phospholipid available for interaction with it (2,6), as well as on the physical properties of the lipids (6,7). It was therefore of interest to try to correlate the restricted cholesterol levels of Acholeplasma laidlawii with the unique lipid components in its cell membrane. Our statistical evaluation of the data obtained with twenty A. laidlawii strains differing in the relative amounts of membrane lipid species revealed a significant negative correlation between certain membrane glycolipids and cholesterol incorporation. MATERIALS

AND

METHODS

Organisms and growth conditions. Twenty A. laidlawii strains were used throughout this study. The strains are presented by their original designation. For convenience of data presentation numerical designations (Sl-S20) were assigned to the strains (in parentheses). Strains 992 (Sl), 7’76 (S5), 1088 (S8), 1023 (S9), 1462 (SlO), 1191 (Sll), 1098 (S13), 815 (S14), 944 (S15), 1150 (S17), 1264 (S18), 1308 (S19), and 1012 (S20) were kindly given to us by Dr. M. F. Barile (Bethesda, Md.). Strains 1305 (S2), GR3 (S3), OR (S4), JAl (S7), GRl (S12), and GR2 (S16) were taken from our collection. The strains were grown in a modified Edward medium (8) containing 0.5% (w/v) bovine serum albumin (Fraction V, Sigma) adjusted to pH 8.5. The medium was supplemented with either a mixture of elaidic and palmitic acids (17.5 and 2.5 pg/ml, respectively), or a mixture of palmitic and elaidic acids (10 pg/ml of each). To label membrane lipids, 0.2 pCi of [1-i4C]palmitic acid (56 Ci/mol, The Radiochemical Centre, Amersham, U.K.) was added to each liter of the growth medium. The organisms were harvested after 20-24 h of incubation at 37”C, at the late exponential phase of growth (Am of 0.30-0.40) by centrifugation at 90008 for 15 min. The sedimented cells were washed once with 0.25 M NaCl and cell membranes were isolated by osmotic lysis of the organisms (9). Cholesterol uptake. Cholesterol/phosphatidylcholine vesicles served as cholesterol donors. The vesicles were prepared by pipetting 2 ml of egg phosphatidylcholine solution (15 mg/ml) in chloroform (Makor Chemicals, Jerusalem, Israel), 1 ml of cholesterol solution (13.5 mg/ml in chloroform, Sigma, St. Louis, MO.), and 10 &i of pH]cholesterol (10 Ci/mmol, The Radiochemical Centre, Amersham, U.K.) into 20-ml test tubes. The solvent was evaporated under nitrogen. The dried lipids were then dispersed in 10 ml of 0.05 M sodium phosphate buffer, pH 7.0, solution by ultrasonic ir-

UPTAKE

BY A.

laidlawii

283

radiation for 45 min at 0°C under nitrogen in a W350 Heat Systems sonicator operated at 50% duty cycle with a large probe at 160 W. The lipid dispersions were then centrifuged at 30,OOOg for 30 min. The sediment containing large liposomes was discarded. The supernatant fluid containing single lamellar vesicles (10) was stored at 4°C and was used within 2 days of preparation. Cholesterol uptake from the vesicles was tested with isolated A. laidlawii membranes. The uptake mixture consisted of 50 mM phosphate buffer, pH 7.9, containing the cholesterol/phosphatidylcholine vesicles (to yield 200 pg/ml) and isolated membranes (1.0 mg protein/ml). The suspension was incubated at 37°C in a shaking bath, 0.9-ml samples were withdrawn at various time intervals, and membranes were collected by centrifugation in an Eppendorf 5414 microfuge for 4 min. Under these conditions 80 f 10% of the membranes were sedimented. The supernatant fluid was removed by suction and the tip of the plastic centrifuge tube (containing the membrane pellet) was cut off with a razor blade, placed in a vial containing 5 ml of scintillation fluid, and counted. Cholesterol uptake was presented as [‘HIcholesterol incorporated per membrane mass (%-labeled lipids) per hour corrected for zero time. As the turnover and/or release of membrane polar lipids of A. laidlawii were found to be low nonexistent, radioactivity derived from :‘C]palmitate remained constant throughout the incubation period. Statistical methods. The correlations between the various polar lipid species of A. luidlawii as well as between the capacity of the membrane preparations to incorporate cholesterol and membrane lipid composition were examined by Pearson Correlation Coefficients and through multiple linear regression analysis (11). Mallous Cp statistic (11) was used to select the subset of lipids highly associated with cholesterol. Computations were carried out by the BMDP-9R computer program. Analytical procedures. Protein was determined according to the method of Lowry et al. (12). Lipids were extracted from the membrane preparations by the method of Bligh and Dyer (13). Polar lipids were estimated according to radioactivity derived from [‘%]palmitate. Thin layer chromatography of polar lipids was performed on silica gel HR (Merck) plates developed with chloroform/methanol/water/acetic acid (65:25:4:1, v/v/v/v) (15). Lipid spots were detected by iodine vapor, phospholipid spots by molybdate spray reagent (16), and glycolipids by anthrone reagent (17). Chemical analysis of lipid species was performed as previously described (15). To determine radioactivity, lipid spots were scraped off the plate into scintillation vials containing 3 ml of toluene scintillation liquor. Radioactivity in membrane preparations was determined using 3 ml of toluene-triton scintillation liquor. Radioactivity was measured in a Model

284

EFRATI,

WAX,

460CD Packard Tri Carb scintillation spectrometer and expressed as decompositions per minute (dpm).

RESULTS

All 20 A. laidlawii strains were found to contain the seven typical major polar lipids previously described (3, 4), i.e., three glycolipids, monoglucosyldiglyceride, diglucosyldiglyceride, and an unidentified glycolipid (glycolipid-X); two phospholipids, phosphatidylglycerol and diphosphatidylglycerol; and two phosphoglucolipids, glycerophosphorylmonoglucosyldiglyceride and glycerophosphoryldiglucosyldiglyceride. Sodium dodecyl sulfate (SDS)polyacrylamide gel electrophoresis of SDSsolubilized membranes revealed very similar polypeptide patterns for all strains throughout the study (not shown). Table I shows the incorporation of [i4C]palmitate into the various lipid species of the strains tested. Since palmitate was shown to be incorporated into the various A. laidlawii lipids to about the same extent (14), the incorporation of [14C]palmitate was taken as a measure to estimate the relative amounts of membrane lipids. The total membrane lipids of the various strains tested was almost identical (0.70 +- 0.05 mg per mg membrane protein). As A. laidlawii membrane lipids are known to be markedly affected by the growth phase of the cells (18,19) and to be dependent on the fatty acid composition of the growth medium (19), all of our comparative experiments were carried out with strains grown in an identical medium and harvested at the same phase of growth. In most experiments the cells were grown with elaidic and palmitic acids (17.5 and 2.5 pg per ml, respectively) to the late exponential phase of growth. Under these conditions the differences in the relative amounts of polar lipids among the various strains were most pronounced. Table II presents the matrix of Pearson correlation coefficients among the polar lipid species. Negative correlations were found within the glycolipid species, phosphoglycolipid species and phospholipid species. However, a systematic pattern was not observed in

AND

ROTTEM

GLYCOLIPIDS

AFFECT

CHOLESTEROL TABLE

UPTAKE

BY

II

PEARSON CORRELATIONCOEFFICIENTSBETWEEN A. l&dkzwiiP~u~L~~~~ ANDBETWEENTHELIPIDSPECIESANDCHOLESTEROL' Lipid species b

GL-X

MGDG

DGDG

GL-X MGDG DGDG GPMGDG GPDGDG PG DPG

-0.517 0.043 -0.410 -0.181 0.115 0.070

-0.189 0.402 -0.296 0.481 -0.481

-0.110 0.353 0.163 -0.502

Cholesterol

-0.793

-0.032

285

hidlutii

A.

SPECIES

GPMGDG

GPDGDG

PG

-0.141 -0.190 -0.339

0.022 -0.296

-0.670

0.424

-0.413

0.137

0.147

DPG

0.194

a Twenty strains of A. luidluwii were grown in a modified Edward medium (8) containing elaidic and palmitic acids and 0.0002 &i/ml [‘%]palmitic acid to a late exponential phase of growth (Am = 0.30-0.40). Membrane isolation, lipid analyses, and correlation coefficients were determined as described under Materials and Methods. *Abbreviations of lipid species are as in Table I.

the correlations between lipids of the various groups. The capacity of the various A. laidlawii strains to incorporate cholesterol from small unilamellar lipid vesicles containing phosphatidylcholine-cholesterol (1:0:9 molar ratio) is also presented in Table I. As is apparent, the rate of cholesterol uptake varied to a large extent among the strains tested. With all strains the uptake was linear for at least 6 h of incubation at 3’7°C. Pearson correlation coefficients between cholesterol uptake and the various membrane lipids revealed significant negative correlation (r = -0.793, P < 0.01) only with glycolipid-X. As significant intercorrelations were shown between glycolipid,

phospholipid, and phosphoglycolipid species (Table II), linear regression models were utilized. The fit of the appropriate linear regression model is summarized in Table III. Two lipid species were highly associated with cholesterol uptake, according to Mallous Cp criterion (11). These two lipids succeeded in explaining 90% (R2 = 0.90) of the cholesterol uptake variation. The regression coefficients for glycolipid-X and monoglucosyldiglyceride are both negative (P < O.OOl), indicating that the capacity of A. laidlawii strain for cholesterol incorporation is inversely proportional to the relative amounts of these glycolipids. Deletion of glycolipid-X from the regression equation decreases the square multiple

TABLE

III

LINEARREGRESSIONANALYSISOFCHOLESTEROL UPTAKEANDLIPIDSPECIES Regression coefficients of each lipida (P values in parentheses) Model 1

GL-X

MGDG

-37.86 (
-14.55 (
2

’ Abbreviations

GPGDG

GPMGDG

species

DGDG

PG

R2 0.89

24.88 (0.006) of lipid

DPG

are as in Table

28.43 (0.024) I.

21.26 (0.027)

20.71 (0.056)

28.74 (0.193)

0.57

286

EFRATI,

WAX,

AND

ROTTEM

decreased from 0.90 mg lipid per mg membrane protein in membranes from cells harvested at the early exponential phase of growth (Am,, = 0.08) to 0.72 mg lipid per mg membrane protein in cells harvested at the late exponential phase of growth (Aa = 0.36). Likewise upon aging, marked differences in membrane lipid composition were found, mainly, a marked decrease in monoglucosyldiglyceride and glycolipid-X with a concomitant increase in diglucosyldiglyceride (Table IV). Phosphatidylglycerol and diphosphatidylglycerol were more stable and showed only small changes. The levels of phosphoglycolipids remained constant. The aging of the cells was associated with a pronounced increase in the capacity of isolated membranes to incorporate cholesterol from the phosphatidylcholinecholesterol vesicles. The rate of cholesterol uptake in late exponential cells (AM0 = 0.36) was three times higher than the rate obtained with early exponential phase cells.

correlation coefficient (R2) by more than twice the decrease associated with the deletion of monoglucosyldiglyceride. The relationships between cholesterol uptake and the remaining membrane lipids (phosphatidylglycerol, diphosphatidylglycerol, diglucosyldiglyceride, glycerophosphorylmonoglucosyldiglyceride, and glycerophosphoryldiglucosyldiglyceride) are also shown in Table III. As expected, each of these lipids is positively associated with cholesterol uptake. It should be noted, however, that when this set of predictions of cholesterol uptake is applied to glycolipid-X and monoglucosyldiglyceride, the square multiple correlation coefficient increases by only 2%. Hence the residual membrane lipids do not contribute to the prediction value for cholesterol uptake beyond the one made by glycolipid-X and monoglucosyldiglyceride. Changes in the content and relative amounts of membrane polar lipids could also be induced in each of the A. laidlawii strains by harvesting the cells at various growth phases. Upon aging of A. laidlawii GR3 the content of membrane lipids was

DISCUSSION

Glycolipids constitute a significant portion of A. laidlawii membrane polar lipids.

TABLE

IV

LIPID COMPOSITIONAND CHOLESTEROLUPTAKEOF A. ~~~~~~~%MEMBRANEs ISOLATED FROM CELLSHARVESTEDATVARIOUSPHASESOFGROWTH" Polar

Lipid

species*

GL-X MGDG DGDG DPG PG GPMGDG GPDGDG Cholesterol

uptake”

lipid

composition (mol%)of from cells harvested

membranes at:

Am = 0.08

Aa,, = 0.20

Aa,, = 0.36

19.4 21.0 10.6 13.7 16.0 11.2 10.8

15.6 14.5 15.0 17.0 14.5 12.6 10.6

11.0 12.6 19.4 19.8 13.0 12.0 11.4

57.5

90.5

175.0

'A. laidlatii strain GR3 (S3) was grown in a modified Edward medium (8) containing 10 pg/ml oleic acid and 10 pg/ml palmitic acid and 0.0002 &i/ml [‘4C]palmitic acid. At three time intervals (19,24, and 41 h of incubation) the cells were harvested and membranes isolated. Lipid analyses and cholesterol uptake were performed as described under Materials and Methods. *Abbreviations of lipid species are as in Table I. ‘Cholesterol uptake was expressed as the ratio of [3H]cholesterol incorporated per h per i4C-labeled membrane polar lipids.

GLYCOLIPIDS

AFFECT

CHOLESTEROL

The glycolipids consist primarily of monoglucosyldiglyceride, diglucosyldiglyceride and an unidentified glycolipid (glycolipidX) that possessed distinctive analytical differences when compared to the two other glycolipids. The RF values of glycolipid-X on thin layer chromatography plates were identical to those of the unidentified glycolipid observed by Wieslander and Rilfors (19). This glycolipid was recently studied by Bhakoo and McElhaney (personal communication) and tentatively identified as l-[O-a-D glucopyranosyl(1 - 15)triisoprenylmenaquinonel(2 - 1)palmitate. All cholesterol uptake experiments were performed with isolated membranes. It was previously shown that the differences in cholesterol incorporation between A. laidlawii and M. cap’colum cells are retained in isolated membranes (20), indicating that the mechanism restricting cholesterol uptake by A. laidlawii is physicochemical. This mechanism probably depends upon specific composition of A. laidlawii membranes rather than on active cholesterol exclusion by an energy dependent process (5). The possibility that cholesterol incorporation is influenced by variations in membrane protein is most unlikely, since changing the lipid to protein ratio by inhibiting protein synthesis by chloramphenicol(21) or intensive digestion of membrane proteins by proteolytic enzymes (6) has no influence on the cholesterol binding capacity of A. laidlawii cells. Furthermore, the 20 A. laidlawii strains utilized in our study had almost identical amounts of membrane protein and similar SDS-polyacrylamide gel electrophoresis patterns. These strains, however, show marked differences in relative amounts of their membrane polar lipids as well as in their capacity to incorporate cholesterol into their membrane. Our statistical data revealed that among membrane polar lipids of A. laidlawii, monoglucosyldiglyceride and glycolipid-X are strongly associated with cholesterol uptake. These glycolipids succeeded in explaining 90% of the variations in cholesterol uptake among the 20 A. laidlawii strains tested. The negative regression coefficients for both glycolipids indicate

UPTAKE

BY A.

hidhwii

287

that the restricted capacity of A. hidlawii for cholesterol incorporation is inversely proportional to the levels of these glycolipids in the membranes. The restricted capacity of A. laidhwii for cholesterol incorporation may depend on the lower affinity of the glycolipids for cholesterol as well as on the induction of nonlamellar phases in the presence of glycolipids forming reversed hexagonal liquid crystals (22, 23). The possibility that glycolipids have lower affinity for cholesterol than phospholipids was previously suggested by McCabe and Green (24), who demonstrated that lipid dispersions containing cerebrosides and gangliosides incorporated smaller amounts of cholesterol than lipid dispersions of choline containing phospholipids. Furthermore, previous data from our laboratory (6) showed that removal of phospholipids (phosphatidylglycerol and diphosphatidylglycerol) which comprise only 30% of the membrane polar lipids caused a decrease of over 55% in cholesterol uptake, suggesting that glycolipids have a lower binding capacity for cholesterol. Our attempts to vary the glycolipid composition by treating membranes with cu-glycosidases, growing the cells with decreasing concentrations of glucose (down to 0.05%) or by replacing glucose with mannose, fructose (0.5%) or glycerol (0.1 M) failed. However, variation in the glycolipid composition was obtained by aging the cells. As expected, the decrease in monoglucosyldiglyceride and glycolipid-X content upon aging was associated with increased capacity to incorporate exogenous cholesterol into their cell membrane. A relationship between cholesterol content and glycolipid composition in growing A. laidlawii cells has been previously described (25), showing that when cholesterol concentration was increased in the growth medium A. laidlawii cells responded by decreasing the monoglucosyldiglyceride to diglucosyldiglyceride ratio. It was therefore suggested that the metabolic response of A. laidlawii counteracts the bilayer-destabilization effect of cholesterol. As the wedge-shaped cholesterol molecule induces nonlamellar phase structures just as monoglucosyldiglyceride does (25, 26), the

288

EFRATI,

WAX,

simultaneous presence of cholesterol and monoglucosyldiglyceride has to be avoided. Our observation on the restricted capacity of isolated A. laidlawii membranes containing monoglucosyldiglyceride and glycolipid-X to incorporate cholesterol may be another consequence of a regulation intended to preserve a stable bilayer. Although the structure of glycolipid-X is as yet undefined, the tentative structure suggested (1) is more prone to form nonlamellar phase structures. This structure, characterized by a small polar head group and a broad hydrophobic part, will tend to form nonbilayer phase structures due to the reduced hydrocarbon-water interfacial area (23, 26). The increase in glycolipids that induces the formation of nonlamellar phase structures will therefore impose disturbances in the packing properties of A. laidlawii membranes that may restrict cholesterol incorporation. ACKNOWLEDGMENTS

Part of this work was supported by an Alberta Heritage Foundation Fellowship (S.R.). We would like to thank P. Knight for typing and proofreading the manuscript. REFERENCES 1. RAZIN, S. (1975) Prog. Surf: Memin-. Sci. 9, 257312. 2. RAZIN, S., AND ROTTEM, S. (1978) Trends B&hem, sci. 3,51-55. 3. SMITH, P. F. (1962) J. Bucterioi. 84,534-538. 4. ROTTEM, S. (1980) Biochim. Biophys. Acta 604,6590.

5. RAZIN, S., WORMSER, M., AND GERSHFELD, N. L. (1974)

B&him

Biophys.

Acta

352,385-396.

6. EFRATI, H., ROTPEM, S., AND RAZIN, S. (1981) Biochim

Biophys.

Acta

691,386-394.

AND

ROTTEM

7. MELCHIOR, D. L. (1982) in Membrane Lipids of Prokaryotes (Razin, S., and Rottem, S., eds.), pp. 263-316, Academic Press, New York. 8. RAZIN, S., AND ROITEM, S. (1976) in. Biochemical Analysis of Membranes (Maddy, A. H., ed.), pp. 3-26, Chapman & Hall, London. 9. RAZIN, S. (1963) J. Germ Microbial 33,471-475. 10. ROTTEM, S., SHINAR, D., AND BITTMAN, R. (1981) B&him

Biophys.

Aeta

649,572-580.

11. WEISBERG, S. (1980) Applied Linear Regression. Wiley, New York. 12. LOWRY, 0. H., ROSEBROUGH,J. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol Chem. 193, 265275. 13. BLIGH, E. G., AND DYER, W. J. (1969) Conad J. Biochem

PhysioL

37,911-917.

14. GROSS,Z., AND RO?TEM, S. (1979) B&him Biophys. Acta 555, 547-552. 15. ROITEM, S., AND MARKOWITZ, 0. (1979) Biochemistry

l&2930-2935.

16. DITTMER, J. C., AND WELLS, M. A. (1969) in Methods in Enzymology (Lowenstein, J. M., ed.), Vol. 14, pp. 482-530, Academic Press, New York. 17. VAN GENT, C. M., ROZELEUR, 0. J., AND VAN DER BIJL, P. (1973) J. Chromatogr. 85,174-176. 18. SMITH, P. F. (1972) Biochim. Biophys. Acta 280, 375-382. 19. WIESLANDER, A., AND RILFORS,L. (1977) Biochim. Biophys.

Acta 466,336-346.

20. RAZIN, S., KUTNER, S., EFRATI, H., AND ROTTEM, S. (1980)

Biochim

Biophys.

Ada

598,528~640.

21. RAZIN, S. (1974) FEBS L&t 47,81-85. 22. WIESLANDER, A., CHRISTIANSSON,A., RILFORS, S., KHAN, A., JOHANSSON,L. B.-A., AND LINDBLOM, G. (1981) FEBS L&t. 124,273-278. 23. WIESLANDER, A., CHRISTIANSSON,A., RILFORS,L., AND LINDBLOM, G. (1980) Biochemistry 19,365@ 3655. 24. MCCABE, P. J., AND GREEN, C. (1977) Chem Phys. Lipids

20,319-330.

25. KHAN, A., RILFORS, L., WIESLANDER, A., AND LINDBLOM, G. (1981) Eur. J. Bioehem. 116,215220. 26. WIESLANDER, A., RILFORS, L., JOHANSSEN, L. B.-A., AND LINDBLOM, G. (1981) Biochemistry 20,730-735.