Lysophospholipase-Catalyzed Hydrolysis of - Journal of Bacteriology

JOURNAL OF BACTERIOLOGY, Sept. 1982, p. 1095-1101 0021-9193/82/091095-07$02.00/0 Copyright © 1982, American Society for Microbiology

Vol. 151, No. 3

Lysophospholipase-Catalyzed Hydrolysis of Lysophospholipids in Mycoplasma gallisepticum Membranes SHIMON GATT,'* BILHA MORAG,W AND SHLOMO ROTrEM2 Laboratory of Neurochemistry, Department of Biochemistry1 and Department of Membrane and Ultrastructure Research,2 Hebrew University-Hadassah Medical School, Jerusalem, Israel Received 29 May 1981/Accepted 14 May 1982

Mycoplasma gallisepticum strains have a membrane-bound lysophospholipase which hydrolyzes lysophospholipid generated in these membranes by treatment with an external phospholipase. This paper studies the hydrolysis of the membranous lysophospholipids by an enzyme residing in the same membrane (intramembrane utilization) or in adjacent membranes (intermembrane utilization). To study intermembrane hydrolysis, the phospholipids of M. gallisepticum were labeled with [3H]oleic acid. Membranes were prepared, heated at 65°C, and subsequently treated with pancreatic phospholipase A2. This resulted in membranes whose enzyme was heat inactivated, but which contained lysophospholipid. When these membranes were mixed with M. gallisepticum cells or membranes, the lysophospholipid was hydrolyzed by the membranous lysophospholipase. To study intramembrane hydrolysis, [3H]oleyl-labeled membranes of M. gallisepticum were treated with pancreatic phospholipase A2 at pH 5.0. At this pH, lysophospholipid was generated but not hydrolyzed. Adjustment of the pH to 7.4 resulted in hydrolysis of the lysophospholipid by the membranous lysophospholipase. These procedures permitted measuring the initial rates of intramembrane and intermembrane hydrolysis of the lysophospholipid, showing that the time course and dependence on endogenous substrate concentration were different in the intramembrane and intermembrane modes of utilization. They also permitted calculation of the molar concentration of the lysophospholipid in the membrane and its rate of hydrolysis, expressed as moles per minute per cell or per square centimeter of cell surface.

Acholeplasma laidlawii strains have a membrane-bound lysophospholipase, which hydrolyzes micellar dispersions of lysophosphatidylcholine (lyso-PC) and lysophosphatidyglycerol (lyso-PG), and lysophospholipid generated by treatment with purified preparation of phospholipase A2 (2, 18). This lysophospholipase is tightly bound to the plasma membrane and cannot be solubilized by treatment with hypotonic buffered solutions. While following the pattern of incorporations of fatty acids into the phospholipids of Mycoplasma species, Rottem and Markowitz (13, 14) suggested that a similar enzyme might also be present in the membrane of M. gallisepticum. This paper documents the presence of a lysophospholipase in the membranes of this organism. The activity of this enzyme was studied using either intact cells or isolated membranes of M. gallisepticum. Most experiments used membrane-bound substrates, thereby pursuing previous studies of Gatt et al. on the utilization of membranous lipid substrates by membranebound enzymes in several diverse systems (1,

12, 15). The lysophospholipid was generated by treating cells or membranes of M. gallisepticum with pancreatic phospholipase A2. Procedures were developed for studying the kinetics of the hydrolysis of the lysophospholipid by enzyme residing in the same membrane (intramembrane utilization) or on external, neighboring membranes (intermembrane utilization). These data permitted calculating initial rates of hydrolysis of the membranous substrate by the membranebound lysophospholipase. MATERIALS AND METHODS

Mycoplasma gallisepticum (strain A5969) cells were grown in 500 ml of Edward medium containing 4% horse serum (11). To label membrane lipids, 0.2 FCi of tritiated oleic or palmitic acid (0.5 to 2 Ci/m mol, Radiochemical Centre, Amersham, England) was added per ml of growth medium. The radioactive fatty acid was added as an ethanolic solution, the final ethanol concentration in the medium did not exceed 0.1%. After 20 to 24 h of incubation at 37°C under static conditions, the culture was harvested at about the midexponential growth phase (absorption at 640

1095

1096

GATT, MORAG, AND ROTTEM

nm about 0.15 to 0.2 absorbance units) by centrifuging for 20 min at 12,000 x g. The sedimented cells were washed once and resuspended in cold NaCl (0.25 M). Membranes were prepared by ultrasonic irradiation of cell suspensions which were diluted with water to a concentration of about 0.1 mg of protein per ml and chilled to 0°C. A W350 Heat Systems cell disruptor was used at 50% of maximal output. The instrument was set on the pulse cycle and intermittent irradiation was maintained for 3 min. The membranes were sedimented for 30 min at 35,000 x g, washed once, resuspended in water, and stored at -20%. Membranes of A. laidlawii were prepared by osmotic lysis (11). Lipid analyses. Lipid analyses were done by the method of Rottem and Markowitz (14). Lipid spots on thin-layer plates of silica gel H were scraped off, transferred to counting vials and stirred with 2 ml of Triton X-100-ethanol, 1/1 (vol/vol). A total of 10 ml of toluene scintillator was added, and the radioactivity was determined in a scintillation spectrometer. Intramembrane hydrolysis of lyso-PG. Lyso-PG was generated in the tritium-labeled membranes as follows. Tubes contained membranes of M. gallisepticum (100 ,ug of protein), suspended in 0.25 M of Britton and Robinson type universal buffer, pH 5.0 (5), containing 2.5 mM CaCl2. Phospholipase A2 from hog pancreas (Boehringer Mannheim Corp., New York, N.Y.), in quantities specified in the legends to the figures was added, followed by water to a volume of 0.2 ml. After 10 min at 37°C, the tubes were cooled to 0°C, and one was removed to determine the zero-time level of fatty acids. To each of the remaining tubes, 45 ,ul of a mixture containing 25 ,umol of Tris-hydrochloride, pH 7.4, and 2 nmol of EDTA, pH 7.4, was added. The tubes were transferred to a 37°C bath and were removed at the desired time intervals for determination of the radioactivity of the fatty acids (8). Intermembrane hydrolysis of lyso-PG. Membranes of M. gallisepticum were suspended in Tris-hydrochloride, pH 7.4 containing 2.5 mM CaCl2 and heated for 15 min at 65°C, thereby inactivating the membranous lysophospholipase. They were then treated with pancreatic phospholipase A2 and phospholipase action was terminated by the addition of EDTA, pH 7.4 (final concentration, 5 to 10 mM). Native unheated membranes, suspended in 0.1 M Tris-hydrochloride, pH 7.4 were added, the tube was incubated at 37°C, and samples were removed for estimation of the radioactivity of the fatty acids (8). When intact cells rather than membranes were used as a source of lysophospholipase, the reaction medium also contained 0.4 M sucrose and DNase (10 ,ug/ml). Hydrolysis of micellar dispersions of lyso-PC. Reaction mixtures, in volumes of 0.2 ml contained 20 ,umol of Tris-hydrochloride, pH 7.4, [3H]choline-labeled lyso-PC (obtained by treating [3H]choline-labeled PC (16) with snake venom phospholipase A2) and membranes (100 ,ug of protein). After 15 to 30 min at 37°C, the reaction was terminated, and [3H]glycerophosphorylcholine was isolated as follows. We added 1 ml each of chloroform and methanol and then 0.8 ml of water. The tube was thoroughly mixed, and the upper phase was washed with 1 ml of chloroform. One milliliter of the aqueous upper phase was transferred to a small counting vial. We added 3.5 ml of toluene scintillator containing 30% Triton X-100 (by volume),

J. BACTERIOL.

and the radioactivity was determined in a scintillation spectrometer. Analytical procedures. Protein content was determined by the method of Lowry et al. (10), and total phosphorus content was determined by the method of Hess and Derr (9).

RESULTS

Rottem and Markowitz (13, 14) demonstrated that the phospholipid fraction of M. gallisepticum is composed of three major compounds: sphingomyelin (incorporated unchanged, from the growth medium), a disaturated PC, and PG, which is synthesized de novo by the organism. This observation was utilized in this study for the selective labeling of membrane PG by growing the cells with [ H]oleic acid. The PG thus produced has a rather unusual positional distribution of fatty acids, the unsaturated fatty acid being present in position 1 and the saturated fatty acid in position 2 of the molecule (13, 14). Therefore, the PG of these cells and the membranes derived from them were labeled with [3H]oleic acid mostly in position 1 of the molecule. This is emphasized by the data of Table 1, which shows that treatment of such cells or membrane preparations of M. gallisepticum with pancreatic phospholipase A2 at pH 7.4 resulted in the release of free 3H-labeled fatty acid rather than of [1-3H]oleyl labeled lyso-PG. In contrast, phospholipase A2 treatment of a lipid dispersion obtained by extracting M. gallisepticum membranes resulted in the release of labeled lyso-PG. These observations suggested the presence of a lysophospholipase activity associated with the cell membrane of M. gallisepticum. Treatment of cells or membranes with phospholipase A2 produced [1-3H]oleyl labeled lyso-PG which presumably was subsequently TABLE 1. Hydrolysis of M. gallisepticum membrane lipids by treatment with pancreatic phospholipase A2 Radioactivity in lipid fractions (cpm) Preparation'

Treatment PG

LysoLyso

free fatty

acids

Untreated Treated

92,000 60,070

76 6,100 50 40,500

Isolated Untreated membranes Treated

98,000 23,000

60 4,060 210 76,700

Intact cells

100 3,750 Untreated 106,000 Lipid dispersions Treated 52,000 44,800 5,200 a Cells were grown with (3H]oleic acid. The various preparations were treated with pancreatic phospholipase A2 (25 jig/ml) for 30 min. Lipids were extracted and analyzed as described in Materials and Methods.

LYSOPHOSPHOLIPASE OF M. GALLISEPTICUM

VOL. 151, 1982

1097

Figure 1 shows the hydrolysis of membranous lyso-PG as a function of increasing concentrations of cells or membranes of M. gallisepticum. When related to the total protein content of the cell, the rate of hydrolysis was higher with the membrane preparation (Fig. 1, curve B) than 290 1-5 3 with intact cells (Fig. 1, curve A). However, recalculation of these data on the basis of the protein content of the cell membrane only suggests that the rates of hydrolysis by isolated membrane preparations and intact cells were practically the same. This is documented in the 0experiment shown in Fig. 2, in which lyso-PGcontaining membranes of M. gallisepticum were oyi f ebaoslyoP yM incubated with cells or with membranes derived FI. 1.04 from the same quantity of cells. The data of Fig. 2 show clearly that the rates of lyso-PG cleavage (jig PROTEIN) were practically the same with cells or memFIG. 1. Hydrolysis of membranous lyso-PG by M. branes as the enzyme source. gallisepticum cells or membrane preparations. The Intramembrane hydrolysis of lyso-PG. To procedure described in Materials and Methods was study the intramembrane utilization of the subfollowed. Lyso-PG was generated by incubating membranes for 30 min with 1 ,ug of pancreatic phospholi- strate, it was necessary to obtain lyso-PG-containing membranes in which the activity of the pase A2. Experiment A: [3H]oleyl-labeled lyso-PGcontaining membranes (100 ,ug of protein) were mixed endogenous lysophospholipase could be conwith increasing concentrations of cells. Experiment B: trolled. Figure 3 shows the effect of pH on the the same membranes (200 ,ug) were mixed with in- hydrolysis of [3H]palmitoyl-labeled PG (with creasing concentrations of M. gallisepticum. After 2 h heat-inactivated membranes of M. galliseptiat 37°C, the reaction was stopped, and radioactivity of cum) by pancreatic phospholipase A2 (Fig. 3, the fatty acids determined (8). curve A) and on the hydrolysis of lyso-PC (a micellar dispersion) by the membranous lysophospholipase (Fig. 3, curve B). It is apparent hydrolyzed by the membranous lysophospholi- that at pH 5.0 the pancreatic phospholipase pase, thereby releasing radioactively labeled provided close to maximal rates, whereas the free fatty acids. Isolation of these fatty acids and activity of the membranous lysophospholipase estimation of their radioactivity provide a value for the rate of hydrolysis of lyso-PG by the membrane-bound enzyme. The lysophospholipase of M. gallisepticum membrane was found 04 0 0 3 to be readily inactivated by heating. Thus, treatfor 5 min ing the membrane preparations at 65°C w resulted in essentially a complete inactivation of 0~~~~~~ the enzyme. Inactivation at 65°C was, therefore, O */I used to obtain membranes containing lysophospholipids but devoid of lysophospholipase activon 0~~~~~ / ity. C,) 2.10 4 Intermembrane interaction. The intermem0 brane interaction of enzyme and substrate 0~0 was achieved by incubating intact cells or memC,, branes (containing lysophospholipase) with heat-inactivated membranes in which [3H]oleyllabeled lyso-PG had been generated with pan20 30 l0 INCUBATION TIME (MIN) creatic phospholipase A2 (see above). In preliminary experiments, hydrolysis of lyso-PG by an FIG. 2. Time course of hydrolysis of membranous intermembrane interaction was followed by usby M. gallisepticum cells or membrane prepaing membranes of A. laidlawii as a source of lyso-PG rations. The procedure was similar to the experiment enzyme and 3H-labeled heated membranes of M. described in the legend to Fig. 1, except that 100 ,ug of gallisepticum as source of substrate. All subse- heat-inactivated [3H]-oleyl-labeled lyso-PG-containing quent experiments were done with M. gallisepti- membranes was mixed with cells (0, 490 Rg protein) cum as the source for the membranous lysoor membranes (0, 200 ,ug protein) and incubated for phospholipase and the substrate (i.e., lyso-PG). the times specified in the figure. B

a.

0

w

0

0

30

CELLS

OR

90

MEMBRANES

150

1098

J. BACTERIOL.

GATT, MORAG, AND ROTTEM

a

0~~~~~~~~~

A~~~ 0

6103

4 660

pH

FIG. 3. pH dependence of hydrolysis of membranous PG by pancreatic phospholipase A2 and a miicellar dispersion of lyso-PG by M. gallisepticum lysophospholipase. In the experiment described by curve A, 100 ,ug of heat-inactivated [3H]palmitoyl-labeled membranes of M. gallisepticum was treated with 1 FLg of pancreatic phospholipase A2 and buffer of various pH values for 30 min. In the experiment described by curve B, a micellar dispersion of lysolecithin was hydrolyzed by the lysophospholipase of M. gallisepticum membranes for 15 min.

square centimeter of cell surface or by cubic centimeter of the bilayered cell membrane (see below). Comparison of intramembrane and intermembrane hydrolysis. Figure 5 compares the time course of hydrolysis of membranous lyso-PG by the intra- and intermembrane routes. In these respective two experiments, the final pH and ionic strength were identical. The initial rates of hydrolysis by the two routes were dissimilar, especially between the 10- and 90-min incubation time (Fig. 5). After longer incubation times, the substrate concentrations were low, and the data were not reliable (Fig. 5). After 2 h, about 97 to 99% of the lyso-PG was hydrolyzed. The intermembrane rate of hydrolysis (Fig. 5, curve B) was practically linear until about 70% of available substrate was split. In comparison, the intramembrane rates (Fig. 5, curve A) of hydrolysis decreased with time when lesser substrate concentrations were present in the membranes. The data of the two respective curves were recalculated and presented as the rate of hydrolysis of lyso-PG as a function of concentration of this compound in the membranes. This was done by calculating, in the beginning of each 5-min interval of the 0- to 90-min incubation time, the concentration of the membranous lysophospholipid and the decrease in the concentration of this lipid within the 5-min period. The latter value was plotted on the ordinate and the former value was plotted on the abcissa of each inset. The two respective v versus S curves (shown in

Figure 4 bears this out by showing the stoichiometric recovery of lyso-PG in membranes of M. gallisepticum which were treated with increasing concentrations of pancreatic was very low.

phospholipase A2 at pH 5.0, again suggesting that at this pH lyso-PG was not degraded by the membranous lysophospholipase. Subsequent adjustment of the pH of the reaction mixture to 7.4 permitted measuring the initial rates of intramembrane hydrolysis of 3H-labeled lyso-PG in membranes of M. gallisepticum. The experimental procedure includes the following steps: (i) treatment of [3H]oleyl-labeled membranes at pH 5 with pancreatic phospholipase A2; (ii termination of the phospholipase reaction with excess EDTA; (iii) addition of Tris buffer, pH 7.4, and immediate removal of samples for extraction of zero-time fatty acid; and (iv) incubation at 37°C for various periods and estimation of the radioactivity released by the lysophospholipase action. The initial rate of intramembrane hydrolysis was expressed as the percentage of total lyso-PG hydrolyzed per minute by the membranous enzyme, or, alternatively, as moles of lyso-PG hydrolyzed in 1 min by one cell by

1.5xl104 0.

0I

0.5104 -

\

250 500 750 PANCREATIC PHOSPHOUPASE (ng)

1000

FIG. 4. Stoichiometric relationship of PG and lysoPG in phospholipase A2-treated membranes of M. gallisepticum. A total of 100 F.g of protein of [3H]oleyllabeled membranes, suspended in calcium-containing buffer (see Materials and Methods), was treated with increasing concentrations of pancreatic phospholipase A2 at 5.0. After 10 min at 37°C, the lipids were extracted with a chloroform-methanol mixture and the radioactivity of PG and lyso-PG was determined as described in Materials and Methods.

VOL. 151, 1982


39 ,10

0

0

wc

N 0

0

>-2XI04

~

0~~~~~~

08

o

o

0a.~ ~ ~ (

Cf)

'I~

0.6

o ~~~~~~~3S a-

1099

0.6-

2

4i

0.

a.

n nemmrn yrlsi fIs-Go ieadsbtaecnetain a-i 5.Dpnec3fita

intranemembrane hydrolysis FIor.Dpedneo For intermembrane n

pryeino-f [Honlieyl-aneldsumbsratecnentratione.

of hydrolysis (curves B and B'), 100 ,ug of protein of heat-inactivated [3H]oleyl-labeled lysoPG-containing membranes was mixed with 100 1g of protein of untreated active membranes of M. gallisepticum. In each case, 25 ,uag of pancreatic phospholipase A2 was used as described in Materials and Methods. At the specified times, samples were removed, and the radioactivity of the fatty acids was estimated (8). The data of the insets (v versus S curves) were calculated from the data of the experiments describing the two respective rates of hydrolysis as a function of time (see text). (uvsAadA)

0

the inset to Fig. 5) are quite dissimilar. Figure 5, curve B, which describes the intermembrane utilization is practically linear (or hyperbolic with a low curvature), suggesting a rather constant value (-6%) for the percentage of membranous lyso-PG hydrolyzed per minute. In comparison, the v versus S curve describing the intramembrane utilization (Fig. 5, curve A) is seemingly parabolic, suggesting a different and more pronounced dependence of the reaction rate on lyso-PG concentration. In this case, the rate varied between 1.3 and 3.5% of membranous lyso-PG hydrolyzed per minute. Calculation of substrate concentration and reaction rates. The concentration of PG in membranes of M. gallisepticum is about 10-7 mol per mg of membrane protein or per 3 mg of cell protein. The cell diameter is assumed to be about 0.5 ,m (3). This provides the following values for PG concentration: 6 x 10-18 mol per cell or 8.5 x 10-10 mol/cm2 of cell surface. Assuming 60A as the thickness of the bilayered membrane (4, 7, 17), the concentration of PG is

u

1.4 x 10-3 mol/cm3 of membrane, namely 1.4 M. This value of PG is also the maximal lyso-PG that can be generated in the membrane. The data reported above (Fig. 1, 2, and 5) provide the following values for the rate of hydrolysis of the membranous lyso-PG by the membrane-bound lysophospholipase: 0.33 nmol/min per mg of cell protein or 1 nmol/min per mg of membrane protein. When recalculated, this equals to 6.7 x 10-20 mol/min per cell or 8.5 x 10-12 mol/min per cm2 of cell surface or 1.4 x 10-5 mol/min per cm3 of cell membrane. DISCUSSION Two previous papers presented studies on the hydrolysis of membranous lipid substrates by membranous enzyme. In the first study (15), diacylglycerol, a neutral lipid which is usually not present in rat brain microsomes, was generated and utilized by the acidophylic lipase present in these membranes. In the second study (12), sphingomyelin of chicken erythrocyte membrane was hydrolyzed by a sphingomyelin-

1100

GATT, MORAG, AND ROYIEM

ase, which is present in this membrane in a seemingly latent form and which is converted to an active enzyme by treating the erythrocytes with a hypotonic medium. In this paper, we present data on another system which resembles the system in the first study. Lysophospholipid, which is normally not present in the cell envelope of M. gallisepticum, was generated by treatment with phospholipase A2 and hydrolyzed by a lysophospholipase present in these membranes. Mycoplasma strains have several advantages for the study of interaction of a membranous enzyme with a membranous lipid substrate. Being prokaryotes, they have a single membrane, with a rather simple lipid composition. When M. gallisepticum is grown in a medium containing horse serum, over 50% of the total phospholipid is PG synthesized de novo. The latter can be labeled in position 1 or 2 by growing the cells in the presence of an unsaturated or saturated radiolabeled fattty acid, respectively (13, 14). Thus, if grown in the presence of radioactively labeled oleic acid, practically all of the radioactivity of the phospholipid is found in position 1 of PG; the other two major lipids, sphingomyelin and phosphatidylcholine, are incorporated from the growth medium and therefore unlabeled. In this paper, we show that M. gallisepticum has a membranebound lysophospholipase. Treatment of cells or isolated membranes with an external phospholipase A2 generates radioactively labeled lyso-PG which is further hydrolyzed by the lysophospholipase. The activity of the latter enzyme can be followed by isolating the fatty acid released, using a simple solvent extraction procedure. The data show that the lysophospholipase can hydrolyze lysophospholipid present in the same membrane (intramembrane utilization) or in other membranes (intermembrane utilization). To study the intermembrane interaction, we used membranes in which radioactively labeled lysoPG was generated, but whose enzyme had been inactivated by heating for several minutes at 65°C. When these were mixed with intact cells or isolated membranes of M. gallisepticum, lyso-PG was hydrolyzed. The fact that intact cells or isolated membranes provided practically equal rates of hydrolysis suggests that the lysophospholipase activity was independent of the viability of the cells. For studying intramembrane utilization, a procedure was developed permitting measurement of initial rates of hydrolysis. Figure 5 shows that although initial rates of intramembrane and intermembrane hydrolysis of lyso-PG were similar at high endogenous substrate concentrations, a more pronounced dependence on substrate concentration was observed in the intramembrane interaction. Thus, although the v versus S curve of the intermem-

J. BACTERIOL.

brane reaction was nearly linear or slightly hyperbolic, the corresponding curve describing the intramembrane hydrolysis was seemingly parabolic. It is of interest that in two other systems investigated previously (the hydrolysis of a diacylglycerol by a microsomal lipase [15] and of sphingomyelin by sphingomyelinase of chicken erythrocyte ghosts [12]), a similar relationship on the membranous substrate concentration was observed by the respective inter- and intramembrane interactions. The data of this paper are the first description of a bacterial system in which membranous enzyme-substrate interaction provides a relationship between rate and substrate concentration. The two different kinetic curves describing the intra- and intermembrane hydrolysis (Fig. 5) strongly suggest dissimilar mechanisms for these two respective modes of utilization of the membranous lysophospholipid by the membranebound lysophospholipase. However, one cannot definitely exclude the possibility of transfer of lysophospholipid molecules from the substratecontaining to the enzyme-containing membrane. This would change the mode of interaction into an intramembrane utilization. The latter would also be true were the two membranes to fuse in the course of the reaction. We tried to test this by mixing intact mycoplasmal cells (used as a source of enzyme) and membranes which had been heated and treated with phospholipase A2 to generate lysophospholipid. These were incubated at 37°C and centrifuged in a 25 to 60% sucrose gradient. However, in the course of the incubation aggregates formed and a complete separation of the cells and membranes could not be achieved. Thus, a clear-cut conclusion cannot yet be reached as to whether some transfer of lysophospholipid or fusion might have occurred in the course of the reaction. ACKNOWLEDGMENT This work was supported in part by National Institutes of Health grant NS02967 and the Israel-U.S. Binational Science Foundation grant no. 1746. LITERATURE CITED 1. Allan, D., P. Thomas, and S. Gatt. 1980. 1,2-Diacylglycerol kinase of human erythrocyte membranes. Assay with endogenously generated substrate. Biochem. J. 191:669-672. 2. Bevers, E. M., S. A. Singal, J. A. F. Op den Kamp, and L. L. M. Van Deenen. 1977. Recognition of different pools of phosphatidyglycerol in intact cells and isolated mem-

branes of Acholeplasma laidlawii by phospholipase A2. Biochemistry 16:1290-1294. 3. Boatman, E. S. 1979. Morphology and ultrastructure of the mycoplasmatales, p. 63-99, In M. F. Barile and S. Razin (ed.), The mycoplasmas. Academic Press, Inc., New York. 4. Carstenses, E. L., J. Maniloff, and E. N. Elnoff, Jr. 1971. Electrical properties and ultrastructure of mycoplasma membranes. Biophys. J. 11:572-581. 5. Dawson, R. M. C. 1968. p. 85. R. M. C. Dawson, D. C.

VOL. 151, 1982

6. 7.

8. 9.

10. 11.

12.


Elliott, W. H. Elliot, and K. M. Jones (ed.), Data for biochemical research, 2nd ed. Oxford Press, England. Dole, V. P. 1956. A relation between non-esterified fatty acids in plasma and the metabolism of glucose. J. Clin. Invest. 35:150-154. Engelman, D. M. 1970. X-ray diffraction studies of phase transitions in the membrane of Mycoplasma laidlawii. J. Mol. Biol. 47:115-117. Gatt, S., and Y. Barenholz. 1969. Phospholipase Al from rat brain specific for a position of lecithin. Methods Enzymol. 14:167-170. Hess, H. H., and J. E. Derr. 1975. Assay of inorganic and organic phosphorous in the 0.1-5 nanomole range. Anal. Biochem. 63:607-613. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. Razin, S. and S. Rottem. 1976. Mycoplasma membranes, p. 3-26. In A. H. Maddy (ed.), Biochemical analysis of membranes. Chapman & Hall, Ltd., London. Record, M., A. Loyter, and S. Gatt. 1980. Utilization of membranous lipid substrates by membranous enzymes. Hydrolysis of sphingomyelin in erythrocyte ghosts and liposomes by the membranous sphingomyelinase of chick-

1101

en erythrocyte ghosts. Biochem. J. 187:115-121. 13. Rottem, S., and 0. Markowitz. 1976. Unusual positional distribution of fatty acids in phosphatidyglycerol of sterolrequiring mycoplasma. FEBS Lett. 107:379-382. 14. Rottem, S. and 0. Markowitz. 1979. Membrane lipids of Mycoplasma gallisepticum: a disaturated phosphatidycholine and a phosphatidyglycerol with an unusual positional distribution of fatty acids. Biochemistry 18:29032935. 15. Rousseau, A., and S. Gatt. 1979. Interaction of membranous enzymes with membranous lipids. Hydrolysis of diacyglycerol by lipase in rat brain microsomes. J. Biol. Chem. 254:7741-7745. 16. Stoffel, W. W. 1975. Chemical synthesis of choline-labeled lecithins and sphingomyelins. Methods Enzymol. 35:533541. 17. Terry, T. M., D. M. Engehna, and H. J. Morowltz. 1967. Characterization of the plasma membrane of Mycoplasma laidlawii. II. Modes of aggregation of solubilized membrane components. Biochim. Biophys. Acta 135:391-3%. 18. Van Golde, L. M. G., R. N. McElhaney, and L. L. M. Van Deenen. 1971. A membrane bound lysophospholipase from Mycoplasma laidlawii strain B. Biochim. Biophys. Acta 231:245-249.