Synthesis of phosphatidylcholine and phosphatidylethanolamine in relation to the concentration of membrane-bound diacylglycerols of rat lung microsomes

Biochimrca

372

et Biophysics

Arm, 793 (1984)

372-378

Elsevier

BBA 51623

SYNTHESIS OF PHOSPHATIDYLCHOLINE AND PHOSPHATIDYLETHANOLAMINE IN RELATION TO THE CONCENTRATION OF MEMBRANE-BOUND DIACYLGLYCEROLS OF RAT LUNG MICROSOMES B. RiiSTOW Institute

and D. KUNZE

of Pathological

and

Clinical

Biochemrsiry,

Humboldt

University,

and

Charitk

Hospital,

Schumannstr.

20/21,

1040

Berlin

(G.D.R.)

(Received August 18th, 1983) (Revised manuscript received December

Key words:

Cholinephosphotransferase;

12th, 1983)

Ethanolaminephosphotransferase;

Diacylglycerol;

Lung

surfactant;

Phospholiprd

synthesis;

(Rat

lung microsome)

Different concentrations of membrane-bound diacylglycerol were generated in vitro in rat lung microsomes by treatment with CMP. Diacylglycerol concentrations of between 16 (endogenous content) and 48 nmol/mg of microsomal protein were obtained. The relative proportion of the disaturated species of diacylglycerol remained constant at all diacylglycerol concentrations. Cholineand ethanolaminephosphotransferase activity was determined in relation to the diacylglycerol concentrations of microsomes. The activity of both phosphotransferases increased. The relative proportion of disaturated phosphatidylcholine synthesized at each diacylglycerol concentration was nearly the same and corresponded to the relative proportion of the disaturated species in the diacylglycerol. Disaturated phosphatidylethanolamine was not formed. The affinities of the choline- and ethanolaminephosphotransferases for the diacylglycerol substrate were different. We conclude that the cholinephosphotransferase is generally non-selective for the diacylglycerol substrate. The available diacylglycerol pattern seems to govern the species pattern of phosphatidylcholine and phosphatidylethanolamine. The kinetics of the phosphotransferases regulate the mass proportion of these phospholipids.

Introduction The lipids of lung surfactant contain a relatively large quantity of dipalmitoylphosphatidylcholine [l]. Recently it was shown that at least part of the disaturated species of the phosphatidylcholine was synthesized via the de novo pathway both in vivo [2] and in vitro in lung microsomes [3,4]. In vitro the disaturated phosphatidylcholine was synthesized in microsomes of rat lung from endogenous and exogenous [3] and also from membrane-bound disaturated diacylglycerol produced by the back reaction of the Abbreviation:

CMP, cytosine

0005-2760/84/$03.00

5’-monophosphate.

0 1984 Elsevier Science Publishers

B.V.

cholinephosphotransferase [4]. During studies on the in vivo incorporation of [3H]glycerol in fetal lung tissue it was shown that the relative distribution of label in disaturated, monoene and diene species of phosphatidylcholine was very similar to that of the corresponding diacylglycerols [2]. This study was carried out using both microsomes from whole lung, which may be derived from many different cell types, and whole lung tissue, whereas the surfactant formation proceeds only in type II alveolar cells [5]. But recent studies with isolated alveolar type II cells have also shown that a part of the disaturated phosphatidylcholine is synthesized by the de novo pathway [6]. Apart from the disaturated species in these experiments, un-

373

saturated fatty acids containing species of phosphatidylcholine were also produced. It is generally accepted that the reaction catalysed by the cytidyl transferase is the rate-limiting step in the synthesis of total phosphatidylcholine [7,8] in lung tissue too [9,10] and type II alveolar cells [ll], but the regulation of the de novo synthesized species pattern of phosphatidylcholine in lung microsomes is unknown. One possibility of regulating the pattern of de novo synthesized species of phosphatidylcholine would be by changing the precursor pool size. In the present study we measured the phosphatidylcholine and phosphatidylethanolamine synthesis of lung microsomes in relation to increasing quantities of membrane bound diacylglycerols. Material and Methods Materials

and Cytidine-5’-diphospho[ methyl-l4 Clcholine cytidine-5’-diphospho[2-‘4C]ethan-l-ol-2-amine were obtained from the Radiochemical Centre at Amersham, U.K. All other chemicals were purchased from Boehringer, Mannheim, F.R.G. The organic solvents were bi-destilled. Methods Isolation of microsomes. Rat lung tissue from adult male animals (body weight 150-200 g) was homogenized in 0.25 M sucrose/l0 mM Tris/l mM EDTA (pH 7.4) yielding a 10% (w/v) homogenate. After centrifugations at 1000 x g for 10 min and at 20000 x g for 20 min, the microsomes were pelleted by centrifugation at 105 000 x g for 60 min. The microsomes were resuspended in 0.25 M sucrose/l0 mM Tris/l mM EDTA (pH 7.4). Production of membrane

of microsomes

with a different content

bound diacylglycerol.

The incubation mixture contained 10 mM Tris-HCl (pH 7.4), 8 mM CMP, 0.25 M sucrose, 1 mM EDTA and microsomal protein with a concentration of 10 mg/ml. The final volume of the incubation mixture was 0.5 ml. After incubation for 0, 3, 5, 7 or 9 mm each incubation mixture was diluted to 1 : 60 with 0.25 M sucrose/l0 mM Tris (pH 7.4)/l mM EDTA at 4’C and centrifuged at 105 000 X g for 60 min. The microsomal pellets were resuspended in 0.5 ml of the sucrose/EDTA/Tris solution.

Determination

of

the

diacylglycerol

content.

Aliquots of the resuspended microsomes (approx. 1 mg protein) were extracted in accordance with the method of Bligh and Dyer [ll]. The lipid extracts were separated by thin-layer chromatography (TLC) on silica gel H (Merck, F.R.G.) with petroleum ether (40-60’C)/ether/formic acid (60 : 40 : 1.5, v/v). The spots containing diacylglycerol were scraped off and extracted with chloroform. The diacylglycerol extracts were dried carefully and acetylated with [‘4C]acetic anhydride (Rossendorf, G.D.R., specific radioactivity 220 dpm/nmol) in pyridine in accordance with published methods [2,12]. In order to determine the contents of the disaturated diacylglycerol, aliquots of the diacylglycerol acetates were separated by TLC on silica gel H-10% AgNO, plates with chloroform/ methanol (99 : 1). The spots containing disaturated diacylglycerol were extracted with chloroform and their radioactivity was determined after evaporation of the chloroform in a stream of N, at 40°C. Dipalmitoylphosphatidylcholine was used as a reference. Cholinephosphotransferase assay. The standard incubation mixture for the cholinephosphotransferase assay with membrane bound diacylglycerol contained 100 mM Tris-HCl (pH S.O), 25 mM MgCl,, 1 mM EDTA, 1 mM cytidine-5’-diphospho[ methyl-i4C]choline (1525 dpm/ nmol) and aliquots of the resuspended microsomes at a protein concentration of 3 mg/ml. The final volume of the incubation mixtures was 0.3 ml. The mixtures were incubated at 37°C for 3 and 6 min. When only total phosphatidylcholine synthesis had to be determined, 30 ~1 of the incubation mixtures after 0, 3 and 6 min at 37°C were applied to filter papers. These were transferred to ice-cold 10% (w/v) trichloroacetic acid solution to precipitate phosphatidylcholine. The filter papers were further processed as described by Goldfine [13]. When both total and disaturated phosphatidylcholine had to be determined the incubation mixtures were extracted in accordance with Bligh and Dyer [ll] after 6 min incubation at 37’C. The quantity of total and disaturated phosphatidylcholine was ascertained as described previously [3]. Ethanolaminephosphotransferase assay. The standard incubation mixture for the ethano-

374

laminephosphotransferase assay contained 100 mM Tris-HCl (pH 7.5) 25 mM MgCl,, 1 mM EDTA, 1 mM CDP[i4C]ethanolamine (2500 dpm/nmol) and aliquots of the resuspended microsomes at a protein concentration of 2.0-2.5 mg/ml. The final volume of the incubation mixture was 0.1 ml. After 0, 3 and 6 min incubation at 37°C 30 ~1 of the incubation mixtures were applied to filter papers and processed as described under cholinephosphotransferase assay. When both total and disaturated phosphatidylethanolamine had to be determined 0.3 ml of the incubation mixture was extracted as described by Bligh and Dyer (111 after 6 min incubation at 37°C. The incorporation of label from CDP[2-‘4C]ethanolamine into disaturated phosphatidylethanolamine was then determined in principle by the procedure of Mason et al. [14]. But in this case the nonoxidized disaturated phosphatidylethanolamine was separated by TLC. After 0~0, oxidation of phosphatidylethanolamine the sample was redissolved in a small volume of CHCl,/CH,OH (2 : 1) and hydrogenated phosphatidylethanolamine was added. The oxidation products were separated from non-oxidized disaturated phosphatidylethanolamine by TLC on silica gel foils (Merck, Darmstadt) with CHCl,/ CH,OH/ cont. NH, (130: 50: 10). The separation of the oxidation products from the non-oxidized phosphatidylethanolamine was completely possible under these conditions as tested by autoradiography of the foils. After oxidation the area of the added hydrogenated phosphatidylethanolamine, which was

TABLE

identified using J,, was free of radioactivity detectable by autoradiography (X-ray film, ORWO, GDR exposition time 10 days). But this area was labelled when the biosynthesized phosphatidylethanolamine (after 0~0, treatment) was mixed with a non-oxidized part of the same phosphatidylethanolamine fraction. Dehydrogenated phosphatidylethanolamine was used as the reference. Analytical methods. Protein was measured as described by Lowry et al. [15]. Results

In order to establish the optimal conditions for lung microsomal ethanolaminephosphotransferase with membrane-bound diacylglycerols, the influences of pH, protein concentration, incubation time and CDPethanolamine concentration were determined in the same manner as previously described for the cholinephosphotransferase (Ref. 3, results not shown). The optimal incubation conditions for determining ethanolaminophosphotransferase as described in Material and Methods were chosen from these results. Table I shows that with longer incubation of lung microsomes with CMP the quantity of membrane-bound diacylglycerols increased but the relative proportion of the disaturated species of diacylglycerol remained constant (26.8 f 5.2%) in all diacylglycerol concentrations. From our experiments we cannot exclude the possibility that active diacylglyceroland triacylglycerol lipase of lung

I

DIACYLGLYCEROL CONTENT MINE OF LUNG MICROSOMES DS, disaturated;

DG, diacylglycerol;

CMP treatment (min)

Total DG content (nmol/mg)

0 3 5 7 9

16.1 + 4.1 19.7 + 3.6 35.Ok7.4 44.6 f 5.9 48.2 f 4.4

a

AND SYNTHESIS OF PHOSPHATIDYLCHOLINE AND PHOSPHATIDYLETHANOLAIN RELATION TO THE INCUBATION TIME WITH CMP PC, phosphatidylcholine;

PE, phosphatidylethanolamine. Total PE synthesis (nmol/min

per mg) a

DSPC synthesis as % of total PC

0.59~0.10 0.65 f 0.08 0.73 kO.12 0.82hO.11 0.90 * 0.07

23.7 26.0 26.0 28.0 25.6

0.21 0.27 0.36 0.38 0.33

DSDG content as % of total DG

Total PC synthesis (nmol/min

23.0 21.8 24.9 29.8 34.4

a The values are means f S.E. for three experiments.

per mg) a & 0.02 i 0.02 k 0.03 * 0.05 i 0.06

375

time of CMP treatment

7

9’

/ / 3

6

0

I -

6

3

Fig. 1. Effect of the time of CMP treatment on the synthesis of phosphatidylcholine and phosphatidylethanolamine in rat lung microsomes. Each point represents the average of two experiments.

ml”

microsomes influenced these results. As the quantity of the total diacylglycerols and their disaturated species increased the de novo synthesis of the total phosphatidylcholine and phosphatidylethanolamine also increased. The relative proportion of disaturated phosphatidylcholine synthesized at each diacylglycerol concentration was nearly the same (25.9 k 1.5) and corresponded to the relative abundance of the disaturated species in the diacylglycerol generated by CMP treatment. The de novo synthesis of disaturated phosphatidylethanolamine was measured in freshly isolated microsomes and in microsomes after 9 min of CMP treatment. The synthesis rate of the disaturated species of phosphatidylethanolamine in both microsomes was

TABLE

lower than 2% of the total phosphatidylethanolamine synthesis. Since in the course of the CMP-treatment the lipid composition of the microsomes changed, we investigated whether the increased de novo synthesis of phosphatidylcholine and phosphatidylethanolamine was related to the increased amount of membrane bound diacylglycerols or to the changed microenvironment of the phosphotransferases. First we recorded the time curve of the phosphatidylcholine and phosphatidylethanolamine synthesis of microsomes after different periods of CMP treatment. As shown in Fig. 1, after each period of CMP treatment the synthesis of phosphatidylcholine and phosphatidylethanolamine was linear with the incubation time. In the next control experiment a large excess of exogenous diacylglycerol was added to microsomes with different quantities of membrane-bound diacylglycerols. This strongly increased the phosphatidylcholine and phosphatidylethanolamine synthesis by phosphotransferases. Under these conditions the relatively small difference in the membrane-bound diacylglycerol concentration has no effect on the activity of the phosphotransferases. As shown in Tables I and II the synthesis of phosphatidylcholine and phosphatidylethanolamine in the absence of exogenous diacylglycerol was increased by about 75% by the highest quantity of membrane-bound diacylglycerol. After addition of exogenous diacylglycerol the phosphatidylcholine and phosphatidylethanolamine synthesis was much higher. Moreover, the rates of phosphatidylcholine and phosphatidylethanolamine

I1

SYNTHESIS OF PC AND PE IN LUNG AFTER TREATMENT WITH CMP

MICROSOMES

WITH

AND

WITHOUT

EXOGENOUS

DIACYLGLYCEROL

Exogenous diacylglycerol (DG) was produced by phospholipase C treatment of egg PC and sonicated as previously described [3]. The synthesis of PC and PE was carried out as described in Material and Methods, the final concentration of the exogenous diacylglycerol was 5 mM. Two experiments were carried out with the same sonicate of exogenous diacylglycerol but with different preparations of lung microsomes. Without

exogenous

CMP time

Membrane-bound DG content

(mm)

(nmol/mg)

PC (nmol/min

0

14.5; 12.8 46.3; 44.1

0.56; 0.60 0.94; 0.89

9

per mg)

DG

With exogenous PE (nmol/min 0.19; 0.20 0.34; 0.36

per mg)

PC (nmol/min 2.43; 2.76 2.7; 2.60

DG

per mg)

PE (nmol/min 1.67; 1.38 1.60; 1.44

per mg)

l--3

‘Ota’ PE

_@-* I

1

10

20

DSPC I

I

I

30 40 50 total DG (nmollmg)

F .t Q2

E

0’

2

V

/ I

I

I

15

10 DSDGhmollmg)

5

USDG (nmol/mg)

Fig. 2. Effect of the concentration of membrane-bound diacylglycerols on the synthesis of total and disaturated phosphatidylcholine and total phosphatidylethanolamine in rat lung microsomes. US, unsaturated; DS, disaturated. USPE corresponds to total PE because no disaturated PE was formed. Each point was calculated from the average of three experiments given in Table I.

TABLE

synthesis were nearly identical for microsomes with a low or high level of membrane-bound diacylglycerols (Table II). This indicates that the generation of membrane-bound diacylglycerol by CMP treatment and not the changes in microsomal phospholipid content introduced by such a procedure has impaired the catalytic capacity of choline- and ethanolaminephosphotransferase’ activity. Consequently, the differences noted with varying amounts of membrane-bound diacylglycerol reflect the influences of the diacylglycerol substrate concentration. Fig. 2 is a graphic representation of the de novo synthesis of total and disaturated phosphatidylcholine and of total phosphatidylethanolamine in relation to the increased concentration of the substrates. Because no disaturated phosphatidylethanolamine was synthesized, not the total but only the unsaturated diacylglycerols were the substrate of the total phosphatidylethanolamine synthesis. It can be shown that the affinities of the two phosphotransferases for membrane-bound diacylglycerol seem to be quite different. In freshly isolated microsomes the endogenous concentration of diacylglycerol under our conditions was nearly half the substrate saturation of the phosphotransferases. Different affinities of the phosphotransferases for their substrate could be an effective regulator of the de novo synthesis of total phosphatidylcholine and phosphatidylethanolamine in lung

III

OF THE APPARENT K, AND V,,,,, TRANSFERASE OF LUNG MICROSOMES S, substrate;

mb-DG,

membrane-bound

CHOLINEPHOSPHOTRANSFERASE

diacylglycerol;

r, coefficient

a Values calculated

S = CDP-base

K,

Vmar

12.8+ 3.9 29.3 + 10.3

1.05*0.11 0.64 f 0.12

r= 0.9173 (n = 4) r = 0.8959 (n = 4)

27.9 + 9.2

0.73 f 0.14

r = 0.9070 (n = 3)

for the unsaturated

ETHANOLAMINEPHOSPHO-

of correlation.

S = mb-DG

PC synthesis PE synthesis

AND

diacylglycerol

substrate.

K,

VI,,

0.14+0.03 1.43 * 0.25

0.98 f 0.05 0.59 + 0.07

r = 0.9052 (n = 6) r = 0.9186 (n = 8)

371

microsomes. In Table III we have given the calculation of the apparent K, and V,,, values for both substrates of the cholinephosphotransferase and ethanolaminephosphotransferase. For both subthis comparison shows that the strates cholinephosphotransferase has a higher V,,, and a higher affinity for its substrates than the ethanolaminephosphotransferase. Discussion The determination of the synthesis rate of phosphatidylcholine and phosphatidylethanolamine as a function of different concentrations of the CDP base is straightforward and poses no experimental problems. The synthesis rate as a function of varying concentrations of diacylglycerols was measured only with exogenous substrate [16,17]. Such values will hardly reflect the in vivo conditions. With CMP treatment of microsomes it is possible to produce different concentrations of membranebound diacylglycerols. We know that these membrane-bound diacylglycerols become re-available for the synthesis of phosphatidylcholine [4]. At all times during the CMP treatment we measured nearly the same relative proportions of the disaturated species in relation to the total concentration of diacylglycerol (Table I). From these results we conclude that the cholinephosphotransferase shows no selectivity for single species of phosphatidylcholine in the back reaction. This conclusion agrees with earlier reports by Van Heusden and Van den Bosch [4]. The measurement of the dependence of the synthesis of phosphatidylcholine on varying quantities of membrane-bound diacylglycerols confirms this conclusion. The percentage of disaturated phosphatidylcholine in the biosynthesized phosphatidylcholine was constant and nearly identical to the percentage of the disaturated species in the diacylglycerol substrate (Table I). This indicates that the cholinephosphotransferase shows no selectivity for single species of diacylglycerol in the forward reaction. This conclusion is in line with the results of Ide and Weinhold [18]. Analysis in whole lung tissue showed a very similar relative content and biosynthesis of disaturated diacylglycerol and phosphatidylcholine [2]. In contrast to these results, other groups [19,20] reported a much lower

level of disaturated diacylglycerol. These differences were discussed in detail by Ishidate and Weinhold [2] in conjunction with the method of diacylglycerol determination. In lung microsomes the biosynthesis of disaturated phosphatidylcholine corresponded to the content of disaturated diacylglycerol but the values were lower than those determined in whole lung tissue [2]. The de novo synthesis of phosphatidylethanolamine also increased with the concentration of membranebound diacylglycerols. In agreement with the recent report of Post et al. [6], no formation of disaturated phosphatidylethanolamine was detectable in freshly isolated lung microsomes. The increased concentration of disaturated diacylglycerol in microsomes after CMP treatment was also not available for the ethanolaminephosphotransferase, which suggests that this enzyme discriminates against disaturated diacylglycerol or that the disaturated species of the substrate is pooled and only available for the cholinephosphotransferase. The endogenous content of diacylglycerols in freshly isolated microsomes was much lower than necessary for the substrate saturation of both phosphotransferases (Fig. 2). If, as we believe, there is no selectivity in phosphatidylcholine synthesis then the molecular composition of diacylglycerol governs the phosphatidylcholine species by de novo synthesis. On the other hand, pool size can determine the relative quantities of phosphatidylcholine and phosphatidylethanolamine. The different affinities of both phosphotransferases for diacylglycerols (Fig. 2) and the corresponding CDP base is seen from the kinetic values of Table III. The calculation of the enzymes apparent K, and V,,, of two-substrate such as the phosphotransferases presents problems, since one of the substrates is water-insoluble and membrane-bound, while the other substrate is water-soluble and localized in the cytosol. Therefore the values in Table III are not characteristic of the enzymes but they are useful for comparison of the substrate affinities of choline- and ethanolaminephosphotransferase. We suggest that the differences in the kinetics of the synthetic enzymes, choline- and ethanolaminephosphotransferase, are the most important factor in regulating the mass proportion of phos-

378

phatidylcholine and phosphatidylethanolamine. On the other hand, the molecular composition of the available diacylglycerols seems to govern the species pattern of these phospholipids as formed by de novo synthesis. References 1 Montfort, A., Van Golde, L.M.G. and Van Deenen, L.L.M. (1971) B&him. Biophys. Acta 231, 335-352 2 Ishidate, K. and Weinhold, P.A. (1981) B&him. Biophys. Acta 664, 133-147 3 Van Heusden, G.P.H., Riistow, B., Van der Mast, M.A. and Van den Bosch (1981) B&him. Biophys. Acta 666, 313-321 4 Van Heusden, G.P.H. and Van den Bosch, H. (1982) Biochim. Biophys. Acta 711, 361-380 5 Goerke, J. (1974) B&him. Biophys. Acta 344, 241-261 6 Post, M., Schuurmans, E.A.J.M., Batenburg, J.J. and Van Golde, L.M.G. (1983) B&him. Biophys. Acta 750, 68-77 7 Sundler, R., Arvidson, G. and Akesson, B. (1972) Biochim. Biophys. Acta 280, 559-568

8 Infante, J.P. (1977) B&hem. J. 167, 847-849 9 Weinhold, P.A. (1968) J. Lipid Res. 9, 262-266 10 Stern, W., Kovac, C. and Weinhold, P.A. (1976) Biochim. Biophys. Acta 441, 280-293 11 Bligh, E.G. and Dyer, W.J. (1959) Can. J. Biochem. 37, 911-917 12 Banschbach, M.W., Geison, R.L. and Hokin-Neaverson, M. (1974) Biochem. Biophys. Res. Commun. 58, 714-718 13 Goldfine, H. (1966) J. Lipid Res. 7, 146-149 14 Mason, R.J., Nellenbogen, J. and Clements, J.A. (1976) J. Lipid Res. 17, 281-284 15 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 16 Kanoh, H. and Ohno, K. (1976) Eur. J. Biochem. 66, 201-210 17 Miller, J.C. and Weinhold, P.A. (1981) J. Biol. Chem. 256, 12662-12665 18 Ide, H. and Weinhold, P.A. (1982) J. Biol. Chem. 257, 14926-14931 19 Okano, G., Kawamoto, T. and Akino, T. (1978) Biochim. Biophys. Acta 528, 385-393 20 Moriya, T. and Kanoh, H. (1974) Tokohu J. Exp. Med. 112, 241-256