Synthesis of disaturated phosphatidylcholine by cholinephosphotransferase in rat lung microsomes

Synthesis of disaturated phosphatidylcholine by cholinephosphotransferase in rat lung microsomes

313 Biochimica et Biophysics Acta, 666 (1981) 313-321 ElsevierlNorth-Holland Biomedical Press BBA 57935 SYNTHESIS OF DISATURATED PHOSPHATIDYLCHOLINE...

921KB Sizes 1 Downloads 17 Views


Biochimica et Biophysics Acta, 666 (1981) 313-321 ElsevierlNorth-Holland Biomedical Press BBA 57935



G.P.H. VAN HEUSDEN, B. RUESTOW *, M.A. VAN DER MAST and H. VAN DEN BOSCH Biochemistry Laboratory, Padualaan8,3584 CH Utrecht (The Netherlands) (Received May 25th, 1981)

Key words: Cholinephosphotransferase;Disaturatedphosphatidylcholine; Lung surfactant; (Rat lung microsome)

1. Incubation of rat lung microsomes with cytidine diphospho[methyl-14C]choline resulted in synthesis of radioactively labeled phosphatidylcholine. 2. lo-15% of this phosphatidylcholine appeared to be disaturated species. In similar experiments with rat liver microsomes only 2-3% of the radioactivity was present in the disaturated species. 3. When de novo synthesis was blocked after 5 mm by addition of Ca” no increase in the proportion of disaturated phosphatidylcholine was observed upon further incubation of lung microsomes. Under these conditions the enzymes involved in a remodelling mechanism, i.e. phospholipase A and acylCoA : lysophosphatidylcholine acyltransferase, remain fully active. 4. Addition of diacylglycerols from egg phosphatidylcholine containing trace amounts of di[ l-14C]pahnitoyl glycerol resulted in direct incorporation of 14C label into phosphatidylcholine. The rate of phosphatidylcholine synthesis measured from incorporation of di[ l-14C]pahnitoyl glycerol equalled that observed with labeled CDPcholine. 5. These results support the conclusion that disaturated phosphatidylcholine in lung can be formed by direct utilization of disaturated diacylglycerol and is not produced exclusively via remodelling of de novo synthesized unsaturated species.

* Present address: Department of Clinical Biochemistry, Humboldt University, Schumannstrasse 20/21, DDR-104 Berlin, G.D.R.

enzyme with either exogenous [5,7], endogenous [5] or generated membrane-bound [6] diacylglycerols, failed to show appreciable incorporation of 1,2-dipalmitoyl-sn-glycerol. This finding has been interpreted to indicate that the dipalmitoylphosphatidylcholine is only synthesized to a minor degree via the cholinephosphotransferase pathway, with most of it being formed via remodelling of de novo synthesized unsaturated phosphatidylcholine. Evidence for the involvement of remodelling in the production of at least part of the total disaturated phosphatidylcholine has been obtained from both in vitro and in vivo experiments (sae Refs. 8-12 for reviews). In this paper the substrate specificity of cholinephosphotransferase was reinvestigated. In contrast to current opinions, we have obtained evidence which indicates that cholinephosphotransferase can utilize both endogenous microsomal and exogenously added 1,2-dipalmitoyl-sn-glycerol.


Biomedical Press

Introduction In lung tissue about 30% of the phosphatidylcholine contains two palmitoyl residues [I]. This disaturated phospholipid has surface-tension lowering properties, and is the major functional component of the lung surfactant [2,3]. The predominant pathway for phosphatidylcholine biosynthesis in lung is the CDPcholine pathway, as described originally for liver by Kennedy [4] and others [5,6]. The final step in this pathway is the formation of phosphatidylcholine and CMP from diacylglycerol and CDPcholine, catalyzed by the enzyme cholinephosphotransferase (EC Studies on the substrate specificity of this

0 1981 Elsevier/North-Holland


Materials and Methods Materials Cytidine 5’-diphospho[methyZ-‘4C]choline, [l14C] palmitic acid, and 1,2-di [ 1-14C]palmitoyl-snglycero-3-phosphocholine were obtained from Radiochemical Centre, Amersham, U.K. l- [ I -14C]Pahnitoyl-sn -glycero-3-phosphocholine was bought from New England Nuclear, Boston, U.S.A. Unlabeled CDPcholine and 1,2-dipalmitoyl-sn-glycero-3-phosphocholine were purchased from Sigma, St. Louis, U.S.A. and Fluka AC, Switzerland, respectively. Egg phosphatidylcholine was isolated by the method of Papahadjopoulos and Miller [ 131. Phospholipase C from Bacillus cereus was purified as described [14] and was a generous gift of Dr. B. Roelofsen from this laboratory. 1,2-Diacyl-sn-glycerols were prepared by phospholipase C degradation of the corresponding 1,2-diacyl-sn-glycero-3-phosphocholines. Unlabeled 1-palmitoyl-sn-glycero-3-phosphocholine was prepared as described previously [ 151. Palmitoyl-CoA was prepared as described by Al-Arif and Blecher

[161. Methods Isolation of microsomes. Rat lung tissue 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 5 min and at 20 000 X g for 20 min, the microsomes were pelleted by centrifugation at 105 000 X g for 60 min. The microsomal fraction was resuspended in 125 mM KCl/ 10 mM Tris (pH 7.4). Cholinephosphotransferase assay. The standard incubation mixture for the cholinephosphotransferase assay with endogenous diacylglycerols contained: 100 mM Tris-maleic acid (pH 8.0), 25 mM MgClz, 10 mM /3-mercaptoethanol, 1 mM EDTA, 0.5 mM CDP[methyl-14C] choline (745 dpm/nmol) and microsomal protein at a concentration of 3 mg/ml. The final volume of the incubation mixture was 0.15 ml. Incubations were performed at 37°C for the times indicated in the legends of tables and figures. When only total phosphatidylcholine synthesis had to be determined, 0.1 ml of the incubation mixture was applied to a filter paper to precipitate phosphatidylcholine in ice-cold 10% (w/v) trichloroacetic acid. The filter papers were further processed as described

by Goldfine [ 171. When both total and disaturated phosphatidylcholine had to determined, the lipids were extracted according to the method of Bligh and Dyer [ 181. An aliquot of the extract was applied on a thin-layer plate to quantitate total phosphatidylcholine synthesis as described previously [ 191. The remainder of the lipid extract was used to isolate phosphatidylcholine by thin-layer chromatography and elution from the silica with 20 ml of CHClJ CH30H (1 : 4, v/v). The percentage disaturated phosphatidylcholine in this preparation was determined as described by Mason et al. [20]. In experiments with exogenous diacylglycerols, the incubation mixture was the same as for endogenous diacylglycerols except that 0.6 mg/ml of microsomal protein was used in a final volume of either 0.15 or 0.3 ml, as indicated in the individual experiments. The concentration of CDPcholine was 1.0 mM. The diacylglycerols were added from 15 mM sonicate in 0.015% (w/v) Tween 20 (sonicated for 6 min at 0°C at 50 W) to yield a final concentration of 10 mM. When CDP- [ 14C] choline was used as radioactive precursor, enzymatic activity was quantitated with the filter paper assay; when 1,2-di[ 1-14C]palmitoyl-snglycerol was the precursor, enzymatic activity was quantitated after extraction and thin-layer chromatography as described before. Acyl-CoA : lysophosphatidylcholine acyltransferase assay. 200 pg of microsomal protein was incubated in 1 ml of a solution containing 200 nmol of l-[ 1-‘4C]palmitoyl-sn-glycero-3-phosphocholine (300 dpm/nmol), 80 E.tmol of Tris-HCl (pH 7.4) and 70 nmol of palmitoyl-CoA for 20 min at 37°C as described previously [2 11. Phosphatidylcholine synthesis was quantitated after extraction and thin-layer chromatography as described [21]. With free palmitate as acyl donor, a 1 ml incubation mixture was used which contained 200 nmol 1-[ 1-14C] palmitoylsn-glycero-3-phosphocholine (300 dpm/nmol)/80 pmol Tris-HCl (pH 7.4)/10 pmol MgClJ 100 pmol KCl/lO pmol ATP/400 nmol coenzyme A/200 nmol palmitate/200 1.18microsomal protein. After 20 min of incubation at 37°C phosphatidylcholine synthesis was quantitated as described above. Analytical methods. Protein was measured as described by Lowry et al. [22]. Lipid phosphorus was determined according to Chen et al. [23] after ashing



Results In order to establish the optimal incubation conditions for lung microsomal choline phosphotransferase with endogenous diacylglycerols, the influence of pH, protein concentration, incubation time and CDPcholine concentration was determined. Fig. 1 shows that the activity was optimal at pH 8.0 (Fig. lA), linear with protein concentration till 3 mg/ml (Fig. 1B) and linear with incubation time for at least 20 min (Fig. 1C). The optimal CDPcholine concentration was 0.5 mM (Fig. 1D). Furthermore, it was found the reaction was absolutely dependent on MgClz (data not shown). Using these optimal conditions, rat lung microsomes were incubated with CDP-[14C]choline and the formation of radioactive phosphatidylcholine was determined (Fig. 2A, upper curve). Aliquots of the biosynthesized phosphatidylcholine were treated with Os04 [20] to determine the amount of


30 50 TIME (mid


1 1.5 2


Fig. 1. Optimal conditions for cholinephosphotransferase reaction with endogenous diacylglycerols in rat lung microsomes. A, Influence of ~JH. Standard incubation mixtures were used with 200 mM Tris-maleic acid of the indicated pH values, 1 mM CDP-[14C]choline and 1 mg/ml microsomal protein. Incubation was for 15 min. B, Influence of protein concentration. Standard incubation mixtures were used with 1 mM CDP-[ 14C] choline and the indicated protein concentrations. Incubation was for 15 min. The activity is expressed as nmol/min per 100 ~1 incubation mixture. C, Influence of the incubation time. Standard incubation mixtures were used with 1 mM CDP-[14C]chol.ine. D, Influence of the CDPcholine concentration. Incubations were performed for 20 min with standard mixtures, except that the CDPcholine ooncentration was varied as indicated.

16 P 2’2 E8 c 4 0






TIME (min.) Fig. 2. Formation of total and disaturated phosphatidylcholine from endogenous diacylglycerols by rat lung (A), liver (B) and brain (C) microsomes during incubation with CDP[ 14C]choline. o, Total phosphatidylcholine; 0, disaturated phosphatidylcholine.

radioactivity in the disaturated phosphatidylcholine. As is shown in Fig. 2A, at all incubation times lo15% of the phosphatidylcholine produced is disaturated. In contrast to lung, liver contains only trace amounts of disaturated phosphatidylcholine [I]. It is expected that when the microsomes from this tissue are incubated with CDP-[‘4C]choline, only unsaturated phosphatidylcholine is formed. Fig. 2B shows that indeed only 2-3% of the phosphatidylcholine formed is disaturated. Furthermore, this experiment shows that the relatively high percentage of disaturated phosphatidylcholine formed in the experiments with lung microsomes represents tissue specificity rather than that it results from insufficient oxidation of unsaturated phosphatidylcholine species during isolation of disaturated phosphatidylcholine. This is further supported by the fatty acid analysis of the isolated disaturated phosphatidylcholine, which showed that it contained less than 2% of unsaturated fatty acids (data not shown). Like lung tissue, brain contains considerable amounts of disaturated phosphatidylcholine species [I]. When the experiment is repeated with brain microsomes, again about lo-15% of the enzymatically formed phosphatidylcholine appears to be disaturated (Fig. 2C). The results given in Fig. 2A show that after incubation of rat lung microsomes with CDP-[‘4C]choline, lo-15% of the formed phosphatidylcholine is disaturated. Most likely, this is a result of a direct reaction of disaturated diacylglyerol with CDP- [’ 4C] choline. However, an alternative explanation could be that first unsaturated phosphatidylcholine is formed,


TABLE I INFLUENCE OF CaCl2 ON ACYLCoA:LYSOPHOSPHATIDYLCHOLINE ACYLTRANSFERASE IN RAT LUNG MICROSOMES Incubation conditions were as described in Materials and Methods. Results arc expressed in nmol phosphatidylcholine synthesized per min per mg protein. 0


2 3 4 5 [CaCLz],mM

Acyl donor


Fig. 3. Inhibition of cholinephosphotransferase activity in rat lung microsomes by CaCl?. Incubations were performed for 10 min.

which is then remodelled into disaturated phosphatidylcholine by the consecutive action of a phospholipase AZ, yielding lysophosphatidylcholine and the acylacyl-CoA : lysophosphatidylcholine enzyme transferase using endogenous saturated acyl-CoA. Subsequent experiments were designed to investigate whether under our incubation conditions such a remodelling did indeed take place. Fig. 3 shows that the presence of 5 mM CaCl, nearly completely inhihits cholinephosphotransferase activity. Therefore, no further phosphatidylcholine synthesis will take place upon addition of 5 mM CaCl? to the incubation mixture. If the production of disaturated phosphatidylcholine before addition of Ca2+ had taken place through remodelling of unsaturated species rather than by direct de novo synthesis it is likely that production of disaturated phosphatidylcholine would continue after addition of 5 mM Ca2+, provided it can be shown that this amount of Ca2+ does not inhibit the enzymes of the remodelling pathway, i.e. phospholipase A2 and acyl-CoA : lysophosphatidylcholine acyltransferase. It is well known that phospholipase A2 of lung microsomes is in fact activated by CaC12 [25,26]. Table I demonstrates that the acyl-CoA : lysophosphatidylcholine acyltransferase in lung microsomes is not influenced at 5 mM was assayed CaCl,, not even when this enzyme coupled

to acyl-CoA


by using palmitate


substrate. These results demonstrate that remodelling of de novo synthesized unsaturated phosphatidylcholine species into fully saturated species could potentially continue after inhibition of de novo synthesis by addition of CaC12. Fig. 4A shows that addition of CaC12, after 5 min of incubation of microsomes with

PalmitoylCoA Palmitate



Without CaClz

With 5 mM CaCl2

1.70 1.03

1.67 1.00

CDP-[ “C]-choline effectively inhibits total phosphatidylcholine synthesis. Fig. 4B demonstrates that also the synthesis of disaturated phosphatidylcholine, which continues in the absence of CaCl,, is blocked by addition of CaC12. These results strongly suggest that also this disaturated phosphatidylcholine is produced by de novo synthesis rather than by remodelling of unsaturated species. Previously reported studies on the substrate specificity of cholinephosphotransferase in lung microsomes have indicated that exogenously added dipalmitoylglycerol is not utilized by the enzyme [5,7, 271. On the other hand, exogenously supplied unsaturated diacylglycerols were effectively used for phosphatidylcholine synthesis in such experiments. Using either exogenous dipalmitoylglycerol or diacylglycerols prepared from egg phosphatidylcholine in the presence of CDP-[14C]choline we have confirmed



5 10 15 20 10 15 20 0 INCUBATION TIME (min.)

Fig. 4. Formation of total (A) and disaturated (B) phosphatidylcholine in rat lung microsomes in the absence and presence of CaC12. The mixtures that were incubated for 10 and 20 min received after 5 min 37.5 bl of either water or 0.1 M CaCl2 and were incubated for another 5 or 15 min, respectively. The original volume was 0.75 ml.


these results (data not shown). As outlined already by Oldenborg and Van Golde [27], the failure of dipalmitoyl glycerol to stimulate cholinephosphotransferase activity could well be the result of the unfavourable physicochemical properties of this lipid in emulsified form, rather than being the consequence of the substrate specificity of cholinephosphotransferase. In order to circumvent the influence of unsuitable dipalmitoyl glycerol emulsions a mixture of egg diacylglycerols (prepared from egg phosphatidylcholine) containing a trace amount of highly radioactive 1,2-di [ l14CIP almitoyl-sn-glycerol was employed in further studies. It is likely that this trace of disaturated diacylglycerol is included in the structures formed by the unsaturated diacylglycerols. When such a mixture is incubated with rat lung microsomes in the presence of unlabeled CDPcholine, radioactive phosphatidylcholine is expected to be formed only when the di[ l14C] palmitoyl glycerol is utilized. Fig. 5A shows that rat lung microsomes, when incubated with 10 mM egg diacylglycerols in the presence of CDP-[ 14C] choline, produce phosphatidylcholine at a rate of 4.8 nmol ’ mine1 . mg-’ protein. When egg diacylglycerols containing 0.125 mol% of 1,2-di-[ l-14C]palmitoyl-sn-glycerol and unlabeled CDPcholine were used as substrates an amount of radioactivity equivalent to 6.0 pmol * min-’ . mg-’ protein of the 1,2-di[ 1-14C]palmitoyl-sn-glycerol was incorporated (Fig. SB, lower line). This demonstrates that the disaturated diacylglycerol is utilized by the cholinephosphotransferase, although at a low rate. However, since the diacylglycerol mixture contains only 0.125 mol% 1 ,2-di[l-14C]palmitoyl-srz-glycerol it can be expected that together with the labeled disaturated phosphatidylcholine a much higher amount of unlabeled unsaturated phosphatidylcholine is formed. When this substrate dilution is considered, by using as specific radioactivity of the diacylglycerol precursor the radioactivity added as 1,2di [ 1-14C] palmitoyl-sn-glycerol divided by the total amount of diacylglycerols, a phosphatidylcholine production of 4.8 nmol . min-’ . mg-’ protein was observed again (Fig. 5B, upper line). The equal rates observed in Figs. 5A and B must mean that rat lung cholinephosphotransferase does not discriminate against dipalmitoyl glycerol, when the latter is present as a trace component in emulsions of unsaturated diacylglycerols.


10 20 0 IO 2Ofminl INCUBATION TIME

Fig. 5. Utilization of egg diacylglycerols and 1,2di[l-14C]palmitoyl-snglycerol by cholinephosphotransferase of rat lung microsomes. A, Incorporation of CDP-[14C] choline into phosphatidylchoIine in the presence of 10 mM diacylglycerols containing 0.125 mol% of dipalmitoyl glycerol. The substrate emulsion was prepared by mixing chloroform solutions containing 7.5 Hmol egg diacylglycerol and 9.4 nmol dipalmitoyl glycerol. The organic solvent was evaporated and 0.5 ml of 0.015% (w/v) Tween 20 was added. After sonication 0.1 ml of the emulsion was used in 0.15 ml incubation mixtures (see Materials and Methods). Phosphatidylcholine synthesis was measured by the filter paper assay. The values were corrected for a zero time incubation. B, Incorporation of 1,2di[ 1-l 4C]palmitoyl-snglycerol into phosphatidylcholine. The incubation was as for Expt. A, except that 9.4 nmol 1,2di[ 1-‘4C]palmitoyl-sn-glycerol (245 000 dpm/nmol) was used in the preparation of the substrate emulsion. Phosphatidylcholine synthesis was measured after extraction and thin-layer chromatography of the lipids, as described under Materials and Methods. Values were corrected for a zero time incubation. 0, incorporation of l,2di[l-14C]palmitoyI-snglycerol (DSPC) based on a spec. act. of 245 000 dpm/nmol. o, incorporation of total diacylglycerols based on a spec. act. of 306 dpm/nmol.

These results stongly support the possibility of de novo synthesis of disaturated phosphatidylcholine, provided the dipalmitoylglycerol is presented to the enzyme in an adequate physical form. The observation that at least 88% of the radioactivity in phosphatidylcho!ine in the experiments of Fig. 5B was present in the disaturated species is in line with a direct utilization of 1,2-di[ l-14C]palmitoyl-sn-glycerol by cholinephosphotransferase. On the other hand, it is well known that lung microsomes contain an active diacylglycerol lipase [6,19], which indeed produced [ l-*4C]pahnitoyl glycerol and [1-14C]palmitate during our incubation experiments (Fig. 6). Any reactivation of this [l- 14C]palmitate to [ l-14C]palmitoylCoA could potentially yield radioactive phosphatidyl-


choline by acylation of endogenous lysophosphatidylcholine by acyltransferases. Fig. 6 shows that this activation of the released [I-‘4C]palmitate is not likely to occur to any appreciable extent. If this were the case, a considerable formation of triacylglycerol from the added diacylglycerol and palmitoyl-CoA is expected. However, very little triacylglycerol is produced in comparison to phosphatidylcholine. Furthermore, when [l-14C]palmitate or l-[14C]palmitoyl-sn-glycero3-phosphocholine were included in the incubation mixtures used to study phosphatidylcholine synthesis from 1,2-di[ l-‘4C]palmitoyl-snglycerol the rate of phosphatidylcholine production from the former two precursors was about 400-fold less than observed from 1,2-di [ 1-l “C] palmitoyl-snglycerol itself (Table II). These results demonstrate that acylation of lysophosphatidylcholine by palmitoylCoA, that could theoretically be produced from pamitate released from dipalmitoyl glycerol, does not contribute significantly to phosphatidylcholine production in incubations with 1,2-di[ l-14C]palmitoy&z-glycerol. Hence, this phosphatidylcholine is

TABLE II INCORPORATION OF VARIOUS PRECURSORS INTO PHOSPHATIDYLCHOLINE BY RAT LUNG MICROSOMES IN THE PRESENCE OF CDPCHOLINE Emulsions of 7.5 Hmol egg diacylglycerol containing either 9.4 nmol 1,2di[ 1-‘4C]palmitoyl-sn-glycero1 (245 000 dpm/ nmol), 6.0 nmol l-[ 1-l 4C]palmitoyl-snglycero-3-phosphocholine (122 300 dpm/nmol) or 7.3 nmol [ 1-14C]palmitate (122 300 dpm/nmol) in 0.5 ml 0.015% (w/v) Tween 20 were prepared as described in the legend to Fig. 5. Incubations were done for 20 min in 0.15 ml mixtures containing 90 fig of microsomal protein and 0.1 ml of the above emulsions. For the calculation of the incorporation of 1,2-di[ l-‘4C]palmitoyl-sn-glycerol the dilution with egg diacylglycerols was taken into consideration, i.e. a specific radioactivity of 306 dpm/nmol was used. The incorporation of [l-14C]palmitate was calculated using a specific radioactivity based on the data of Fig. 6; i.e. assuming that after 20 min 169 nmol fatty acid was formed per mg protein. This yields a minimal value for the specific radioactivity of the fatty acid at the end of the incubation of 11 740 dpm/nmol. Using this number a maximal value for the nmol amount of palmitatc incorporated into phosphatidylcholine is obtained. Precursor

Incorporation into phosphatidylcholine (nmol/mg per 20 min)

1,2di[ 1-‘4C]Palmitoyl-sn-glycerol 1-I 1-14C]Palmitoyl-sn-glycero-3phosphocholine [lJ4C]Palmitate

95.6 0.22 0.28

synthesized by direct utilization of dipalmitoyl glycerol by cholinephosphotransferase, in line with the results obtained in experiments using endogenous diacylglycerols (Figs. 2 and 4).




15 20 TIME (mid

Fig. 6. Recovery of [ l-14C]palmitate in various lipids after incubation of rat lung microsomes with CDPcholine and egg diacylglycerols containing a trace of 1,2di[ l-‘4C]palmitoylsnglycerol. Incubation mixtures were identical to those described in Fig. SB. After the incubation, lipids were extracted and aliquots of the extract were used to quantitate radioactivity in phosphatidylcholine, phosphatidylethanolamine and neutral lipids as described [ 191. The amount of each lipid was calculated assuming a specific radioactivity of 306 dpm/nmol diacylglycerol or 153 dpm/nmol fatty acid. Values were corrected for a zero time incubation. 0, fatty acid; X, monoacylglycerol; . , phosphatidylethanolamine; 0, triacylglycerol; A, phosphatidylcholine.

Discussion As mentioned in the Introduction the lack of utilization of disaturated diacylglycerols under a variety of conditions by cholinephosphotransferase from lung microsomes has been interpreted as one of the arguments against de novo synthesis of disaturated phosphatidylcholine by the CDPcholine pathway [5-73. Consequently, remodelling mechanisms for the conversion of de novo synthesized unsaturated phosphatidylcholines into the functionally essential disaturated phosphatidylcholine have been proposed. In these mechanisms the unsaturated acylchain at the


sn-2-position is thought to be removed by a phospholipase As to yield saturated lysophosphatidylcholine [25,26,28]. The latter can be converted into saturated phosphatidylcholine either by acylation with palmitoyl-CoA [8,29,30] or by a transacylation reaction between two molecules of lysophosphatidylcholine [3 l-361. However, recent results from both in vitro [21,37] and in vivo 121,381 studies have provided evidence for acylCoA-dependent acylation as the exclusive pathway for conversion of lysophosphatidylcholine into disaturated phosphatidylcholine in vivo . The substrate specificity of lung microsomal acylCoA : lysophosphatidylcholine acyltransferase has been investigated by several research groups [30,39441. For this pathway to be the major synthetic route to disaturated phosphatidylcholine one would expect the acyltransferase to show a preference for palmitoyl-CoA when compared with unsaturated acylCoA’s. In this respect somewhat conflicting results have been reported, but in general the highest enzymatic activities were observed with unsaturated-, especially arachidonoyl-CoA [39,41-431. The predominant incorporation of intravenously injected lysophosphatidylcholine into unsaturated phosphatidylcholine in lung [21,34,38], is in line with this substrate specificity of acyl-CoA : lysophosphatidylcholine acyltransferase. Similar results were obtained in experiments with lung slices [33]. Even when isolated type II cells, i.e. the cell type in which surfactant phosphatidylcholine is thought to be synthesized, were incubated with labeled lysophosphatidylcholine most of the radioactivity in phosphatidylcholine was recovered in unsaturated species [45,46]. On the other hand, the acyl-CoA : lysophosphatidylcholine acyltransferase from type II cells showed both a higher specific activity than the enzyme from whole lung and a preference for palmitoylCoA over oleoylCoA [43,44]. Further evidence for the involvement of acylCoA : lysophosphatidylcholine acyltransferase in the synthesis of at least part of the total disaturated phosphatidylcholine pool has been obtained from experiments in which labeled palmitate was used. With this precursor it has been found that injection into rats and rabbits [9,47], perfusion of rat lungs [48] and incubation of rat lung slices [49], adenoma alveolar type II cells [50,51] or isolated type II cells [43,44,

461 produced disaturated phosphatidylcholine that contained about 75% of its palmitate content at the sn-2-position. Since at first assumption one would expect that palmitate would be incorporated equally into positions 1 and 2 by de novo synthesis these results are easiest explained by a significant contribupalmitoylCoA : lysophosphatidylcholine tion of acyltiansferase to total disaturated phosphatidylcholine production. However, this acyl-CoA-dependent remodelling mechanism need not to be the exclusive pathway for disaturated phosphatidylcholine synthesis. In order to study the possible contribution of de novo synthesis to disaturated phosphatidylcholine formation, the substrate specificity of cholinephosphotransferase was investigated in some detail. In the first approach endogenous microsomal diacylglycerols were used as substrate. After incubation of microsomes with CDP-[14C]choline 10-l 5% of the radioactivity in phosphatidylcholine was recovered in the disaturated species (Fig. 2A). Since remodelling of phosphatidylcholine species under the incubation conditions was excluded (Fig. 4), this result indicates that disaturated endogenous diacylglycerols can be utilized by cholinephosphotransferase. In a similar experiment reported in the literature [5] only 1 .l% of the phosphatidylcholine appeared to be disaturated. The reason for this discrepancy is not apparent, but may be due to differences in the methods applied to isolate disaturated phosphatidylcholine, i.e. cryochromatography vs. 0s04 oxidation as described by Mason et al. [20] in our procedure. We checked our isolation procedure in two ways. First, it was shown that the isolated disaturated phosphatidylcholine contained only about 2% of residual unsaturated fatty acids. Secondly, when the complete isolation procedure was applied to an incubation with liver rather than lung microsomes only 2-3% of the labeled phosphatidylcholine was recovered in the disaturated fraction. In a second approach exogenous diacylglycerols were used as substrate. Previous studies on the substrate specificity of cholinephosphotransferase in rat [5], rabbit [7] and mouse [27] lung microsomes have indicated that 1,2-dipalmitoyl-sn-glycerol emulsions functioned as a poor substrate for the enzyme. However, when an emulsion of egg diacylglycerols containing a trace amount of highly radioactive 1,2-


dipalmitoyl-sn-glycerol was employed the disaturated diacylglycerol was utilized at a rate which equalled that of unsaturated diacylglycerols (Fig. 5). This result is at variance with data reported by Sarzala and Van Golde [6] who showed that of membrane-bound diacylglycerols, generated by phospholipase C treatment of microsomes, only unsaturated diacylglycerols were used by the cholinephosphotransferase. A possible explanation might be that the considerable degradation of microsomal phospholipids by phospholipase C caused a phase-separation which made the disaturated diacylglycerols inaccessible to the cholinephosphotransferase. Several in vivo studies aimed at a determination of the relative amounts of the various phosphatidylcholine species produced in lung by de novo synthesis. Moriya and Kanoh [9] observed that 2 min after the injection of [9,10-3H,]palmitate into rats 60% of the labeled phosphatidate and 57% of the labeled diacylglycerols consisted of disaturated species. Also, 5 min after injection of [2-3H]glycerol 33% of the radioactivity in diacylglycerols was in disaturated molecules. Similar results were obtained when rat lung slices were incubated with [2-3H]glycerol [52]. These results suggested that dipalmitoyl glycerol is effectively formed via de novo synthesis and is available for cholinephosphotransferase. Nevertheless, Moriya and Kanoh [9] concluded from turnover rates that disaturated phosphatidylcholine is only partly synthesized by the CDPcholine pathway. Vereijken et al. [8] injected [1(3)-3H]glycerol in rats and observed that after 5.5 and 60 min 20.9 and 31.8%, respectively, of labeled phosphatidylcholine consisted of disaturated species. These results were interpreted to indicate that disaturated phosphatidylcholine was mainly synthesized via remodelling of unsaturated species. In contrast, Jobe [47] noticed that up to 120 mm after injection of [‘4C]choline into rabbits the label appeared with equal specific activities in both unsaturated and disaturated phosphatidylcholine. This suggested that disaturated phosphatidylcholine is synthesized either de novo or that remodelling of unsaturated molecules is fast compared to the de novo synthesis. Such a rapid remodelling is not supported by the results obtained by Vereijken et al. [8]. While this manuscript was in preparation Ishidate and Weinhold [53] reported on the metabolic heterogeneity of diacylglycerols and phosphatidylcholine

during rat lung development. The relative incorporation of [2-3H]glycerol into disaturated, monoene and diene species of phosphatidylcholine in fetal lung was very similar to that for the corresponding diacylglycerol species. The rate of the reaction from disaturated diacylglycerol to disaturated phosphatidylcholine in comparison to this rate for other species indicated considerable potential for the synthesis of disaturated phosphatidylcholine via this route. The results of Ishidate and Weinhold [53], as well as those of others in combination with the data [8,93,47,=1, reported in this paper strongly suggest that dipalmitoyl phosphatidylcholine can be produced by the cholinephosphotransferase pathway. The relative contribution of this direct de novo synthesis and remodelling of unsaturated species to disaturated phosphatidylcholine production remains to be evaluated. Acknowledgements This study was carried out under the auspices of the Netherlands Foundation for Chemical Research (S.O.N.) with financial aid from the Netherlands Organization for the Advancement of Pure Research (Z.W.O.). References 1 Montfoort, A., Van Golde, L.M.G. and Van Deenen, L.L.M. (1971) Biochim. Biophys. Acta 231, 335-352 2 Klaus, M.H., Clements, J.A. and Havel, R. (1961) Proc. Natl. Acad. Sci.U.S.A.47,1858-1859 3 King, R.J. and Clements, J.A. (1972) Am. J. Physiol. 223,715-726 4 Kennedy, E.P. (1961) Fed. Proc. 20,934-940 5 Possmayer, F., Duwe, G., Hahn, M. and Buchnea, D. (1977) Can. J. Biochem. 55,609-617 6 Sarzala, M.G. and Van Golde, L.M.G. (1976) Biochim. Biophys. Acta 441,423-432 7 Rooney, S.A. and Wai-Lee, T.S. (1977) Lung 154,201211 8 Vereijken, J.M., Montfoort, A. and Van Golde, L.M.G. (1972) Biochim. Biophys. Acta 260,70-81 9 Moriya, T. and Kanoh, H. (1974) Tohoku J. Exp. Med. 112,241-256 10 Van Golde, L.M.G. (1976) Am. Rev. Resp. Dis. 114, 977-1000 11 Frosolono, M.F. (1977) in Lipid Metabolism in Mammals (Snyder, F., ed.) Vol. 2, pp. l-38, Plenum Press, New York and London 12 Ohno, K., Akino, T. and Fujiwara, T. (1978) in Reviews in Perinatal Medicine (Scarpelli, E.M. and Cosmi, E.V.,

321 eds.) Vol. 2, pp. 227-318, Raven Press, New York 13 Papahadjopoulos, D. and Miller, N. (1967) Biochim. Biophys. Acta 135,624-638 14 Zwaal, R.F.A., Roelofsen, B., Comfurius, P. and Van Deenen, L.L.M. (1971) Biochim. Biophys. Acta 233, 474-479 15 Brumley, G. and Van den Bosch, H. (1977) J. Lipid Res. 18,523-532 16 Al-Arif, A. and Blecher, M. (1969) J. Lipid Res. 10, 344345 17 Goldfine, H. (1966) J. Lipid Res., 7, 146-149 18 Bligh, E.G. and Dyer, W.J. (1959) Can. J. Biochem. 37, 911-917 19 Aarsman, A.J. and Van den Bosch, H. (1980) Biochim. Biophys. Acta 620,410-417 20 Mason, R.J., Nellenbogen, J. and Clements, J.A. (1976) J. Lipid Res. 17,281-284 21 Van Heusden, G.P.H., Vianen, G.M. and Van den Bosch, H. (1980) J. Biol. Chem. 255,9312-9318 22 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193,265-275 23 Chen, P.S., Torribara, T.Y. and Warner, H. (1956) Anal. Chem. 28,1756-1758 24 Ames, B.N. and Dubin, D.T. (1960) J. Biol. Chem. 235, 769-775 25 Garcia, A., Newkirk, J.D. and Mavis, R.D. (1975) Biothem. Biophys. Res. Commun. 64,128-135 26 Longmore, W.J., Oldenborg, V. and Van Golde, L.M.G. (1979) Biochim. Biophys. Acta 572,452-460 27 Oldenborg, V. and Van Golde, L.M.G. (1976) Biochim. Biophys. Acta 441,433-442 28 Ohta, M., Hasegawa, H. and Ohno, K. (1972) Biochim. Biophys. Acta 280,552-558 29 Lands, W.E.M. (1960) J. Biol. Chem. 235,2233-2237 30 Frosolono, M.F., Slivka, S. and Charms, B.L. (1971) J. Lipid Res. 12,96-103 31 Erbland, J.F. and Marinetti, G.V. (1965) Biochim. Biophys. Acta 105,128-138 32 Van den Bosch, H., Bonte, H. and Van Deenen, L.L.M. (1965) Biochim. Biophys. Acta 98,648-651 33 Akino, T., Abe, M. and Arai, T. (1971) Biochim. Biophys. Acta 248,274-281

34 Akino, T., Yamazaki, J. and Abe, M. (1972) Tohoku J. Exp. Med. 108,133-139 35 Hallman, M. and Raivio, K. (1974) Pediat. Res. 8, 874879 36 Wykle, R.L., Malone, B., Blank, M.L. and Snyder, F. (1980) Arch. Biochem. Biophys. 199,526-537 37 Van Heusden, G.P.H., Reutelingsperger, C.P.M. and Van den Bosch, H. (1981) Biochim. Biophys. Acta 663, 2233 38 Van Heusden, G.P.H., Noteborn, H.P.J.M. and Van den Bosch, H. (1981) Biochim. Biophys. Acta 664,49-60 39 Hasegawa-Sasaki, H. and Ohno, K. (1975) Biochim. Biophys. Acta 380,486-495 40 Engle, M.J., Sanders, R.L. and Longmore, W.J. (1976) Arch. Biochem. Biophys. 173,586-595 41 Yamada, K. and Okuyama, H. (1979) Arch. Biochem. Biophys. 196,209-219 42 Holub, B.J., Piekarski, J. and Possmayer, F. (1980) Can. J. Biochem. 58,434-439 43 Tansey, F.A. and Frosolono, M.F. (1975) Biochem. Biophys. Res. Commun. 67,1560-1566 44 Batenburg, J.J., Longmore, W.J., Klazinga, W. and Van Golde, L.M.G. (1979) Biochim. Biophys. Acta 573,136144 45 Smith, F.B. and Kikkawa, Y. (1978) Lab. Invest. 38,4551 46 Mason, R.J. and Dobbs, L.G. (1980) J. Biol. Chem. 255, 5101-5107 47 Jobe, A. (1979) Biochim. Biophys. Acta 574,268-279 48 Batenburg, J.J., Post, M., Oldenborg, V. and Van Golde, L.M.G. (1980) Exp. Lung Res. 1,57-65 49 Akino, T., Kawamoto, T. and Ohno, K. (1978) Tohoku J. Exp. Med. 124,307-322 50 Wykle, R.L., Malone, B., Blank, M.L. and Snyder, F. (1980) Arch. Biochem. Biophys. 199,526-537 51 Voelker, D.R. and Snyder, F. (1979) J. Biol. Chem. 254, 8628-8633 52 Akino, T. and Abe, M. (1977) J. Jap. Med. Sot. Interfac. 8,2-18 53 Ishidate, K. and Weinhold, P.A. (1981) Biochim. Biophys. Acta 664,133-147