- Email: [email protected]
Comparison of the HPLC-separated species patterns of phosphatidic acid, CDP-diacylglycerol and diacylglycerol synthesized de novo in rat liver microsomes (a new method)
364
Biochimicu
et Biophysics
Acta, 961 (1988) 364-369 Elsevier
BBA 52888
Comparison of the HPLC-separated CDP-diacylglycerol
species patterns of phosphatidic acid,
and diacylglycerol synthesized de novo in rat liver microsomes (a new method)
B. Riistow
a, Y. Nakagawa
b, H. Rabe a, G. Reichmann and K. Waku b
a, D. Kunze
’
’ Institute of Pathological and Clinical Biochemisny, Charite Hospital, Humboldt University, Berlin ID. D. R.) and ’ Faculty of Pharmaceutical Sciences, Teikyo University, Kanagawa (Japan) (Received
Key words:
Phosphatidic
18 January
acid; Diacylglycerol;
1988)
CDPdiacylglycerol;
Microsome;
(Rat liver)
The species pattern of phosphatidic acid was compared with that of CDP-diacylglycerol and diacylglycerol synthesized de novo by glycerol 3-phosphate acylation in a CoA ester-generating system in liver microsomes. The similarity of the species patterns of phosphatidic acid and CDP-diacylglycerol indicated that the CTP-phosphatidyl cytidylyltransferase showed no selectivity for individual species of its phosphatidic acid substrate. Since the species pattern of diacylglycerol deviated from that of phosphatidic acid, a slight acyl selectivity of the phosphatidic acid phosphohydrolase or a slight inhomogeneity of its substrate pool might be assumed. For the determination of the molecular species of CDP-diacylglycerol, a new method was developed. By incubation of CDP-diacylglycerol with oligonucleate 5’-nucleotidohydrolase (phosphodiesterase), phosphatidic acid was produced. The CDP-diacylglycerol-derived phosphatidic acid was methylated with diazomethane and then separated by reverse-phase HPLC in 15 molecular species.
Introduction Phosphatidic acid is a key intermediate of the glycerolipid biosynthesis which can either be dephosphorylated to yield diacylglycerol or activated to CDP-diacylglycerol. It has been thought that this diacylglycerol represents a common substrate pool for the formation of phosphatidylcholine, phosphatidylethanolamine and triacylglycerol. CDP-diacylglycerol acts as the precursor for the biosynthesis of the acidic phospholipids phosphatidylinositol and phosphatidylglycerol [l].
Correspondence: B. R&tow, Clinical Biochemistry, Charite 1040 Berlin, D.D.R.
0005-2760/88/$03.50
Institute Hospital,
0 1988 Elsevier
of Pathological and Humboldt University,
Science Publishers
Modern species analysis may be a helpful approach for the investigation of the relative participation of different pathways in the biosynthesis of molecular species of lipids which are important in physiological reactions. Sensitive and well practicable HPLC methods [2] exist for species analysis of microsomal and de novo synthesised diacylglycerol, but the species analysis of de novo-formed CDP-diacylglycerol was performed by argentation thin-layer chromatography only [3,41. A more detailed species analysis of CDP-diacylglycerol might be helpful to differentiate between the CDP-diacylglycerol formed in the de novo path of phosphatidylinositol and that formed via the phosphatidylinositol cycle. It has been supposed that the biosynthesis of these CDP-di-
B.V. (Biomedical
Division)
365
acylglycerols might be located within cells differentially (for reviews see Refs. 1 and 5). It may be assumed that the CDP-diacylglycerol produced in the phosphatidylinositol cycle is highly enriched in the stearoylarachidonyl species [5], whereas de novo-formed CDP-diacylglycerol seemed to contain a much lower proportion of this species [3,6]. Therefore, an extended remodeling of CDP-diacylglycerol [7] and/or phosphatidylinositol ]8,9] producing the stea~larac~donyl species up to the level which was determined in membrane-bound phosphatidylinositol has to be assumed. Despite the fact that the pathways for the formation of complex lipids are well established fl], very little is known about the regulation of these processes. Van Heusden and Van den Bosch [lOI investigated one of the quantitative aspects of the regulation and showed that both phosphatidic acid-consuming enzymes, phosphatidic acid phosphatase and CTP-phosphatidyl cytidylyltransferase, are operating in endoplasmic reticulum far below their maximal capacity. We investigated the selectivity of these phosphatidic acid-consuming enzymes using a newly developed method for the species analysis of CDP-~acylglycerol and previously published HPLC methods for the species analysis of diacylglycerol [2] and phosphatidic acid [ll]. In this way, we described an important qualitative aspect of this branchpoint and found no selectivity for the CTP-phosphatidyl cytidylyltransferase, whereas the phosphatidic acid phosphatase showed weak selectivity for its substrate. Materials and Methods
sn-[i4C]Glycerol 3-phosphate (specific activity 170 mCi/mmol) and [5-3H]CTP (specific activity 25.6 Ci/mmol) were purchased from Amersham International, U.K. CDP-diacylglycerol from egg phosphatidylcholine was obtained from Seidary Research Laboratories, London, Ontario, Canada. CDP-dipalmitoylglycerol, phosphatidic acid and phosphatidylcholine were from Sigma (St. Louis, U.S.A.). Phosphodiesterase from Crotulus durissus (EC 3.1.4.1) was purchased from Boehringer, Man~eim. All other enzymes were purchased
from Sigma (St. Louis, U.S.A.). The chemicals used were of analytical grade.
other
Isolation of microsomes. Rat livers from male animals (body weight 200-250 g) were homogenized in 0.25 M sucrose/l0 mM Tris-HCl (pH 7.4)/l mM EDTA, yielding a 10% (w/v) homogenate. After cent~fugation at 1000 X g for 10 min, the supematant was centrifuged at 200~ x g for 20 min. The microsomes were pelleted from the 20000 x g supernatant by centrifugation at 105000 x g for 60 min. The microsomes were resuspended in the buffer used for the homogenization. Incubation mixture. The incubation mixture contained 0.1 mM EDTA, 3 mM Mg2+, 10 mM cysteine, 0.167 mM CoA, 150 mM KCl, 3.5 mM ATP, 50 mM Hepes (pH 7.8) 2.5 mM [3H]CTP (specific activity 5000-6~0 dpm/nmol), 0.75 mM [ “C]glycerol 3-phosphate (specific activity 10 000 -20000 dpm/nmol) and 1 mg microsomal protein/ml in a final volume of 5 ml. The incubation time at 37 o C was 10 min. Extraction and separafion of lipids. Lipids were extracted according to Bligh and Dyer [12] with ~~oroform/methanol, 0.1 M HCl and 0.05 M HCl instead of water. For the detection of the trace components on the plates after thin-layer chromatography, phosphatidic acid and CDP-diacylglycerol were added as carriers to the lipid extracts before separation. Separations by two-dimensional TLC were performed on silica gel HR (Merck, Darmstadt)/Florisil (Serva) 99 : 1 (w/w) with chloroform/methanol/25% NH, (65 : 35 : 5, v/v) for the first direction and chloroform/ methanol/ acetic acid/ water (50 : 25 : 8 : 3, v/v) for the second direction. Species analysis. Labeled phosphatidic acid and CDP-diacylglycerol were extracted from silica gel by stirring overnight with chloroform/methanol 0.1 M HCl (1: 2, v/v). Then chloroform and 0.05 M HCl were added, yielding a final mixture containing c~orofo~/methanol/O.O5 M HCl in the proportions 1: 1: 1 (v/v). The organic solvents were evaporated in a stream of N, at 40” C. Diazomethane in diethyl ether was added to phosphatidic acid and incubated for 30 min at room temperature. The dimethylphosphatidic acid was
366
then separated by HPLC according to the method of Nakagawa and Waku [ll]. Diacylglycerol extracted from silica gel was converted to the naphthylurethane derivatives, which were separated by HPLC as previously described [2]. Species analysis of CAP-d~acy~g~cero~. In CDPdiacylglycerol, a nucleotide is linked with a lipid moiety by phosphate. Therefore, in principle, two groups of enzymes might be able to cleave this compound, yielding the nucleotide and the lipid moiety: (i) the lipid moiety recognizing phospholipases C and/or D and (ii) the nucleotide recognizing nucleases and/or nucleotidases. Phospholipases C and D of different sources did not cleave this compound. In oligonucleate 5’nucleotidohydrolase (phosphodiesterase), we found an activity which split off phosphatidic acid from CDP-diacylglycerol. This enzyme did not cleave other phospholipids, such as phosphatidylcholine and phosphatidylinositol. CDP-diacylglycerol was mixed with 200 nmol phosphatidylcholine from egg and dried under a stream of nitrogen. After addition of 1 ml 0.05 M Tris-HCl (pH 8.5), we sonicated the mixture twice for lo-15 s at 50 W using a Braun sonicator type Labsonic 1510. Then 0.02 ml phosphodiesterase from C. durissus (1 mg protein/O.5 ml; specific activity 1.5 U/mg protein) and 3 ml diethyl ether were added and stirred for 18 h at room temperature. The incubation mixture was extracted with chloroform/methanol 0.1 M HCI as described above. The lipid extract was separated by one-dimensional TLC on silica gel/Florisil (99 : 1, w/w), and developed with chloroform/ methanol/ acetic acid/ water (50 : 25 : 8 : 3, v/v). Phosphatidic acid derived from CDP-diacylglycerol was extracted from the silica gel, and methylated and separated by HPLC as described [ll]. Using this method, 70-80% of CDP-diacylglycerol was degradated yielding phosphatidic acid. The above-described conditions of the CDP-diacylglycerol degradation were evaluated using mixtures of CDP-diacylglycerol from egg phosphatidylcholine and ~DP-dipalmitoylglycero~. Quantity and quality of the CDP-diacylglycerol degradation were determined from gas-chromato-
TABLE
I
FATTY ACID PATTERN OF CDP-DIACYLGLYCEROL (CDP-DC?) AND CDP-DC&DERIVED PHOSPHATIDIC ACID (PA) AFTER DEGRADATION OF CDP-DG BY PHOSPHODIESTERASE For the 3.5 h and 16 h incubations, different mixtures of CDP-dipal~toylglyceroi and CDP-d~acylglycerol from egg phosphatidyichol~ne were used. n = number of independent incubations; the values are means; SD. is given in parentheses; the values are mass%; fatty acids below 1% are not listed. Fatty acid
16:0 16:l 18:O 1X:1 18:2 20:4 nmol Reaction rate
3.5 h incubation
76 h incubation
CDP-DG fn=4)
CDP-DG frr=5)
41.2 1.o 16.7 26.0 12.6 2.3 214
PA (.i? = 4)
(0.3) (0.3) (0.4) (0.6) (0.1) (0.2)
40.9 1.l 15.4 24.6 11.6 3.5
(0.8) (0.1) (1.8) (2.0) (0.9) (0.5)
(6)
42.4 (8)
56.8 0.8 11.6 20.6 8.9 1.2 192
(0.2) (0.1) (0.1) (0.5) (0.4) (0.3) (2)
19.8%
PA (12 = 5) 55.8 1.1 10.2 21.6 10.4 1.6 128
(1.4) (0.4) (0.3) (1.1) (1.1) (0.4) (18)
66.7%
graphic analyses of the fatty acid patterns of CDP-diacylglycerol mixtures and CDP-diacylglycerol-derived phosphatidic acids using heptadecanoic acid as internal standard [2]. Table I shows that the fatty acid patterns of the CDP-diacylglycerol mixtures used and the phosphatidic acid derived from them were the same. The same fatty acid patterns were obtained when, using different incubation times, the degradation rate was 19 or 66%, respectively. From these results, we concluded that the phosphodiesterase showed no selectivity for individual species of CDP-diacylglycerol. Analytical methods. Protein was determined according to Lowry et al. [13]. The radioactivity was measured using a Philips scintillation counter type PW 4700 equipped with automatic compensation of quenching. For the gas-cbromatographic analysis, we used a Varian 2100 equipped with a flame ionization detector and a Shimadzu C-R 3 A [2]. Result. and Discussion In CTP.
a CoA-ester-generating system we used the microsomal fatty
containing acids as a
367
substrate for the acylation of [‘4C]glycerol 3-phosphate, and this yielded labeled phosphatidic acid, CDP-diacylglycerol and diacylglycerol. This experimental approach seemed to be closely related to in vivo conditions and enabled a direct comparison of the distribution of radioactivity among the molecular species of the lipids of the first branchpoint of glycerolipid biosynthesis. In our system, from the total label incorporated into the lipid fractions, we found 70-808 in the phosphatidic acid, 59% in CDP-diacylglycerol, lo-13% in diacylglycerol and l-2% in the triacylglycerol fraction. The labelling of other lipids separated by TLC was below 1%.
TABLE II SPECIES PATTERNS OF PHOSPHATIDIC ACID (PA) AND CDP-DIACYLGLYCEROL (CDP-DG) SYNTHESIZED DE NOVO IN MICROSOMES OF RAT LIVER Liver microsomes were incubated in a CoA ester-generating system without CTP or in presence of 2.5 mM CTP: n = number of independent incubations; the values are means; S.D. is given in parentheses; the values are percent of label incorporated into individual species from total; n.i. = not identified. Major species of the HPLC fraction
n.i. 18:2-20:4 16:0-20:5
PA without CTP (n=3) 2.2 (0.3) 2.5 (0.5) 2.0 (0.4)
with CTP (n=3) 2.2 (0.4) 2.2 (0.3) 1.7 (0.4)
CDP-DG with CTP (n=3) 2.7 (0.6) 1.9 (0.5) 1.8 (0.2)
18:1-22:6 18:2-18:2
3.1 (0.5)
3.0 (0.6)
3.4 (0.5)
16:0-22:6 18: l-20:4 16:0-20:4 1X:1-18:2 16:0-18:2 18:0-22:6 ni. 18:0-20:4 16:0-18:l
5.5 (0.6) 2.3 (0.1) 4.0 (0.2) 8.3 (1.0) 15.8 (1.7) 7.0 (2.1) 2.0 (0.5) 5.0 (0.8) 14.6 (1.1)
5.1 (0.7) 2.4 (0.4) 4.0 (0.9) 8.5 (0.9) 16.7 (2.4) 5.8 (1.1) 2.1 (0.3) 5.1 (0.9) 15.7 (0.8)
5.2 (2.0) 2.2 (0.1) 5.0 (2.1) 6.7 (1.0) 15.4 (2.3) 5.6 (0.8) 2.0 (0.6) 4.7 (1.2) 14.9 (2.4)
18:0-18:2 16:0-16:O
14.7 (1.9)
16.4 (1.5)
16.0 (1.7)
18:0-20:3 18:0-16:0 lS:O-18:l
4.7 (0.7) 1.7 (0.6) 1.0 (0.7)
5.6 (1.0) 1.6 (0.5) 1.9 (0.8)
4.3 (0.9) 0.9 (0.3) 1.6 (0.4)
TABLE III SPECIES PATTERN OF DIACYLGLYCEROL (DG) SYNTHESIZED DE NOVO IN LIVER MICROSOMES For explanation, see Table II legend. Major species of the HPLC fraction
Without CTP (n=3)
With CTP (n=3)
n.i. n.i. n.i.
9.8 (2.0)
10.0 (0.9)
18:2-22:6 18:2-20:4 16:0-20:5
5.3 (1.7)
5.1 (0.2)
18:2-18:2 16:1-16:l 16:0-22:6 16:0-20:4 16:0-18:2
4.5 2.8 6.4 3.0 10.1
(0.8) (0.7) (0.9) (0.2) (2.9)
4.4 2.5 1.2 3.6 12.1
(0.7) (0.5) (0.3) (0.7) (2.4)
18:1-18:2 16:0-20:5
4.4 (1.6)
5.1 (0.9)
16:0-20:3 16:0-16:O 16:0-18:l
7.3 (1.0) 11.4 (3.2) 10.9 (2.1)
6.8 (0.5) 11.1 (2.1) 9.7 (1.1)
18:0-22:6 18:1-18:l
4.5 (0.5)
3.8 (0.6)
18:0-20:4 18:0-18:2 18:0-22:5 18:0-20:3 18:0-16:0 18:0-18:l
3.6 2.3 0.8 2.9 4.2 2.9
4.0 2.6 0.8 2.7 4.3 3.4
(0.4) (0.1) (0.3) (0.7) (0.8) (0.6)
(0.5) (0.3) (0.4) (0.4) (1.2) (0.6)
Table II shows that phosphatidic acid synthesized de novo contained three major fractions representing the 16 : O-18 : 1, 16 : O-18 : 2 and 18 : O-18 : 2/16 : O-16 : 0 species, respectively. The label incorporated into these species accounted for about 48%. Nearly 20% of the label was found in molecular species containing two unsaturated fatty acids, whereas the radioactivity incorporated into the 16 : O-20 : 4 plus 18 : O-20 : 4 species accounted for less than 10% only. In a rough appro~mation, we were able to compare our results with those obtained after separation of de novo formed phosphatidic acid by argentation TLC [3]. In comparison with the results of Holub and Piekarski [3], we found that the sum of label incorporated into the saturated, monoene and diene species classes was
368
much lower, which is in favour of the label being incorporated into tetraene and polyene species classes. The addition of CTP to the incubation mixture was, on the one hand, necessary when labeled CDP-diacylglycerol had to be produced parallel to phosphatidic acid. On the other hand, it might be possible that higher concentrations of CTP were inhibiting for the glycerol 3-phosphate acylation as shown in brain microsomes [14]. Such inhibition of the acylation of glycerol 3-phosphate might be combined with alterations of the species pattern of de novo synthesized phosphatidic acid. Following this assumption, we measured the effect of CTP on the species patterns of phosphatidic acid and diacylglycerol generated de novo. As shown in Tables II and III, no differences were found. CTP did not change the species patterns of phosphatidic acid or diacylglycerol synthesized de novo in liver microsomes. Further, we compared the species pattern of phosphatidic acid with that of CDP-diacylglycer01. No significant differences were measured (Table 11) between the species patterns of phosphatidic acid and CDP-diacylglycerol synthesized parallel in the same incubation mixture. Because liver microsomes contain CDPdiacylglycerol hydrolase [15,16], it may be assumed that the species pattern of CDP-diacylglycerol is the result of a balanced action of the synthesizing and hydrolyzing enzyme. If this hydrolase showed a predominant hydrolysis of individual species of CDP-diacylglycerol as was shown in brain microsomes [17], some deviation of the species pattern of CDP-diacylglycerol in comparison with that of phosphatidic acid should be expected. From the similarity of both species patterns, it may be assumed that the CDP-diacylglycerol hydrolase of liver microsomes did not follow the acyl species selectivity reported for this enzyme in brain microsomes [17]. According to previously published results [18], an acyl-specific remodeling of the de novo-formed CDP-diacylglycerol seemed excludable. Hence, the CTP-phosphatidyl cytidylyltransferase showed no selectivity for individual species of its substrate. This conclusion is in agreement with that of Holub and Piekarski [3], but different from the results of Thompson [19], who found a small but significant difference between the distribution
of label among the species classes of the phosphatidic acid substrate and CDP-diacylglycerol synthesized de novo from sonicated dispersions of labeled phosphatidic acid by liver microsomes. The experimental approach used in our investigation allowed, on principle, comparison of the species patterns of phosphatidic acid, diacylglycerol and CDP-diacylglycerol. However, it has to be taken into account that a comparison of the species analysis based on different HPLC methods may be troublesome because the elution sequence of some species is different in each chromatogram (compare Tables II and III). A comparison of the species pattern of phosphatidic acid (Table II) with that of diacylglycerol (Table III) showed that the relative proportions of label incorporated into 16 : O-18 : 1, 16 : O-18 : 2 and 18 : l-18 : 2 species of diacylglycerol were somewhat lower. It might be assumed that these deviations ranged beyond what has been caused by the methodical problems. This assumption seems to be justified when the sum of the neighboring fraction 16 : O-18 : 2 plus 18 : l-18 : 2 (PA: 25.2%; DG: 14.5%) and 16 : O-18 : 1 plus 18:0-18:2/16:0-16:0(PA: 32.1%; DG: 24.6%) of phosphatidic acid (PA) and diacylglycerol (DG) were compared. In this way, one minimized the effect of possible cross-contamination of neighboring fractions which might be caused by the peak collection. From these results, it might be assumed that phosphatidic acid phosphatase discriminated against the 18 : l-18 : 2, 16 : O-18 : 2 and 16 :0-18: 1 species of phosphatidic acid. Using argentation TLC, Akesson et al. [20] measured the initial incorporation of [ 3HIglycerol into species classes of hepatic phosphatidic acid and diacylglycerol after intraportal injection of the labeled precursor. They assumed from the results of this in vivo experiment the non-selectivity of the phosphatidic acid phosphatase. The non-selectivity of the phosphatidic acid phosphatase was assumed also from the results of investigations using exogenous phosphatidic acid as substrate [21,22]. The species pattern of diacylglycerol might be changed by diacylglycerol-consuming reactions which may be considered to prefer individual species. The label incorporated into triacylglycerol in our system was 10-20s of that incorporated
369
into diacylglycerol. The low conversion rate of diacylglycerol to triacylglycerol and the assumption of random diacylglycerol utilization for the triacylglycerol formation [23,24] seemed to exclude a change of species pattern of diacylglycerol by diacylglycerol acyltransferase. Microsomal diacylglycerol lipase is another enzyme which may be assumed to select individual acylspecies, although this idea could not be supported by experiments using lung microsomes [25]. Selective action of the diacylglycerol kinase might be another possibility to produce differences between the species patterns of phosphatidic acid and diacylglycerol over the time-course of incubation. It seems that this enzyme of lung microsomes showed a slight preference for individual species of diacylglycerol as exogenous substrate [26], but it is questionable as to whether such findings may be relevant to de novo-synthesized membrane-bound diacylglycer01s. Besides the properties of the enzymes included in the diacylglycerol metabolism, the availability of molecular species of phosphatidic acid might be different for the phosphatidic acid phosphatase. Assuming a compartmentation of phosphatidic acid species, CTP-phosphatidylcytidylyltransferase, in contrast to the phosphatidic acid phosphatase, has to be randomly distributed among the compartments of phosphatidic acid because the species pattern of CDP-diacylglycerol showed no differences from that of phosphatidic acid. Summarizing our results, we conclude that one of the phosphatidic acid-consuming enzymes - the CTP-phosphatidyl cytidylyltransferase - showed no selectivity, and the other one - the phosphatidic acid phosphatase - showed a slight selectivity for its substrate, or the substrate pool exhibited some inhomogeneity for this enzyme. If a species selectivity of the phosphatidic acid phosphatase exists at all, it is certainly of minor physiological importance for the formation of individual species patterns of glycerolipids synthesized from diacylglycerol. It has to be taken into account that phosphatidic acid, CDP-diacylglycerol and also diacylglycerol [27] synthesized de novo represent intermediates which are immediately converted in vivo. Therefore, an in vitro system forcing enrichment of phosphatidic acid over its steady state
level in vivo by relatively high concentrations of glycerol 3-phosphate might cause artificial results. References 1 Kennedy, E.P. (1986) in Lipids and Membranes (Op den Kamp, J.A.F., Roelfson, B. and Wirtz, K.W.A.. eds.), pp. 171-207, Elsevier, Amsterdam. Rtistow, B., Rabe, H. and Kunze, D. (1987) J. Chromatogr. 37, 191-222. Holub, B.J. and Piekarski, J. (1975) Lipids 4, 251-257. Bishop, H.H. and Strickland, K.P. (1976) Can. J. Biochem. 54, 249-260. 5 Downes, C.D. and Michell, R.H. (1985) in Molecular Mechanisms of Transmembrane Signalling, Elsevier Science Publishers, pp. 3-56. 6 Akino, T. and Shimojo, T. (1970) B&hem. Biophys. Acta 210, 343-346. 7 Thompson, W. and MacDonald, G. (1979) J. Biol. Chem. 254, 3311-3314. 8 Holub, B.L. and Kuksis, A. (1972) Lipids 7, 78-89. 9 Baker, R.R. and Thompson, W. (1973) J. Biol. Chem. 248, 7060-7066. 10 Van Heusden, G.P.H. and Van den Bosch. H. (1978) Eur. J. Biochem. 84, 405-412. 11 Nakagawa, Y. and Waku, K. (1986) J. Chromatogr. 381, 225-231. 12 Bligh, E.G. and Dyer, W.J. (1959) Can. J. Biochem. 37, 911-917. 13 Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 14 Possmeyer, F., Meiners, B. and Mudd, J.B. (1973) B&hem. J. 132, 381-394. 15 Sturton, R.G. and Brindley, D.N. (1977) B&hem. J. 162, 25-32. 16 Zborowski, J. and Brindley, D.N. (1983) Biochim. Biophys. Acta 751, 473-483. 17 Murthy, P.P.N. and Agranoff, B.W. (1982) Biochim. Biophys. Acta 712, 473-483. 18 Thompson, W. and Zuk, R.T. (1983) J. Biol. Chem. 258, 9623. 19 Thompson, W. (1978) in Cyclitols and Phosphoinositides (Wells, W.W. and Eisenberg, F., eds.), pp. 215-221. Academic Press, New York. 20 Akesson, B., Elovson, J. and Arvidson, G. (1970) B&him. Biophys. Acta 210, 15-27. 21 McCaman, R.E., Smith, M. and Cook. K. (1965) J. Biol. Chem. 240, 3513-3517. 22 Mitchell, M.P., Brindley, D.N. and Hiibscher, G. (1971) Eur. J. Biochem. 18, 214-220. 23 Hill, E.E., Lands, W.E.K. and Slakey, Sr. P.M. (1968) Lipids 3, 411-419. 24 Akesson, B. (1969) Eur. J. B&hem. 9, 406-414. 25 Riistow, B., Kunze, D., Rabe, H. and Reichmann, 0. (1985) Biochim. Biophys. Acta 835, 465-476. 26 Ide, H. and Weir&told, P.A. (1982) B&him. Biophys. Acta 713, 547-554. 27 Rtistow, B. and Kunze, D. (1987) Biochim. Biophys. Acta 921, 552-558.
Biochimicu
et Biophysics
Acta, 961 (1988) 364-369 Elsevier
BBA 52888
Comparison of the HPLC-separated CDP-diacylglycerol
species patterns of phosphatidic acid,
and diacylglycerol synthesized de novo in rat liver microsomes (a new method)
B. Riistow
a, Y. Nakagawa
b, H. Rabe a, G. Reichmann and K. Waku b
a, D. Kunze
’
’ Institute of Pathological and Clinical Biochemisny, Charite Hospital, Humboldt University, Berlin ID. D. R.) and ’ Faculty of Pharmaceutical Sciences, Teikyo University, Kanagawa (Japan) (Received
Key words:
Phosphatidic
18 January
acid; Diacylglycerol;
1988)
CDPdiacylglycerol;
Microsome;
(Rat liver)
The species pattern of phosphatidic acid was compared with that of CDP-diacylglycerol and diacylglycerol synthesized de novo by glycerol 3-phosphate acylation in a CoA ester-generating system in liver microsomes. The similarity of the species patterns of phosphatidic acid and CDP-diacylglycerol indicated that the CTP-phosphatidyl cytidylyltransferase showed no selectivity for individual species of its phosphatidic acid substrate. Since the species pattern of diacylglycerol deviated from that of phosphatidic acid, a slight acyl selectivity of the phosphatidic acid phosphohydrolase or a slight inhomogeneity of its substrate pool might be assumed. For the determination of the molecular species of CDP-diacylglycerol, a new method was developed. By incubation of CDP-diacylglycerol with oligonucleate 5’-nucleotidohydrolase (phosphodiesterase), phosphatidic acid was produced. The CDP-diacylglycerol-derived phosphatidic acid was methylated with diazomethane and then separated by reverse-phase HPLC in 15 molecular species.
Introduction Phosphatidic acid is a key intermediate of the glycerolipid biosynthesis which can either be dephosphorylated to yield diacylglycerol or activated to CDP-diacylglycerol. It has been thought that this diacylglycerol represents a common substrate pool for the formation of phosphatidylcholine, phosphatidylethanolamine and triacylglycerol. CDP-diacylglycerol acts as the precursor for the biosynthesis of the acidic phospholipids phosphatidylinositol and phosphatidylglycerol [l].
Correspondence: B. R&tow, Clinical Biochemistry, Charite 1040 Berlin, D.D.R.
0005-2760/88/$03.50
Institute Hospital,
0 1988 Elsevier
of Pathological and Humboldt University,
Science Publishers
Modern species analysis may be a helpful approach for the investigation of the relative participation of different pathways in the biosynthesis of molecular species of lipids which are important in physiological reactions. Sensitive and well practicable HPLC methods [2] exist for species analysis of microsomal and de novo synthesised diacylglycerol, but the species analysis of de novo-formed CDP-diacylglycerol was performed by argentation thin-layer chromatography only [3,41. A more detailed species analysis of CDP-diacylglycerol might be helpful to differentiate between the CDP-diacylglycerol formed in the de novo path of phosphatidylinositol and that formed via the phosphatidylinositol cycle. It has been supposed that the biosynthesis of these CDP-di-
B.V. (Biomedical
Division)
365
acylglycerols might be located within cells differentially (for reviews see Refs. 1 and 5). It may be assumed that the CDP-diacylglycerol produced in the phosphatidylinositol cycle is highly enriched in the stearoylarachidonyl species [5], whereas de novo-formed CDP-diacylglycerol seemed to contain a much lower proportion of this species [3,6]. Therefore, an extended remodeling of CDP-diacylglycerol [7] and/or phosphatidylinositol ]8,9] producing the stea~larac~donyl species up to the level which was determined in membrane-bound phosphatidylinositol has to be assumed. Despite the fact that the pathways for the formation of complex lipids are well established fl], very little is known about the regulation of these processes. Van Heusden and Van den Bosch [lOI investigated one of the quantitative aspects of the regulation and showed that both phosphatidic acid-consuming enzymes, phosphatidic acid phosphatase and CTP-phosphatidyl cytidylyltransferase, are operating in endoplasmic reticulum far below their maximal capacity. We investigated the selectivity of these phosphatidic acid-consuming enzymes using a newly developed method for the species analysis of CDP-~acylglycerol and previously published HPLC methods for the species analysis of diacylglycerol [2] and phosphatidic acid [ll]. In this way, we described an important qualitative aspect of this branchpoint and found no selectivity for the CTP-phosphatidyl cytidylyltransferase, whereas the phosphatidic acid phosphatase showed weak selectivity for its substrate. Materials and Methods
sn-[i4C]Glycerol 3-phosphate (specific activity 170 mCi/mmol) and [5-3H]CTP (specific activity 25.6 Ci/mmol) were purchased from Amersham International, U.K. CDP-diacylglycerol from egg phosphatidylcholine was obtained from Seidary Research Laboratories, London, Ontario, Canada. CDP-dipalmitoylglycerol, phosphatidic acid and phosphatidylcholine were from Sigma (St. Louis, U.S.A.). Phosphodiesterase from Crotulus durissus (EC 3.1.4.1) was purchased from Boehringer, Man~eim. All other enzymes were purchased
from Sigma (St. Louis, U.S.A.). The chemicals used were of analytical grade.
other
Isolation of microsomes. Rat livers from male animals (body weight 200-250 g) were homogenized in 0.25 M sucrose/l0 mM Tris-HCl (pH 7.4)/l mM EDTA, yielding a 10% (w/v) homogenate. After cent~fugation at 1000 X g for 10 min, the supematant was centrifuged at 200~ x g for 20 min. The microsomes were pelleted from the 20000 x g supernatant by centrifugation at 105000 x g for 60 min. The microsomes were resuspended in the buffer used for the homogenization. Incubation mixture. The incubation mixture contained 0.1 mM EDTA, 3 mM Mg2+, 10 mM cysteine, 0.167 mM CoA, 150 mM KCl, 3.5 mM ATP, 50 mM Hepes (pH 7.8) 2.5 mM [3H]CTP (specific activity 5000-6~0 dpm/nmol), 0.75 mM [ “C]glycerol 3-phosphate (specific activity 10 000 -20000 dpm/nmol) and 1 mg microsomal protein/ml in a final volume of 5 ml. The incubation time at 37 o C was 10 min. Extraction and separafion of lipids. Lipids were extracted according to Bligh and Dyer [12] with ~~oroform/methanol, 0.1 M HCl and 0.05 M HCl instead of water. For the detection of the trace components on the plates after thin-layer chromatography, phosphatidic acid and CDP-diacylglycerol were added as carriers to the lipid extracts before separation. Separations by two-dimensional TLC were performed on silica gel HR (Merck, Darmstadt)/Florisil (Serva) 99 : 1 (w/w) with chloroform/methanol/25% NH, (65 : 35 : 5, v/v) for the first direction and chloroform/ methanol/ acetic acid/ water (50 : 25 : 8 : 3, v/v) for the second direction. Species analysis. Labeled phosphatidic acid and CDP-diacylglycerol were extracted from silica gel by stirring overnight with chloroform/methanol 0.1 M HCl (1: 2, v/v). Then chloroform and 0.05 M HCl were added, yielding a final mixture containing c~orofo~/methanol/O.O5 M HCl in the proportions 1: 1: 1 (v/v). The organic solvents were evaporated in a stream of N, at 40” C. Diazomethane in diethyl ether was added to phosphatidic acid and incubated for 30 min at room temperature. The dimethylphosphatidic acid was
366
then separated by HPLC according to the method of Nakagawa and Waku [ll]. Diacylglycerol extracted from silica gel was converted to the naphthylurethane derivatives, which were separated by HPLC as previously described [2]. Species analysis of CAP-d~acy~g~cero~. In CDPdiacylglycerol, a nucleotide is linked with a lipid moiety by phosphate. Therefore, in principle, two groups of enzymes might be able to cleave this compound, yielding the nucleotide and the lipid moiety: (i) the lipid moiety recognizing phospholipases C and/or D and (ii) the nucleotide recognizing nucleases and/or nucleotidases. Phospholipases C and D of different sources did not cleave this compound. In oligonucleate 5’nucleotidohydrolase (phosphodiesterase), we found an activity which split off phosphatidic acid from CDP-diacylglycerol. This enzyme did not cleave other phospholipids, such as phosphatidylcholine and phosphatidylinositol. CDP-diacylglycerol was mixed with 200 nmol phosphatidylcholine from egg and dried under a stream of nitrogen. After addition of 1 ml 0.05 M Tris-HCl (pH 8.5), we sonicated the mixture twice for lo-15 s at 50 W using a Braun sonicator type Labsonic 1510. Then 0.02 ml phosphodiesterase from C. durissus (1 mg protein/O.5 ml; specific activity 1.5 U/mg protein) and 3 ml diethyl ether were added and stirred for 18 h at room temperature. The incubation mixture was extracted with chloroform/methanol 0.1 M HCI as described above. The lipid extract was separated by one-dimensional TLC on silica gel/Florisil (99 : 1, w/w), and developed with chloroform/ methanol/ acetic acid/ water (50 : 25 : 8 : 3, v/v). Phosphatidic acid derived from CDP-diacylglycerol was extracted from the silica gel, and methylated and separated by HPLC as described [ll]. Using this method, 70-80% of CDP-diacylglycerol was degradated yielding phosphatidic acid. The above-described conditions of the CDP-diacylglycerol degradation were evaluated using mixtures of CDP-diacylglycerol from egg phosphatidylcholine and ~DP-dipalmitoylglycero~. Quantity and quality of the CDP-diacylglycerol degradation were determined from gas-chromato-
TABLE
I
FATTY ACID PATTERN OF CDP-DIACYLGLYCEROL (CDP-DC?) AND CDP-DC&DERIVED PHOSPHATIDIC ACID (PA) AFTER DEGRADATION OF CDP-DG BY PHOSPHODIESTERASE For the 3.5 h and 16 h incubations, different mixtures of CDP-dipal~toylglyceroi and CDP-d~acylglycerol from egg phosphatidyichol~ne were used. n = number of independent incubations; the values are means; SD. is given in parentheses; the values are mass%; fatty acids below 1% are not listed. Fatty acid
16:0 16:l 18:O 1X:1 18:2 20:4 nmol Reaction rate
3.5 h incubation
76 h incubation
CDP-DG fn=4)
CDP-DG frr=5)
41.2 1.o 16.7 26.0 12.6 2.3 214
PA (.i? = 4)
(0.3) (0.3) (0.4) (0.6) (0.1) (0.2)
40.9 1.l 15.4 24.6 11.6 3.5
(0.8) (0.1) (1.8) (2.0) (0.9) (0.5)
(6)
42.4 (8)
56.8 0.8 11.6 20.6 8.9 1.2 192
(0.2) (0.1) (0.1) (0.5) (0.4) (0.3) (2)
19.8%
PA (12 = 5) 55.8 1.1 10.2 21.6 10.4 1.6 128
(1.4) (0.4) (0.3) (1.1) (1.1) (0.4) (18)
66.7%
graphic analyses of the fatty acid patterns of CDP-diacylglycerol mixtures and CDP-diacylglycerol-derived phosphatidic acids using heptadecanoic acid as internal standard [2]. Table I shows that the fatty acid patterns of the CDP-diacylglycerol mixtures used and the phosphatidic acid derived from them were the same. The same fatty acid patterns were obtained when, using different incubation times, the degradation rate was 19 or 66%, respectively. From these results, we concluded that the phosphodiesterase showed no selectivity for individual species of CDP-diacylglycerol. Analytical methods. Protein was determined according to Lowry et al. [13]. The radioactivity was measured using a Philips scintillation counter type PW 4700 equipped with automatic compensation of quenching. For the gas-cbromatographic analysis, we used a Varian 2100 equipped with a flame ionization detector and a Shimadzu C-R 3 A [2]. Result. and Discussion In CTP.
a CoA-ester-generating system we used the microsomal fatty
containing acids as a
367
substrate for the acylation of [‘4C]glycerol 3-phosphate, and this yielded labeled phosphatidic acid, CDP-diacylglycerol and diacylglycerol. This experimental approach seemed to be closely related to in vivo conditions and enabled a direct comparison of the distribution of radioactivity among the molecular species of the lipids of the first branchpoint of glycerolipid biosynthesis. In our system, from the total label incorporated into the lipid fractions, we found 70-808 in the phosphatidic acid, 59% in CDP-diacylglycerol, lo-13% in diacylglycerol and l-2% in the triacylglycerol fraction. The labelling of other lipids separated by TLC was below 1%.
TABLE II SPECIES PATTERNS OF PHOSPHATIDIC ACID (PA) AND CDP-DIACYLGLYCEROL (CDP-DG) SYNTHESIZED DE NOVO IN MICROSOMES OF RAT LIVER Liver microsomes were incubated in a CoA ester-generating system without CTP or in presence of 2.5 mM CTP: n = number of independent incubations; the values are means; S.D. is given in parentheses; the values are percent of label incorporated into individual species from total; n.i. = not identified. Major species of the HPLC fraction
n.i. 18:2-20:4 16:0-20:5
PA without CTP (n=3) 2.2 (0.3) 2.5 (0.5) 2.0 (0.4)
with CTP (n=3) 2.2 (0.4) 2.2 (0.3) 1.7 (0.4)
CDP-DG with CTP (n=3) 2.7 (0.6) 1.9 (0.5) 1.8 (0.2)
18:1-22:6 18:2-18:2
3.1 (0.5)
3.0 (0.6)
3.4 (0.5)
16:0-22:6 18: l-20:4 16:0-20:4 1X:1-18:2 16:0-18:2 18:0-22:6 ni. 18:0-20:4 16:0-18:l
5.5 (0.6) 2.3 (0.1) 4.0 (0.2) 8.3 (1.0) 15.8 (1.7) 7.0 (2.1) 2.0 (0.5) 5.0 (0.8) 14.6 (1.1)
5.1 (0.7) 2.4 (0.4) 4.0 (0.9) 8.5 (0.9) 16.7 (2.4) 5.8 (1.1) 2.1 (0.3) 5.1 (0.9) 15.7 (0.8)
5.2 (2.0) 2.2 (0.1) 5.0 (2.1) 6.7 (1.0) 15.4 (2.3) 5.6 (0.8) 2.0 (0.6) 4.7 (1.2) 14.9 (2.4)
18:0-18:2 16:0-16:O
14.7 (1.9)
16.4 (1.5)
16.0 (1.7)
18:0-20:3 18:0-16:0 lS:O-18:l
4.7 (0.7) 1.7 (0.6) 1.0 (0.7)
5.6 (1.0) 1.6 (0.5) 1.9 (0.8)
4.3 (0.9) 0.9 (0.3) 1.6 (0.4)
TABLE III SPECIES PATTERN OF DIACYLGLYCEROL (DG) SYNTHESIZED DE NOVO IN LIVER MICROSOMES For explanation, see Table II legend. Major species of the HPLC fraction
Without CTP (n=3)
With CTP (n=3)
n.i. n.i. n.i.
9.8 (2.0)
10.0 (0.9)
18:2-22:6 18:2-20:4 16:0-20:5
5.3 (1.7)
5.1 (0.2)
18:2-18:2 16:1-16:l 16:0-22:6 16:0-20:4 16:0-18:2
4.5 2.8 6.4 3.0 10.1
(0.8) (0.7) (0.9) (0.2) (2.9)
4.4 2.5 1.2 3.6 12.1
(0.7) (0.5) (0.3) (0.7) (2.4)
18:1-18:2 16:0-20:5
4.4 (1.6)
5.1 (0.9)
16:0-20:3 16:0-16:O 16:0-18:l
7.3 (1.0) 11.4 (3.2) 10.9 (2.1)
6.8 (0.5) 11.1 (2.1) 9.7 (1.1)
18:0-22:6 18:1-18:l
4.5 (0.5)
3.8 (0.6)
18:0-20:4 18:0-18:2 18:0-22:5 18:0-20:3 18:0-16:0 18:0-18:l
3.6 2.3 0.8 2.9 4.2 2.9
4.0 2.6 0.8 2.7 4.3 3.4
(0.4) (0.1) (0.3) (0.7) (0.8) (0.6)
(0.5) (0.3) (0.4) (0.4) (1.2) (0.6)
Table II shows that phosphatidic acid synthesized de novo contained three major fractions representing the 16 : O-18 : 1, 16 : O-18 : 2 and 18 : O-18 : 2/16 : O-16 : 0 species, respectively. The label incorporated into these species accounted for about 48%. Nearly 20% of the label was found in molecular species containing two unsaturated fatty acids, whereas the radioactivity incorporated into the 16 : O-20 : 4 plus 18 : O-20 : 4 species accounted for less than 10% only. In a rough appro~mation, we were able to compare our results with those obtained after separation of de novo formed phosphatidic acid by argentation TLC [3]. In comparison with the results of Holub and Piekarski [3], we found that the sum of label incorporated into the saturated, monoene and diene species classes was
368
much lower, which is in favour of the label being incorporated into tetraene and polyene species classes. The addition of CTP to the incubation mixture was, on the one hand, necessary when labeled CDP-diacylglycerol had to be produced parallel to phosphatidic acid. On the other hand, it might be possible that higher concentrations of CTP were inhibiting for the glycerol 3-phosphate acylation as shown in brain microsomes [14]. Such inhibition of the acylation of glycerol 3-phosphate might be combined with alterations of the species pattern of de novo synthesized phosphatidic acid. Following this assumption, we measured the effect of CTP on the species patterns of phosphatidic acid and diacylglycerol generated de novo. As shown in Tables II and III, no differences were found. CTP did not change the species patterns of phosphatidic acid or diacylglycerol synthesized de novo in liver microsomes. Further, we compared the species pattern of phosphatidic acid with that of CDP-diacylglycer01. No significant differences were measured (Table 11) between the species patterns of phosphatidic acid and CDP-diacylglycerol synthesized parallel in the same incubation mixture. Because liver microsomes contain CDPdiacylglycerol hydrolase [15,16], it may be assumed that the species pattern of CDP-diacylglycerol is the result of a balanced action of the synthesizing and hydrolyzing enzyme. If this hydrolase showed a predominant hydrolysis of individual species of CDP-diacylglycerol as was shown in brain microsomes [17], some deviation of the species pattern of CDP-diacylglycerol in comparison with that of phosphatidic acid should be expected. From the similarity of both species patterns, it may be assumed that the CDP-diacylglycerol hydrolase of liver microsomes did not follow the acyl species selectivity reported for this enzyme in brain microsomes [17]. According to previously published results [18], an acyl-specific remodeling of the de novo-formed CDP-diacylglycerol seemed excludable. Hence, the CTP-phosphatidyl cytidylyltransferase showed no selectivity for individual species of its substrate. This conclusion is in agreement with that of Holub and Piekarski [3], but different from the results of Thompson [19], who found a small but significant difference between the distribution
of label among the species classes of the phosphatidic acid substrate and CDP-diacylglycerol synthesized de novo from sonicated dispersions of labeled phosphatidic acid by liver microsomes. The experimental approach used in our investigation allowed, on principle, comparison of the species patterns of phosphatidic acid, diacylglycerol and CDP-diacylglycerol. However, it has to be taken into account that a comparison of the species analysis based on different HPLC methods may be troublesome because the elution sequence of some species is different in each chromatogram (compare Tables II and III). A comparison of the species pattern of phosphatidic acid (Table II) with that of diacylglycerol (Table III) showed that the relative proportions of label incorporated into 16 : O-18 : 1, 16 : O-18 : 2 and 18 : l-18 : 2 species of diacylglycerol were somewhat lower. It might be assumed that these deviations ranged beyond what has been caused by the methodical problems. This assumption seems to be justified when the sum of the neighboring fraction 16 : O-18 : 2 plus 18 : l-18 : 2 (PA: 25.2%; DG: 14.5%) and 16 : O-18 : 1 plus 18:0-18:2/16:0-16:0(PA: 32.1%; DG: 24.6%) of phosphatidic acid (PA) and diacylglycerol (DG) were compared. In this way, one minimized the effect of possible cross-contamination of neighboring fractions which might be caused by the peak collection. From these results, it might be assumed that phosphatidic acid phosphatase discriminated against the 18 : l-18 : 2, 16 : O-18 : 2 and 16 :0-18: 1 species of phosphatidic acid. Using argentation TLC, Akesson et al. [20] measured the initial incorporation of [ 3HIglycerol into species classes of hepatic phosphatidic acid and diacylglycerol after intraportal injection of the labeled precursor. They assumed from the results of this in vivo experiment the non-selectivity of the phosphatidic acid phosphatase. The non-selectivity of the phosphatidic acid phosphatase was assumed also from the results of investigations using exogenous phosphatidic acid as substrate [21,22]. The species pattern of diacylglycerol might be changed by diacylglycerol-consuming reactions which may be considered to prefer individual species. The label incorporated into triacylglycerol in our system was 10-20s of that incorporated
369
into diacylglycerol. The low conversion rate of diacylglycerol to triacylglycerol and the assumption of random diacylglycerol utilization for the triacylglycerol formation [23,24] seemed to exclude a change of species pattern of diacylglycerol by diacylglycerol acyltransferase. Microsomal diacylglycerol lipase is another enzyme which may be assumed to select individual acylspecies, although this idea could not be supported by experiments using lung microsomes [25]. Selective action of the diacylglycerol kinase might be another possibility to produce differences between the species patterns of phosphatidic acid and diacylglycerol over the time-course of incubation. It seems that this enzyme of lung microsomes showed a slight preference for individual species of diacylglycerol as exogenous substrate [26], but it is questionable as to whether such findings may be relevant to de novo-synthesized membrane-bound diacylglycer01s. Besides the properties of the enzymes included in the diacylglycerol metabolism, the availability of molecular species of phosphatidic acid might be different for the phosphatidic acid phosphatase. Assuming a compartmentation of phosphatidic acid species, CTP-phosphatidylcytidylyltransferase, in contrast to the phosphatidic acid phosphatase, has to be randomly distributed among the compartments of phosphatidic acid because the species pattern of CDP-diacylglycerol showed no differences from that of phosphatidic acid. Summarizing our results, we conclude that one of the phosphatidic acid-consuming enzymes - the CTP-phosphatidyl cytidylyltransferase - showed no selectivity, and the other one - the phosphatidic acid phosphatase - showed a slight selectivity for its substrate, or the substrate pool exhibited some inhomogeneity for this enzyme. If a species selectivity of the phosphatidic acid phosphatase exists at all, it is certainly of minor physiological importance for the formation of individual species patterns of glycerolipids synthesized from diacylglycerol. It has to be taken into account that phosphatidic acid, CDP-diacylglycerol and also diacylglycerol [27] synthesized de novo represent intermediates which are immediately converted in vivo. Therefore, an in vitro system forcing enrichment of phosphatidic acid over its steady state
level in vivo by relatively high concentrations of glycerol 3-phosphate might cause artificial results. References 1 Kennedy, E.P. (1986) in Lipids and Membranes (Op den Kamp, J.A.F., Roelfson, B. and Wirtz, K.W.A.. eds.), pp. 171-207, Elsevier, Amsterdam. Rtistow, B., Rabe, H. and Kunze, D. (1987) J. Chromatogr. 37, 191-222. Holub, B.J. and Piekarski, J. (1975) Lipids 4, 251-257. Bishop, H.H. and Strickland, K.P. (1976) Can. J. Biochem. 54, 249-260. 5 Downes, C.D. and Michell, R.H. (1985) in Molecular Mechanisms of Transmembrane Signalling, Elsevier Science Publishers, pp. 3-56. 6 Akino, T. and Shimojo, T. (1970) B&hem. Biophys. Acta 210, 343-346. 7 Thompson, W. and MacDonald, G. (1979) J. Biol. Chem. 254, 3311-3314. 8 Holub, B.L. and Kuksis, A. (1972) Lipids 7, 78-89. 9 Baker, R.R. and Thompson, W. (1973) J. Biol. Chem. 248, 7060-7066. 10 Van Heusden, G.P.H. and Van den Bosch. H. (1978) Eur. J. Biochem. 84, 405-412. 11 Nakagawa, Y. and Waku, K. (1986) J. Chromatogr. 381, 225-231. 12 Bligh, E.G. and Dyer, W.J. (1959) Can. J. Biochem. 37, 911-917. 13 Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 14 Possmeyer, F., Meiners, B. and Mudd, J.B. (1973) B&hem. J. 132, 381-394. 15 Sturton, R.G. and Brindley, D.N. (1977) B&hem. J. 162, 25-32. 16 Zborowski, J. and Brindley, D.N. (1983) Biochim. Biophys. Acta 751, 473-483. 17 Murthy, P.P.N. and Agranoff, B.W. (1982) Biochim. Biophys. Acta 712, 473-483. 18 Thompson, W. and Zuk, R.T. (1983) J. Biol. Chem. 258, 9623. 19 Thompson, W. (1978) in Cyclitols and Phosphoinositides (Wells, W.W. and Eisenberg, F., eds.), pp. 215-221. Academic Press, New York. 20 Akesson, B., Elovson, J. and Arvidson, G. (1970) B&him. Biophys. Acta 210, 15-27. 21 McCaman, R.E., Smith, M. and Cook. K. (1965) J. Biol. Chem. 240, 3513-3517. 22 Mitchell, M.P., Brindley, D.N. and Hiibscher, G. (1971) Eur. J. Biochem. 18, 214-220. 23 Hill, E.E., Lands, W.E.K. and Slakey, Sr. P.M. (1968) Lipids 3, 411-419. 24 Akesson, B. (1969) Eur. J. B&hem. 9, 406-414. 25 Riistow, B., Kunze, D., Rabe, H. and Reichmann, 0. (1985) Biochim. Biophys. Acta 835, 465-476. 26 Ide, H. and Weir&told, P.A. (1982) B&him. Biophys. Acta 713, 547-554. 27 Rtistow, B. and Kunze, D. (1987) Biochim. Biophys. Acta 921, 552-558.