1,3-Diacylglycero-2-phosphocholines - Synthesis, aggregation behaviour and properties as inhibitors of phospholipase D

Article abstract of DOI:10.1016/S0009-3084(99)00120-6

A series of 1,3-diacylglycero-2-phosphocholines (1,3-PCs) with acyl chain lengths of C₈-C₁₈ were synthesised by chemical introduction of the phosphocholine moiety into the regioisomerically pure 1,3-diacylglycerols, which were obtained from glycerol and the vinyl esters of fatty acid by means of lipase from Rhizomucor mihei. The 1,3-PCs being regioisomers of the natural glycerophospholipids were studied with respect to their aggregation behaviour in the absence and in the presence of sodium dodecylsulfate (SDS) as well as their properties as substrates and inhibitors of phospholipase D (PLD) from cabbage. While the main structures of the pure 1,3-PCs were micelles (C₈), liposomes (C₁₀, C₁₂) or planar bilayers (C₁₄, C₁₆, C₁₈), the addition of SDS resulted in the formation of mixed micelles (C₈, C₁₀) and mixed liposomes (C₁₂, C₁₄, C₁₆, C₁₈). None of the 1,3-PCs was found to be hydrolysed by PLD, whereas all of them showed inhibitory properties in the standard assay for PLD. The inhibitory power was strongest with 1,3-didecanoylglycero-2-phosphocholine (IC₅₀=43 μM). Copyright (C) 2000 Elsevier Science Ireland Ltd.

Full text of DOI:10.1016/S0009-3084(99)00120-6

Chemistry and Physics of Lipids
104 (2000) 57–66
www.elsevier.com/locate/chemphyslip
1,3-Diacylglycero-2-phosphocholines — synthesis,
aggregation behaviour and properties as inhibitors of
phospholipase D
Regine Haftendorn, Gabriele Schwarze, Renate Ulbrich-Hofmann *
Institut fu¨r Biotechnologie, Martin-Luther-Uni6ersita¨t Halle-Wittenberg, Kurt-Mothes-Strasse 3, D-06120 Halle, Germany
Received 19 April 1999; received in revised form 3 September 1999; accepted 21 September 1999
Abstract
A series of 1,3-diacylglycero-2-phosphocholines (1,3-PCs) with acyl chain lengths of C8–C18 were synthesised by
chemical introduction of the phosphocholine moiety into the regioisomerically pure 1,3-diacylglycerols, which were
obtained from glycerol and the vinyl esters of fatty acid by means of lipase from Rhizomucor mihei. The 1,3-PCs being
regioisomers of the natural glycerophospholipids were studied with respect to their aggregation behaviour in the
absence and in the presence of sodium dodecylsulfate (SDS) as well as their properties as substrates and inhibitors of
phospholipase D (PLD) from cabbage. While the main structures of the pure 1,3-PCs were micelles (C8), liposomes
(C10, C12) or planar bilayers (C14, C16, C18), the addition of SDS resulted in the formation of mixed micelles (C8, C10)
and mixed liposomes (C12, C14, C16, C18). None of the 1,3-PCs was found to be hydrolysed by PLD, whereas all of
them showed inhibitory properties in the standard assay for PLD. The inhibitory power was strongest with
1,3-didecanoylglycero-2-phosphocholine (IC50=43 mM). © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: 1,3-Diacylglycero-2-phosphocholines; Inhibition; Liposomes; Micelles; Phospholipase D; Synthesis
Abbre6iations: 1,2-dC8PC, 1,2-dioctanoyl-sn-glycero-3-phosphocholine; 1,2-dC10PC, 1,2-didecanoyl-sn-glycero-3-phosphocholine;
1,2-PC, 1,2-diacylglycero-3-phosphocholine; 1,3-PC, 1,3-diacylglycero-2-phosphocholine; 1,3-dC8PC, 1,3-dioctanoylglycero-2-phos-
phocholine; 1,3-dC10PC, 1,3-didecanoylglycero-2-phosphocholine; 1,3-dC12PC, 1,3-dilauroylglycero-2-phosphocholine; 1,3-dC14PC,
1,3-dimyristoylglycero-2-phosphocholine; 1,3-dC16PC, 1,3-dipalmitoylglycero-2-phosphocholine; 1,3-dC18PC, 1,3-distearoylglycero-
2-phosphocholine; CC, column chromatography; cmc, critical micelle concentration; Cn, acyl residue of 1,3- or 1,2-PC; DPH,
1,6-diphenyl-1,3,5-hexa-triene; PC, phosphatidylcholine (egg yolk); PLD, phospholipase D; SDS, sodium dodecylsulfate.
* Corresponding author. Fax: +49-345-5527013.
E-mail address: ulbrich-hofmann@biochemtech.uni-halle.de (R. Ulbrich-Hofmann)
0009-3084/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S0009-3084(99)00120-6

58
R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
dilauroylglycero-2-phosphocholine (1,3-diC12PC)
1. Introduction
(Haftendorn and Ulbrich-Hofmann, 1995). These
compounds were studied with respect to their
tendency to form reverse micelles in organic me-
dia showing great similarities to the correspond-
ing 1,2-diacyl-sn-glycero-3-phosphatides (Frense
et al., 1995) and with respect to their properties as
effectors in the catalysis by phospholipase D
(PLD) (Dittrich et al., 1998).
While 1,2-diacyl-sn-glycero-3-phosphocholines
(1,2-PCs) isolated from natural sources or pro-
duced by chemical syntheses have been widely
studied under different aspects of interest (Cevc,
1993), there are only a few reports on the re-
gioisomeric 1,3-diacylglycero-2-phosphocholines
(1,3-PCs). De Haas and van Deenen (van Deenen
and de Haas, 1963; de Haas and van Deenen,
1964, 1965) studied several 1,3-PCs with respect
to their cleavability by phospholipase A. Using
1,3-PCs as substrates, the inhibitory power of two
PLD has been recently recognised as an impor-
tant participant in cellular signal transduction
processes in plants (Munnik et al., 1998) and
animals (Exton, 1997) but little is known about
the action mechanism of PLD and its effectors.
The main substrate of most types of PLD is
phosphatidylcholine (PC) yielding phosphatidic
acid and choline (Heller, 1978). In the presence of
an additional alcohol the enzyme is also able to
transfer the phosphatidic moiety of the substrate
to this alcohol, which made PLD an important
tool in phospholipid synthesis (d’Arrigo and
Servi, 1997). Detergents are known to strongly
influence PLD activity. Anionic detergents like
sodium dodecylsulfate (SDS) (Dawson and Hem-
ington, 1967; Heller, 1978) were found to stimu-
late PLD activity, whereas cationic detergents
were described as inhibitors (Dawson and Hem-
ington, 1967). In the experiments with the phos-
pholipid analogous 2-modified 1,3-diacylglycerols,
the anionic compounds 1,3-dilauroylglycerol 2-
phosphate and 1,3-dilauroylglycerol 2-sulfate also
acted as activators, whereas 1,3-dilauroylglycero-
2-phosphocholine inhibited the enzyme (Dittrich
et al., 1998).
In the present paper these studies are extended
to a series of 1,3-diacylglycero-2-phosphocholines
with different chain lengths of fatty acids (C8–
C18). The synthesis of these compounds according
to the newly established chemoenzymatic strategy
has been described. The compounds have been
characterised with respect to their aggregation
behaviour in aqueous medium by dynamic laser
light scattering and determination of the critical
micelle concentration (cmc). Furthermore, their
efficiency as potential substrates or inhibitors of
PLD (from cabbage) has been examined.
stereoisomeric
monoacylamino-glycero-2-phos-
phocholines on PLA2 was investigated (de Haas et
al., 1990). In some papers the phase transition
temperature (Smith, 1981; Stu¨mpel et al., 1983;
Nuhn et al., 1986), and the bilayer characteristics
(Seelig et al., 1980; Kunitake et al., 1984) of 1,3-
and 1,2-PCs were compared.
The main reason for the poor knowledge of the
1,3-PCs may result from difficulties in the synthe-
sis of regioisomerically pure 1,3-diacylglycerols
from which these compounds are obtainable. 1,3-
Diacylglycerols were mostly synthesised by acyla-
tion
of
glycerol
(Rose,
1947)
or
2-O-benzylglycerol (Hartmann, 1958) with fatty
acid chlorides or anhydrides. Alternative routes
are the reactions of fatty acids with glycidol (Pal-
tauf and Hermetter, 1994) or 1,3-benzylidene glyc-
erol (Nuhn et al., 1986). Starting from the
1,3-diacylglycerols, 1,3-PCs were mostly synthe-
sised via 2-bromoethyl dichlorophosphate as orig-
inally described by Hirth and Berchtold (1958),
and subsequent replacement of the bromine by
trimethylamine (Bonsen et al., 1972; Hansen et
al., 1982; Nuhn et al., 1986).
Recently, however, a new elegant method has
been described for the synthesis of regioisomeri-
cally pure 1,3-diacylglycerols (Berger et al., 1992),
which exploits the 1,3-selective action of lipase
(Mucor mihei ) to catalyse the reaction of glycerol
with vinyl esters of fatty acids. We used 1,3-dia-
cylglycerols obtained in this way to synthesise
several anionic and neutral 2-modified 1,3-diacyl-
glycerols such as 1,3-dilauroylglycerol 2-phos-
phate, 1,3-dilauroylglycerol 2-sulfate or 1,3-

R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
59
2. Materials and methods
under vigorous stirring and the reaction mixture
was kept at 50°C in a tightly closed vessel
overnight. Thin layer chromatography indicated
the complete conversion of the bromoethyl phos-
phate. The solvent was removed under reduced
pressure. The residue was dissolved in 40 ml chlo-
roform and extracted with the following mixtures:
twice with 35 ml formic acid–methanol (5:7),
twice with 35 ml 0.1 M sodium acetate–methanol
(5:7) and finally with 20 ml 1 M NaCl–methanol
(5:7). The remaining organic solution was dried
with Na2SO4, and the solvent was removed by
evaporation. The products (yellow oils) were
purified by CC on silica gel using a gradient
elution with chloroform–methanol–water. After
evaporation of the solvent, the products were
dried in a vacuum dessiccator with P4O10.
2.1. Materials
Silica gel 60 (230–400 mesh) for column chro-
matography (CC) and aluminium sheets silica 60
for thin-layer chromatography were from Merck,
Darmstadt (Germany). PC (egg yolk), 98.8%, was
a gift from Lipoid (Ludwigshafen, Germany).
SDS (special grade) was a product of Boehringer
(Mannheim, Germany). PLD was prepared from
fresh cabbage leaves and partially purified by heat
precipitation and gel filtration on Sephadex G50
(Pharmacia, Sweden) according to Hirche et al.
(1997). All other reagents were of the purest grade
commercially available.
All compounds gave single spots in thin layer
chromatography with chloroform–methanol–am-
monia 25% (65:25:5, v/v/v). 1H NMR and 13C
NMR spectra were obtained in CDCl3 using a
Bruker spectrometer ARX 500. 31P NMR spectra
were obtained in CHCl3 with D2O-capillary using
a Bruker spectrometer AC 80. Chemical shifts
were referred to H3PO4 at 0 ppm.
2.2. Chemical synthesis (general procedure)
1,3-Diacylglycerols were prepared according to
Berger et al. (1992). Some of them were gifts from
the laboratory of Professor Schneider, Bergische
Universita¨t-GH-Wuppertal.
2-Bromoethyl
dichlorophosphate was synthesised according to
Hirth and Berchtold (1958). 2-Bromoethyl
dichlorophosphate (2.42 g, 10 mmol) was dis-
solved in 70 ml of absolute diethyl ether cooled in
an ice-bath. After the addition of dry pyridine
(7.1 g, 90 mmol) in 15 ml of diethyl ether, the
phosphorylation mixture was stirred at room tem-
perature. A solution of 1,3-diacylglycerol (3
mmol) in 25 ml of diethyl ether was added drop-
wise. Stirring was continued at room temperature
until the conversion of the 1,3-diacylglycerol was
completed, usually after 24–48 h. The
dichlorophosphate was hydrolysed by addition of
25 ml ice-water under vigorous stirring for 1 h.
Then the phases were separated and the solvent
was removed by evaporation. The residue was
dissolved in 60 ml of chloroform–methanol (2:1)
and extracted twice with 15 ml of water acidified
to pH 5.0 by HCl. The organic layer was dried
with Na2SO4 and the solvent was evaporated un-
der reduced pressure yielding 2%-bromoethyl 1,3-
diacylglycerol 2-phosphate. The product was
dissolved in 40 ml chloroform and 40 ml acetoni-
tril–isopropanol (2:1). To this, 2.36 g of 33%
trimethylamine (40 mmol) in ethanol were added
2.2.1. 1,3-Dioctanoylglycero-2-phosphocholine
(1,3-diC8PC)
1,3-Dioctanoylglycerol (1.05 g, 3 mmol) was
converted according to the general procedure. CC
with 500 ml chloroform–methanol (9:1), 1000 ml
chloroform–methanol (8:2), 3400 ml chloroform–
methanol–H2O (8:2:0.2) gave 0.37 g of 1,3-
diC8PC (yield 24%).
1
Rf 0.22, H NMR (500 MHz, CDCl3): l 0.86 (t,
6H, CH3, 3J(HH) 6.98 Hz); 1.26 (m, 16H,
(CH2)4); 1.57 (m, 4H, COCH2CH2
6
); 2.28 (t, 4H,
3
COCH2
6 , J(HH) 7.66 Hz); 3.36 (s, 9H (CH3)3);
3.79 (m, 2H, CH2-N); 4.23-4.26 (m, 4H, CH2 of
glycerol); 4.33 (m, 2H, POCH2); 4.49–4.54 ppm
(m, 1H, CHOP).
13C NMR (125.71 MHz, CDCl3); l 14.02
(CH3); 22.58 (CH3–CH2
28.95-29.12 ((CH2)2); 31.66 (CH3–CH2–C
34.15 (C
(POCH2, J(COP) 4.9 Hz); 63.00 (CH2 of glycerol,
J(CCOP) 5.03 Hz); 66.43 (CH2–N, J(CCOP) 5.16
Hz); 70.33 (CHOP, J(COP) 5.03 Hz); 173.46 ppm
(C.O).
6
); 24.89 (COCH2C
6
H2);
6
H2);
6
H2OCOR); 54.45 (N+(CH3)3); 59.39

60
R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
31P NMR (32.44 MHz, CHCl3; D2O-capillary,
(POCH2, J(COP) 4.9 Hz); 62.99 (CH2 of glycerol,
J(CCOP) 3.61 Hz); 66.49 (CH2–N, J(CCOP) 7.46
Hz); 70.44 (CHOP); 173.44 ppm (CꢀO).
31P NMR (32.44 MHz, CHCl3; D2O-capillary,
(H3PO4 (0 ppm)): −1.9 ppm.
(H3PO4 (0 ppm)): −1.42 ppm.
2.2.2. 1,3-Didecanoylglycero-2-phosphocholine
(1,3-diC10PC)
1,3-Didecanoylglycerol (1.17 g, 3 mmol) was
converted according to the general procedure. CC
with 400 ml chloroform–methanol (9:1), 700 ml
chloroform–methanol (8:2), 1200 ml chloroform–
methanol–H2O (8:2:0.2) and 700 ml chloroform–
2.2.4. 1,3-Dimyristoylglycero-2-phosphocholine
(1,3-diC14PC)
1,3-Dimyristoylglycerol (1.54 g, 3 mmol) was
converted according to the general procedure. CC
with 500 ml chloroform–methanol (9:1), 900 ml
chloroform–methanol (8:2), 1500 ml chloroform–
methanol–H2O (8:2:0.2) and 1000 ml chloro-
form–methanol–H2O (7:3:0.3) gave 0.55 g of
1,3-diC14PC (yield 24%).
methanol/H2O (7:3:0.3) gave 0.33
g
of
1,3-diC10PC (yield 20%).
1
Rf 0.21, H NMR (500 MHz, CDCl3): l 0.86 (t,
6H, CH3, 3J(HH) 6.86 Hz); 1.24 (m, 24H,
(CH2)6); 1.57 (m, 4H, COCH2CH2
6
); 2.28 (t, 4H,
1
3
Rf 0.26, H NMR (500 MHz, CDCl3): l 0.86 (t,
COCH2
6 , J(HH) 7.66 Hz); 3.36 (s, 9H (CH3)3);
6H, CH3, 3J(HH) 6.87 Hz); 1.24 (m, 40H,
3.79 (m, 2H, CH2–N); 4.21–4.27 (m, 4H, CH2 of
glycerol); 4.34 (m, 2H, POCH2); 4.51–4.55 ppm
(m, 1H, CHOP).
(CH2)10); 1.56 (m, 4H, COCH2CH2
6
); 2.28 (t, 4H,
3
COCH2
6 , J(HH) 7.67 Hz); 3.36 (s, 9H (CH3)3);
3.81 (m, 2H, CH2–N); 4.23 (m, 4H, CH2 of
glycerol); 4.32 (m, 2H, POCH2); 4.47–4.51 ppm
(m, 1H, CHOP).
13C NMR (125.71 MHz, CDCl3); l 14.06
(CH3); 22.65 (CH3–C
29.18–29.68 ((CH2)4); 31.86 (CH3–CH2–C
34.19 (C
6
H2); 24.91 (COCH2C
6
H2);
H2);
6
13C NMR (125.71 MHz, CDCl3); l 14.05
6
H2OCOR); 54.52 (N+(CH3)3); 59.32
(CH3); 22.65 (CH3–C
6
H2); 24.91 (COCH2C
6
H2);
(POCH2, J(COP) 4.9 Hz); 63.14 (CH2 of glycerol,
J(CCOP) 5.03 Hz); 66.55 (CH2–N, J(CCOP) 3.01
Hz); 70.26 (CHOP, J(COP) 5.03 Hz); 173.47 ppm
(C.O).
29.21–29.69 ((CH2)8); 31.90 (CH3–CH2–C
6
H2);
34.16 (CH2OCOR); 54.46 (N+(CH3)3); 59.39
(POCH2, J(COP) 4.9 Hz); 63.08 (CH2 of glycerol,
J(CCOP) 3.89 Hz); 66.49 (CH2–N, J(CCOP) 4.78
Hz); 70.32 (CHOP, J(COP) 5.91 Hz); 173.44 ppm
(CꢀO).
31P NMR (32.44 MHz, CHCl3; D2O-capillary,
(H3PO4 (0 ppm)): −1.35 ppm.
31P NMR (32.44 MHz, CHCl3; D2O-capillary,
(H3PO4 (0 ppm)): −1.35 ppm.
2.2.3. 1,3-Dilauroylglycero-2-phosphocholine
(1,3-diC12PC)
1,3-Dilauroylglycerol (1.37 g, 3 mmol) was con-
verted according to the general procedure. CC
with 500 ml chloroform–methanol (9:2), 1000 ml
chloroform–methanol (8:2), 3000 ml chloroform–
methanol–H2O (8:2:0.2) gave 0.56 g of 1,3-
diC12PC (yield 30%).
2.2.5. 1,3-Dipalmitoylglycero-2-phosphocholine
(1,3-diC16PC)
1,3-Dipalmitoylglycerol (1.70 g, 3 mmol) was
converted according to the general procedure. CC
with 300 ml chloroform–methanol (9:1), 600 ml
chloroform–methanol (8:2), 1100 ml chloroform–
methanol–H2O (8:2:0.2) and 1000 ml chloro-
form–methanol–H2O (7:3:0.3) gave 0.29 g of
1,3-diC16PC (yield 13%).
1
Rf 0.22, H NMR (500 MHz, CDCl3): l 0.86 (t,
3
6H, CH3, J(HH) 6.4 Hz); 1.23 (m, 32H, (CH2)8);
1.56 (m, 4H, COCH2CH6 2); 2.28 (t, 4H, COCH6 2,
3J(HH) 7.5 Hz); 3.34 (s, 9H (CH3)3); 3.76 (m, 2H,
CH2–N); 4.21–4.23 (m, 4H, CH2 of glycerol);
4.31 (m, 2H, POCH2); 4.55 ppm (m, 1H, CHOP).
13C NMR (20.149 MHz, CDCl3); l 14.06
1
Rf 0.30, H NMR (500 MHz, CDCl3): l 0.86 (t,
6H, CH3, 3J(HH) 6.87 Hz); 1.24 (m, 48H,
(CH2)12); 1.56 (m, 4H, COCH2CH2
6
); 2.28 (t, 4H,
3
COCH6 2, J(HH) 7.67 Hz); 3.36 (s, 9H (CH3)3);
(CH3); 22.67 (CH3–C
29.43–29.63 ((CH2)6); 31.91 (CH3–CH2–C
34.19 (C
6
H2); 24.92 (COCH2C
6
H2);
H2);
3.79 (m, 2H, CH2–N); 4.23-4.32 (m, 4H, CH2 of
glycerol); 4.49 (m, 2H, POCH2); 4.52 ppm (m, 1H,
CHOP).
6
6
H2OCOR); 54.48 (N+(CH3)3); 59.31

R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
61
13C NMR (125.71 MHz, CDCl3); l 14.05
shaken by a Vortex mixer and homogenised by
ultrasonic treatment with 300 Ws ml−1 using an
ultrasonic disintegrator 60 W Vibra cell (Avantec,
France) with a microtip probe of 3 mm. After
dilution of the stock solutions with aqua bidest.
1:1 (v/v), the size of the 1,3-PC aggregates in
aqueous solution was determined by dynamic
laser light scattering (scattering angle 90°) using
the Autosizer 2C (Malvern Instruments, operating
at He–Ne-laser at 633 nm) and fitted software.
The size of the aggregates is expressed as the
mean diameter of at least three measurements.
(CH3); 22.66 (CH3–C
29.22–29.71 ((CH2)10); 31.91 (CH3–CH2–C
34.17 (C
6
H2); 24.92 (COCH2C
6
H2);
H2);
6
6
H2OCOR); 54.48 (N+(CH3)3); 59.33
(POCH2, J(COP) 3.94 Hz); 63.02 (CH2 of glyc-
erol, J(CCOP) 4.05 Hz); 66.51 (CH2–N, J(CCOP)
7.04 Hz); 70.31 (CHOP, J(COP) 4.02 Hz); 173.44
ppm (CꢀO).
31P NMR (32.44 MHz, CHCl3; D2O-capillary,
(H3PO4 (0 ppm)): −1.39 ppm.
2.2.6. 1,3-Distearoylglycero-2-phosphocholine
(1,3-diC18PC)
1,3-Distearoylglycerol (1.90 g, 3 mmol) was
converted according to the general procedure. CC
with 700 ml chloroform–methanol (9:1), 1200 ml
chloroform–methanol (8:2), 2200 ml chloroform–
methanol–H2O (8:2:0.2) gave 0.62 g of 1,3-
diC18PC (yield 26%).
2.3.2. Determination of cmc
Cmcs for 1,3-diC8PC, 1,3-diC8PC/SDS (2:1)
and 1,3-diC10PC/SDS (2:1) were determined by
measuring the fluorescence of 1,6-diphenyl-1,3,5-
hexatriene (DPH) as a function of the phospho-
lipid concentration according to Chattopadhyay
and London (1984). DPH was excited at 358 nm
and emission was recorded at 430 nm.
1
Rf 0.29, H NMR (500 MHz, CDCl3): l 0.86 (t,
6H, CH3, 3J(HH) 6.96 Hz); 1.24 (m, 56H,
(CH2)14); 1.56 (m, 4H, COCH2CH2
6
); 2.28 (t, 4H,
3
COCH2
6
, J(HH) 7.56 Hz); 3.36 (s, 9H (CH3)3);
2.4. Enzymatic studies
3.81 (m, 2H, CH2–N); 4.23 (m, 4H, CH2 of
glycerol); 4.32 (m, 2H, POCH2); 4.48–4.50 ppm
(m, 1H, CHOP).
2.4.1. Substrate stock solutions
For the preparation of standard substrate PC/
SDS (2:1), 50 mg (65 mmol) PC was dissolved in
500 ml chloroform. The solvent was removed in a
vacuum evaporator. The PC film obtained was
suspended in 5 ml aqua bidest. containing 32.5
mmol SDS. The mixture was heated to 40°C,
shaken by a Vortex mixer and sonicated in an
ultrasonic bath. Additional homogenisation of the
vesicle preparation was performed by ultrasonic
treatment with 300 Ws ml−1 using an ultrasonic
disintegrator 60 W Vibra cell (Avantec, France)
with a microtip probe of 3 mm. The resulting lipid
dispersion was an opalescent solution containing
13 mM PC/6.5 mM SDS.
13C NMR (125.71 MHz, CDCl3); l 14.03
(CH3); 22.63 (CH3–C
29.19–29.68 ((CH2)12); 31.88 (CH3–CH2–C
34.14 (C
6
H2); 24.88 (COCH2C
6
H2);
H2);
6
6
H2OCOR); 54.44 (N+(CH3)3); 59.29
(POCH2, J(COP) 4.99 Hz); 63.00 (CH2 of glyc-
erol, J(CCOP) 3.99 Hz); 66.47 (CH2–N, J(C-
COP) 6.98 Hz); 70.25 (CHOP, J(COP) 5.98 Hz);
173.42 ppm (CꢀO).
31P NMR (32.44 MHz, CHCl3; D2O-capillary,
(Hd3PO4 (0 ppm)): −1.45 ppm.
2.3. Physicochemical experiments
2.3.1. Laser light scattering
For the experiments with 1,3-PCs, 13 mM stock
solutions of these compounds were prepared as
described in Section 2.3.1.
Stock solutions (13 mM) of 1,3-PCs were pre-
pared as follows: 1,3-PCs (26 mmol) were dis-
solved in chloroform and the solvent was removed
in a vacuum evaporator. The resulting lipid film
was suspended in 2 ml aqua bidest. to obtain the
pure lipid dispersions or with 2 ml 6.5 mM SDS
solution for 1,3-PC/SDS (2:1). The mixtures were
heated to their phase transition temperature,
2.4.2. Assay for PLD acti6ity
PLD activity was determined from the pH
change caused by formation of phosphatidic acid
as described by Allgyer and Wells (1979), and by
Lambrecht and Ulbrich-Hofmann (1992).

62
R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
Substrate stock solution (200 ml) was diluted
tendorn and Ulbrich-Hofmann, 1995), 1,3-PCs
with different chain length of the acyl residues (C8,
C10, C12, C14, C16, and C18) were produced. In the
first step the corresponding 1,3-diacylglycerols
were synthesised by the method of Berger et al.
(1992) (Fig. 1a). These compounds were converted
to the 1,3-diacylglycero-2-phosphoric acid-2%-bro-
moethylesters by reaction with 2-bromoethyl-
with 2 mM sodium acetate buffer, pH 5.7, con-
taining 40 mM CaCl2 to give the desired substrate
concentration in 3 ml total volume. In the stan-
dard assay it was 0.87 mM PC/0.43 mM SDS.
After thermal equilibration at 30°C, the reaction
was started by the addition of 10 ml PLD (1 mg
protein/ml, determined according to Bradford
(1976), in 2 mM sodium acetate buffer, pH 5,7,
containing 40 mM CaCl2) and followed by pH-stat
titration on an Autotitrator VIT 90 (Radiometer,
Denmark) with 10 mM NaOH. From the linear
increase of consumption of NaOH the reaction
rate was estimated using the integrated pro-
gramme Titra File II. In the standard assay PLD
activity amounted to 0.26 mkat mg−1. The S.D.
were 56% (n=3). The detection limit was 0.003
phosphoric acid dichloride according to
a
modified method by Hansen et al. (1982) (Fig. 1b).
The crude products were converted to the 1,3-dia-
cylglycero-2-phosphocholines by the replacement
of bromine by trimethylamine. After column chro-
matography the regioisomerically pure products
were obtained in yields up to 30%. They were
1
identified by H NMR, 13C NMR, and 31P NMR.
The advantage of this method consists in the easy
accessibility of the regioisomerically pure 1,3-dia-
cylglycerols. In comparison with conventional
chemical methods (Eibl et al., 1967; Nuhn et al.,
1986) two reaction steps could be saved.
mkat mg−1
.
2.4.3. Inhibition of PLD by 1,3-PC
(determination of IC50 and k50)
Standard substrate stock solution (PC/SDS=
2:1) (200 ml) and 10–200 ml stock solution of
1,3-PC were unified and homogenised by sonifica-
tion with 300 Ws ml−1. This mixture (200–400 ml)
was diluted with 2 mM sodium acetate buffer, pH
5,7, containing 40 mM CaCl2 to 3 ml. The reaction
was started with 10 ml PLD (1 mg protein/ml in
the same buffer) and enzyme activity was assayed
by pH-stat titration as described above. Kinetic
data were fitted to the exponential equation:
3.2. Aggregation beha6iour
The aggregation behaviour of the synthesised
1,3-PCs in aqueous solution was characterised by
dynamic laser light scattering. Since SDS, as an
effective activator of PLD, was used in the enzy-
matic studies described below, its influence on the
light scattering behaviour of the 1,3-PC aggregates
was also studied. In the absence of SDS, particle
sizes could be only determined for 1,3-diC10PC
and 1,3-diC12PC suggesting that these compounds
are organised as liposomes (Table 1). The short
chain 1,3-diC8PC was assumed to form micelles,
since the particle size is smaller than 20 nm, which
is the lowest detectable size for the method used.
In order to confirm the presence of micelles, the
fluorescence of DPH was measured proving the
formation of micelles with a cmc of 0.165 mM.
Also, no reasonable results of particle size were
obtained for the long chain 1,3-PCs (C14, C16, and
C18) by laser light scattering. The high polydisper-
sity index (PI) of 1.0, indicating a wide distribu-
tion of the particle sizes beyond the registration
limit, suggests the existence of planar bilayers as
the main structure of the long chain 1,3-PCs.
These results are in accordance with the findings
6i=60×exp (−ax)+b,
(1)
where 6i and 60 are the initial rates in the presence
and absence of inhibitor, x is the concentration (or
the mole fraction) of the inhibitor, and a and b are
constants. The inhibitor concentration, IC50, (or
the mole fraction, k50) where 6i=60/2, were calcu-
lated from Eq. (2):
IC50 (k50)=ln [260/(60−2b)]/a.
(2)
3. Results and discussion
3.1. Synthesis
In analogy to the synthesis of 1,3-diC12PC (Haf-

R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
63
of Kunitake et al. (1984) who characterised the
structure of sonicated samples of 1,3-diC12PC and
1,3-diC14PC as multi-walled vesicles and that of
1,3-diC16PC and 1,3-diC18PC as lamellar aggre-
gates by freeze fracture electron microscopy.
When the aggregation behaviour of the 1,3-PCs
is compared with that of the naturally configured
occurs in the form of bilayers (Marsh, 1990). For
the longer chain 1,2-PCs bilayer structures are
also known (Cevc, 1993).
In the presence of SDS remarkable changes in
aggregate structures were observed (Table 1). In
the case of 1,3-diC10PC, SDS is obviously able to
solubilise the lipid to give mixed micelles similar
to 1,3-diC8PC. The cmc-values of the mixed mi-
celle systems 1,3-diC8PC/SDS (2:1) and 1,3-
diC10PC/SDS (2:1) were determined to be 0.120
1,2-diacyl-sn-glycero-3-phosphocholines,
great
similarities can be realised. 1,2-diC8PC forms mi-
celles above 0.27 mM (Marsh, 1990). 1,2-diC10PC
Fig. 1. Scheme of chemoenzymatic synthesis of 1,3-diacylglycero-2-phosphocholines: (a) enzymatic synthesis of 1,3-diacylglycerols;
(b) chemical introduction of the phosphocholine group.
Table 1
Aggregate structures of 1,3-diacylglycero-2-phosphocholines
1,3-PC
Pure 1,3-PC
1,3-PC/SDS (2:1)
Particle size (nm) PI
Suggested aggregation
form
Particle size (nm) PI
Suggested aggregation form
1,3-dC8PC
1,3-dC10PC
1,3-dC12PC
1,3-dC14PC
1,3-dC16PC
1,3-dC18PC
B20
157
174
–
Micelles
B20
B20
77
79
101
185
–
–
Mixed micelles
Mixed micelles
0.21 Liposomes
0.15 Liposomes
1.0
1.0
1.0
0.19 Mixed liposomes
0.29 Mixed liposomes
0.29 Mixed liposomes
0.29 Mixed liposomes
n.d.a
n.d.
n.d.
Planar bilayer
Planar bilayer
Planar bilayer
a n.d.: Not detectable by this method.

64
R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
3.3. 1,3-PCs as potential substrates and inhibitors
of PLD
All 1,3-PCs (C8–C18) were tested as substrates
for PLD. The experiments were performed in
the absence and in the presence of 0.25, 0.5 and
1 equivalent of SDS, which is a potent activator
of PLD from cabbage (Dawson and Hemington,
1967). In contrast to the corresponding 1,2-PCs
in no case any activity could be detected. Since
there are no evident differences in the aggrega-
tion behaviour of 1,3-PCs and 1,2-PCs as
demonstrated above, it has to be assumed that
the molecular geometry of the 1,3-PCs does not
fit into the active site of PLD. In contrast, 1,3-
PCs were shown to be substrates of phospholi-
Fig. 2. Inhibition of phospholipase D as function of mole
fraction (k) of 1,3-PCs. 6i/60 are the initial reaction rates in
presence and absence of inhibitor. k=[1,3-PC]/([1,3-PC]+
[PC]).
pase
A
from Crotalus adamanteus which
hydrolyses their 1-ester bonds albeit with a
lower rate than the 2-ester bonds in the natu-
rally configured substrates (de Haas and van
Deenen, 1964).
mM and 7.04 mM, respectively. With 1,3-
diC12PC, liposomes as the main structure were
formed even in the presence of SDS but the
particle sizes decreased compared with those in
the absence of SDS (Table 1). For the long
chain 1,3-PCs (C14, C16, and C18) light scattering
experiments indicate a transformation of planar
bilayer structures by SDS to mixed liposomes.
This transformation is presumably due to the
cone shape of SDS (Israelachvili et al., 1980),
which might force a stronger curvature of the
lipid bilayer and lead to a decrease of particle
sizes.
The effector properties of the different 1,3-
PCs (C8–C18) were studied in the standard assay
for PLD from cabbage using PC/SDS (2:1). Due
to homogenisation by sonification, it can be as-
sumed that 1,3-PCs are incorporated into the
mixed liposomes of PC/SDS. In all cases 1,3-
PCs showed an inhibitory effect on PLD activ-
ity. Fig. 2 shows the relative reaction rates 6i/60
as a function of the mole fraction of the effec-
tors 1,3-diC8PC, 1,3-diC12PC and 1,3-diC18PC,
and Table 2 gives the IC50- and k50-values for
all the compounds tested. A remarkable depen-
dence of the inhibitory power on the chain
Table 2
IC50 and k50-values for the inhibition of PLD by 1,3-PCs in comparison to its inhibition by alkylphosphocholines of different chain
lengths
a
1,3-PC
IC50 (mM)
k50
Alkylphosphocholine
IC50 (mM)a
k50
1,3-dC8PC
1,3-dC10PC
1,3-dC12PC
1,3-dC14PC
1,3-dC16PC
1,3-dC18PC
50
43
108
185
181
192
0.053
0.045
0.098
0.173
0.172
0.183
C10
C12
C14
C16
C18
142
102
41
13
6.4
0.141
0.106
0.045
0.015
0.007
a Values according to Dittrich et al., 1996.

R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
65
Acknowledgements
length of the acyl chains can be observed. Short
chain 1,3-PCs (C8 and C10) are more effective as
inhibitors than the long-chain 1,3-PCs (C14, C16,
and C18). This tendency is opposite to the in-
hibitory effects of the alkyl phosphocholines
(Table 2), which have been studied recently
(Dittrich et al., 1996). Interestingly, the maxi-
mum inhibition effects are much larger with the
one-chain alkyl phosphocholines than with the
glycerol derivatives. The different organisation
of the substrate aggregates containing inhibitor
and detergent molecules and/or the different
affinities of the inhibitor molecules to the PLD
might be reasons for this behaviour. With both
types of inhibitors the inhibitory effect decreases
with increasing substrate concentration. There-
fore, a competitive type of inhibition is sug-
gested although a classical kinetic analysis is not
possible because of the sigmoidal curves of ini-
tial rates versus substrate concentration (Dittrich
et al., 1998). Further studies are necessary to
elucidate the inhibition mechanism.
The financial support by the Ministerium des
Landes Sachsen-Anhalt is gratefully acknowl-
edged. The authors also wish to thank Yvonne
Markert for the determination of the cmc-values
of the 1,3-PCs, and Gerd Scherer for performing
the NMR spectra.
References
Allgyer, T.T., Wells, M.A., 1979. Phospholipase D from savoy
cabbage: purification and preliminary kinetic characteriza-
tion. Biochemistry 18, 5348–5353.
D’Arrigo, P., Servi, S., 1997. Using phospholipases for phos-
pholipid modification. Trends BioTech. 15, 90–96.
Berger, M., Laumen, K., Schneider, M.P., 1992. Enzymatic
esterification of glycerol. I. Lipase-catalyzed synthesis of
regioisomerically pure 1,3-sn-diacylglycerols. J. Am. Oil
Chem. Soc. 69, 955–960.
Bonsen, P.P.M., Burbach-Westerhuis, G.J., de Haas, G.H.,
van Deenen, L.L.M., 1972. Chemical synthesis of some
lecithin analogues. Potential inhibitors of phospholipase A.
Chem. Phys. Lipids 8, 199–220.
Bradford, M.M., 1976. A rapid and sensitive method for the
quantification of microgram quantities of protein utilizing
the principle of protein-dye binding. Anal. Biochem. 72,
255–260.
4. Conclusions
Cevc, G., 1993. Phospholipid Handbook. Marcel Dekker,
New York.
As demonstrated for six 1,3-PCs the chemoen-
zymatic synthesis route opens a simple access to
the regioisomers of naturally configured glycero-
phospholipids. These compounds show aggrega-
tion properties similar to the natural
phospholipid structures as demonstrated for
aqueous milieu (this paper) or for organic media
(Frense et al., 1995). Obviously, the position of
the hydrophilic phosphatidylcholine head group
has no or only little effect on the formation of
suprastructures. In contrast, enzymes such as
PLD are able to differentiate the 2- or 3-posi-
tion of the head group. Therefore, 1,3-PCs are
interesting candidates as substitutes for natural
phospholipids in pharmaceutical drug delivery
systems or vector materials possessing the same
physicochemical properties but higher biological
stability. Furthermore, they might become im-
portant as effectors of phospholipid-transform-
ing enzymes in medicine.
Chattopadhyay, A., London, E., 1984. Fluorimetric determi-
nation of critical micelle concentration. Avoiding interfer-
ence from detergent charge. Anal. Biochem. 139, 408–412.
Dawson, R.M.C., Hemington, N., 1967. Some properties of
purified phospholipase D and especially the effect of am-
phiphatic substances. Biochem. J. 102, 76–86.
van Deenen, L.M., de Haas, G.H., 1963. The substrate specifi-
city of phospholipase A. Biochim. Biophys. Acta 70, 538–
553.
Dittrich, N., No¨ssner, G., Kutscher, B., Ulbrich-Hofmann, R.,
1996. Alkylphosphate esters as inhibitors of phospholipase
D. J. Enzym. Inhib. 11, 67–75.
Dittrich, N., Haftendorn, R, Ulbrich-Hofmann, R., 1998.
Hexadecyl-phosphocholine and 2-modified 1,3-diacylglyc-
erols as effectors of phospholipase D. Biochim. Biophys.
Acta 1391, 265–272.
Eibl, H, Arnold, D., Weltzien, H.U., Westphal, C., 1967. Zur
8
Synthese von a-und b-Lecithinen und ihren Atheranaloga.
Liebigs Ann. Chem. 709, 226–230.
Exton, J.H., 1997. Phospholipase D: enzymology, mechanism
of regulation, and function. Phys. Rev. 77, 303–320.
Frense, D., Haftendorn, R., Ulbrich-Hofmann, R., 1995. 2-
Modified 1,3-diacylglycerols as new surfactants for the
formation of reverse micelles. Chem. Phys. Lipids 78,
81–87.

66
R. Haftendorn et al. / Chemistry and Physics of Lipids 104 (2000) 57–66
de Haas, G.H., van Deenen, L.M., 1964. The site action of
phospholipase A on b-lecithins. Biochim. Biophys. Acta
84, 469–471.
de Haas, G.H., van Deenan, L.M., 1965. Structural identifica-
tions of isomeric lysolecithins. Biochim. Biophys. Acta 106,
315–325.
de Haas, G.H., Dijkman, R., van Oort, M.G., Verger, R.,
1990. Competitive inhibition of lipolytic enzymes. III.
Some acylamino analogues of phospholipids are potent
competitive inhibitors of porcine pancreatic phospholipase
A2. Biochim. Biophys. Acta 1043, 75–82.
Kunitake, T., Okahata, Y., Tawaki, S.I., 1984. Bilayer charac-
teristics of 1,3-dialkyl- and 1,3-diacyl-rac-glycero-2-phos-
phocholines. J. Colloid Interface Sci. 103, 190–201.
Lambrecht, R., Ulbrich-Hofmann, R., 1992. A facile purifica-
tion procedure of phospholipase D from cabbage and its
characterization. Biol. Chem. Hoppe-Seyler 373, 81–88.
Marsh, D., 1990. Handbook of Lipid Bilayers. CRC, Boca
Raton, FL, p. 276.
Munnik, T., Irvine, R.F., Musgrave, A., 1998. Phospholipid
signalling in plants. Biochim. Biophys. Acta 1389, 222–
272.
Haftendorn, R., Ulbrich-Hofmann, R., 1995. Synthesis of
2-modified 1,3-diacylglycerol. Tetrahedron 51, 1177–1186.
Hansen, W.J., Murari, R., Wedmid, Y., Baumann, W.J., 1982.
An improved procedure for the synthesis of choline phos-
pholipids via 2-bromoethyl dichlorophosphate. Lipids 17,
453–459.
Nuhn, P., Brezesinski, G., Dobner, B, Fo¨rster, G., Gutheil,
M., Do¨rfler, H., 1986. Synthesis, calorimetry and X-ray
diffraction of lecithins containing branched fatty acid
chains. Chem. Phys. Lipids 39, 221–236.
Paltauf, F., Hermetter, A., 1994. Strategies for the synthesis of
glycerophospholipids. Prog. Lipid. Res. 33, 239–328.
Rose, W.G., 1947. Synthesis of cephalin. J. Am. Chem. Soc.
69, 1384–1387.
Hartmann, L., 1958. Advances in the synthesis of glycerides of
fatty acids. Chem. Rev. 58, 845–867.
Heller, M., 1978. Phospholipase D. Adv. Lipid Res. 16, 267–
326.
Seelig, J., Dijkmann, R., de Haas, G.H., 1980. Thermody-
namic and conformational studies on sn-2-phosphatidyl-
cholines in monolayers and bilayers. Biochemistry 19,
2215–2219.
Smith, N.B., 1981. Effects of temperature on the NMR spectra
of sonicated dispersions of isomers of dipalmitoylphos-
phatidylcholine. Chem. Phys. Lipids 29, 277–282.
Stu¨mpel, J., Eibl, H., Niksch, A., 1983. X-ray analysis and
calorimetry on phosphatidylcholine model membranes.
The influence of length and position of acyl chains upon
structure and phase behaviour. Biochim. Biophys. Acta
727, 246–254.
Hirche, F., Koch, M.H.J., Ko¨nig, S., Wadewitz, T., Ulbrich-
Hofmann, R., 1997. The influence of organic solvents on
phospholipid transformation by phospholipase D in emul-
sion systems. Enzym. Microb. Techol. 20, 453–461.
Hirth, R., Berchtold, R., 1958. Zur Synthese der Phosphatide.
Eine neue Synthese der Lecithine. Mitteilung. Pharm. Acta
Helv., 349–356.
Israelachvili, J.N., Marcelja, S., Horn, R.G., 1980. Physical
principles of membrane organization. Q. Rev. Biophys. 13,
112–120.
.
.