γ-Ray irradiation of liposomes of polymerizable phospholipids containing octadeca-2,4-dienoyl groups and characterization of the irradiated liposomes

Article abstract of DOI:10.1039/a908732j

The synthesis of a variety of polymerizable phospholipids containing the octadeca-2,4-dienoyl moiety on 2-acyl chains and the characteristics of liposomes containing those phospholipids of the γ-irradiation are described. We synthesized three different polymerizable phosphocholines that have different 1-acyl chain lengths with the octadeca-2,4-dienoyl moiety on the 2- acyl chain: myristoyl (MODPC), palmitoyl (PODPC) and stearoyl (SODPC). The liposomes were prepared by extrusion through polycarbonate filters with a pore size of 0.2 μm, and were polymerized by γ-irradiation with various dose rates. The polymerization rate increased in the order SODPC>MODPC>PODPC. The mechanism of the polymerization of SODPC was the same as that of 1,2-bis- [(E,E)-octadeca-2,4-dienoyl]-sn-glycero-3-phosphocholine (DODPC), but differed from that of MODPC and PODPC. Freeze-thaw testing was used to evaluate the stability of the polymerizable liposomes. The MODPC liposome was more stable than other monofunctional liposomes. For similar irradiation, the polymerization behavior of the liposomes was significantly affected by the 1- acyl length.

Full text of DOI:10.1039/a908732j

View Article Online / Journal Homepage / Table of Contents for this issue
J O U R N A L O F
c-Ray irradiation of liposomes of polymerizable phospholipids
containing octadeca-2,4-dienoyl groups and characterization of the
irradiated liposomes
a
a
a
b
Kazuhiro Akama,* { Yoshihiro Yano, Satoru Tokuyama, { Fumio Hosoi § and
Hideki Omichi }
b
C H E M I S T R Y
a
Tsukuba Research Laboratory, NOF Corporation, 5-10 Tokodai, Tsukuba, Ibaraki,
300-2635, Japan
Takasaki Radiation Chemistry Research Establishment, Japan Atomic Energy Research
Institute, 1233, Watanuki-machi, Takasaki, Gunma, 370-1207, Japan
b
Received 29th October 1999, Accepted 18th January 2000
The synthesis of a variety of polymerizable phospholipids containing the octadeca-2,4-dienoyl moiety on 2-acyl
chains and the characteristics of liposomes containing those phospholipids of the c-irradiation are described.
We synthesized three different polymerizable phosphocholines that have different 1-acyl chain lengths with the
octadeca-2,4-dienoyl moiety on the 2-acyl chain: myristoyl (MODPC), palmitoyl (PODPC) and stearoyl
(
SODPC). The liposomes were prepared by extrusion through polycarbonate ®lters with a pore size of 0.2 mm,
and were polymerized by c-irradiation with various dose rates. The polymerization rate increased in the order
SODPCwMODPCwPODPC. The mechanism of the polymerization of SODPC was the same as that of 1,2-
bis-[(E,E)-octadeca-2,4-dienoyl]-sn-glycero-3-phosphocholine (DODPC), but differed from that of MODPC and
PODPC. Freeze±thaw testing was used to evaluate the stability of the polymerizable liposomes. The MODPC
liposome was more stable than other monofunctional liposomes. For similar irradiation, the polymerization
behavior of the liposomes was signi®cantly affected by the 1-acyl length.
encapsulated substances, and polymerizes the liposomes at
temperature as low as 4 ³C. The c-ray polymerization of
DODPC liposomes yielded high polymerization conversion
(70±80%) and the liposomes were stable against physical
Introduction
Liposomes are well known as biomimetic membranes that self-
organize when phosphocholine is dispersed in water. Lipo-
somes can encapsulate drugs, such as anticancer agents,
proteins, and other bioactive materials. Much attention has
been paid to applications of carriers for drug delivery systems.
However, conventional, unmodi®ed liposomes without addi-
tives (e.g. DPPC) have low stability against physical and
chemical stimuli, resulting in changes of diameter and leakage
1
8
stimuli, such as repeated freeze±thawing.
We developed previously a liposome type of arti®cial red
blood cell (called arti®cial red cells, or ARC), which
encapsulates hemoglobin with DODPC. For the c-ray poly-
merization of the membrane of ARC, we observed no change in
diameter and no leakage of hemoglobin, even after repeated
1
19
of inner substances caused by fusion and ®ssion in vitro and in
freeze±thaw cycles. The polymerized ARC was stable in the
blood stream without aggregating, and showed no toxicity or
2
vivo. Liposomes can be stabilized by the polymerization of
,3
4
phospholipids contained in the membrane. Diacetylenic,
5,6
18
side effects. In contrast to the attention given to the stability
of the liposomes, little attention has been given to the
5
,7±9
10
vinylic, methacryloylic
7,11
12,13
butadienic,
and thiolic
1
7
units have been used as polymerizable units in acyl chains.
Since Akimoto et al. synthesized 1,2-bis[(E,E)-octadeca-2,4-
dienoyl]-sn-glycero-3-phosphocholine (DODPC), which con-
tains two butadienic moieties in the acyl chains attached to
mechanism of polymerization using c-irradiation. Recently,
we reported the polymerization of DODPC liposomes using c-
irradiation and found that polymerization of DODPC lipo-
somes by c-rays gave selective polymerization between the 1-
9
20
phosphocholine, researchers have studied various methods for
and 2-acyl chains.
1
the polymerization of liposomes, by UV irradiation, organic
4
We report here (a) the synthesis of polymerizable phospho-
lipids that have different 1-acyl chain lengths with the same
polymerizable moieties on 2-acyl chains of phosphocholine, (b)
characteristics of polymerization by c-radiation compared with
those of DODPC liposomes, and (c) the stability of
polymerized liposomes against freeze±thawing.
1
5,16
17
redox initiators, and c-irradiation.
14,17
initiators,
Ohno et
al. reported selective polymerization of 1- or 2-acyl chains of
1
5,16
DODPC by using appropriate radical initiators.
In
particular, c-ray polymerization is one of the best methods
for encapsulating a drug in liposomes, because it requires no
residual initiators, produces less degradation and denaturing of
Results
{Present address: Medical Materials Business Dept., Oleo & Speciality
Chemicals Division, NOF Corporation, 4-20-3 Ebisu, Shibuya-ku,
Tokyo 150-6019, Japan. Tel z81-3-5424-6692; Fax z81-3-5424-6810;
E-mail: kazuhiro_akama@nof.co.jp
Synthesis of polymerizable phospholipids
Scheme 1 shows the method for synthesizing the chemical
structures of the polymerizable phospholipids (2±5). We
synthesized DODPC 2, which has two polymerizable groups
{
Present address: Oleochemicals Research Laboratory, NOF Corpora-
tion, 1-56 Ohama-cho, Amagaski, Hyogo 660-0095, Japan.
Deceased March 13, 1998.
Present address: Japan Atomic Energy Research Institute Washington
§
in the 1- and 2-acyl chains, from
GPC) and (E,E)-octadeca-2,4-dienoyl anhydride 1. We also
synthesized MODPC 3, PODPC 4, and SODPC 5 from
L-a-glycerophosphorylcholine
}
(
Of®ce, 1825 K Street, N.W., Suite 508, Washington, DC 20006-1202,
USA.
DOI: 10.1039/a908732j
J. Mater. Chem., 2000, 10, 1047±1059
This journal is # The Royal Society of Chemistry 2000
1047

View Article Online
Scheme 1 Synthetic scheme for preparation of the polymerizable phospholipids.
hydrolyzed diacyl phosphatidylcholine, corresponding to 1-
acyl lysophosphorylcholine and (E,E)-octadeca-2,4-dienoyl
anhydride. The phospholipids were puri®ed twice using
chromatography on silica gel and freeze-dried. They were
isolated as white colorless powders without polymers.
easily compressed under these conditions, i.e. a temperature of
4 ³C. The monolayers of DODPC and other monofunctional
phospholipids, MODPC and PODPC, collapsed before reach-
ing a less compressible region. For monofunctional polymeriz-
able phospholipids, the order of compressibility was
SODPCwPODPCwMODPC.
Surface pressure±area isotherms
Fig. 1 shows the surface pressure±area isotherms for various
polymerizable phospholipids. The monolayers of SODPC were
Preparation of liposomes
The liposomes were prepared by an extrusion method. The
c
phase-transition temperatures (T ) of the polymerizable lipo-
somes of MODPC, PODPC and SODPC were determined by
DSC (see Table 1). The T for MODPC, PODPC, and SODPC
c
liposomes, which contain saturated fatty acyl chains in the 1-
acyl moiety, were lower than those of the saturated diacylphos-
phatidylcholines (e.g., the T values for 1,2-dimyristoyl-sn-
c
glycero-3-phosphocholine (DMPC), 1,2-dipalmitoyl-sn-gly-
cero-3-phosphocholine (DPPC), and 1,2-distearoyl-sn-gly-
cero-3-phosphocholine (DSPC) are 23, 41, and 58 ³C,
2
1
respectively ), which corresponds to the length of the fatty
-acyl chain in the polymerizable phospholipids. The measured
increased with increasing length of the methylene units of 1-
1
T
c
acyl chains. The values of lmax and the molecular absorption
coef®cient (emax) corresponding to maximum absorption for
MODPC, PODPC, and SODPC liposomes are also given in
Table 1. Above T , lmax~259 nm, and below T ,
c
c
lmax~253 nm. The values of lmax obtained for PODPC
liposomes were identical with those reported by Takeoka et
2
2
al. For SODPC liposomes which contain two more methylene
units than PODPC, and for MODPC which contains two
methylene units fewer than PODPC, our measured values of
Fig. 1 Surface pressure±area isotherms of various polymerizable
phospholipids measured at 3.8 ³C.
lmax were the same as those for the PODPC liposomes. The
1048
J. Mater. Chem., 2000, 10, 1047±1059

View Article Online
c
Table 1 Phase-transition temperatures (T ), lmax and emax of polymerizable phospholipids containing octadeca-2,4-dienoyl groups as liposomes
4
21
21
lmax/nm
emax610 /L mol cm
Compound
c
T /³C
Below T
c
Above T
c
Below T
c
Above T
c
b
MODPC
PODPC
16.6
28.2
253.5
253.5
253.0
253.3
Ð
259.9
259.5
259.3
259.1
256.0
2.22
2.17
Ð
2.10
Ð
2.44
2.36
Ð
2.28
4.60
c
a
a
a
2
8.0
d
SODPC
DODPC
34.0
18.9
e
a
a
b
From ref. 21. The diameter was 197.0¡30.7 nm, SD~15.6%. The diameter was 164.0¡44.7 nm, SD~27.2%. The diameter was
c
d
e
77.4¡51.1 nm, SD~28.8%. The diameter was 203.2¡59.4 nm, SD~28.8%.
1
value of emax for each monofunctional polymerizable phos-
pholipid below T
c c
was about 10% less than that above T .
Polymerization of liposomes
Fig. 2 shows the UV spectra of irradiated and monomeric
DODPC and MODPC liposome solutions. The lmax of
absorbance of monomeric DODPC and MODPC liposomes
was at 256 nm and 259 nm, respectively. The lmax of both
DODPC and MODPC liposomes decreased with increasing
irradiation time. To ensure 100% yield of polymerization
conversion, we directly irradiated the quartz cell containing the
diluted monomeric liposome solution. The absorption com-
pletely disappeared on irradiation with a UV lamp at a
wavelength of 254 nm. This baseline agreed with 100% yield of
UV-irradiated liposomes (1 wt%) from other experiments,
which also did not indicate any residual monomers. The
width of the absorption curves of MODPC liposomes
decreased to those of the DODPC liposomes.
Fig. 3 Effect of dose rate on polymerization conversion of DODPC
21
21
liposomes. Dose rate: ($); 1.65 kGy h , (©); 3.3 kGy h , (#);
2
1
21
5.0 kGy h and (%); 10.0 kGy h
.
Fig. 3 shows changes in the polymerization conversion when
DODPC liposomes were c-irradiated at various dose rates. The
polymerization conversion increased with increasing irradia-
tion time and increasing dose rate. Furthermore, the poly-
merization conversions rapidly increased up to 50% and then
increased more slowly with additional irradiation. However,
under these irradiation conditions the polymerization did not
proceed beyond about 90% conversion (dose rate was
tion time, similar to the DODPC liposomes. For similar dose
levels, the order of increasing polymerization conversion of
liposomes was SODPCwMODPCwPODPC.
DODPC liposomes were hydrolyzed so that DODPC
liposome polymers, after irradiation by c-rays, were insoluble
in any organic solvent. To extract the soluble polymer from
DODPC polymers, methanolysis of DODPC polymers has
2
1
1
7
3
.3 kGy h , total irradiation time 16 h).
Fig. 4 shows changes in the polymerization conversion of
been used. However, the methanolysis of DODPC requires
relatively high temperatures of 100 ³C and long reaction times
of 2 days. Therefore, we developed a simpler method. Under
hydrolysis conditions of 70 ³C and 24 h, we observed a visual
change of the homogeneous liposome solution to a turbid
solution after hydrolysis. Then, we con®rmed the complete
hydrolysis of the DODPC polymers by using IR spectroscopy.
The measured molecular weights were reproducible for this
protocol. Under these conditions the molecular weight of the
polymers might not have an impact on the effectiveness of the
hydrolysis. Therefore, the obtained poly (octadeca-2,4-dienoic
acid)s were determined by using gel permeation chromato-
graphy (GPC). Fig. 5 shows a chromatogram of hydrolyzed
poly(octadeca-2,4-dienoic acid)s and octadeca-2,4-dienoic
acid. The degrees of polymerization were estimated to be
between 15 and 20, corresponding to a molecular weight for
octadeca-2,4-dienoic acid of 557, using polystyrene standards.
Fig. 6 shows molecular weights of octadeca-2,4-dienoyl poly-
mers that were obtained by hydrolysis of DODPC liposomes.
The molecular weight of octadeca-2,4-dienoyl polymers
obtained from DODPC liposomes decreased rapidly to an
asymptotic limit with irradiation up to 1 h.
MODPC, PODPC, and SODPC liposomes. For all liposomes,
polymerization conversion increased with increasing irradia-
Fig. 7 shows the molecular weights of octadeca-2,4-dienoyl
polymers obtained by hydrolysis of SODPC liposomes
irradiated under various conditions. The molecular weights
of the octadeca-2,4-dienoyl polymers obtained from SODPC
liposomes were less than those of the DODPC liposomes. The
decrease in molecular weight with irradiation was the same as
that observed for the DODPC liposomes. The degrees of
polymerization of SODPC liposomes were estimated to be
Fig. 2 Change of UV spectra for liposomes during c-irradiation at
4 ³C. (a) DODPC liposome, dose rate 10 kGy h , (A) before c-
irradiation, (B) 15 min, (C) 30 min, (D) 45 min, (E) 60 min, (F)
21
1
5
2
20 min, and (G) after UV irradiation. (b) MODPC liposome, dose rate
21
kGy h , (A) before c-irradiation, (B) 0.5 h, (C) 1 h, (D) 1.5 h, (E)
h, (F) 3 h, (G) 5 h, and (H) after UV irradiation.
J. Mater. Chem., 2000, 10, 1047±1059
1049

View Article Online
Fig. 5 Chromatogram of hydrolysis products by gel permeation
chromatography. (a) hydrolysis products of DODPC liposome,
2
1
which was irradiated at dose rate 1.65 kGy h for 5 h, (b) hydrolysis
product of non-irradiated DODPC liposome, and (c) octadeca-2,4-
dienoic acid. Column: G2500HXLzG2000HXL, range of molecular
4
2
weights 3.5610 ±3610 .
liposome. Fig. 9 shows the change of diameter for MODPC
and PODPC liposomes after freezing and thawing, for various
irradiation levels. The diameters of non-irradiated MODPC
and PODPC liposomes changed in all cases after a freeze±thaw
cycle. After freeze±thawing of non-irradiated and short time
irradiation (e.g., 0.5 h) of the MODPC and PODPC liposomes,
the sizes increased to around 1000±2000 nm with a wide size
distribution (Fig. 9). The liposome solution became turbid due
to either aggregation or fusion of the particles. The change in
diameter as a function of the polymerization conversion
(Fig. 10) clearly shows that the diameter of MODPC liposomes
did not change signi®cantly even for low polymerization
conversion levels around 40%. For PODPC liposomes, it was
dif®cult to stabilize the membrane of vesicles even for
polymerization conversion of 60%.
Fig. 4 Effect of dose rate on polymerization conversion of various
liposomes. (a) MODPC liposome, (b) PODPC liposome, (c) SODPC
2
1
21
liposome. Dose rate: (&); 1.5 kGy h , ($); 1.65 kGy h , (#);
2
1
21
5
.0 kGy h , and (%); 10.0 kGy h
.
between 10 and 14, as described above. On the other hand, the
hydrolyzed PODPC and MODPC liposomes were insoluble
not only in THF but also in all other polar solvents, such as
DMF and DMSO. However, it was observed that the IR
spectra of hydrolyzed MODPC and PODPC polymers were
identical to those for the SODPC polymers (data not shown).
The IR spectra of polymers obtained from the SODPC
liposomes were also similar to those of the DODPC liposomes.
Stability of polymerized liposomes
We evaluated the physical stability of the polymerizable
liposomes by using freeze±thaw tests. Freeze±thawing is
commonly used for trapping drugs in liposomes. However,
liposomes composed of a phosphatidylcholine can not retain
their structure following freeze±thawing cycles, because either
the liposome structure or the diameter of the liposomes is
2
3
affected by the freezing and thawing. Thus, we used freeze±
thawing tests as one evaluation for the stability of liposome
membranes.
Fig. 8 shows the changes of mean diameter of polymerized
liposomes after one freeze±thaw cycle. The error bar for each
point indicates the standard deviation of the mean diameter.
Within statistical uncertainty, the diameter of DODPC
liposomes did not change after freezing and thawing. On the
other hand, a small change was observed in the diameter of
monofunctional polymerized liposomes except for MODPC
Fig. 6 Relationship between number-average molecular weight and
irradiation time, for various dose rates. The number-average molecular
weights were obtained from hydrolysis of irradiated DODPC
2
1
21
liposomes. Dose rate: ($); 1.65 kGy h , (©); 3.3 kGy h , (#);
2
1
21
5.0 kGy h , and (%); 10.0 kGy h
.
1050
J. Mater. Chem., 2000, 10, 1047±1059

View Article Online
Fig. 7 Relationship between number-average molecular weight and
irradiation time, for various dose rates. The number-average molecular
weights were obtained from hydrolysis of irradiated SODPC liposomes.
2
1
,
21
Dose rate: ($); 1.65 kGy h
(#); 5.0 kGy h
,
and (%);
2
1
1
0.0 kGy h .
Discussion
Fig. 8 Change of diameter before and after freeze±thawing for various
irradiated liposomes with polymerization conversion near 50%. Error
bars represent standard deviation of mean diameter for liposomes, (%)
before freeze±thawing and (&) after freeze±thawing. For SODPC, the
liposome was extruded through a ®lter of 0.1 mm pore size.
Surface pressure±area isotherms
All of our synthesized polymerizable phospholipids contain the
octadeca-2,4-dienoyl groups in the 2-position of glycerophos-
phatidylcholine units, only fatty acyl chains of the 1-acyl chains
are different (i.e., saturated or unsaturated, different lengths of
methylene units). The monolayer of DODPC had a lower
compressibility at 4 ³C. On the other hand, the monolayer of
SODPC was easy to compress, although the length of the 1-acyl
chain in SODPC is the same as that in the DODPC molecule.
The difference of compressibility between DODPC and
SODPC molecules might be due to intermolecular interaction
of liposomes by creating hydroxyl radicals, which are generated
when the c-rays decompose the water surrounding the
liposomes. The concentration of hydroxyl radicals during c-
irradiation remained constant. The conversion curve showed
increased convexities with irradiation and molecular weight
decreased gradually with irradiation (Figs. 3 and 6). From the
result of the polymerization conversions, the polymerization
kinetics approximate a stationary state of polymerization in
2
5
(
van der Waals interaction) of the DODPC molecules, which is
which initiation is equal to termination. This polymerization
behavior was classi®ed as a slow initiation and a stationary
weaker than the intermolecular interaction between the
SODPC molecules.
2
system. The mechanism for c-ray polymerization of the
6
On the other hand, in the monolayer of monofunctional
phospholipids, the ease of compression increased with
increasing length of the fatty 1-acyl chains. By increasing the
length of the fatty 1-acyl chains, the chain±chain interactions
between neighboring molecules would increase. Regarding the
effect of the fatty acyl chain, by measuring surface pressure±
area isotherms and by using DSC, Ogino et al. reported
liposomes is presented in 4 steps as described in the following
0
2
elementary reactions.
Initiation
kdi
.
.
H O A H z OH
2
R ~k rH OI
2
d
di
(1)
(2)
(3)
.
.
kdt
.
.
H z OH A H O
2
R ~k ‰H Š‰ OHŠ
dt
dt
interactions between water molecules and L-a-phosphatidyl-
cholines, which are saturated diacylphosphatidylcholines of
different acyl chain lengths (e.g. DMPC, DPPC, and DSPC
.
ki
.
.
OHzM A R
R ~k ‰ OHŠ‰MŠ
i
i
2
4
Propagation
etc.). They concluded that the length of the acyl chains was
affected by the ®lm phase of the monolayer (i.e., either
condensed or expanded ®lm) due to the intermolecular
interaction of the acyl chains of each phosphatidylcholine.
By increasing the length of the fatty acyl chain, the ®lm changes
from an expanded to a condensed ®lm. Furthermore, they
concluded that both monolayer and bilayer membranes of
liposomes might experience the same horizontal intermolecular
interaction by the chain±chain interaction. Although we used
different phospholipids from those used in the study by Ogino
.
kp
.
n
R zM A R
(4)
kp
.
.
.
Rn zM A Rnz1
Rp~kp‰R Š‰MŠ
n
Termination
.
n
kt
.
R zZ A Pn
Rt~kt‰R Š‰ZŠ
(5)
(6)
n
Chain transfer
2
4
et al., these ®ndings can also be applied to our polymerizable
phospholipids, which contain the polymerizable fatty acyl
chain as the 2-acyl chain and different chain lengths in the 1-
acyl chain (myristoyl, palmitoyl and stearoyl). Therefore, the
chain±chain interaction of polymerizable phospholipids in
liposomes is probably similar to the interactions of monolayers.
.
n
ktr
.
.
R zM A P zR
n
R ~k ‰R Š‰MŠ
tr
tr
n
from eqns. (2) and (3)
.
di
2
k rH OI
‰ OHŠ~
.
~KI
(7)
kdt‰H Š
where
Polymerization of liposomes
kdirH2O
A kinetic study was carried out for the polymerization of
DODPC liposomes by c-rays. The c-rays cause polymerization
K~
.
(8)
kdt‰H Š
J. Mater. Chem., 2000, 10, 1047±1059
1051

View Article Online
Fig. 9 Change of diameter before and after freeze±thawing for MODPC and PODPC liposomes, for various dose rates and irradiation times. Error
bars represent standard deviation of mean diameter for liposomes. (a) MODPC liposome and (b) PODPC liposome; (%) before freeze±thawing and
(
&) after freeze±thawing.
?
?
n
are monomeric active and
where M is the monomer, R and R
growing radicals, respectively, Z is a substance by which R
deactivated. R , Rdt, R , R , R and Rtr denote the rate of
?
n
is
d
i
p
t
radical formation from the water, its reverse reaction,
initiation, propagation, termination, and chain transfer,
respectively. The corresponding rate constants are denoted
by k , k , k , k , k and k , respectively. rH O and I are the
di
dt
i
p
t
tr
2
density of water and dose rate, respectively.
For the initiation rate equal to the termination rate (i.e.,
~R ), from eqns. (3), (5) and (8),
R
i
t
.
n
kiKI
kt‰ZŠ
‰
R Š~
‰MŠ
(9)
Eqn. (4) can be written by using eqn. (9) as
d‰MŠ
dt
kiKI
kt‰ZŠ
2
Rp~{
~kp
‰MŠ
(10)
From eqn. (10), the relationship of monomer concentration
and irradiation time can be expressed by eqn. (11):
1
kiKI
kt‰ZŠ
1
~
{kp
tz
(11)
‰
MŠt
‰MŠ0
where [M] and [M] are the monomer concentration at time 0
0
t
and t, respectively.
Fig. 10 Diameter changes vs. conversion for MODPC and PODPC
2
1
21
The kinetic treatment was applied to DODPC liposome. For
DODPC liposomes, the conversion curve (see Fig. 3) shows a
rapid rise to 50%, followed by a slower increase. Therefore, the
liposomes; MODPC liposome: ($) 1.5 kGy h , (+) 5 kGy h , and
2
1
21
21
(
&) 10 kGy h . PODPC liposome: (#) 1.51 kGy h , (©) 5 kGy h
,
21
and (%) 10 kGy h
.
1052
J. Mater. Chem., 2000, 10, 1047±1059

View Article Online
reaction can be divided into two steps: reactions governing the
conversion up to 50% (region I) and reactions in the region
above a conversion of 50% (region II). Eqn. (11) applies to each
region. For region I, the plot of the irradiation time vs. the
inverse of the monomer concentration gave a linear relation-
ship at various dose rates, with an intercept of 1 (Fig. 11).
Region II also gave a linear relationship, with an x-intercept
1
7
corresponding to the start time of region II. The slope of this
line corresponds to the overall polymerization rate constant,
although polymerization in reaction regions I and II proceeds
independently. Fig. 13 shows the plot of the obtained overall
rate constants vs. dose rate for each region. We found that
eqns. (3)±(6) and (8) satis®ed this mechanism.
Using the degree of polymerization and monomer concen-
tration, we con®rmed that this mechanism is reasonable. If the
termination is ®rst-order, the degree of polymerization can be
expressed as eqn. (12):
Fig. 12 Plot of inverse of degree of polymerization vs.
„
„
t
t
.
Rpdt
1=Mp
ꢀRp=‰MŠ†dt for each region of DODPC liposome. ($);
0
0
Pn~ „
„
(12)
21
21
21
21
t
t
1.65 kGy h
10.0 kGy h
,
.
(©); 3.3 kGy h
,
(#); 5.0 kGy h
,
and (%);
Ridtz
Rtrdt
0
0
step in region II is smaller than that of region I. Consequently,
in this system the rate of polymerization is affected by the
initiation step.
From eqns. (4) to (6), eqn. (12) can be expressed as a
function of the relationship between the degree of polymeriza-
tion and the monomer concentration as eqn. (13):
ꢀ
The two acyl chains of phosphatidylcholine provide two
different environments in bilayer membranes, such as lipo-
somes. The diene moiety in the 2-acyl chain of DODPC might
be associated with the aqueous phase. The diene moiety in the
t
1
kt‰ZŠ
1
Rp
ktr
kp
.
~
dtz
(13)
Pn
kp Mp 0 ‰MŠ
„
t
where Pn is the degree of polymerization, Mp~ Rpdt, i.e.,
concentration of consumed monomer at time t.
0
1
-acyl chain of DODPC is located in a more hydrophobic
environment than that of the 2-acyl chain of DODPC. This
hypothesis is demonstrated by using different radical initiators
Fig. 12 shows plots of eqn. (13) as the inverse of the degree of
polymerization vs. monomer concentration in each region. A
linear relationship was obtained in each region, with region-
dependent slope. These results can be reasonably explained by
using the kinetic treatment of degree of polymerization by a
two-step polymerization with different reactivities. The differ-
ence in reactivity might be explained as follows: The slope and
the y-intercept of the lines in Fig. 12 correspond to k
in eqn. (13), respectively. The slopes of the lines within
each region were nearly the same, and the y-intercepts of all
1
5,16,27
for DODPC or PODPC liposomes.
During c-ray
polymerization, the 2-acyl chain on DODPC liposomes is
easily attacked by the hydroxyl radical, which acts as an
2
8
initiator. The 2-acyl chain apparently polymerizes ®rst,
followed by the 1-acyl chain. The polymerization of each
region proceeds independently of the other. The reason is that
it is dif®cult for hydroxyl radicals to penetrate the diene groups
t p
[Z]/k and
ktr/k
p
lines were also nearly the same (Fig. 12). These results show
that the rate constant (k ) in each region is nearly constant in
these systems. On the other hand, the overall rate constant
p
(
k
p
k
i
KI/k
the rate constant for region
polymerization (R ) of region II was about 10% smaller than
t
[Z]) for region II decreased by a factor about 10 from
I
(Fig. 13). The rate of
p
that of region I. Inserting k
decrease of R in region II resulted in a decrease of k
[Z]/k is constant within each region. The rate of the initiation
t
[Z]/k
p
into eqn. (10), we see that the
p
i
, because
k
t
p
p
Fig. 13 Overall rate constant (k ) vs. dose rates for various polymeriz-
Fig. 11 Plot of 1/[M]
t
vs. irradiation time for DODPC liposomes. Dose
able liposomes. (%) DODPC liposome on region I, (&) DODPC
liposome on region II, ($); SODPC liposome, (©); MODPC liposome,
and (#); PODPC liposome.
2
1
21
21
rate: ($); 1.65 kGy h , (©); 3.3 kGy h , (#); 5.0 kGy h , and (%);
2
1
1
0.0 kGy h .
J. Mater. Chem., 2000, 10, 1047±1059
1053

View Article Online
of the 1-acyl chain, which is located in a hydrophobic
environment. Therefore, polymerization of regions I and II
corresponds to polymerization of the 2- and 1-acyl chains,
respectively. Consequently, the diffusion of hydroxyl radicals is
suf®ciently slow that the initiation reaction of the 1-acyl chain
is slower than that of the 2-acyl chain. Kinetic studies suggest
that the polymerizable 1- and 2-acyl chains of DODPC
liposome have different reactivities when irradiated with c-
rays. The difference can be explained by the diffusion of the
hydroxyl radical being the rate-determining step during
polymerization induced by c-irradiation.
Although we used a homogeneous polymerization model to
describe 2-dimensional polymerization with restricted mobility,
the mechanism of polymerization is the same for both 1- and 2-
i
acyl chains. The only difference is the value of k , which is a
parameter of the initiation reaction. The polymerization of the
octadeca-2,4-dienoyl group in the bilayer membrane proceeds
by 1,4-addition, which requires conformational rearrangement
before polymerization. Ringsdorf and Schupp reported poly-
2
9
merization of octadeca-2,4-dienoic acid in the monolayer.
The polymerization of octadeca-2,4-dienoic acid proceeds only
in monolayers that are in a regular formation. This ®nding also
applies to bilayer membranes in the liposome that consist of
polymerizable phospholipids with an octadeca-2,4-dienoyl
moiety. At low temperatures, such as 4 ³C, the mobility of
the DODPC molecules on the membrane was already
restricted, but might be stacked and arranged for easy
polymerization. The ®rst stage of polymerization occurs at
the 2-acyl chain, because hydroxyl radicals readily attack the 2-
acyl chains that face the aqueous phase. After the initiation
reaction, the rearranged molecules on the membrane easily
react with 1,4-addition at the 2-acyl chain. In this case,
polymerization only occurs between the 2-acyl chain and the 2-
acyl chains of the neighboring molecules. It is dif®cult to cause
either intramolecular or intermolecular polymerization
between the 1- and 2-acyl chains, because the 1- and 2-acyl
chains are far apart and do not rearrange to permit 1,4-
addition. After the ®rst stage of polymerization of the
monolayer is completed, the 1-acyl chain of DODPC still
retains its regular formation. For the second stage, 1-acyl chain
polymerization, the 1-acyl chain easily rearranges to a
polymerizable conformation. If the initiator attacks the diene
group of the 1-acyl chain polymerized with the 2-acyl chain,
polymerization occurs immediately. The activated radicals
react with the neighboring molecules even without ®nding a
polymerizable monomer. The 1-acyl chain successively poly-
merizes, although the rate of polymerization is slower than that
of the 2-acyl chain, due to the initiation step being rate-
determining. Therefore, this polymerization behavior can be
explained by using a homogeneous polymerization model.
To explain the reactivity of DODPC liposomes, we
synthesized non-polymerizable saturated fatty acyl chains as
the 1-acyl chains and octadeca-2,4-dienoyl groups as the 2-acyl
chains, by polymerization of the liposomes using c-irradiation.
1
t
=2
Fig. 14 Plot of 2=‰MŠ vs. irradiation time. (a) MODPC liposome and
21
2
1
21
,
(
b) PODPC liposome. Dose rate: (&); 1.5 kGy h , (#); 5.0 kGy h
and (%); 10.0 kGy h . (c) Plot of 1/[M]t vs. irradiation time for
21
21
,
SODPC liposome. Dose rate: ($); 1.65 kGy h , (#); 5.0 kGy h
2
1
and (%); 10.0 kGy h
.
inverse of the degree of polymerization vs. the monomer
concentration in SODPC liposomes was linear with a slope
t p
(k [Z]/k ) for SODPC, a two-fold increase compared with the
3
0
slope for DODPC. When k [Z]/k was included in the overall
t
p
rate constant (k
p
k
i
KI/k
t i
[Z]), the k KI of SODPC decreased by
1
-Stearoyl-2-[(E,E)-octadeca-2,4-dienoyl]-sn-glycero-3-phos-
about 50% compared with that of DODPC. Furthermore, the
decrease in k was because the decrease in the overall rate
constant was larger than the decrease in k . On the other hand,
our results showed that because k is equal to k , k did not
increase, but decreased in this polymerization system. There-
fore, the decrease in the overall rate constant of SODPC is due
phocholine (SODPC), which has octadeca-2,4-dienoyl groups
at its 1-position and the same number of carbon atoms in each
acyl chain as DODPC, was polymerized by c-rays under the
same conditions as for DODPC liposomes. The polymerization
behavior of SODPC liposomes was similar to that of DODPC
liposomes. Fig. 14c shows a plot of the inverse monomer
concentration vs. irradiation time, and as indicated by
eqn. (11), yields a straight line with a y-intercept of 1. This
suggests that SODPC liposomes are polymerized by the same
p
i
t
i
t
i p
to the decrease in both k and k . The difference in the
polymerization rate between SODPC and DODPC might be
explained as follows. The decrease in the initiation of SODPC
indicated that the diene moiety of the octadeca-2,4-dienoyl
group of SODPC is in a different environment from that of
DODPC. The diene moiety of the 2-acyl chain of SODPC
might be further from the water phase than that of DODPC, or
SODPC molecules might have a somewhat denser packing than
DODPC molecules. In either case it would be more dif®cult for
3
0
mechanism as DODPC liposomes. However, the polymeriza-
tion rate of SODPC liposomes decreased to 20 to 25% of the
polymerization rate of DODPC liposomes (Fig. 13). This
decrease in the overall rate constant (k k KI/k [Z]) of SODPC
p
i
t
p i t
suggests a decrease in both k and k , and an increase in k [Z]
compared with DODPC liposomes. A plot of eqn. (13) as the
1054
J. Mater. Chem., 2000, 10, 1047±1059

View Article Online
hydroxyl radicals to penetrate to the reactive diene moiety.
Thus, the rate of initiation decreases due to the decrease in the
diffusion of hydroxyl radicals. For the decrease in propagation,
the reaction requires rearrangement of the conformation of the
octadeca-2,4-dienoyl moiety because this polymerization
proceeds by 1,4-addition of the octadeca-2,4-dienoyl moiety.
For SODPC liposomes, it is more dif®cult to change the
conformation of the monomer for polymerization compared
with DODPC. Consequently, the rate of polymerization
decreased with decreasing rates of both initiation and
propagation.
On the other hand, the polymerization behavior of MODPC
and PODPC liposomes differed from that of SODPC
liposomes. The rates of polymerization were slower than
those of SODPC liposomes. The rate of polymeriza-
tion decreased in the following order (Fig. 13);
SODPCwMODPCwPODPC. The molecular weight could
not be measured, because of insolubility in THF. Though the
rates of polymerization of MODPC and PODPC were slower
than those for SODPC liposomes, the molecular weights of
poly(octadeca-2,4-dienoyl acid)s, which are hydrolyzed
MODPC and PODPC liposomes, were higher than that of
SODPC. It is estimated that this behavior differs from that of
SODPC liposomes. If termination occurs, causing recombina-
tion of the end of the chain transfer, then the termination
reaction is represented by eqn. (14).
Stabilities of polymerized liposomes
Fig. 8 indicates that the polymerized DODPC liposomes are
more stable than any other monofunctional polymerizable
phospholipids. We also observed that the molecular weights of
poly(octadeca-2,4-dienoyl acid)s hydrolyzed to DODPC and
SODPC liposomes are similar. However, DODPC, which
contains two polymerizable groups in a molecule, is cross-
linked due to the polymerization of the 1-acyl chains following
the polymerization of the 2-acyl chains. Consequently, the
overall molecular weight was higher than the measured
molecular weight. Furthermore, it is estimated that the
molecular weight of poly(octadeca-2,4-dienoyl acid)s that are
hydrolyzed to MODPC liposomes is as high as that of PODPC
liposomes, because the stability of polymerized MODPC
liposomes is higher than that of PODPC liposomes. As a
control, the DPPC liposome was used to compare the stability
of the liposomes for the freeze±thaw tests. The diameter of the
DPPC liposome changed from 201¡53 nm to 234¡70 nm.
The changes of the mean diameter of DPPC were relatively
small compared with the non-polymerized, shortly irradiated
(
The monolayer of DPPC was a condensed ®lm at this
0.5 h) MODPC and PODPC liposomes (see Figs. 8 and 9).
24
temperature.
The intermolecular interaction of DPPC
molecules in the liposome might be stronger than those of
non-irradiated MODPC and PODPC molecules in the lipo-
somes. However, the polymerized MODPC and PODPC
liposomes were more stable than the DPPC liposome during
the freeze±thaw cycle.
.
n
.
n
kt
.
n
2
R zR A Pn
Rt~kt‰R Š
Eqn. (14) can be written by using eqns. (3), (5), and (8)
(14)
ꢁ
ꢂ
1=2
.
kiKI
Conclusions
1=2
‰
R Š~
‰MŠ
(15)
n
kt
The polymerization of liposomes is initiated by hydroxyl
radicals generated by c-irradiation that decomposes the water
surrounding the liposome membrane surfaces. For the
DODPC liposome, the two diene groups in the 1- and 2-acyl
From eqns. (4) and (15), the relationship of monomer
concentration and irradiation time can be expressed by
eqn. (16).
chains see different environments. That is, as Tsuchida
15,16,27
ꢁ
ꢂ
1=2
2
kiKI
kt
2
1=2
‰MŠ0
et al.
reported, the diene group of the 2-acyl chains
~kp
tz
(16)
1
=2
might face the water phase and the diene group of the 1-acyl
chains might exist in a hydrophobic environment. Kinetic
studies suggest that the c-ray polymerization of DODPC
liposomes can be represented as a two-step polymerization: the
2-acyl chains are polymerized ®rst, followed by the 1-acyl
chains. The reason is that hydroxyl radicals readily attack the
diene groups of the 2-acyl chain, which is located in a
hydrophilic environment. Furthermore, it is dif®cult for
hydroxyl radicals to penetrate the diene groups of the 1-acyl
chain, which is located in a hydrophobic environment. The
difference can be explained by the diffusion of the hydroxyl
radical being the rate-determining step during polymerization
induced by c-irradiation. Therefore, DODPC liposomes
containing two polymerizable groups in a molecule selectively
polymerize each other even though irradiated with c-rays.
Furthermore, the polymerization behavior of SODPC lipo-
somes with monofunctional polymerizable lipids is the same as
that of DODPC liposomes. However, the fact that the
initiation and propagation rates were different might be due
to the decrease in hydroxyl radicals and the susceptibility of the
conformational rearrangement. On the other hand, the
polymerization rates of MODPC and PODPC liposomes
differ from that of SODPC liposomes, which have the same
polymerizable groups on the 2-acyl chains in phosphatidylcho-
line. The polymerized MODPC liposomes are more resistant to
freeze±thaw damage than are PODPC liposomes. The termina-
tion of the MODPC and PODPC liposomes is preferable to
recombination. Therefore, the molecular weight of the poly-
(octadeca-2,4-dienoyl acid)s polymer on hydrolysis to MODPC
and PODPC liposomes might be larger than that of the
SODPC liposome. In general, the polymerized liposomes
consist of monofunctional polymerizable lipids that are less
‰
MŠt
1/2
Fig. 14a and 14b show plots of 2/[M] vs. irradiation time,
and according to eqn. (16), yield straight lines. The x-intercept
1
/2
of 2/[M] indicates the time of the induction period as shown
from the polymerization conversion curves (Fig. 4a and 4b).
Fig. 13 shows the slope of eqn. (16), which indicates the rate of
polymerization, vs. the dose rate, and yields a straight line
whose slope is 0.7. This suggested that for MODPC and
PODPC liposomes, the termination step preferentially occurs
by recombination. The intermolecular interaction of MODPC
and PODPC decreases relative to that of SODPC, because the
surface pressure±area isotherms of MODPC and PODPC are
different from that of SODPC. The intermolecular interactions
of MODPC and PODPC are also probably different from that
of SODPC. This recombination±termination behavior of
MODPC and PODPC liposomes is due to the decreased
intermolecular interaction of neighboring molecules, which in
turn increases the mobility of the monomer and growing
radical. This is also suggested from results on differences in the
ability of the molecules to change conformation in response to
polymerization; the mobility of MODPC and PODPC
molecules is greater than that of SODPC molecules. Therefore,
the termination reaction prefers recombination to unimolecu-
lar termination, and the molecular weight increases over that of
SODPC. Moreover, the intermolecular interaction of MODPC
decrease relative to that of PODPC, so that the chain length of
the 1-position of MODPC is as short as that of PODPC. The
mobility of MODPC molecules in bilayer membranes is greater
than that of the PODPC molecules. Therefore, the rate of
polymerization of MODPC that we measured was as rapid as
that of PODPC.
J. Mater. Chem., 2000, 10, 1047±1059
1055

View Article Online
stable than liposome membranes. In fact, the stability of
SODPC and PODPC liposomes is lower than the stability of
DODPC liposomes. The stability of irradiated MODPC
liposomes, however, is similar to that of DODPC liposomes.
Thus, even though polymerizable phospholipids may contain
the same polymerizable groups for the 2-acyl chain, for
different hydrocarbon lengths of its 1-acyl chain, different
polymerization behavior and characteristics are observed for c-
ray irradiation under similar polymerization conditions (i.e.
although the polymerization conditions using c-irradiation are
the same, the polymerization behaviour and characteristics
differ for each phospholipid). The reason for this is that the
mobility of the monomer and the growing chains are different
among the packing and conformation of phospholipids in
bilayer membranes. The packing density and conformation of
phospholipids are affected by the length of the acyl chain of the
phospholipids. The design of polymerizable phospholipids is
therefore important for obtaining a stabilizable liposome by
polymerization.
of (E,E)-octadeca-2,4-dienoic anhydride was used without
further puri®cation in the next acylation reaction.
1,2-Bis[(E,E)-octadeca-2,4-dienoyl]-sn-glycero-3-
phosphocholine (DODPC) 2
L-a-Glycerophosphorylcholine (16.5 g, 64.15 mmol) was dried
twice by evaporation of dry benzene. A solution of (E,E)-
octadeca-2,4-dienoic anhydride (ca. 17.83 mmol) and 11 g of
DMAP (90.04 mmol) were added and stirred mechanically for
2 days at 30 ³C under a nitrogen atmosphere. The progress of
the reaction was monitored by TLC (CHCl ±MeOH±
3
H O~65 : 25 : 4). After the reaction, the solvent was evapo-
2
rated in vacuo below 30 ³C. The residue was chromatographed
on 650 g of silica gel with eluted CHCl ±MeOH gradient
3
solvents (10 : 0, 9 : 1, 8 : 2, 7 : 3, 5 : 5) using an open column (i.d.
of 10 cm). The elute-containing PC fraction was combined and
the solutions evaporated. The residue was puri®ed by using a
middle-pressure chromatograph on 100 g of silica gel (i.d. of
3
3
0 mm, length of 22 cm, elute was CHCl ±MeOH±
2
2
2
H O~80 : 19 : 1, 80 : 18 : 2, pressure was 0.2 kgf cm , ¯ow
2
1
Experimental
rate was 35 mL min , pump was Model-540 Yamazen). The
elute was passed through a column of ion-exchange resin to
remove the residual DMAP. The resin was washed with 1 L of
Materials
Unless otherwise stated all reagents and chemicals (e.g., diethyl
ether, ethanol, methanol, tetrahydrofuran (THF), and calcium
dichloride) were obtained commercially and used without
further puri®cation. 4-Dimethylaminopyridine (DMAP) and
N,N'-dicyclohexylcarbodiimide (DCC) were obtained from
Wako Pure Chemical Industries, Ltd.; and 1,2-dimyristoyl-
sn-glycero-3-phosphocholine (DMPC), 1,2-dipalmitoyl-sn-gly-
cero-3-phosphocholine (DPPC) and 1,2-distearoyl-sn-glycero-
3
the MeOH and CHCl before use. The combined eluents were
evaporated and the residue was suspended in 200 mL of ultra-
pure water. The suspension was frozen by liquid nitrogen and
the solid was freeze-dried in vacuo for 24 h. The pure DODPC
was obtained in 28.9% yield (14.5 g, 18.54 mmol): R
f
~0.2
) d 0.88 (t,
), 1.35±1.50 (m,
CLC), 3.32 (s, 9 H,
N), 4.01 (m, 2 H, CH OCO), 4.20±
N, CH OCO), 5.22±5.35 (m, 1 H,
1
(CHCl
J~6.6 Hz, 6 H, CH
4 H, CH CH
CLC), 2.15 (s, 4H, CH
N(CH ), 3.75 (s, 2 H, CH
4.50 (m, 4 H, CH CH
±MeOH±H
3
O~65 : 25 : 4); H NMR (CDCl
2
3
3
), 1.20±1.35 (m, 40 H, CH
2
2
2
2
3
-phosphocholine (DSPC) were obtained from NOF Corpora-
3
)
3
2
2
tion. Chloroform was distilled with phosphorus pentoxide
before use in order to remove alcohol. Reagent-grade methanol
and chloroform were used as eluents without further puri®ca-
tion. Ultra-pure water was puri®ed with a TORAYPURE LV-
2
2
2
CHOCO), 5.74 and 5.76 (d, J~15 Hz, 2 H, CLCHCOO), 6.10±
6.20 (m, 4 H, CH
1
CHLCH), 7.15±7.30 (m, 2 H, CHLCCOO);
) d 13.99, 22.56, 28.64, 29.19, 29.25, 29.37,
2
3
C NMR (CDCl
3
0
8 system (TORAY Co., Ltd.). Other materials used were ion-
29.50, 29.57, 31.82, 32.97, 54.28, 59.20, 62.96, 63.41, 66.19,
70.74, 118.37, 118.50, 128.13, 145.20, 145.43, 145.74, 145.96,
2
exchange resin (AMBERLITE IRC-50, Organo Co., Ltd.),
silica gel (C-200, Wako Pure Chemical Industries, Ltd.), and
thin layer chromatography (TLC) plates (Silica gel 60 F254,
Merck).
166.39, 166.78; IR (KBr) 3397.1 (nOH), 2923.5 (nasCH
2853.1 (n CH
), 1716.9 (nCLO unsat), 1644.5 (vasCLC), 1468.0
(dasCH
), 1248.1 (nPLO), 1091.8 and 825.6 (nP±O), 997.3,
969.4, 723.4 cm
),
2
s
2
2
2
1
Phospholipase A was puri®ed by heating 19 mg of Snake
2
.
Venom (from Najanaja kaquthia, Wako Pure Chemical
Industries, Ltd.) in 10 mL of acetic acid buffer (CH COOH±
3
1-Myristoyl-2-[(E,E)-octadeca-2,4-dienoyl)]-sn-glycero-3-
phosphocholine (MODPC) 3
CH COONa, pH 5.0) at 90 ³C for 5 min. The mixture was then
3
allowed to stand at room temperature for 30 min, and then
centrifuged at 3000 rpm for 5 min. The upper layer of the
Into 2 L ¯asks was placed DMPC (45 g, 66.38 mmol) in
centrifuge tube was used for the phospholipase A solution.
2
600 mL of Et
buffer (pH 8.0), 100 mL of 100 mM CaCl
phospholipase A solution. The reaction mixture was sus-
pended and stirred at room temperature for 19 h. The progress
of the reaction was monitored by TLC (CHCl ±MeOH±
O~65 : 25 : 4). Upon completion of the reaction, the solvent
was evaporated under reduced pressure. The residual water
phase was combined, and extracted three times with 500 mL of
chloroform. To the water phase was added 500 mL of
chloroform and 100 mL of methanol and the organic phase
was separated; this procedure was repeated three times. Finally,
the water phase was extracted three times with 500 mL of
chloroform. The combined organic phase (ca. 4.5 L) was dried
2
O±EtOH (95 : 5), 300 mL of 100 mM Tris±HCl
The (E,E)-octadeca-2,4-dienoic acid was synthesized as
follows. Ethyl octadeca-2,4-dienoate was prepared by coupling
triethyl 4-phosphonocrotonate with tetradecanal by using the
Wittig reaction. The ester was hydrolyzed, and the precipitate
, and 15 mL of the
2
2
3
was recrystallized twice from n-hexane.
phorylcholine (GPC) was obtained from egg yolk lecithin by
L-a-Glycerophos-
H
2
3
1
the method of Hanahan.
(
E,E)-Octadeca-2,4-dienoic anhydride 1
solution of 100 g of (E,E)-octadeca-2,4-dienoic acid
0.3566 mol) in 500 mL of chloroform was placed in a 1 L
A
(
round-bottomed ¯ask equipped with a dropping funnel and
nitrogen bubbler, and stirred mechanically at 4 ³C for an hour
under a nitrogen atmosphere. To the mixture was added
dropwise 44.1 g of DCC (0.2140 mol) dissolved in 50 mL of
chloroform within 10 min at 4 ³C, and then the reaction
mixture was stirred at room temperature for 2 h. After the
reaction, precipitated dicyclohexylurea was ®ltered off under
vacuum, then the precipitate was washed twice with 50 mL of
cold chloroform. The ®ltrates were combined, and the solution
2 4
over 150 g of anhydrous Na SO and evaporated. The residue
was dried seven times by evaporation of 100 mL of dry
benzene. The residue was puri®ed by chromatography on 650 g
silica gel eluted with CHCl ±MeOH gradient solvents (10 : 0,
3
9 : 1, 8 : 2, 7 : 3, 5 : 5). The
H O~65 : 25 : 4) fractions were collected and evaporated.
R
f
3
0.07 (CHCl ±MeOH±
2
The pure lysomyristoyl phosphatidylcholine gave a 78.6%
yield (48.8 g, 0.1043 mol). The TLC (CHCl ±MeOH±
3
2 f
H O~65 : 25 : 4) showed one spot (R 0.07).
1
056
J. Mater. Chem., 2000, 10, 1047±1059

View Article Online
In a 1 L ¯ask was placed 66 wt% of (E,E)-octadeca-2,4-
dienoic anhydride solution (ca. 0.1177 mol). Lysomyristoyl
phosphatidylcholine (40.5 g, 86.66 mmol) and 8.44 g of DMAP
(69.08 mmol) were added, and the gas in the reaction ¯ask was
replaced with nitrogen and the ¯ask sealed. The reaction
52.16 mmol). The TLC (CHCl ±MeOH±H O~65 : 25 : 4)
3
showed one spot (R 0.07).
f
2
In a 1 L ¯ask was placed 44 wt% of (E,E)-octadeca-2,4-
dienoic anhydride solution (ca. 78.50 mmol). Lysopalmitoyl
phosphatidylcholine (20.8 g, 41.79 mmol) and 6.38 g of DMAP
(52.20 mmol) were added, and the gas in the reaction ¯ask was
replaced with nitrogen and the ¯ask sealed. The reaction
mixture was stirred for 3 days at 30 ³C. The progress of the
mixture was stirred for 2 days at 30 ³C. The progress of the
reaction was monitored by TLC (CHCl
3
±MeOH±
2
H O~65 : 25 : 4). After the reaction, the solvent was evapo-
rated in vacuo. The residue was divided into two portions, then
each portion was chromatographed on 650 g silica gel with
eluted CHCl ±MeOH gradient (10 : 0, 9 : 1, 8 : 2, 7 : 3, 5 : 5)
3
reaction was monitored by TLC (CHCl
O~65 : 25 : 4). After the reaction, the solvent was evapo-
rated in vacuo. The residue was chromatographed on 650 g
silica gel with eluted CHCl ±MeOH gradient (10 : 0, 9 : 1, 8 : 2,
7 : 3, 5 : 5) by using an open column (i.d. of 10 cm). The
combined eluate of the R 0.2 fractions by TLC were
evaporated. The residue was puri®ed on 100 g silica gel with
eluted CHCl ±MeOH±H O gradient (80 : 19 : 1, 80 : 18 : 2) by
using medium-pressure chromatography (i.d. of 30 mm, length
of 22 cm, eluent was CHCl ±MeOH±H O~80 : 19 : 1,
80 : 18 : 2, pressure was 0.2 kgf cm ¯ow rate was
5 mL min , pump was Model-540 Yamazen). The elute
was poured through a column of ion-exchange resin to remove
the DMAP. The resin was washed with 1 L of the MeOH and
3
CHCl before use. The combined eluates were evaporated, and
the residue was suspended in 200 mL of ultra-pure water. The
suspension was frozen by liquid nitrogen at 2196 ³C, and the
solid was freeze-dried in vacuo for 24 h. The pure PODPC was
3
±MeOH±
H
2
using an open column (i.d. of 10 cm). The combined eluate of
the R 0.2 fractions of TLC was evaporated. The residue was
3
f
puri®ed on 100 g silica gel with eluted CHCl ±MeOH±H O
3
f
2
gradient (80 : 19 : 1, 80 : 18 : 2) by using medium-pressure
chromatography (i.d. of 30 mm, length of 22 cm, eluent was
3
2
3 2
CHCl ±MeOH±H O~80 : 19 : 1, 80 : 18 : 2, pressure was
2
2
21
0
5
.2 kgf cm , ¯ow rate was 35 mL min , pump was Model-
40 Yamazen). The eluate was poured through a column of
3
2
2
2
,
2
1
3
ion-exchange resin to remove the DMAP. The resin was
washed with 1 L of the MeOH and CHCl before use. The
3
combined eluates were evaporated, and the residue was
suspended in 200 mL of ultra-pure water. The suspension
was frozen by liquid nitrogen at 2196 ³C, and the solid was
freeze-dried in vacuo for 24 h. The pure MODPC gave a 39.5%
yield (25 g, 34.24 mmol): R ~0.19 (CHCl ±MeOH±
f
3
1
obtained in 62.8% yield (20 g, 26.38 mmol): R
f
~0.19 (CHCl
NMR (CDCl3) 0.88 (t,
), 1.00±1.40 (m, 44 H, CH ), 1.40±1.50
3
±
H
2
O~65 : 25 : 4); H NMR (CDCl
H, CH ), 1.00±1.35 (m, 40 H, CH
CH CH
CLC), 1.50±1.60 (m, 2 H, CH
m, 2 H, CH
CLC), 2.27 (t, J~ 7 Hz, 2 H, ±CH
.35 (s, 9 H, N(CH ), 3.80 (s, 2 H, CH N), 3.90±4.10 (m, 2 H,
OCO), 4.15±4.48 (m, 4 H, CH CH N, CH OCO), 5.26±
3
) d 0.88 (t, J~6.6 Hz, 6
1
MeOH±H O~65 : 25 : 4);
H
d
), 1.35±1.50 (m, 2 H,
CH COO), 2.10±2.20
CH COO),
2
3
2
J~6.6 Hz, 6 H, CH
(
3
2
2
2
2
2
m, 2 H, CH CH CLC), 1.50±1.65 (m, 2 H, CH CH COO),
(
3
CH
2
2
2
2
2
2
2
2
CH CH COO), 3.31 (s, 9 H, N(CH ) ), 3.80 (s, 2 H, CH N),
.10±2.20 (m, 2 H, CH
2
CLC), 2.27 (t, J~6.6 Hz, 2 H,
3
)
3
2
2
2
3 3
2
2
2
2
2
3
.70±3.83 (m, 2 H, CH
CH OCO), 5.20±5.32 (m, 1 H, CHOCO), 5.75 (d, J~15.5 Hz, 1
2
2 2
OCO), 4.15±4.45 (m, 4 H, CH CH N,
5
.28 (m, 1 H, CHOCO), 5.75 (d, J~15 Hz, 1 H, CLCHCOO),
.15 (m, 2 H, CH
6
2
±CHLCH), 7.18±7.30 (m, 1 H, CHLCCOO);
2
1
3
H, CLCHCOO), 6.15 (2 H, CH CHLCH), 7.18±7.30 (m, 1 H,
C NMR (CDCl ) d 14.10, 22.69, 24.92, 28.75, 29.16, 29.33,
2
3
13
CHLCCOO); C NMR (CDCl ) 14.12, 22.71, 24.94, 28.81,
2
5
1
9.49, 29.52, 29.61, 29.69, 31.93, 33.10, 34.16, 54.48, 59.28,
9.35, 63.02, 63.50, 66.47, 70.65, 70.76, 118.54, 128.19, 145.67,
46.09, 166.50, 173.58; IR (KBr) 3397 (nOH), 2955.3, 2917.7
3
2
5
1
9.2, 29.40, 29.52, 29.60, 29.72, 29.76, 31.95, 33.14, 34.18,
4.36, 59.33, 59.41, 63.05, 63.50, 63.58, 66.27, 70.62, 70.73,
18.54, 128.21, 145.68, 146.16, 166.52, 173.62; IR (KBr) 2918.7
(
n
asCH
2
), 2850.2 (n
s
CH
2
), 1735.2 (nCLO sat), 1713.9 (nCLO
), 1262.6
nPLO), 1092.8, 1065.8 and 832 (nP±O), 1169.0, 1008.9,
(
n CH ), 2850.2 (n CH ), 1735.2 (nCLO sat), 1715.9 (nCLO
as
2
s
2
unsat), 1645.5 (vasCLC±CLC), 1472.8 (dasCH
(
9
2
unsat), 1643.6 (vasCLC±CLC), 1467.0 (dasCH
2
), 1257.7
2
1
.
(nPLO), 1089.9 and 1062.9 and 826 (nP±O), 1003.1, 969.4,
71.3, 720.5 cm
Anal. Calcd for
40 76 8 2
C H O PN?H O
2
21.5 cm . Anal. Calcd for C42
1
7
80 8 2
H O PN?1.5H O (785.06): C,
(
1
748.00): C, 64.22; H, 10.51; N, 1.87. Found: C, 63.97; H,
0.76; N, 1.53%. MS (FAB): m/z 731 (Mz1).
64.52; H, 10.66; N, 1.78. Found: C, 64.30; H, 10.95; N, 1.50%.
MS (FAB): m/z 758 (Mz1).
1-Palmitoyl-2-[(E,E)-octadeca-2,4-dienoyl]-sn-glycero-3-
phosphocholine (PODPC) 4
1-Stearoyl-2-[(E,E)-octadeca-2,4-dienoyl]-sn-glycero-3-
phosphocholine (SODPC) 5
Into a 2 L ¯ask was placed DPPC (50 g, 68.12 mmol) in 530 mL
of Et O±EtOH (95 : 5), 300 mL of 100 mM Tris±HCl buffer
Into a 2 L ¯ask was placed DSPC (50 g, 63.28 mmol) in 680 mL
of Et O±EtOH (95 : 5), 340 mL of 100 mM Tris±HCl buffer
2
2
(
pH 8.0), 100 mL of 100 mM CaCl , and 5 mL of the
2
(pH 8.0), 100 mL of 100 mM CaCl , and 17 mL of the
2
phospholipase A
pended and stirred at room temperature for 7 h. The progress
of the reaction was monitored by TLC (CHCl ±MeOH±
O~65 : 25 : 4). Upon completion of the reaction, the solvent
was evaporated under reduced pressure. The residual water
phase was extracted three times with 250 mL of chloroform. To
the water phase was added 250 mL of chloroform and 50 mL of
methanol and the organic phase was separated; this procedure
was repeated three times. Finally, the water phase was
extracted three times with 250 mL of chloroform. The
combined organic phases (ca. 2 L) were dried over 70 g of
2
solution. The reaction mixture was sus-
phospholipase A2 solution. The reaction mixture was sus-
pended and stirred at room temperature for 19 h. The progress
of the reaction was monitored by TLC (CHCl ±MeOH±
2
H O~65 : 25 : 4). Upon completion of the reaction, the solvent
was evaporated under reduced pressure. The residual water
phase was extracted three times with 250 mL of chloroform. To
the water phase was added 250 mL of chloroform and 50 mL of
methanol and the organic phase was separated; and this
procedure was repeated three times. Finally, the water phase
was extracted three times with 250 mL of chloroform. The
combined organic phases (ca. 2 L) were dried over 70 g of
3
3
H
2
anhydrous Na SO and evaporated. The residue was dried
seven times by evaporation of 100 mL of dry benzene. The
residue was puri®ed by chromatography on 650 g silica gel
2
4
anhydrous Na SO4 and evaporated. The residue was dried
2
seven times by evaporation of 100 mL of dry benzene. The
residue was puri®ed by chromatography on 650 g silica gel
eluted with CHCl
: 3, 5 : 5). The R
fractions were collected and evaporated. The pure lysopalmi-
toyl phosphatidylcholine gave 77.0% yield (26.0 g,
3
±MeOH gradient solvents (10 : 0, 9 : 1, 8 : 2,
eluted with CHCl ±MeOH gradient solvents (10 : 0, 9 : 1, 8 : 2,
3
7
f
0.07 (CHCl ±MeOH±H O~65 : 25 : 4)
3
2
7 : 3, 5 : 5). The Rf 0.07 (CHCl ±MeOH±H O~65 : 25 : 4)
3
2
fractions were collected and evaporated. The pure lysostearoyl
phosphatidylcholine was obtained in 77.5% yield (25.6 g,
a
J. Mater. Chem., 2000, 10, 1047±1059
1057

View Article Online
52.6 mmol). The TLC (CHCl ±MeOH±H O~65 : 25 : 4)
showed one spot (R 0.07).
In a 1 L ¯ask was placed 34 wt% of (E,E)-octadeca-2,4-
dienoic anhydride solution (ca. 60.62 mmol). Lysostearoyl
phosphatidylcholine (25.6 g, 48.86 mmol) and 4.76 g of DMAP
Preparation of liposomes. All of the liposomes in this study
were prepared by an extrusion method according to the
following procedure. Lipids (2 g) were added to a polyethylene
bag (1006150 cm) and hydrated with 40 mL of ultra-pure
water. The dispersions were kept at a temperature above the
3
2
f
(
replaced with nitrogen and the ¯ask sealed. The reaction
mixture was stirred for 7 days at 30 ³C. The progress of
38.96 mmol) were added, and the gas in the reaction ¯ask was
c
phase transition temperature (T ) of each phospholipid. The
dispersions were homogenized by using a Stomacher (UAC
HOUSE, Model: BA6020, SEWARD MEDICAL) for 20 min.
The solutions were extruded stepwise through polycarbonate
®lters with pore size of 5, 1, 0.8, 0.4, and 0.2 mm. To reduce the
diameter distribution, the liposomes were passed three times
through the 0.2 mm polycarbonate ®lters. The liposomes were
adjusted to 1.8±2.0 wt% lipid concentration by diluting with
ultra-pure water.
the reaction was monitored by TLC (CHCl
O~65 : 25 : 4). After the reaction, the solvent was evapo-
rated in vacuo. The residue was chromatographed on 650 g
silica gel eluted with CHCl ±MeOH gradient (10 : 0, 9 : 1, 8 : 2,
: 3, 5 : 5) by using an open column (i.d. of 10 cm). The
3
±MeOH±
H
2
3
7
f
combined eluate of the R 0.2 fractions of TLC was evaporated.
The residue was puri®ed on 100 g silica gel with eluted CHCl
MeOH±H O gradient (80 : 19 : 1, 80 : 18 : 2) by using medium-
pressure chromatography (i.d. of 30 mm, length of 22 cm,
eluate was CHCl ±MeOH±H O~80 : 19 : 1, 80 : 18 : 2, pressure
was 0.2 kgf cm , ¯ow rate was 35 mL min , pump was
Model-540 Yamazen). The eluate was poured through a
column of ion-exchange resin to remove DMAP. The resin was
3
±
Measurement of diameter. In a 1 cm quartz cuvette equipped
with a Te¯on stopper was placed 3 mL of ultra-pure water that
was ®ltered through a 0.2 mm ®lter by using a syringe. The
liposome solution (ca. 1 mL) was added, and the solution was
dispersed homogeneously. The average diameter of the
liposomes was measured by using a Nicomp model-370HPL
particle sizer instrument (Paci®c Scienti®c) with a He±Ne laser
light source. The intensity of scattered light was recorded and
converted to size distributions as a gaussian mode. The
standard deviation (SD) of the mean diameter was also
recorded as the scattered-light intensity.
2
3
2
2
2
21
3
washed with 1 L of the MeOH and CHCl before use. The
combined eluates were evaporated, and the residue was
suspended in 200 mL of ultra-pure water. The suspension
was frozen by liquid nitrogen at 2196 ³C, and the solid was
freeze-dried in vacuo for 24 h. The pure SODPC was obtained
in 52.0% yield (20 g, 25.44 mmol): R
f
~0.19 (CHCl
) d 0.88 (t, J~6.6 Hz, 6 H,
), 1.00±1.35 (m, 48 H, CH ), 1.35±1.50 (m, 2 H,
3
±MeOH±
1
Differential scanning calorimetry (DSC). The liposomes
2
2
H O~65 : 25 : 4); H NMR (CDCl
3
1
(
60 mL, 20 mg mL
lipid concentration) were placed in
CH
3
2
large silver pans and the pans were sealed tightly. The phase
transition temperatures of the liposomes were measured with a
DSC-120 instrument (Seiko Instrument Inc.) at a heating rate
CH CH CLC), 1.50±1.65 (m, 2 H, CH CH COO), 2.10±2.20
2
2
2
2
(
m, 2 H, CH CLC), 2.27 (t, J~7 Hz, 2 H, CH CH COO), 3.35
2
2
2
(
CH OCO), 4.18±4.43 (m, 4 H, CH CH N, CH OCO), 5.22±
s, 9 H, N(CH ) ), 3.66±3.90 (s, 2 H, CH N), 3.90±4.10 (m, 2 H,
3
3
2
2
1
of 1 ³C min
.
2
2
2
2
5
6
.30 (m, 1 H, CHOCO), 5.75 (d, J~15.5 Hz, 1 H, CLCHCOO),
Polymerization of liposomes. Into a 110 mL glass vial were
placed 100 mL of the 2 wt% liposomal solution, and the vials
were sealed with an aluminium-covered rubber cap. Then, the
solution was bubbled with argon for 30 min at room
temperature. Using a 10 mL injection syringe, 5 mL of
liposomal solution was transferred to a 10 mL sealed glass
vial that contained an argon atmosphere. The liposomal
solutions were stirred magnetically at 4 ³C and irradiated with
.15 (m, 2 H, CH CHLCH), 7.18±7.30 (m, 1 H, CHLCCOO);
2
1
3
C NMR (CDCl ) 14.12, 22.71, 24.94, 28.77, 29.2, 29.38,
3
2
5
1
9.51, 29.56, 29.63, 29.74, 31.95, 33.12, 34.18, 54.43, 59.32,
9.37, 63.01, 63.50, 66.32, 66.43, 70.78, 70.78, 118.54, 128.21,
45.71, 146.13, 166.50, 173.58; IR (KBr) 3413.5 (nOH), 2919.6
(
n CH ), 2851.1 (n CH ), 1735.2 (nCLO sat), 1718.8 (nCLO
as
2
s
2
unsat), 1642.6 (v CLCCLC), 1468.0 (d CH ), 1252.0 (nPLO),
1
Calcd for C H O PN?1.5H O (795.12): C, 64.99; H, 10.78; N,
as
as
2
2
1
092.8 and 1065.8 and 823.7 (nP±O), 970.3, 721.5 cm . Anal.
6
0
c-rays from a Co source (25 000 Ci). The dose rate was
4
4
84
8
2
2
adjusted between 0.5 to 10 kGy h by varying the distance
1
1
(
.72. Found: C, 65.27; H, 11.19; N, 1.42%. MS (FAB): m/z 786
Mz1).
6
0
between the Co source and the liposomal solution. Fresh
liposomal solutions were prepared for each measurement.
Measurement of polymerization conversion. No description
of detailed method for measuring the polymerization conver-
Methods
Measurement. Infrared spectra were obtained with a JASCO
FT/IR-7300 spectrometer. Nuclear magnetic resonance spectra
NMR) were recorded with a JEOL JNM-EX270 (270 MHz)
spectrometer. The chemical shifts are given in parts per million
ppm) relative to tetramethylsilane as an internal standard.
Mass spectra were obtained with a JEOL DIMS- spectrometer.
Elemental analyses were performed by Ibaraki Environment
Research Center, Inc.
1
7
sion of the polymerized liposome (e.g., DODPC, PODPC
22
etc.) has been reported in the open literature. The baseline (i.e.,
100% polymerization) might be estimated by using a freehand
drawing. We developed a method for drawing these polymer-
izable phospholipids by using simple UV-irradiation in a cell.
The monomeric liposomal solutions (DODPC liposome
14 mL, other liposomes 28 mL) were diluted with 10 mL of
ultra-pure water by using a transfer pipette. The solution was
dispersed homogeneously and transferred to a 1 cm quartz cell.
Using a UV spectrometer (UV-240, Shimadzu Corporation),
the diluted solutions were measured at room temperature for
the absorption of each liposome at the absorption maximum
(lmax) (DODPC 254 nm, MODPC and PODPC 259 nm,
SODPC 253 nm), based on the diene group. The irradiated
liposomal solutions were measured by the same procedure. A
polymerization yield of 100% was obtained by the following
procedure: The diluted monomeric liposome was irradiated by
a UV lamp (MODEL ENF-260C/J, Spectronics Corporation)
at 254 nm until absorption at lmax disappeared. The poly-
merization conversions were calculated from the difference
between the absorption of the irradiated liposome solution at
(
(
Surface pressure±area isotherm. The surface pressure±area
isotherms of polymerizable phospholipids were measured by a
Wilhelmy plate method using a moving-wall type Langmuir±
Blodgett membrane trough (Nippon Laser Electric). The
trough dimensions were 400 mm626 mm. For the subphase
we used ultra-pure water, which has a resistivity of 18 MV. A
2
1
chloroform solution of the lipids (0.7 mg mL ) was spread on
the water surface using a microsyringe, and the solvent was
allowed to evaporate for 20 min. The temperature of the
subphase was maintained at 3.8 ³C. The monolayer was
2
21
21
compressed at a rate of 13.0 cm min (50 mm min ).
1
058 J. Mater. Chem., 2000, 10, 1047±1059

View Article Online
lmax and that of the monomeric liposome. The polymerization
conversions were calculated from eqn. (17).
3
4
5
6
7
E. Mayhew and D. Papahajopoulos, in Liposomes, ed. M. J. Ostro,
Marcel Dekker, Inc., New York, 1983, p. 289.
H. Ringsdorf, B. Schlarb and J. Venzmer, Angew. Chem., Int. Ed.
Engl., 1988, 27, 113.
B. Hupfer, H. Ringsdorf and H. Schupp, Chem. Phys. Lipids,
Polymerisation conversion ꢀ%†~ꢀAI{AT†=ꢀAI{A0†|100
(
17)
1983, 33, 355.
D. S. Johnston, L. R. McLean, M. A. Whittam, A. D. Clark and
D. Chapman, Biochemistry, 1983, 22, 3194.
K. Dorn, R. T. Klingbiel, D. P. Specht, P. N. Tyminski,
H. Ringsdorf and D. F. O'Brien, J. Am. Chem. Soc., 1984, 106,
T I
where A is the absorption intensity at lmax after irradiation, A
is the absorption intensity at lmax before irradiation, A is the
0
absorption intensity for 100% polymerized by UV irradiation.
1627.
8
9
0
A. Anikin, V. Chupin, M. Anikin and G. Serebrennikova,
Makromol. Chem., 1993, 194, 2663.
A. Akimoto, K. Dorn, L. Gros, H. Ringsdorf and H. Schupp,
Angew. Chem., Int. Ed. Engl., 1981, 20, 90.
Y. Matsushita, E. Hasegawa and K. Eshima, Makromol. Chem.,
Rapid Commun., 1987, 8, 1.
Molecular weight measurement. The molecular weights of
the poly(phospholipid) were determined by measuring the
octadecadienoic acid polymer obtained from hydrolysis of the
irradiated liposome, because the polymerized liposome was
insoluble in organic solvents. To 5 mL of irradiated liposomes
was added 5 mL of 2 M NaOH and the temperature of the
mixture was maintained at 70 ³C for 24 h. The reaction mixture
was cooled to room temperature, neutralized with 5 mL of 2 M
HCl, and further cooled to 4 ³C. The precipitate was ®ltered
through a 5G glass ®lter, washed with water, and dried in vacuo
for 12 h. The obtained solids (10 mg) were dissolved in 5 mL of
THF. The solutions containing octadecadienoic acid polymer
1
11 S. L. Regen, A. Singh, G. Oehme and M. Singh, J. Am. Chem.
Soc., 1982, 104, 791.
12
S. L. Regen, K. Yamaguchi, N. K. P. Samuel and M. Singh, J. Am.
Chem. Soc., 1983, 107, 42.
N. K. P. Samuel, M. Singh, K. Yamaguchi and S. L. Regen, J. Am.
Chem. Soc., 1985, 107, 42.
13
14 H. Ohno, Y. Ogata and E. Tsuchida, J. Polym. Sci., Polym. Chem.
Ed., 1986, 24, 2959.
1
1
1
5
6
7
H. Ohno, S. Takeoka and E. Tsuchida, Bull. Chem. Soc. Jpn.,
987, 60, 2945.
H. Ohno, Y. Ogata and E. Tsuchida, Macromolecules, 1987, 20,
29.
E. Tsuchida, E. Hasegawa, N. Kimura, M. Hatashita and
C. Makino, Macromolecules, 1992, 25, 207.
(
10 mL) were measured by using gel permeation chromato-
graphy (HLC-802A and HLC-802UR, column: G2500HXL
G2000HXL and G5000HXLzG3000HXL, range of molecular
1
z
9
4
2
5
3
weight 3.5610 ±3610 and 2610 ±1610 , respectively, Toyo
Soda Co., Ltd.) with THF as eluent at a ¯ow rate of
1
from the calibration curve by using standard polystyrenes.
2
1
mL min at 40 ³C. The molecular weights were calculated
18 K. Akama, K. Awai, S. Tokuyama, T. Satoh, F. Hosoi and
H. Omichi, Radiat. Phys. Chem., 1995, 46, 257.
1
9
0
K. Morizawa, K. Akama, Y. Kawakami and E. Tsuchida,
Biomater. Artif. Cells Immobil. Biotechnol., 1992, 20, 641.
F. Hosoi, H. Omichi, K. Akama, K. Awai, S. Tokuyama and
T. Satoh, Nucl. Instrum. Methods Phys. Res., Sect. B, 1995, 105,
318.
Physical stability of liposomes (freeze±thawing test). To
measure the stability of liposomes, a freeze±thaw test was
used as follows. First, the irradiated liposome solution (2 mL)
was transferred to a 10 mL sample tube and frozen with liquid
nitrogen for 5 min. Then the frozen liposome was thawed
gradually at room temperature. After the sample was
completely thawed, the average diameters of the liposome
were measured by using the same particle sizer instrument
previously described. The changes in average diameter were
calculated by comparing the diameter measurements taken
before freezing with those taken after the freeze±thaw cycle.
2
21 D. Chapman, in Form and Function of Phospholipids, ed.
G. B. Ansell, J. N. Hawthore and R. H. C. Dawson, Elsevier
Scienti®c, Amsterdam, 1973, p. 117.
2
2
S. Takeoka, H. Iwai, H. Ohno and E. Tsuchida, Bull. Chem. Soc.
Jpn., 1989, 62, 102.
L. D. Mayer, M. J. Hope, P. R. Cullis and A. S. Janoff, Biochim.
Biophys. Acta, 1985, 817, 193.
2
3
24 K. Ogino, M. Goto and M. Abe, J. Jpn. Oil Chem. Soc., 1988, 37,
40.
T. Kagiya, M. Izu, S. Machi and K. Fukui, Bull. Chem. Soc. Jpn.,
967, 40, 1049.
T. Kagiya, M. Izu and K. Fukui, Bull. Chem. Soc. Jpn., 1967, 40,
045.
6
2
2
5
6
1
Acknowledgements
1
The authors wish to thank Dr Yoshishige Murata and Mr
Yoshio Nakano (Tsukuba Research Laboratory, NOF Cor-
poration) for their helpful discussions and encouragement. We
express deepest condolences and gratefully acknowledge Mr
Fumio Hosoi, who died on March 13, 1998.
27 H. Ohno, S. Takeoka, H. Iwai and E. Tsuchida, J. Polym. Sci.
Polym. Chem. Ed., 1987, 25, 2737.
28
29
30
H. Sakai, S. Takeoka, H. Yokohama, H. Nishide and E. Tsuchida,
Polym. Adv. Technol., 1992, 3, 389.
H. Ringsdorf and H. Schupp, Am. Chem. Soc., Div. Org. Coat.
Plast. Chem., Pap. Meet., 1980, 42, 379.
F. Hosoi, H. Omichi, K. Akama, K. Awai, S. Endo and
Y. Nakano, Nucl. Instrum. Methods Phys. Res., Sect. B, 1997,
References
131, 329.
D. J. Hanahan, Biochem. Prep., 1962, 9, 55.
31
1
2
J. H. Fendler, Acc. Chem. Res., 1980, 13, 7.
F. C. Szoka and D. Papahadjopoulos, in Liposomes, ed.
C. G. Knight, Elsevier, Amsterdam, 1981, p. 51.
J. Mater. Chem., 2000, 10, 1047±1059
1059