Synthesis and Spectroscopic Analysis of Chromophoric Lipids Inducing pH-Dependent Liposome Fusion

Article abstract of DOI:10.1021/ja037796x

We design novel chromophoric amphiphiles 6a-c, which lead to pH-dependent membrane fusion of egg phosphatidylcholine (eggPC) liposome containing them. Lipids 6a-c comprise double alkyl chains, a single chain with a 2-nitrophenol group as a pH trigger, and dipeptide (Asp-Asp) between them. The pKa values of 2-nitrophenol groups of 6a-c in liposome are larger than that of hydrophilic compound 9 in an aqueous solution. Absorption spectra indicate that the fields around 2-nitrophenol of 6a-c situated in liposome membranes are more hydrophobic than that of 9 in an aqueous solution, whereas the environments around deprotonated 2-nitrophenolate of 6b and 6c are not so hydrophobic as that of 6a. This means that protonated 2-nitrophenol groups of 6a-c are embedded in bilayer membranes. Deprotonated 2-nitrophenol groups of 6b and 6c must be located in less hydrophobic circumstances, while that of 6a is still embedded in bilayer membranes because of its larger hydrophobicity. Absorption spectra and ¹H NMR spectra respectively suggest that protonated 2-nitrophenol groups of 6a and those of 6c might take face-to-face associations in bilayer membranes.

Full text of DOI:10.1021/ja037796x

Published on Web 11/08/2003
Synthesis and Spectroscopic Analysis of Chromophoric
Lipids Inducing pH-Dependent Liposome Fusion
Takenori Tomohiro,*, †,‡ Yoshikatsu Ogawa,† Hiroaki Okuno,†,§ and
Masato Kodaka*,†
Contribution from the Institute for Biological Resources and Functions, National Institute of
AdVanced Industrial Science and Technology (AIST), Central 6, 1-1-1 Higashi,
Tsukuba, Ibaraki 305-8566, Japan, and Faculty of Pharmaceutical Sciences, Toyama Medical
and Pharmaceutical UniVersity, 2630 Sugitani, Toyama 930-0194, Japan
Received August 7, 2003; E-mail: ttomo@ms.toyama-mpu.ac.jp; m.kodaka@aist.go.jp.
Abstract: We design novel chromophoric amphiphiles 6a-c, which lead to pH-dependent membrane fusion
of egg phosphatidylcholine (eggPC) liposome containing them. Lipids 6a-c comprise double alkyl chains,
a single chain with a 2-nitrophenol group as a pH trigger, and dipeptide (Asp-Asp) between them. The pKa
values of 2-nitrophenol groups of 6a-c in liposome are larger than that of hydrophilic compound 9 in an
aqueous solution. Absorption spectra indicate that the fields around 2-nitrophenol of 6a-c situated in
liposome membranes are more hydrophobic than that of 9 in an aqueous solution, whereas the environments
around deprotonated 2-nitrophenolate of 6b and 6c are not so hydrophobic as that of 6a. This means that
protonated 2-nitrophenol groups of 6a-c are embedded in bilayer membranes. Deprotonated 2-nitrophenol
groups of 6b and 6c must be located in less hydrophobic circumstances, while that of 6a is still embedded
in bilayer membranes because of its larger hydrophobicity. Absorption spectra and 1H NMR spectra
respectively suggest that protonated 2-nitrophenol groups of 6a and those of 6c might take face-to-face
associations in bilayer membranes.
Introduction
pH induces membrane interaction by insertion of the fusion
peptide. There are some observations that pores are formed in
Compared with a viral vector, liposomes are good candidates
as efficient carriers of oligonucleotides or drugs into a variety
of cells in the fields of gene therapy1 or chemotherapy.2 One
of problems is, however, the stability of liposomes in vivo due
to their rapid uptake into the mononuclear phagocyte system.
Various types of lipid derivatives and thus liposomes have been
designed to solve this problem and furthermore to improve the
targeting efficiency toward tissues.3 Some challenges are focused
on the artificial mimetics of the biological uptake systems
through endocytosis or phagocytosis pathways, e.g., the DNA
introduction of influenza virus into a cell via membrane fusion.4
The fusion mechanism is not clear yet, whereas it is known
that conformational change of influenza hemagglutinin at low
membranes after the perturbation is caused by insertion of the
fusion peptide, which leads to complete fusion.5 A similar
phenomenon was recently reported for protein Bax that pro-
moted apoptosis in neurons.6 One of the attractive methods is
to use pH-sensitive liposomes.7 Therefore, interaction between
these liposomes and cells has been tested in vitro, which showed
possibilities for the application of the liposomes to gene or drug
delivery.8 Their applications are nevertheless limited, since most
of these liposomes consist of only natural lipids or simply
modified lipids. To achieve biomimetic pH-dependent lipo-
somes, we designed new artificial lipids with a carboxyl group9
or a 2-nitrophenol group,10 which act as adhesion triggers toward
cell membranes. In the lipids bearing a carboxylate trigger, the
efficiency of liposome fusion in an acidic solution is clearly
* To whom inquiries about the paper should be addressed.
† National Institute of Advanced Industrial Science and Technology.
‡ Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceuti-
cal University.
(5) (a) Bonnafous, P.; Stegmann, T. J. Biol. Chem. 2000, 275, 6160. (b)
Kanaseki, T.; Kawasaki, K.; Murata, M.; Ikeuchi, Y.; Ohnishi, S. J. Cell
Biol. 1997, 137, 1041.
(6) Epand, R. F.; Martinou, J. C.; Montessuit, S.; Epand, R. M.; Yip, C. M.
Biochem. Biophys. Res. Commun. 2002, 298, 744.
(7) (a) Wang, C.-Y.; Huang, L. Biochemistry 1989, 28, 9508. (b) Ropert, C.;
Malvy, C.; Couvreur, P.; In Targeting of Drugs; Gregoriadis, G., McCor-
mack, B., Eds.; Plenum: New York, 1996; Vol. 5, pp 151-162. (c)
Venugopalan, P.; Jain, S.; Sankar, S.; Singh, P.; Rawat, A.; Vyas, S. P.
Pharmazie 2002, 57, 659.
§ Present address: School of Pharmaceutical Sciences, Toho University.
(1) (a) Dass, C. R. J Pharm Pharmacol. 2002, 54, 593. (b) Garcia-Chaumont,
C.; Seksek, O.; Grzybowska, J.; Borowski, E.; Bolard, J. Pharmacol. Ther.
2000, 87, 255. (c) Agrawal, S.; Ikeuchi, T.; Sun, D.; Sarin, P. S.; Konopka,
A.; Maizel, J.; Zamecnik, P. C. Proc. Natl. Acad. Sci. U. S. A. 1989, 86,
7790.
(2) (a) Kim, C. K.; Lim, S. J. Arch. Pharm. Res. 2002, 25, 229. (b) Sidwell,
R. W. Huffman, J. H. Schafer, T. W.; Shipman, C. Chemotherapy 1990,
36, 58.
(3) (a) Bendas, G. Biol. Drugs 2001, 15, 215. (b) Maruyama, K. Biol. Pharm.
Bull. 2000, 23, 791.
(4) (a) Wilson, I. A.; Skehel, J. J.; Wiley: D. C. Nature 1981, 289, 366. (b)
Bullough, P. A.; Hughson, F. M.; Skehel, J. J.; Wiley, D. C. Nature 1994,
371, 37. (c) Blumenthal, R.; Clague, M. J.; Durell, S. R.; Epand R. M.
Chem. ReV. 2003, 103, 53.
(8) (a) Drummond, D. C.; Zignani, M.; Leroux, J.-C. Prog. Lipid Res. 2000,
39, 409. (b) Dass C. R. Drug DeliV. 2002, 9, 169.
(9) (a) Ogawa, Y.; Tomohiro, T.; Yamazaki, Y.; Kodaka, M.; Okuno, H. J.
Chem. Soc., Chem. Commun. 1999, 823. (b) Ogawa, Y.; Kodaka, M.;
Okuno, H. Chem. Phys. Lipids 2002, 119, 51.
(10) Shah, S. N.; Tomohiro, T.; Ogawa, Y.; Kodaka, M.; Okuno, H. Lipids 2000,
35, 689.
9
10.1021/ja037796x CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 14733-14740
14733

A R T I C L E S
Tomohiro et al.
Scheme 1
dependent on the length of aliphatic trigger chains; that is, the
balance of hydrophobicity and hydrophilicity of the trigger part
is important for efficient liposome fusion.9 We also used a
2-nitrophenol group, because it normally has pKa at neutral pH
and therefore we can expect direct transfer of liposome contents
into cells at the physiological pH, in analogy with a vector in
vitro. Actually the lipid (6a) with a 2-nitrophenol trigger can
induce liposome fusion at pH lower than 8.10
In the present study, a series of lipids bearing a 2-nitrophenol
group, 6a and its analogues (6b and 6c), is synthesized and
analyzed in detail by spectroscopic methods to evaluate their
microscopic environments in bilayer membranes.
since its pKa value is normally around 7 and thus liposome
fusion is possible at physiological pH. In addition, this is a good
chromophore for spectroscopic analyses of its microscopic
environments, since the absorption wavelength is sensitive to
surrounding polarity, molecular interaction, and so on. Three
kinds of linkers, n-alkyl, polyethylene-type, and protic chains
between aspartic acid and 2-nitrophenol were designed to
evaluate the effect of hydrophobicity of trigger chains. In the
previous study using lipids bearing a carboxylate trigger,
efficiency of lipid mixing was largely dependent on the length
of trigger alkyl chains.9 Namely, when the pH of the solution
was changed from the neutral region toward the acidic side,
hydrophobicity of the trigger chain was increased by protonation
of the carboxylate anion. The trigger chain should be inserted
into the bilayer membrane of another liposome, which resulted
in membrane perturbation.13 These results indicated that the
balance of hydrophilic and hydrophobic character should be
important for membrane fusion. To reveal the effect of
hydrophobicity of the trigger chains on the microscopic
structures of lipids, two other types of analogous lipids (6b,c)
with less hydrophobicity were designed and synthesized as well
as 6a.
The liposome solution was prepared by sonication using a
probe-type sonicator. The liposomes were composed of eggPC
and the synthetic lipids (5-20 mol % of the whole lipids).
Purification of the liposomes was achieved by gel filtration on
Sephadex G75 (eluant: 10 mM Taps buffer, 0.1 M NaCl, pH
9.0) with monitoring the absorbance at 415 nm due to 2-nitro-
phenolate moiety. There was one peak with the same retention
time as the peak of liposomes detected by light scattering at
420 nm (irradiated at 350 nm) as shown in Figure 1. This
Results
Preparation of Lipids and Liposomes. Synthetic routes of
lipids bearing a 2-nitrophenol group (6a-c) derived from 1 are
shown in Scheme 1. Compound 2 composed of two aspartic
acids and double long alkyl chains was prepared and character-
ized as previously described.9
The synthetic lipids consist of two aspartic acids, double long
alkyl chains, and the additional third chain with a 2-nitrophenol
group at the end. These fragments connect by amide bonds that
are stable under ordinary acidic and basic conditions. This
property is important for further combination between the lipids
and peptides, and we have directly synthesized lipopeptides by
a peptide synthesizer on a solid phase after preparing pep-
tides.11,12 The hydrophilic carboxyl group of the aspartic residue
has a role to maintain its position on the surface of the liposome
membrane. The 2-nitrophenol group is employed as a pH trigger,
(11) Ogawa, Y.; Kawahara, H.; Yagi, N.; Kodaka, M.; Tomohiro, T.; Okada,
T.; Konakahara T.; Okuno, H. Lipids 1999, 34, 387.
(12) Yagi, N.; Ogawa, Y,; Kodaka, M.; Okada, T.; Tomohiro, T.; Konakahara,
T.; Okuno, H. Lipids 2000, 35, 673.
(13) Ogawa, Y.; Epand, R. M. Biosci. Rep. 1997, 17, 401.
9
14734 J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003

Spectroscopic Analysis of Liposome Fusion
A R T I C L E S
Table 1. pKa Values of Compounds 6a-c, 8, and 9
compound
conditiona
pKa
6a
6b
6c
8
S10% EggPC liposome
S10% EggPC liposome
S10% EggPC liposome
S10% EggPC liposome
H2O
>9
8.97
8.55
8.73
6.78
9
a The pKa values were measured in Fusion-lip solution at 20 °C for
compound 6 and 8 and in an aqueous solution for compound 9.
the supporting information), where the absorbance was measured
after the pH of the solution became steady. As summarized in
Table 1, all the pKa values of synthetic lipids 6a-c and 8 in
liposome were larger than 8.5, which were quite different from
the pKa of 9 (6.78) in an aqueous solution. In 6a, nitrophenolic
protons were not completely dissociated even at pH 11.5 (see
the supporting information).
Electronic Absorption Spectra. Some spectroscopic methods
are known to be useful to study hydrophobic environments in
membranes.15 In the present study, the microscopic environment
around the 2-nitrophenol trigger parts were analyzed by absorp-
tion spectra of the protonated and deprotonated species in
aqueous solutions, organic solvents, and lipid bilayer mem-
branes.
Figure 1. Gel filtration of liposome using a Sephadex G75 column eluted
with 10 mM Taps buffer (pH 9.0, 0.1 M NaCl). Left: absorbance at 415
nm (solid line). Right: light scattering (detection 420 nm/irradiation 350
nm, dotted line).
Absorption maxima of 9 were observed at 367 nm (λp) due
to the protonated form (pH 3.1) and at 427 nm (λd) due to the
deprotonated one (pH 11.3) in an aqueous solution. These
absorption bands showed a red shift (379 and 430 nm,
respectively) in MeOH, where the protonated 2-nitrophenol form
gave a particularly large red shift. This suggests that the change
(viz., red shift) in the absorption maxima (λp and λd) is a good
measure of the hydrophobic environment around 2-nitrophenol
group. The absorption maxima of 8 in MeOH-CHCl3 appeared
at 383 nm (λp) for the protonated form and at 433 nm (λd) for
the deprotonated one, which were close to those of 9 in MeOH
(379 and 430 nm, respectively). In CHCl3, λp and λd of 8
showed a further red shift up to 390 and 446 nm, respectively.
In the liposomes, 8 gave λp at 384 nm and λd at 449 nm, which
were very close to those in CHCl3. This obviously indicates
that 8 should be located in a lipid bilayer membrane whose
hydrophobic strength is comparable to that of CHCl3. Since there
seems to be a good relation between the red shift and the
hydrophobic strengths, we measured absorption spectra of the
synthetic lipids 6a-c in a similar manner (Table 2). Though
6a-c in liposomes all exhibited appreciable red shift as
compared to 9 in an aqueous solution, the behavior of 6a is
clearly different from those of 6b and 6c. In fact, 6a shows a
small red shift (374 nm) in λp and a large red shift (448 nm) in
λd, while 6b and 6c show large red shifts (389 and 396 nm) in
λp and small red shifts (431 and 436 nm) in λd.
Figure 2. Time course of lipid mixing (%) of Control-lip and S10% Fusion-
lip including 6c (pH 7.4, 37 °C).
indicates that the synthetic lipids bearing 2-nitrophenol groups
are surely incorporated into eggPC liposomes.
pKa of the 2-Nitrophenol Group. There are some methods
to estimate pKa values of compounds embedded in mem-
branes.14 Here we investigated the properties of synthetic lipids
in the liposome membrane by measuring pKa of the 2-nitro-
phenol group. Compound 9 was derived from glyceric acid as
a water-soluble standard derivative. In addition, lipid 8 was
synthesized by coupling with palmitic acid as a hydrophobic
2-nitrophenolate probe with a single alkyl chain to compare the
hydrophobic strengths of organic solvents and lipid bilayer
membranes. The pKa value of the 2-nitrophenol group of water-
soluble derivative 9 was determined by a pH-metric titration in
an aqueous solution as 6.78. In synthetic lipids 6a-c and 8,
the pH-metric titration was performed by addition of citric acid
into the liposome solution containing 10 mol % of each synthetic
lipid (S10% Fusion-lip). The pH-titration was monitored by the
absorbance of the 2-nitrophenolate moiety at 20°C (e.g., see
NMR Spectra of Liposome. The liposome solutions contain-
ing synthetic lipids were prepared in deuterium oxide (D2O),
and then the pD was adjusted to around 10. The chemical shifts
of aromatic protons (Ha, Hb, Hc) in 6c were compared under
both basic and acidic conditions, namely, for deprotonated and
protonated states of the nitrophenol group (Table 3). Whereas
aromatic protons of 6a, 6b, 6c, 8, and 9 in protonated form
showed their signals at comparable positions in 10% CD3OD-
CDCl3, the corresponding signals of lipid 6c obviously moved
(15) Epand, R. F.; Kraayenhof, R.; Sterk, G. J.; Wong, F.; Sang, H. W.; Epand,
(14) Bailey, A. L.; Cullis, P. R. Biochemistry 1997, 33, 112573.
R. M. Biochim. Biophys. Acta 1996, 1284, 191.
9
J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003 14735

A R T I C L E S
Tomohiro et al.
Table 2. Electronic Absorption Maxima of Lipids 6a-c, 8, and 9
absorption maxima
a
b
compound
condition
λp, nm
λd, nm
6a
50% MeOH-CHCl3
S10% liposome
S10% liposome
S10% liposome
10% MeOH-CHCl3
CHCl3
S10% liposome
H2O
MeOH
384
374
389
396
383
390
384
367
379
368
433
448
432
438
433
446
449
427
430
428
6b
6c
8
9
10 mM Hepes
a The absorption maxima of protonated species in each condition at 20
°C. b The values of deprotonated species.
Table 3. Chemical Shifts (ppm) of Aromatic Protons in
Compounds 6a-c, 8, and 9
Figure 3. Time-resolved electronic absorption spectra of lipid mixing
solution of Control-lip and S10% Fusion-lip including 6b (pH 4.0, 20 °C).
contained 2 mol % each of fluorescent lipids. To a Hepes buffer
with appropriately adjusted pH at 37 °C in a quartz cell was
added an eggPC liposome solution (Control-lip) followed by
addition of Label-lip, then the lipid mixing was initiated by the
addition of Fusion-lip (eggPC liposomes containing certain
amount of synthetic lipids). Figure 2 showed the lipid mixing
of Control-lip and S10% Fusion-lip including 6c at pH 7.4. After
Control-lip and Label-lip were mixed, fluorescent intensity at
530 nm was not changed during 5 min before Fusion-lip was
added. All runs were done in duplicate or more. It was confirmed
in a control experiment, where Control-lip was added instead
of Fusion-lip, that the lipid mixing did not occur during the
whole measurement.
CDCl3-CD OD
3
solvent
compound
CDCl3
8
D O
9
2
6a
50
8.32
7.84
7.10
6b
50
6c
8
9
temp, °C
Ha
Hb
50
rt
rt
rt
rt
8.45
7.87
7.14
8.43
7.81
7.11
8.41
7.79
7.10
8.52
7.81
7.15
8.29
7.77
7.13
8.18
7.60
7.11
Hc
liposomea
DMF-d7
CDCl3−CD OD
3
6c
protonated
deprotonated
protonated
protonated
deprotonated
Ha
Hb
Hc
8.16
7.59
7.08
7.85
7.21
6.66
8.43
7.81
7.11
8.73
8.00
7.20
7.75
7.60
7.30
Time-Resolved Electronic Absorption Spectra. The initial
fusion process was detected by a multichannel spectrophotom-
eter. Absorbance at 432 nm due to 2-nitrophenolate anion was
immediately decreased within 100 ms after the addition of citric
acid to a stirring Fusion-lip (10 mol % of 6b) solution and an
absorption band due to the protonated species emerges at 390
nm (Figure 3). After several minutes, the absorption was
gradually increased throughout the range from 350 to 600 nm,
and the solution became turbid. This suggests that the fusion
or the phase transition of lipid membranes takes place after the
rapid protonation of 2-nitrophenolate anion. Kinetic analysis of
the initial decrease of absorbance at 432 nm gave the apparent
a S10% Fusion-lip was prepared in deuterized 10 mM Hepes buffer
followed by pD adjustment to around 10 with NaOD, and the solution was
acidified with deuterized citric acid.
to higher magnetic field in liposome. These signals of 6c are
also situated at higher magnetic field compared with that in
DMSO-d6. In lipid 9, on the other hand, the chemical shifts
moved to higher magnetic field when the solvent was changed
from 10% CD3OD-CDCl3 to D2O. These two tendencies are
not consistent, since the hydrophobicity of the liposome
membrane should be as large as that of CHCl3 in view of the
results of the absorption spectra in Table 2. The NMR shift to
higher magnetic field on deprotonation of 6c in liposome and
in DMF-d7 is reasonably understood, because electron density
on the hydrogen atoms of 2-nitrophenolate should be increased
by deprotonation. Contrary to 6c, 6a, and 6b did not give NMR
spectra in liposome that are suitable for quantitative analysis
of chemical shift.
Lipid Mixing. The lipid mixing assay was performed by
monitoring FRET (fluorescence resonance energy transfer)
between two fluorescent labeled lipids, N-(7-nitro-2,1,3-ben-
zoxadiazol-4-yl)-1,2-dipalmitoyl-sn-phosphatidylethanolamine
(NBD-PE) and N-(lissamine rhodamine B sulfonyl)-1,2-dipalmi-
toyl-sn-phosphatidylethanolamine (Rh-PE), using a method of
Struck et al.16,17 Fluorescent labeled liposome (Label-lip)
first-order rate constant of protonation to be 5.2 s-1
.
Discussion
Perturbation of membranes is known to be induced by Ca2+ 18
,
fusion peptides, and proteins such as hemagglutinin and Bax,
as well as by cooperation of these proteins and some lipid such
as cardiolipin.19 In the previous studies, we found that the
balance of hydrophilic and hydrophobic character of trigger
chains is important for liposome membrane fusion.9 To initiate
lipid mixing, the trigger chain should penetrate into the
membrane or disturb the surface. However, there is a hydro-
phobic barrier of the phospholipid bilayer membrane.20 Actually,
(18) Sundler, R.; Papahadjopoulos, D. Biochim. Biophys. Acta 1981, 649, 743.
(19) (a) Epand, R. F.; Martinou, J.-C.; Montessuit, S.; Epand, R. M. Biochem.
J. 2002, 367, 849. (b) Kuwana, T.; Mackey, M. R.; Perkins, G.; Ellisman,
M. H.; Latterich, M.; Schneiter, R.; Green, D. R.; Newmeyer, D. D. Cell
2002, 111, 331.
(16) Struck, D. K.; Hoekstra, D.; Pagano, R. E. Biochemistry 1981, 20, 4093.
(17) Hoekstra, D.; de Boer, T.; Klppe, K.; Wilschut, J. Biochemistry 1984, 36,
5675.
9
14736 J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003

Spectroscopic Analysis of Liposome Fusion
A R T I C L E S
Figure 4. Dependence of lipid mixing (%) on content of synthetic lipids (a) and on pH (b).
The efficiency of lipid mixing in this series (6a-c) was lower
compared with that using fusion peptide or proteins10 but
apparently increased with the amount of the synthetic lipids from
5 to 20 mol % (Figure 4a). The data of lipid mixing used here
were the values 20 min after the addition of Fusion-lip in the
lipid mixing experiment. There was no large difference in the
efficiency between lipids 6b and 6c; viz., the lipid mixing was
around 5-8%, which is rather small in comparison with those
of 6a (ca. 10%)10 and 7 (9-13% on using S10% Fusion-lip
including 7).9 The efficiency of lipid mixing also depends on
the solution pH; viz., the efficiency is increased on protonation
of a 2-nitrophenol group (Figure 4b). Molecular shape of these
synthetic lipids might affect the stability of the liposome
membrane. The result that the shape of the synthetic lipids in
minimum-energy states calculated by MM3 (CAChe system)
was not very different from that of eggPC suggests that on
protonation the hydrophobic trigger chain induces membrane
perturbation leading to lipid mixing between liposomes.
Figure 5. Typical parallel arrangements between two 2-nitrophenol
derivative molecules.
Absorption spectra give useful information on the microscopic
environments around the deprotonated and protonated 2-nitro-
phenol groups. Table 2 shows the relation between λd and λp,
in which the both wavelengths on the whole show red shift as
the environments around the 2-nitrophenol become more
hydrophobic. It should be noted, however, that the red shift of
λp for 8, 9, 6b, and 6c situated in MeOH, MeOH-CHCl3 or
liposome is as a whole larger (∆λ ) 12-29 nm) than that of
∆λ (∆λ ) 3-22 nm). Here ∆λ is defined as the increase in the
wavelength (red shift) from that of 9 in an aqueous solution.
Red shift is also observed when the individual lipid exists in
different solvents. For example, the red shift of λd in 8 is 13
nm between CHCl3 (446 nm) and MeOH (433 nm) and 16 nm
between liposome in Hepes buffer (449 nm) and MeOH (433
nm). These comparable values clearly demonstrate that the
hydrophobic fields surrounding 2-nitrophenolate anions of 8
resemble each other in liposome and CHCl3. Another example
is that 9 exhibits the red shift of λp by 12 nm when the solvent
is changed from H2O (367 nm) to MeOH (379 nm). In 6a, a
similar phenomenon was observed in two lipid conditions; that
is, λd is 448 nm for liposome, which is close to λd of 8 in
liposome (449 nm), and 433 nm for MeOH-CHCl3. This
suggests that the trigger of 6a exists in a liposome membrane
even when the 2-nitrophenol headgroup takes an anionic form.
In contrast to 6a, 6b and 6c in liposomes gave λd (432 and 438
Figure 6. Dependence of calculated wavelength on angle θ.
efficiency of lipid mixing largely depends on the length of the
trigger chain, n-alkylcarboxylate (7, 2 e n e 12, as described
in Scheme 1); that is, the short trigger (n ) 2, 4) chain has
only a little effect owing to the poor hydrophobicity. In this
report, 2-nitrophenolate was chosen as a pH-sensitive and
chromophoric group that is more hydrophobic than carboxylate.
(20) Wisniewska, A.; Subczynski, W. K. Biochim. Biophys. Acta 1998, 1368,
235.
9
J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003 14737

A R T I C L E S
Tomohiro et al.
nm) comparable to 9 in an aqueous solution (427 nm), possibly
because the triggers of 6b and 6c exist out of the liposome
membrane. The fact that the pKa values of lipid 6a-c and 8
(>8.55) in liposome solutions were much larger than that of 9
(6.78) definitely proves that the former lipids in the protonated
state are located in a more hydrophobic field than the latter,
because electrostatic attraction is stronger and thus deprotonation
is more depressed in higher hydrophobic surroundings. The
anionic chromophore can probably form hydrogen bonds with
MeOH in 6a (50% MeOH-CHCl3) and 8 (10% MeOH-
CHCl3), which makes λd smaller than expected (∆λ ) 6 nm).
Actually, in CHCl3 8 gives large red shifts of λd (∆λ ) 19 nm)
and λp (∆λ ) 23 nm). Also, in liposome the 2-nitrophenol group
of 8 is probably buried inside the bilayer membrane and thus
hydrogen bonding with water is thought to be prohibited.
Contrary to all these compounds, 6a in liposome shows
obviously unusual behavior in that the red shift of λp (∆λ ) 7
nm) is too small compared with that of λd (∆λ ) 21 nm).
Though the large red shift of λd (∆λ ) 21 nm) in 6a indicates
that the anionic chromophore is situated in the hydrophobic
region, the small red shift of λp (∆λ ) 7 nm) in 6a cannot be
explained by small hydrophobicity of the fields, since the
protonated form should be more stable in hydrophobic environ-
ments than the deprotonated form. Therefore, one has to find
another explanation for the above unusual result. Since there is
a possibility that 6a forms aggregates in bilayer membranes,
we evaluated the effect of association on the shift of λp with a
molecular orbital calculation method (INDO/1 in ZINDO of
CAChe system). We calculated λp for typical face-to-face
associations between two 2-nitro-4-acetamidophenol molecules
(Figure 5). The angle (θ) between the two nitro groups is
changed from 0 deg to 180 deg by 60 deg, where there are two
arrangements for θ ) 60 deg. The distance between the two
benzene rings is fixed to 3.7 Å. Dependence of the calculated
λp is illustrated in Figure 6, which proves that λp shows blue
shift from the wavelength of the monomer when the aggregate
is formed and that the blue shift is especially large when θ is
near zero. These results might account for the observation that
6a in liposome gives λp smaller than 6b and 6c in liposome.
It is known that interaction between aromatic lipids such as
azobenzene- or pyren-modified lipids in ordinal aliphatic lipid
membrane results in phase separation.21 However, since elec-
trostatic repulsion is stronger than the aromatic attractive
interaction, aromatic lipids are separated from each other when
they have charge with the same sign.21a There is a possibility
that the present synthetic lipids bearing 2-nitrophenol groups
should show a similar phenomenon. Actually, the absorption
spectra and the 1H NMR spectra, respectively, showed that 6a
and 6c in the protonated state might associate in the membrane
through the aromatic π-π interaction. This association can
change their electronic states, resulting in the blue shift of the
absorption spectra and the NMR shift toward higher magnetic
field due to diamagnetic shielding of each aromatic ring. The
reason 6c does not have so small a λp as 6a is not clear. One
of the possible reasons might be that two 6c molecules take an
arrangement near θ ) 120 deg as shown Figures 5 and 6, while
6a has a possibility to take an arrangement near θ ) 0 deg.
The 2-nitrophenolate trigger chain of 6c should penetrate into
a membrane, and thus 6c becomes a so-called tridentate-type
lipid, which can interact with each other to generate a micro-
domain. This might destabilize membranes and induce mem-
brane fusion via lipidic particle formation (inverted micelles)
in a joining part of liposomes, such as the cases of phosphati-
dylserine (PS)-PC22 or cardiolipin with a calcium metal ion.23
Conclusions
Liposome membrane fusion was investigated using the
chromophoric synthetic lipids having 2-nitrophenol derivatives
as trigger chains. The lipid mixing assay applying FRET showed
that the liposome fusion proceeded in the pH region lower than
the pKa of 2-nitrophenol groups. The pKa values of the
chromophoric lipids in a lipid bilayer membrane were larger
than 8.5, which were quite different from that of 9 (6.78) in an
aqueous solution. The protonation of 2-nitrophenolate anions
in membranes was not rapid, and membrane fusion gradually
took place after protonation. Electronic absorption of these
synthetic lipids indicated different environments around the
chromophoric groups, depending on the hydrophobicity of the
trigger chains. The deprotonated hydrophilic trigger chains could
be out of the membrane and then penetrate into the membrane
of another liposome after protonation. The higher hydrophobic
trigger chains, however, could not remain outside the membrane
even in the deprotonated form. The synthetic lipids having three
chains (6a-c) led to membrane fusion, but the one having a
single chain (8) did not induce membrane fusion. In addition,
membrane fusion was not clearly observed for the Fusion-lip
including 6a in the deprotonated form. These data suggested
that the synthetic lipid might form a tripodal lipid and they might
interact each other after penetration of the protonated trigger
chains into the membrane.
Experimental Section
Materials. Egg phosphatidylcholine (PC) was obtained from Kawa-
hara Petrochemical Co., Ltd. N-(7-Nitro-2,1,3-benzoxadiazol-4-yl)-1,2-
dipalmitoyl-sn-phosphatidylethanolamine (NBD-PE) and N-(lissamine
rhodamine B sulfonyl)-1,2-dipalmitoyl-sn-phosphatidylethanolamine
(Rh-PE) were commercially available from Molecular Probes (Eugene,
OR). N-tert-Butyloxycarbonyl-L-aspartic acid (Boc-Asp), N-9-fluore-
nylmethoxycarbonyl-L-aspartic acid R-tert-butyl ester (Fmoc-Asp-
(RBut)), glycylglycine, 1-hydroxybenzotriazole hydrate (HOBt‚H2O),
and 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium tetrafluorobo-
rate (TBTU) were purchased from Watanabe Chemical Industries, Ltd.
(Hiroshima, Japan). 1-Hexadecylamine, 4-amino-2-nitrophenol, and 3,6-
dioxaoctanedioic acid were obtained from Aldrich Chemical Co., Inc.
(St. Louis, MO). 4-Dimethylaminopyridine (DMAP), 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide monohydrochloride (WSCI), seba-
cic acid, and DL-tartaric acid were purchased from Wako Pure Chemical
Industries, Ltd. (Osaka, Japan).
Instruments. Fluorescent studies were performed by a JASCO FP-
777 spectrofluorometer. Stationary electronic absorption spectra and
time-resolved electronic absorption spectra were, respectively, measured
by JASCO V-550 UV/ Vis and OTSUKA ELECTRONICS MCPD-
1000 spectrophotometers. IR spectra were measured on a JASCO FT/
IR-7300 spectrometer, and NMR spectra were recorded at 270 MHz
on a JEOL JNM EX-270 FT-NMR spectrometer.
(21) (a) Singh, A.; Tsao, Li-I.; Markowitz, M.; Gaber, B. P. Langmuir 1992,
8, 1570. (b) Kawasaki, T.; Tokuhiro, M.; Kimizuka, N.; Kunitake, T. J.
Am. Chem. Soc. 2001 123, 6792. (c) Sasaki, D. Y.; Shnek, D. R.; Pack, D.
W.; Arnold, F. H. Angew. Chem., Int. Ed. Engl. 1995, 34, 905.
(22) Ohnishi, S. AdV. Biophys. 1976, 8, 35.
(23) De Kruijff, B.; Verkleij, A. J.; Leunissen-Bijvelt, J.; Van Echteld, C. J.;
Hille, J.; Rijnbout, H. Biochim. Biophys. Acta 1982, 693, 1.
9
14738 J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003

Spectroscopic Analysis of Liposome Fusion
A R T I C L E S
Synthesis of Lipids. The synthetic route of the lipids bearing a
2-nitrophenol group derived from N-(tert-butoxycarbonyl)-L-aspartic
acid (1) via 2-amino-N-(1,2-bis-hexadecylcarbamoylethyl)succinamic
acid tert-butyl ester (2) is shown in Scheme 1.
eluted with 2-5% CH3CN-H2O. The yellow solid was washed with
20% THF-CH2Cl2 and dried under vacuum. Yield: 55%. H NMR
(DMSO-d6) δ 9.88 (s, 1 H), 8.48 (d, 1 H, J ) 2.6 Hz), 8.34 (dd, 1 H,
J ) 2.6, 8.9 Hz), 7.09 (d, 1 H, J ) 8.9 Hz), 4.26 (s, 1 H), 4.15 (s, 1
H), 3.82 (s, 2 H).
1
2-[2-(2-Aminoacetylamino)acetylamino]-N-(1,2-bishexadecylcar-
bamoylethyl)succinamic acid butyl ester (3). FmocGlyGly (57 mg,
0.16 mmol) was dissolved in 1 mL of DMF followed by addition of
20 mL CHCl3 solution of 2 (100 mg, 0.13 mmol) and HOBt‚H2O (25
mg, 0.16 mmol) at room temperature. A solution of TBTU in 5 mL of
DMF was added to the reaction mixture at room temperature and was
stirred overnight under nitrogen. The product was purified by column
chromatography on silica gel eluted by 10% MeOH-CHCl3. Yield:
5a. Compounds 2 (0.25 g, 0.33 mmol) and 4a (0.11 g, 0.32 mmol)
were dissolved in 25 mL of CHCl3. HOBt‚H2O (0.05 g, 0.33 mmol in
0.9 mL of DMF and 9.1 mL of CHCl3) and DMAP (0.04 g, 0.33 mmol
in 5 mL of CHCl3) were added to the reaction mixture followed by
addition of TBTU (0.11 g, 0.33 mmol in 0.9 mL of DMF and 9.1 mL
of CHCl3). The reaction mixture was stirred overnight at room
temperature. After washing with 1 N HCl 3 times and brine, the solution
was dried over Na2SO4 and evaporated in vacuo. Purification was
performed by column chromatography on silica gel eluted with 3%
1
42%. H NMR (CDCl3+ CD3OD) δ 7.77 (d, 2 H, J ) 7.6 Hz), 7.63
(d, 2 H, J ) 7.9 Hz), 7.38 (t, 2 H, J ) 7.9 Hz), 7.31 (t, 2 H, J ) 7.6
Hz), 4.6-4.7 (m, 2 H, CH), 4.43 (d, 2 H, J ) 6.6 Hz), 4.23 (t, 1 H, J
) 7.9 Hz), 3.93 (s, 2 H), 3.87 (s, 2 H), 3.1-3.3 (m, 4 H), 2.4-2.8 (m,
4 H), 1.47 (s br, 4 H), 1.44 (s, 9 H), 1.26 (s br, 52 H), 0.88 (t, 6 H, J
) 6.6 Hz). The product (25 mg) was treated with 5% piperidine-
CHCl3 (5 mL) at room temperature overnight, and the solvent was
removed. The residue was washed with EtOAc, and 3 was obtained in
MeOH-CHCl3. Yield: 77%. IR (KBr): 3285, 1740, 1648, 1537 cm-1
.
1H NMR (CDCl3) δ 10.4 (s, 1 H), 8.38 (s, 1 H), 8.32 (d, 1 H, J ) 2.3
Hz), 7.96 (dd, 1 H, J ) 2.6 and 2.6 Hz), 7.51 (d, 1 H, J ) 7.3 Hz),
7.42 (t, 1 H, J ) 5.8 Hz), 7.11 (d, 1 H, J ) 9.2 Hz), 6.59 (d, 1 H, J
) 8.3 Hz), 5.99 (t, 1 H, J ) 5.7 Hz), 4.79 (ddd, 1 H, J ) 6.3, 6.6, 6.6
Hz), 4.66 (ddd, 1 H, J ) 3.6, 7.2 Hz), 3.21 (m, 4 H), 1.46 (s, 9 H),
1.25 (br s, 60 H), 0.87 (t, 6 H, J ) 6.6 Hz). Anal. Calcd for
C60H106N6O10: C, 67.25; H, 9.97; N, 7.84. Found: C, 67.36; H, 10.20;
N, 7.63.
1
65% yield. H NMR (CDCl3+ CD3OD) δ 10.41 (s, 1 H), 8.97 (s, 1
H), 8.42 (d, 1 H, J ) 2.6 Hz), 7.96 (dd, 1 H, J ) 2.6, 9.2 Hz), 7.85 (d,
1 H, J ) 8.2 Hz), 7.35-7.45 (m, 2 H), 7.12 (d, 1 H, J ) 9.2 Hz), 6.00
(t, 1 H, J ) 5.6 Hz), 4.81 (ddd, 1 H, J ) 5.3, 5.6, 8.2 Hz), 4.59 (ddd,
1 H, J ) 2.6, 7.3, 7.3 Hz), 4.22 (d, 1 H, J ) 15.7 Hz), 4.21 (d, 1 H,
J ) 15.7 Hz), 4.09 (d, 1 H, J ) 16.0 Hz), 4.06 (d, 1 H, J ) 16.0 Hz),
3.81 (s, 4 H), 3.1-3.3 (m, 4 H), 2.7-2.9 (m, 3 H), 2.38 (dd, 1 H, J )
7.3, 15.3 Hz), 1.47 (s br, 4 H), 1.40 (s, 9 H), 1.25 (s br, 52 H), 0.88 (t,
6 H, J ) 6.6 Hz).
9-(4-Hydroxy-3-nitrophenylcarbamoyl)nonanoic Acid (4a). To a
solution of sebacic acid (1.0 g, 5.0 mmol in 3 mL DMF) and 4-amino-
2-nitrophenol (0.765 g, 5.0 mmol in 87 mL CHCl3) was added WSC
(1.0 g, 5.0 mmol in 10 mL of CHCl3), and the mixture was stirred at
room temperature overnight. The reaction mixture was washed with
50 mL of dilute HCl twice and brine and was then dried over MgSO4.
Purification of the product was performed by column chromatography
on silica gel eluted with 50% EtOAc-hexane, followed by recrystal-
lization from the same solvent. Yield: 27%. IR (KBr) 3305, 1697,
1667, 1542 cm-1. 1H NMR (CDCl3) δ 8.34 (d, 1 H, J ) 2.3 Hz), 7.82
(dd, 1 H, J ) 2.6, 8.9 Hz), 7.11 (d, 1 H, J ) 9.2 Hz), 2.37-2.26 (m,
4 H), 1.70 (t, 2 H, J ) 6.9 Hz), 1.61 (t, 2 H, J ) 6.9 Hz), 1.33 (br s,
8 H). Anal. Calcd for C16H22N2O6: C, 56.80; H, 6.55; N, 8.28. Found:
C, 57.31; H, 6.22; N, 8.08.
5b. A solution of TBTU (128 mg, 0.40 mmol) in 25 mL of DMF
was added to the mixture of 2 (0.2 g, 0.27 mmol), 4b (84 mg, 0.27
mmol), and HOBt‚H2O (61 mg, 0.40 mmol) in 50 mL of CHCl3 at
room temperature, and the reaction mixture was stirred overnight under
nitrogen. After removal of the solvent, the residue was dissolved in
hot EtOAc and CHCl3 and washed with 3 N HCl and brine and then
dried over MgSO4. The solvent was evaporated in vacuo, and yellow
solid was filtrated and washed with MeOH, EtOAc, and ether. Yield:
1
0.159 g (57%). H NMR (CDCl3) δ 10.41 (s, 1 H, COOH), 8.97 (s, 1
H), 8.42 (d, 1 H, J ) 2.6 Hz), 7.96 (dd, 1 H, J ) 2.6, 9.2 Hz), 7.85 (d,
1 H, J ) 8.2 Hz), 7.35-7.45 (m, 2 H), 7.12 (d, 1 H, J ) 9.2 Hz), 6.00
(t, 1 H, J ) 5.6 Hz), 4.81 (ddd, 1 H, J ) 5.3, 5.6, 8.2 Hz), 4.59 (ddd,
1 H, J ) 2.6, 7.3, 7.3 Hz), 4.22 (d, 1 H, J ) 15.7 Hz), 4.21 (d, 1 H,
J ) 15.7 Hz), 4.09 (d, 1 H, J ) 16.0 Hz), 4.06 (d, 1 H, J ) 16.0 Hz),
3.81 (s, 4 H), 3.1-3.3 (m, 4 H), 2.7-2.9 (m, 3 H), 2.38 (dd, 1 H, J )
7.3, 15.3 Hz), 1.47 (s br, 4 H), 1.40 (s, 9 H), 1.25 (s br, 52 H), 0.88 (t,
6 H, J ) 6.6 Hz). Anal. Calcd for C58H102N6O12: C, 64.77; H, 9.56;
N, 7.81. Found: C, 64.58; H, 9.77; N, 7.69.
5c. Compounds 3 (100 mg, 0.12 mmol), 4c (33 mg, 0.12 mmol),
and HOBt‚H2O (21 mg, 0.14 mmol) were dissolved in 15 mL of DMF
and 5 mL of CHCl3. TBTU (45 mg, 0.14 mmol) was added to the
solution followed by stirring at room temperature under nitrogen
overnight. The solvent was removed in vacuo, and the yellow precipitate
was washed with MeOH, EtOAc, CH2Cl2, and 10% MeOH-CH2Cl2
3-{2-[2-(4-Hydroxy-3-nitrophenylcarbamoyl)ethoxy]ethoxy}-
propionic Acid (4b). WSC (0.90 g, 5.0 mmol) was added to a solution
of 3,6-dioxaoctanedioic acid (0.99 g, 5.0 mmol) and HOBt‚H2O (0.77
g, 5.0 mmol) in 50 mL of DMF followed by addition of 4-amino-2-
nitrophenol (0.77 g, 5.0 mmol), and the reaction mixture was stirred
overnight at room temperature under nitrogen. After removal of the
solvent, EtOAc was added to the residue. The products were extracted
with 5% aqueous NaOH solution, which was washed with EtOAc twice.
The aqueous phase was acidified with concentrated HCl and extracted
with EtOAc 3 times. The organic phase was washed with water and
brine and was dried over MgSO4. The crude product was purified by
short ODS column chromatography eluted with 20% MeOH-H2O after
1
to give 58 mg (44%) of the product. H NMR (CDCl3) δ 10.38 (s, 1
H), 8.41 (s, 1 H), 8.32 (d, 1 H, J ) 2.6 Hz), 7.99 (dd, 1 H, J ) 2.6, 9.0
Hz), 7.52 (d, 1 H, 7.3 Hz), 7.44 (t, 1 H, 5.6 Hz), 7.12 (d, 1 H, J ) 9.0
Hz), 6.55 (d, 1 H, J ) 8.1 Hz), 5.88 (s, 1 H), 4.82 (ddd, 1 H, J ) 4.7,
6.8, 8.1 Hz), 4.68 (ddd, 1 H, J ) 3.4, 6.4, 7.3 Hz), 3.13-3.30 (m, 4
H), 2.88 (dd, 1 H, J ) 3.4, 15.2 Hz), 2.80 (dd, 1 H, J ) 4.7, 14.8 Hz),
2.66 (dd, 1 H, J ) 6.8, 14.8 Hz), 2.45 (dd, 1 H, J ) 6.4, 15.2 Hz),
2.29-2.37 (m, 2 H), 2.15-2.29 (m, 2 H), 1.68-1.75 (m, 2 H), 1.54-
1.67 (m, 4 H), 1.44-1.54 (m, 13 H), 1.25 (s br, 52 H), 0.88 (t, 6 H, J
) 7.1 Hz).
1
1% HCl. Yield: 0.50 g (32%). H NMR (CDCl3) δ 10.41 (s, 1 H),
8.89 (S, 1 H), 8.32 (d, 1 H, J ) 2.6 Hz), 7.96 (dd, 1 H, J ) 2.6, 9.2
Hz), 7.13 (d, 1 H, J ) 9.2 Hz), 4.26 (s, 2 H), 4.15 (s, 2 H), 3.82 (s, 4
H).
6a. Compound 5a (74 mg, 0.069 mmol) was dissolved in 8 mL of
30% TFA-CHCl3 and stirred overnight at room temperature. After
the solvent was evaporated, the residue was suspended in MeOH and
the precipitate was collected by filtration and washed with MeOH, dried
under vacuum for overnight. Yield: 87%. IR (KBr): 3284 (OH), 1717,
2,3-Dihydroxy-N-(4-hydroxy-3-nitrophenyl)succinamic Acid (4c).
DL-Tartaric acid (2.25 g, 15.0 mmol) and 4-amino-2-nitrophenol (0.77
g, 5.0 mmol) were dissolved in 50 mL of DMF. HOBt‚H2O (0.77 g,
5.0 mmol) was added to the solution at 0 °C under nitrogen followed
by addition of WSC (0.90 g, 5.0 mmol). The reaction mixture was
stirred at room temperature under nitrogen overnight. After removal
of the solvent, the residue was purified by ODS column chromatography
1
1647, 1542 cm-1. H NMR (CDCl3 + CD3OD at 50 °C) δ 8.32 (d, 1
H, J ) 2.5 Hz), 7.84 (dd, 1 H, J ) 9.0, 2.5 Hz), 7.10 (d, 1 H, J ) 9.0
Hz), 4.76 (t, 1 H, J ) 5.6 Hz, NHCHCH2CO), 4.63 (dd, 1 H, J ) 5.6,
6.7 Hz), 3.17-3.22 (m, 4 H), 2.78 (dd, 1 H, J ) 5.6, 15.1 Hz), 2.77
9
J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003 14739

A R T I C L E S
Tomohiro et al.
(dd, 1 H, J ) 5.6, 15.1 Hz), 2.67 (dd, 1 H, J ) 5.1, 14.9 Hz), 2.49 (dd,
1 H, J ) 6.7, 14.9 Hz), 2.34 (t, 2 H, J ) 7.6 Hz), 2.22 (t, 2 H, J ) 7.6
Hz), 1.40-1.54 (m, 4 H), 1.26 (s br, 60 H), 0.88 ppm (t, 6 H, J ) 6.6
Hz). Anal. Calcd for C56H98N6O10: C, 66.24; H, 9.73; N, 8.28. Found:
C, 66.34; H, 9.72; N, 8.10.
Preparation of Liposomes. The lipids were dissolved in MeOH-
CHCl3 at the desired molar ratio. Small amount of DMF or DMSO
was used to dissolve 6b and 6c. The solvent was evaporated by nitrogen
gas stream and completely dried under vacuum pressure to deposit the
lipid on the wall of glass test tube as a thin film. One milliliter of
Hepes buffer (10 mM Hepes, 0.1 M NaCl, pH 9) was then added and
vortexed. The resulting suspension of multilameller vesicles (MLVs)
was sonicated for 5-10 min with a probe-type sonicator (Branson
Sonifier 250) until the solution became clear with luster. The liposome
solutions were directly used for the experiments. The liposome solutions
containing the synthetic lipids (Fusion-lip) were freshly prepared before
use and kept at 0 °C, whose pH was adjusted to around 10 with aqueous
NaOH.
pH-Metric Titration of Chromophoric Lipids. After preparing
S10% Fusion-lip, 2 mL of the Hepes buffer was added and the pH of
the solution was adjusted to around 11 with aqueous NaOH in a quartz
cell at 20 °C. To the stirred liposome solution, 0.5 M HCl was added
stepwise with a syringe, and the pH and electronic absorption due to
each nitrophenolate anion were measured at 20 °C. In the case of 8, it
was directly dissolved in water and the solution was acidified with
concentrated HCl, and 1N NaOH was then added into the stirring
solution followed by measurement of pH at 20 °C.
6b. Compound 5b (50 mg, 0.048 mmol) was dissolved in 20 mL of
30% TFA-CHCl3, and the reaction mixture was stirred overnight at
room temperature. After the solvent was evaporated, the residue was
suspended in MeOH, and the precipitate was collected by filtration,
washed with MeOH and CHCl3, and dried under vacuum for overnight.
1
Yield: 71%. H NMR (CDCl3 + CD3OD at 50 °C) δ 8.45 (d, 1 H, J
) 2.6 Hz), 7.87 (dd, 1 H, J ) 2.6, 9.2 Hz), 7.14 (d, 1 H, J ) 9.2 Hz),
4.82 (dd, 1 H, J ) 5.0, 5.3 Hz), 4.60 (dd, 1 H, J ) 5.6, 6.6 Hz), 4.18
(s, 2 H), 4.08 (s, 2 H), 3.75-3.85 (m, 4 H), 3.05-3.20 (m, 4 H), 2.85
(dd, 1 H, J ) 5.0, 15.6 Hz), 2.83 (dd, 1 H, J ) 5.3, 15.6 Hz), 2.61 (dd,
1 H, J ) 5.6, 14.9 Hz), 2.48 (dd, 1 H, J ) 6.6, 14.9 Hz), 1.46 (s br,
4 H), 1.27 (s br, 52 H), 0.89 (t, 6 H, J ) 6.6 Hz).
6c. Compound 5c (20 mg, 18 mmol) was dissolved in 6 mL of CHCl3
and 2 mL of TFA, and the reaction mixture was stirred at room
temperature overnight. The solvent and TFA were evaporated. To the
residue was added MeOH, and the yellow precipitate was filtrated and
washed with MeOH, EtOAc, CHCl3, and 10% MeOH-CHCl3 to obtain
18 mg (95%) of the yellow solid. 1H NMR (CDCl3 + CD3OD) δ 8.43
(d, 1 H, J ) 2.6 Hz), 7.81 (dd, 1 H, J ) 2.6, 9.0 Hz), 7.11 (d, 1 H, J
) 9.0 Hz), 4.77 (t, 1 H, J ) 6.0 Hz), 4.63 (dd, 1 H, J ) 6.0, 6.2 Hz),
3.09-3.20 (m, 4 H), 2.78 (dd, 1 H, J ) 6.0, 15.4 Hz), 2.76 (dd, 1 H,
J ) 6.0, 15.4 Hz), 2.64 (dd, 1 H, J ) 6.0, 15.0 Hz), 2.51 (dd, 1 H, J
) 6.2, 15.0 Hz), 2.34 (t, 2 H, 7.7 Hz), 2.20-2.26 (m, 2 H), 1.66-1.73
(m, 2 H), 1.58-1.65 (m, 2 H), 1.47 (s br, 4 H), 1.26 (s br, 52 H), 0.88
(t, 6 H, J ) 6.8 Hz).
Measurement of NMR Spectra. A dried lipid film of eggPC and
synthetic lipid was suspended in a Hepes buffer prepared using
deuterium oxide. After sonication of the solution to prepare the
liposome, the solution was directly used for the NMR measurement.
The solution was adjusted at pD around 10 with an aqueous NaOD
solution and transferred to a NMR tube. The measurement was
performed under basic and acidic conditions at 20 °C.
Lipid Mixing Assay. Label-lip containing both NBD-PE and Rh-
PE was freshly prepared before use. Hepes buffer (1.89 mL, 10 mM)
containing 0.1 M NaCl was adjusted at appropriate pH with aqueous
NaOH or citric acid in a quartz cell at 37 °C. To the solution was
added 90 µL of Control-lip solution followed by addition of 10 µL of
Label-lip solution; then lipid mixing was initiated by the addition of
10 µL of Fusion-lip solution containing certain amount of synthetic
lipids. The time course of lipid mixing was measured on a fluorescent
spectrometer (λex ) 470 nm, λem ) 530 nm). After 20 min, the
liposomes were completely collapsed by adding Triton X-100 to give
maximum fluorescence intensity, Fmax. The percentage of lipid mixing
at a given time was calculated by [(Ft - F0)/(Fmax - F0)]100, where
F0, Ft, and Fmax are the fluorescent intensity at time 0, t, and after
collapsing the liposomes. The assay was carried out at various pHs
using S10% Fusion-lip and was also studied using Fusion-lip containing
5, 10, and 20 mol % of the corresponding synthetic lipid at pH 7.4.
Time-Resolved Electronic Absorption Spectra. S10% Fusion-lip
containing 6b was transferred into a quartz cell at 4 °C under nitrogen
and the pH was adjusted to around 10. The solution was mixed with a
Control-lip at a molar ratio of 9:1 (Control-lip/Fusion-lip). To start
membrane fusion, the pH of the solution was acidified to pH 4 at 20
°C by addition of 2 M citric acid. Absorption spectra were measured
every 100 ms for several minutes by a multichannel spectrophotometer.
Hexadecanoic Acid 4-Hydroxy-3-nitrophenylamide (8). Palmitic
acid (0.256 mg, 1 mmol), 4-amino-2-nitrophenol (0.154 mg, 1 mmol),
and DMAP (0.134 mg, 1 mmol) were dissolved in 10 mL of CHCl3.
HOBt (0.184 mg, 1.2 mmol) was dissolved in 10 mL of DMF and
CHCl3 (1:9), and the solution was added dropwise to the above reaction
mixture followed by addition of WSC (0.212 mg, 1.1 mmol) dissolved
in 5 mL of CHCl3. The reaction mixture was stirred overnight. The
solvent was evaporated and the residue was dissolved in EtOAc. The
solution was then washed with 1N HCl 3 times and brine, and dried
over Na2SO4. Purification by preparative thin-layer chromatography
with 30% EtOAc-hexane gave the product in 87% yield. IR (KBr):
1
3310, 1669, 1542, 1314 cm-1. H NMR (CDCl3) δ 10.40 (s, 1 H),
8.29 (d, 1 H, J ) 2.6 Hz), 7.77 (dd, 1 H, J ) 2.6, 8.9 Hz), 7.13 (d, 1
H, J ) 8.9 Hz), 2.36 (t, 2 H, J ) 7.3 Hz), 1.73 (t, 2 H, J ) 7.3 Hz),
1.26 (s, 24 H), 0.87 (t, 6 H, J ) 6.6 Hz). Anal. Calcd for C22H36N2O4:
C, 67.32; H, 9.24; N, 7.14. Found: C, 67.64; H, 9.12; N, 6.93.
2,3-Dihydroxy-N-(4-hydroxy-3-nitrophenyl)propionamide (9). DL-
Glyceric acid (0.577 mg, 3.76 mmol), 4-amino-2-nitrophenol (0.400
mg, 3.76 mmol), DMAP(0.460 mg, 3.76 mmol), HOBt (0.577 mg, 3.76
mmol), and WSC (0.751 mg, 3.77 mmol) were dissolved in 30 mL of
DMF, and the mixture was stirred overnight. The solvent was
evaporated in vacuo, and the residue was dissolved in CHCl3 followed
by washing with dilute HCl and was extracted with 1N NaOH and
water. After reducing the solution volume, the product was purified
by ODS column eluted with 5% MeCN-H2O. IR (KBr) 3305, 1666,
Supporting Information Available: Additional experimental
details, pH-metric titration for 6b and 6a in the liposomes. This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
1550, 1330 cm-1. H NMR (CD3CN) δ 10.14 (s, 1 H), 8.96 (br s, 1
H), 8.60 (d, 1 H, J ) 2.6 Hz), 7.78 (dd, 1 H, J ) 2.6, 8.9 Hz), 7.14 (d,
1 H, J ) 8.9 Hz), 4.16 (t, 1 H, J ) 4.2 Hz), 3.76 (d, 2 H, J ) 6.9, 4.2
Hz).
JA037796X
9
14740 J. AM. CHEM. SOC. VOL. 125, NO. 48, 2003