Anchoring cationic amphiphiles for nucleotide delivery significance of DNA release from cationic liposomes for transfection

Article abstract of DOI:10.1248/bpb.30.1117

We have designed and synthesized lithocholic acid-based cationic amphiphile molecules as components of cationic liposomes for gene transfection (lipofection). To study the relationship between the molecular structures of those amphiphilic molecules, particularly the extended hydrophobic appendant (anchor) at the 3-hydroxyl group, and transfection efficiency, we synthesized several lithocholic and isolithocholic acid derivatives, and examined their transfection efficiency. We also compared the physico-chemical properties of cationic liposomes prepared from these derivatives. We found that isolithocholic acid derivatives exhibit higher transfection efficiency than the corresponding lithocholic acid derivatives. This result indicates that the orientation and extension of hydrophobic regions influence the gene transfection process. Isolithocholic acid derivatives showed a high ability to encapsulate DNA in a compact liposome-DNA complex and to protect it from enzymatic degradation. Isolithocholic acid derivatives also facilitated the release of DNA from the liposome-DNA complex, which is a crucial step for DNA entry into the nucleus. Our results show that the transfection efficiency is directly influenced by the ability of the liposome complex to release DNA, rather than by the DNA-encapsulating ability. Molecular modeling revealed that isolithocholic acid derivatives take relatively extended conformations, while the lithocholic acid derivatives take folded structures. Thus, the efficiency of release of DNA from cationic liposomes in the cytoplasm, which contributes to high transfection efficiency, appears to be dependent upon the molecular shape of the cationic amphiphiles.

Full text of DOI:10.1248/bpb.30.1117

June 2007
Biol. Pharm. Bull. 30(6) 1117—1122 (2007)
1117
Anchoring Cationic Amphiphiles for Nucleotide Delivery
Significance of DNA Release from Cationic Liposomes for Transfection
Naohide HIRASHIMA, ,a Kazuhiro MINATANI,a Yoshifumi HATTORI,a,b Tomohiko OHWADA, ,b and
Mamoru NAKANISHI
*
*
a
a Graduate School of Pharmaceutical Sciences, Nagoya City University; Tanabe-dori, Mizuho-ku, Nagoya 467–8603,
Japan: and b Graduate School of Pharmaceutical Sciences, The University of Tokyo; Hongo, Bunkyo-ku, Tokyo 113–0033,
Japan. Received February 8, 2007; accepted March 9, 2007; published online March 19, 2007
We have designed and synthesized lithocholic acid-based cationic amphiphile molecules as components of
cationic liposomes for gene transfection (lipofection). To study the relationship between the molecular structures
of those amphiphilic molecules, particularly the extended hydrophobic appendant (anchor) at the 3-hydroxyl
group, and transfection efficiency, we synthesized several lithocholic and isolithocholic acid derivatives, and ex-
amined their transfection efficiency. We also compared the physico-chemical properties of cationic liposomes
prepared from these derivatives. We found that isolithocholic acid derivatives exhibit higher transfection effi-
ciency than the corresponding lithocholic acid derivatives. This result indicates that the orientation and exten-
sion of hydrophobic regions influence the gene transfection process. Isolithocholic acid derivatives showed a high
ability to encapsulate DNA in a compact liposome–DNA complex and to protect it from enzymatic degradation.
Isolithocholic acid derivatives also facilitated the release of DNA from the liposome–DNA complex, which is a
crucial step for DNA entry into the nucleus. Our results show that the transfection efficiency is directly influ-
enced by the ability of the liposome complex to release DNA, rather than by the DNA-encapsulating ability. Mol-
ecular modeling revealed that isolithocholic acid derivatives take relatively extended conformations, while the
lithocholic acid derivatives take folded structures. Thus, the efficiency of release of DNA from cationic liposomes
in the cytoplasm, which contributes to high transfection efficiency, appears to be dependent upon the molecular
shape of the cationic amphiphiles.
Key words transfection; cationic amphiphile; DNA relese; lipofection; lithocholic acid
Establishment of an efficient and safe method for introduc- brane and endosomal membrane.21—25) The released DNA in
ing exogenous nucleotides into mammalian cells is critical the cytoplasm is translocated into the nucleus, while DNA re-
for both basic sciences and clinical applications, such as maining in the endosomes is degraded in lysosomes. There-
gene therapy.1) Among various methods for gene transfec- fore, the efficiency of cellular uptake of the liposome–DNA
tion, lipofection using cationic liposomes is considered to be complex and the efficiency of release of DNA from both
a promising way to deliver a foreign gene to target cells.2—4) the endosomes and the liposomes are postulated to be major
Although many kinds of cationic liposomes have been devel- factors that determine the transfection efficiency, which can
oped for lipofection,5—10) further studies are still required to be estimated by means of luciferase assay. In the present
achieve a transfection efficiency comparable with that of study, we prepared cationic liposomes from lithocholic
viral vectors. The results of a recent structural study indi- acid–polyamine conjugates (Chart 1), and investigated two
cated that a multilamellar or inverted hexagonal structure chemico-physical characteristics that may affect the effi-
with alternating lipid bilayer and DNA monolayer might be ciency of the processes mentioned above. Firstly, we exam-
favorable.11—13) Therefore, the structure of the lipid bilayers ined the encapsulation of DNA and compaction of the lipo-
of cationic liposomes are probably conserved upon complex- some–DNA complex, since the liposome–DNA interaction is
ation with DNA. The width of the bilayer is estimated to be thought to facilitate the uptake of liposome–DNA complexes
36 Å.11—13) Recently, we have developed several cationic am- by target cells and also to protect the plasmid DNA from en-
phiphiles for nucleotide delivery.14,15) These are lithocholic zymatic degradation. Secondly, we examined the ability of
acid-based molecules (see Chart 1) which bear cis-decalin cationic liposomes to release DNA upon membrane fusion
structures, while the frequently used cholestane-based mole- with anionic liposomes, which mimic the endosomal mem-
cules (such as DC-Chol) have trans-decalin structures. The brane.21—25) This release allows DNA to translocate to the nu-
additional hydrophobic region at the 3-hydroxyl group of cleus, where it becomes accessible to the transcription appa-
lithocholic acid derivatives may enforce hydrophobic interac- ratus. We also examined the relationship between these prop-
tion in the assembly, resulting in anchoring of the am- erties. Transfection efficiency was positively correlated with
phiphiles to the bilayer. We found that the hydrophobic ap- the ability of liposomes to release DNA, which is apparently
pendant at the 3-position and the orientation and extension of dependent upon the molecular shape of the cationic am-
the hydrophobic regions around the ether linkage both signif- phiphiles.
icantly influence the gene transfection. However, the mecha-
nism of the dependence of the transfection efficiency on mo- MATERIALS AND METHODS
lecular structure was not elucidated. In liposome-mediated
transfection, a liposome–DNA complex is taken into target
Materials 3b-[N-(Dimethylaminoethane)carbamoyl]-
cells by endocytosis.16—20) The internalized exogenous DNA cholesterol (DC-Chol), 1,2-dioleoyl-sn-glycero-3-phospha-
is released by disruptive interaction between liposomal mem- tidylethanolamine (DOPE), 1,2-dioleoyl-sn-glycero-3-phos-
∗ To whom correspondence should be addressed. e-mail: hirashim@phar.nagoya-cu.ac.jp; ohwada@mol.f.u-tokyo.ac.jp © 2007 Pharmaceutical Society of Japan

1118
Vol. 30, No. 6
phatidylcholine (DOPC), and 1,2-dioleoyl-sn-glycero-3-phos- whole was allowed to warm to rt, stirred for 3 h, and poured
phatidylglycerol (DOPG) were purchased from Sigma. N- into water. The mixture was extracted with diethyl ether, and
(7-Nitrobenz-2-oxa-1,3-diazol-4-yl)-1,2-dihexadecanoyl-sn- the organic phase was dried over Na2SO4. The residue, ob-
glycero-3-phosphoethanol-amine (NBD-PE) was purchased tained after evaporation of the solvent, was flash-chro-
from Molecular Probes. Plasmid DNA coding luciferase matographed (silica gel; CHCl3 : iPrNH2ꢀ24 : 1) to give a
(pGL3) was purchased from Promega.
pale yellow oil (15).
Synthesis of the Isolithocholic Acid–Polyamine Conju-
(v) Isolithocholic Acid Methyl Ester (16) (Chart 4): A
gates (1—3) and the Lithocholic Acid–Polyamine Con- mixture of isolithocholic acid (10) (3.38 g (8.65 mmol)) and
jugates (4—6). Synthesis of Isolithocholic Acid (Chart 1.5 ml of concentrated H2SO4 in 40 ml of methanol was
2) (i) Lithocholic Acid Methyl Ester (8): A mixture of stirred at rt for 3 h. The whole was diluted with 150 ml of
lithocholic acid (7) and concentrated (98%) sulfuric acid in methylene chloride. The mixture was washed with water
MeOH was stirred at room temperature (rt) for 2.5 h. The (100 mlꢁ2) and brine, and the organic phase was dried over
whole was diluted with CH2Cl2, and washed with water and Na2SO4. The residue, obtained after evaporation of the
brine. The organic phase was dried over Na2SO4, and the solvent, was column-chromatographed (silica gel; n-hexane :
solvent was evaporated. The residue was column-chro- AcOEtꢀ4 : 1) to give a colorless solid (16).
matographed (silica gel; n-hexane : AcOEtꢀ4 : 1) to give a
colorless solid (8).
(vi) O-Allyl Isolithocholic Acid Methyl Ester (17): To a
solution of 16 in DMF, N,N-diisopropylethylamine was
(ii) Isolithocholic Acid p-Nitrobenzoate (9): To a solu- added, and the whole was heated at reflux at 160 °C for 2 h.
tion of 8, triphenylphosphine and p-nitrobenzoic acid in dry Then, allyl bromide was added, and the whole was heated at
THF, a solution of diethyl azodicarboxylate in dry THF was reflux for 16 h. After removal of the DMF by distillation, the
added at rt under an argon atmosphere. The whole was resultant residue was diluted with water, and the whole was
stirred for 2 h. After evaporation of the solvent, the residue extracted with methylene chloride. The organic phase was
was extracted with CH2Cl2 and water, and the organic phase washed with brine, and dried over Na2SO4. The residue, ob-
was washed with brine, then dried over Na2SO4. The residue tained after evaporation of the solvent, was flash-chro-
obtained after evaporation of the solvent was column-chro- matographed (silica gel; n-hexane : AcOEtꢀ19 : 1 to 4 : 1) to
matographed (silica gel; n-hexane : CH2Cl2ꢀ4 : 1 to 3 : 2 to give a pale yellow solid 17.
1 : 1) to give the p-nitrobenzoate (9).
(vii) O-Isoheptenyl Isolithocholic Acid Methyl Ester
(iii) Isolithocholic Acid (10): A suspension of 9 in a (18a): To a solution of benzylidene-bis(tricyclohexylphos-
mixture of 40% aqueous KOH and methanol was heated at phine)dichlororuthenium (the Grubbs catalyst) (55.9 mg
reflux at 100 °C for 14 h. The whole was diluted with water (8.5 mol%)) in 3 ml of dry chloroform, a solution of 16
and acidified with aqueous HCl, and the precipitate was fil- (344.7 mg (0.80 mmol)) and 4-methyl-1-pentene (485.7 mg
tered off. The filtrate was extracted with CHCl3. The organic (7.08 eq)) in 3 ml of dry chloroform was added at rt under an
layer was washed with brine, dried over Na2SO4, and evapo- argon atmosphere. The whole was stirred at rt for 37 h. The
rated. The residue was flash-chromatographed (silica gel; n- residue, obtained after evaporation of the solvent, was flash-
hexane : AcOEtꢀ9 : 1 to 3 : 1) to give a colorless solid, 10.
chromatographed (silica gel; n-hexane : CH2Cl2ꢀ1 : 1) to
Synthesis of Polyamine Derivatives (Chart 3) (i) give a colorless solid (17).
Mono-N-Boc-amine (12): To a solution of 1,4-diaminobu-
(viii) O-Isoheptenyl Isolithocholic Acid (19a): To a solu-
tane (11) in methylene chloride, a solution of di-t-butyl di- tion of the ester 18a in THF, a solution of lithium hydroxide
carboxylate in methylene chloride was added dropwise over monohydrate in water was added at rt. The whole was stirred
20 min at rt. The whole was stirred at rt for 6.5 h, then diluted for 28 h, diluted with water and acidified with 2 N aqueous
with methylene chloride, and washed with water and brine. HCl. This mixture was extracted with ethyl acetate. The or-
The organic phase was dried over Na2SO4. Evaporation of ganic phase was washed with brine, and dried over Na2SO4.
the solvent gave 12 as a colorless oil.
The residue, obtained after evaporation of the solvent, was
(ii) Mono-N-Boc-amine-nitrile (13): To a solution of 12 flash-chromatographed (silica gel; CHCl3 : AcOEtꢀ9 : 1) to
in methanol, acrylonitrile was added at 5 °C. The whole was give a colorless wax 19a.
allowed to warm to rt, and stirred for 12 h. The residue, ob-
(ix) O-Isoheptenyl Isolithocholic Acid-N-Boc-spermi-
tained after evaporation of the solvent, was flash-chro- dine Conjugate (20a): To a solution of the carboxylic acid
matographed (silica gel; CHCl3 : MeOHꢀ19 : 1) to give a 19a, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hy-
pale yellow oil (13).
drochloride and N-hydroxysuccinimide in methylene chlo-
(iii) Bis-N-Boc-nitrile (14): To a solution of 13 in meth- ride, a solution of the protected polyamine 15 in methylene
ylene chloride, a solution of di-t-butyl dicarboxylate in meth- chloride was added at rt. The whole was stirred for 40 h, and
ylene chloride was added at rt. The whole was stirred for diluted with methylene chloride. The organic phase was
4.5 h. Water was added, and the whole was extracted with washed with aqueous saturated NaHCO3, water and brine,
methylene chloride. The organic layer was washed with and dried over Na2SO4. The residue, obtained after evapora-
brine. The organic phase was dried over Na2SO4, and the sol- tion of the solvent, was flash-chromatographed (silica gel; n-
vent was evaporated. The residue was flash-chromatographed hexane : AcOEtꢀ3 : 2) to give a colorless oil (20a).
(silica gel; CHCl3 : AcOEtꢀ19 : 1) to give a pale yellow oil
14.
(x) O-Isoheptenyl Isolithocholic Acid–Spermidine Con-
jugate (1): To a solution of 20a in methylene chloride, triflu-
(iv) Bis-N-Boc-spermidine (15): To a suspension of oroacetic acid was added at 5 °C. The mixture was stirred at
LiAlH4 in dry diethyl ether (argon-bubbled), a solution of 14 5 °C for 1.5 h, then the TFA was distilled off. The residue
in dry diethyl ether was added at 5 °C over 25 min. The was flash-chromatographed (silica gel; CHCl3 : MeOH :

June 2007
1119
iPrNH2ꢀ18 : 1 : 1) to give a yellow oil (1).
Anionic liposomes were prepared from DOPC, DOPE and
Other polyamine derivatives (2—6) were synthesized simi- DOPG (DOPC : DOPE : DOPGꢀ1 : 2 : 1 (molar ratio)). The
larly as shown in Chart 3. fluorescence intensity of NBD (535 nm) excited at 465 nm
Liposome Preparation DOPE (dioleoyl L-a-phospha- was monitored with a spectrofluorometer (RF-5300PC; Shi-
tidylethanolamine) (Sigma) in chloroform was mixed with a madzu).
solution of a steroid polyamine conjugate in chloroform to
Molecular Modeling The energy-minimum structures
prepare a mixture of DOPE/synthetic conjugate in a molar under three environmental conditions (gas-phase, in CHCl3,
ratio of 3 : 2.10,14) The mixture was evaporated under reduced and in water) were obtained by conformation searches with
pressure to give a thin film of lipids. Phosphate-buffered pure LowMode methods27) and subsequent multiple-mini-
saline (PBS) was added to the lipid film, and the sample was mization. All the calculations were done with the OPLS-AA
allowed to stand for 1 h. Hydrated samples were sonicated force field.28) The calculations were carried out with Macro-
for 2—3 min in a bath-type sonicator (model 5210J, Bran- model, version 8.1, Schrödinger Inc.
son), followed by sonication with a probe-type sonicator
(Sonifier 250, Branson) for 10 min to give small unilamellar RESULTS AND DISCUSSION
vesicles (SUVs). Plasmid DNA (pGL3) was mixed with
cationic liposomes in a charge ratio (ꢂ/ꢃ) of 1.4.
Synthesis We synthesized three different polyamine–
Cell Culture and Transfection NIH3T3 and COS-7 lithocholic acid conjugates and their isolithocholic acid
cells were cultured in DMEM (Gibco) supplemented with counterparts (Chart 1). Isolithocholic acid was prepared
10% FBS (Bio-Whittaker). Transfection was carried out fol- by Mitsunobu inversion reaction (p-nitrobenzoic acid/
lowing the procedure described in our previous paper.14,15) In Ph3P/DEAD) of the 3-hydroxyl group of lithocholic acid (in
brief, plasmid pGL3 DNA (5.4 mg) in TE buffer was com- the form of the methyl ester) (Chart 2). The synthesis of O-
plexed with the above liposomes in SFM101 (Nissui) at room allyl isolithocholic (17) and lithocholic acid (22) derivatives
temperature by pipetting well, and then incubated at 37 °C was carried out as reported previously.15) Synthesis of the
for 15 min to allow formation of the DNA–liposome com- polyamine conjugates 1—6 was accomplished by olefin
plex. This DNA–liposome complex was incubated with the metathesis of the O-allyl derivatives and a terminal olefin in
cells (1ꢁ106 cells/60 mm dish) in the SFM101 medium for the presence of the ruthenium-alkylidene catalyst developed
4 h at 37 °C. The cells were washed and cultured in growth by Grubbs.29—31) Preparation of polyamines (Chart 3) and
medium (DMEM) for another 40 h at 37 °C. Luciferase assay amide coupling (Chart 4) were also described previously.15)
1
was carried out after cell lysis.
All new compounds gave satisfactory H-NMR and high-
Luciferase Assay Transfection efficiency was evaluated resolution mass spectra (HR-MS, EIꢂ) or FAB mass spectra
by means of luciferase assay using a Picagene luciferase (FAB-MS).
assay kit (Toyo Ink) as previously described.14,24) The cells
Transfection Efficiency of Polyamine–Lithocholic Acid
were washed three times with PBS and lysed in a cell lysis Conjugates The transfection efficiency of the three differ-
buffer for 15 min at room temperature. The cell lysate was
centrifuged at 12000 rpm at 4 °C for 3 min and the super-
natant was subjected to luciferase assay. Light emission was
measured with a luminometer (TD-20/20, Turner Designs)
and normalized to the amount of protein of each sample, de-
termined by means of BCA assay.
Ethidium Bromide Intercalation Assay The interaction
between liposomes and DNA was examined by means of
ethidium bromide (EtBr) intercalation assay.21,26) Fluores-
cence of EtBr at 595 nm (excited at 520 nm) was monitored
with a spectrofluorometer (RF-5300PC; Shimadzu) linked
with a personal computer. Values were expressed as the frac-
tion of the maximum intensity obtained when EtBr was
added to free plasmid DNA without liposomes.
FRET Assay To examine the efficiency of the release Chart 1. Structures of Cationic Amphiphiles Used in This Study
process, the dissociation of DNA from cationic liposomes
caused by the addition of anionic membrane, which mimics
the endosomal membrane, was monitored by means of fluo-
rescent resonance energy transfer (FRET) assay using NBD-
labeled liposomes and rhodamine-labeled DNA.10,22) Fluores-
cence-labeled liposomes were prepared by adding NBD-PE
(0.5% mol) to the liposomes described above, and c-myc
antisense oligonucleotides (phosphorothioate; CACGTTG-
AGGGGCAT) were labeled with rhodamine (Nihon Gene
Research Laboratories Inc., Sendai, Japan). After formation
of liposome–DNA complexes with a charge ratio (ꢂ/ꢃ) of
1.4, dissociation of DNA from liposomes was induced by the
a) conc-H2SO4, MeOH, rt, 2.5 h, yield 98%, b) p-nitrobenzoic acid, PPh3, DEAD,
THF, rt, 2 h, yield 97%, c) 40% aq. KOH, MeOH, reflux, 5 h, 90%.
addition of anionic liposomes in citrate buffer (pH 5.0).
Chart 2

1120
Vol. 30, No. 6
ent polyamine–lithocholic acid conjugates and their isolitho- used to detect DNA. When it intercalates between the base
cholic acid counterparts (Chart 1) was estimated by means of pairs of the DNA double helix, its fluorescence is greatly en-
luciferase assay (Fig. 1). Cationic liposomes were prepared hanced. Aqueous solutions of DNA stained with EtBr conse-
by mixing of the polyamine–lithocholic acid conjugate and quently have a high initial level of fluorescence. Tight inter-
DOPE (dioleoyl L-a-phospatidylethanolamine), followed by action of liposomes with DNA results in encapsulation of the
mixing of a plasmid DNA, pGL3, which encodes luciferase. DNA and prevents intercalation of EtBr with the DNA, re-
NIH3T3 cells were transfected with the liposome–DNA sulting in a decrease in the fluorescence intensity of EtBr.
complex and the transfection efficiency was evaluated by The efficiency of encapsulation estimated with the EtBr in-
means of luciferase assay. As shown in Fig. 1, the transfec- tercalation assay is shown in Fig. 2. All three isolithocholic
tion efficiencies of the isolithocholic acid conjugates (1—3) acid conjugates showed higher efficiency than DC-Chol.
were higher than those of the corresponding lithocholic acid However, the efficiencies of these isolithocholic acid conju-
conjugates (4—6). The transfection efficiencies of the gates (1—3) were roughly constant at about 50%, and a
isolithocholic acid conjugates were higher than that of the higher magnitude of encapsulation was observed with the
commercially available cationic cholesterol derivative (DC- lithocholic acid derivatives (4—6), which nevertheless
Chol). The efficiency of the derivative 3 was about 6 times showed much less potent gene transfection activities than the
higher than that of DC-Chol. On the other hand, the litho- isolithocholic acid conjugates. Thus, the efficiency of encap-
cholic acid conjugates had lower efficiencies than DC-Chol. sulation of DNA by cationic amphiphiles does not directly
Similar results were obtained in COS-7 cells (data not parallel the transfection efficiency in this system.
shown). Thus, the stereochemistry of the hydrophobic appen-
dant at the 3-position, i.e., the orientation and extension of
the hydrophobic regions around the ether linkage at the 3-po-
sition, is crucial for the transfection efficiency. The present
results are consistent with previous observations.12)
Efficiency of DNA Release from Cationic Liposomes
Efficiency of Encapsulation of DNA As described
above, the isolitocholic acid conjugates had higher transfec-
tion efficiency than the lithocholic acid conjugates, so we
further investigated the isolithocholic acid conjugates. Elec-
trostatic interaction between the cationic liposomes and the
anionic DNA results in encapsulation and compaction of the
liposome–DNA complexes. This interaction facilitates the
uptake of the liposome–DNA complexes by the target cells
and the complexation protects the plasmid DNA from enzy-
matic degradation. Thus, we evaluated the strength of the in-
teraction of the cationic liposomes of the isolithocholic acid
conjugates with DNA by using the ethidium bromide (EtBr)
intercalation assay. EtBr is a fluorescent label that is widely
Fig. 1. Transfection Efficiency of Lithocholic Acid-Based Molecules
Transfection activities of cationic liposomes prepared with DC-Chol or lithocholic
acid derivatives were determined by means of luciferase assay. Each value represents
the meanꢄS.E. (nꢀ3).
a) Boc2O, CH2Cl2, rt, 6.5 h, yield 76%, b) acrylonitrile, MeOH, 0 °C—rt, 12 h, yield
90%, c) Boc2O, CH2Cl2, rt, 4.5 h, yield 92%, d) LiAlH4, Et2O, 0 °C—rt, 3 h, yield 65%.
Fig. 2. DNA Encapsulation Ability of Cationic Liposomes
The efficiency of formation of liposome–DNA complex was determined by means of
Chart 3
EtBr assay. Each value represents the meanꢄS.E. (nꢀ3).
a) (i-C3H5)2EtN, DMF, reflux, 2 h, then allyl bromide, reflux, 14 h, yield 83%, b) Cy2(PPh3)2RuꢀCHPh, H2CꢀCHR1, CH2Cl2, rt, (18a, 37 h, 51%; 18b, 26 h, 39%; 18c, 22 h,
54%; 23a, 17 h, 62%; 23b, 36 h, 39%; 23c, 25 h, 48%), c) 40% aq. KOH, MeOH, rt, or LiOH–H2O, THF, rt (19a, 22 h, 61%; 19b, 22 h, 74%; 19c, 32 h, 75%; 24a, 24 h, 50%; 24b,
19 h, 78%; 24c, 22 h, 97%), d) 15, N-hydroxysuccinimide, EDC, CH2Cl2, rt, (20a, 22 h, 61%; 20b, 22 h, 74%; 20c, 32 h, 75%; 25a, 24 h, 50%; 25b, 19 h, 78%; 25c, 22 h, 97%), e)
TFA, CH2Cl2, 0 °C (1, 1.5 h, 77%; 2, 1 h, 85%; 3, 1 h, 86%; 4, 1 h, 70%; 5, 1.5 h, 35%; 6, 1 h, 78%).
Chart 4

June 2007
1121
Fig. 3. Efficiency of DNA Release from Cationic Liposomes
Time course of dissociation reaction of cationic liposomes and DNA. Relative fluo-
rescence intensity of NBD was plotted against time (meanꢄS.E., nꢀ3). Anionic lipo-
somes were added at 0 min. Values of relative fluorescence intensity (RFI) were calcu-
lated as follows; RFI(t)ꢀ(F(t)ꢃF(0))/F(0), where F(t) is the fluorescence intensity of
NBD at the indicated time and F(0) is the fluorescence intensity of NBD before addi-
tion of anionic liposomes.
Next, we examined the efficiency of release of exogenous
DNA from the endosomes and the cationic liposomes. Lipo-
some–DNA complexes are taken up into the cells by endocy-
tosis, and transported to lysosomes, where the complexes are
degraded.19) Therefore, for high transfection activity, DNA
must be released from the endosomes and liposomes before
the lysosomal degradation, so that it can be transported intact
to the nucleus. To evaluate the influence of the molecular
structure of the 3-axial appendant (1, 2, 3) on the efficiency
Fig. 4. Energy-Minimum Molecular Structures of Isolithocholic Acid
(1—3) and Lithocholic Acid (4—6) Derivatives in Water, Obtained by Com-
putational Modeling
of this release process, the dissociation of DNA from the endosomal escape of DNA.32,33) Using small-angle X-ray
cationic liposomes caused by the addition of anionic lipo- spectroscopy (SAXS), important roles of non-lamellar struc-
somes containing phosphatidylglycerol (PG), which mimics ture of lipoplex, such as micellar cubic and hexagonal, in dis-
the endosomal membrane, was monitored by means of fluo- ruption of endosomal membrane and DNA release from
rescent resonance energy transfer (FRET) assay using rho- lipoplex.34,35) High transfection efficiencies of the isolitho-
damine-labeled DNA and NBD-labeled liposomes. The en- cholic acid conjugates might relate to their ability to induce
dosomal membrane contains anionic lipids, since the endo- non-lamellar structure in lipoplex.
some is formed by pinching off of invaginated plasma mem-
Molecular Structure of Lithocholic Acid Derivative–
brane that contains anionic phospholipids. Anionic lipid is Polyamine Conjugates To understand the relationship be-
reported to be critical for efficient fusion between cationic tween the molecular structure and transfection efficiency, we
and anionic liposomes, whose lipid composition is similar to carried out molecular modeling of the compounds used in
that of endosomes.21,23) Anionic lipid mixing neutralizes the this study. Energy-minimized structures of these compounds
positive charge of the cationic liposome–DNA complex and (1—6) were obtained for three different environmental con-
displaces DNA from the complex. Lipid mixing between ditions (gas-phase, in CHCl3 and in water), and the energy-
cationic and anionic liposomes also causes membrane fusion, minimum structures in water are shown in Fig. 4. There was
and this allows the displaced DNA to be released from endo- little difference among the preferred structures in the differ-
some.21,23) FRET assays are widely used to monitor dynamic ent environments. The polyamine–lithocholic acid conjugates
changes in the distance between two fluorescent probes. In took folded conformations, while the polyamine–isolitho-
the present system, the fluorescence of the NBD fluorophore cholic conjugates took relatively extended conformations.
is quenched by the neighboring rhodamine fluorophore The former structural trend was also found in the superposi-
through the FRET mechanism in the DNA–cationic liposome tion of 100 frames of N-protonated structures of other related
complex. When the rhodamine-labeled DNA is released from polyamine–lithocholic acid conjugates previously reported.14)
the cationic liposomes following interaction with the anionic Among the polyamine–isolithocholic conjugates, derivative 3
liposomes containing DOPE, DOPC and DOPG, the FRET took the most stretched-out conformation. Interestingly, the
from excited NBD to rhodamine was diminished, and the flu- derivative 3 exhibited the highest efficiency of both transfec-
orescence intensity of the NBD chromophore was increased. tion and DNA release.
Figure 3 shows the time course of the fluorescence intensity
of NBD at 535 nm, where the increase of the fluorescence re- CONCLUSIONS
flects dissociation of DNA from the liposomes. The effi-
ciency of the DNA release was highest when cationic lipo-
We investigated the mechanism causing the different trans-
somes containing the derivative 3 were used, while the lowest fection efficiencies of cationic liposomes prepared from
release was observed when the derivative 2 was used. These lithocholic acid– and isolithocholic acid–polyamine conju-
results are consistent with the observed transfection effi- gates. As we reported previously,14) the orientation and exten-
ciency. Several lines of evidence suggest that lipid phase for- sion of the hydrophobic regions around the ether linkage at
mation in DNA–liposome complex (lipoplex) is crucial for the 3-position of the lithocholic acid-based molecules signifi-

1122
Vol. 30, No. 6
M., FEBS Lett., 397, 207—209 (1996).
9) Okayama R., Noji M., Nakanishi M., FEBS Lett., 408, 232—234
cantly affect the gene transfection efficiency. To elucidate the
factors involved, we examined two crucial chemico-physical
characteristics of the cationic liposomes. One is the robust-
ness of the liposome–DNA complex. The other is the ease of
(1997).
10) Hasegawa S., Hirashima N., Nakanishi M., Bioorg. Med. Chem. Lett.,
12, 1299—1302 (2002).
release of DNA from the liposome–DNA complex and endo- 11) Rädler J. O., Koltover I., Salditt T., Safinya C. R., Science, 275, 810—
814 (1997).
somes. Compaction of the liposome–DNA complex is re-
quired for efficient uptake into the target cells via endocyto-
sis, and also for protection of the exogenous DNA from en-
12) Koltover I., Salditt T., Radler J. O., Safinya C. R., Science, 281, 78—
81 (1998).
13) Ewert K. K., Evans H. M., Zidovska A., Bouxsein N. F., Ahmad A.,
zymatic degradation. Once the DNA is internalized into the
target cells by endocytosis, the exogenous DNA must be re-
leased from the complex and the resultant endosomes for
transport into the nucleus. This release process is thought to
be induced by disruptive membrane interaction between the
endosome membrane and the liposome membrane of the
complex.21—25) In the present study, we found that isolitho-
cholic acid derivatives enhance transfection efficiency by fa-
cilitating the release of DNA from the liposome–DNA com-
plex and endosomes.
Safinya C. R., J. Am. Chem. Soc., 128, 3998—4006 (2006).
14) Fujiwara T., Hasegawa S., Hirashima N., Nakanishi M., Ohwada T.,
Biochim. Biophys. Acta, 1468, 396—402 (2000).
15) Yoshimura T., Hasegawa S., Hirashima N., Nakanishi M., Ohwada T.,
Bioorg. Med. Chem. Lett., 11, 2897—2901 (2001).
16) Legendre J.-Y., Szoka F. C., Jr., Pharm. Res., 9, 1235—1242 (1992).
17) Zabner J., Fasbender A. J., Moninger T., Poellinger K. A., Welsh M. J.,
J. Biol. Chem., 270, 18997—19007 (1995).
18) Hasegawa S., Hirashima N., Nakanishi M., Gene Ther., 8, 1669—1673
(2001).
19) Prasad T. K., Rangaraj N., Rao N. M., BFEBS Lett., 579, 2635—2642
(2005).
20) Medina-Kauwe L. K., Xie J., Hamm-Alvarez S., Gene Ther., 12,
1734—1751 (2005).
21) Xu Y., Szoka F. C., Jr., Biochemistry, 35, 5616—5623 (1996).
22) Zelphati O., Szoka F. C., Jr., Proc. Natl. Acad. Sci. U.S.A., 93, 11493—
11498 (1996).
Acknowledgments This work was supported in part by a
Grant-in-Aid for Scientific Research (Category S, No.
17109001) to TO from the Ministry of Education, Science,
Sports, Culture and Technology. A part of the calculations
23) Bailey A. L., Cullis P. R., Biochemistry, 36, 1628—1634 (1997).
was carried out at the Computer Center of the Institute for 24) Noguchi A., Furuno T., Kawaura C., Nakanishi M., FEBS Lett., 433,
169—173 (1998).
Molecular Science. We thank this computational facility for
generous allotments of computer time.
25) Bhattacharya S., Mandal S. S., Biochemistry, 37, 7764—7777 (1998).
26) Sakurai F., Inoue R., Nishino F., Okuda A., Matsumoto O., Taga T.,
Yamashita F., Takakura Y., Hashida M., J. Control. Release, 66, 255—
269 (2000).
REFERENCES
27) Kolossvary I., Guida W. C., J. Am. Chem. Soc., 118, 5011—5019
1) Crystal R. G., Science, 270, 404—410 (1995).
(1996).
2) Felgner P. L., Gadek T. R., Holm M., Roman R., Chan H. W., Wenz 28) Jorgensen W. L., Maxwell D. S., Tirado-Rives J., J. Am. Chem. Soc.,
M., Northrop J. P., Ringold G. M., Danielsen M., Proc. Natl. Acad. Sci.
U.S.A., 84, 7413—7417 (1987).
3) Ulrich A. S., Biosci. Rep., 22, 129—150 (2002).
4) Lasic D. D., Trends Biotechnol., 16, 307—321 (1998).
118, 11225—11235 (1996).
29) Miller S. J., Blackwell H. E., Grubbs R. H., J. Am. Chem. Soc., 118,
9606—9614 (1996).
30) Review: Grubbs R., Chang S., Tetrahedron, 54, 4413—4450 (1998).
5) Farhood H., Bottega R., Epand R. M., Huang L., Biochim. Biophys. 31) Review: Füstner A., Angew. Chem. Int. Ed., 39, 3012—3043 (2000).
Acta, 1111, 239—246 (1992).
32) Zuhorn I. S., Bakowsky U., Polushkin E., Visser W. H., Stuart M. C.
A., Engberts J. B. F. N., Hoekstra D., Mol. Ther., 11, 801—810 (2005).
33) Rao N. M., Gopal V., Biosci. Rep., 26, 301—324 (2006).
34) Koltover I., Salditt T., Rädler J. O., Safinya C. R., Science, 281, 78—
81 (1998).
6) Byk G., Dubertret C., Escriou V., Frederic M., Jaslin G., Rangara R.,
Pitard B., Couzet J., Wils P., Schwartz B., Scherman D., J. Med.
Chem., 41, 224—235 (1998).
7) Gao X., Huang L., Biochem. Biophys. Res. Commun., 179, 280—285
(1991).
35) Tarahovsky Y. S., Koynova R., MacDonald R. C., Biophys. J., 87,
1054—1064 (2004).
8) Takeuchi K., Ishihara M., Kawaura C., Noji M., Furuno T., Nakanishi