Gene Therapy (2000) 7, 764–768
 2000 Macmillan Publishers Ltd All rights reserved 0969-7128/00 $15.00
www.nature.com/gt
NONVIRAL TRANSFER TECHNOLOGY
BRIEF COMMUNICATION
Structural basis of DOTMA for its high intravenous
transfection activity in mouse
T Ren, YK Song, G Zhang and D Liu
Department of Pharmaceutical Sciences, School of Pharmacy, University of Pittsburgh, 527 Salk Hall, Pittsburgh, PA 15261, USA
Eleven structural analogues of two known cationic lipids, N[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA) and N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTAP) were synthesized and utilized to evaluate the structural characteristics of DOTMA for
its high intravenous transfection activity. Using a CMV-driven
expression system and luciferase gene as a reporter, the
transfection activity of these analogues was evaluated in
mice using tail vein injection. Results concerning the structure–activity relationship with regard to the influence of the
backbone, relative position between head group and the
hydrophobic chains on the backbone, linkage bonds, as well
as the composition of the aliphatic chains revealed that cationic lipids which give a higher in vivo transfection activity
share the following structural characteristics: (1) cationic
head group and its neighboring aliphatic chain being in a
1,2-relationship on the backbone; (2) ether bond for bridging
the aliphatic chains to the backbone; and (3) paired oleyl
chains as the hydrophobic anchor. Cationic lipids without
these structural features had lower in vivo transfection
activity. These structural characteristics, however, did not
significantly influence their in vitro transfection activity. The
contribution that cationic lipids make to the overall in vivo
transfection activity is likely to be determined by the structure
of DNA/lipid complexes and by the outcome of the interaction between the DNA/lipid complexes and blood components upon intravenous administration. Gene Therapy
(2000) 7, 764–768.
Keywords: gene therapy; cationic liposome; gene transfer; transfection
Successful demonstration of cationic lipid-mediated gene
transfer into a variety of cells in vitro has generated considerable interest in achieving an optimal condition for
in vivo transfection. After the initial studies by Brigham
et al1 and Zhu et al,2 who demonstrated successful transfection of cells in the lung, heart, liver, spleen and kidney
by tail vein injection of DNA/lipid complexes, significant
efforts have been directed at improving the transfection
activity of cationic lipid-based gene carriers in vivo
through intravenous administration. It has been shown3–19
that the level of transgene expression after intravenous
injection of DNA/lipid complexes can be significantly
increased through optimization of the physicochemical
properties of DNA/lipid complexes which include: (1)
the cationic lipid to DNA ratio, (2) the diameter of lipid
particles, and (3) the inclusion of helper lipids. While the
in vivo results indicated the importance of physical
properties of DNA/lipid complexes in transfection, an
equally important conclusion derived from these studies
is that the structure of cationic lipid also has an important
influence in the in vivo transfection activity.
A good example for demonstrating the importance
of lipid structure to the in vivo transfection comes from
a previous study,4 in which N-[1-(2,3-dioleyloxy)
propyl]-N,N,N-trimethylammonium chloride (DOTMA)
and N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium (DOTAP), the two most commonly used cationic
Correspondence: D Liu
Received 27 September 1999; accepted 22 December 1999
lipids for transfection,17–19 were systematically evaluated
for their in vivo transfection activity. It was found that
DOTMA and DOTAP, which differ in the linkage bonds
(two ether bonds in DOTMA and two ester bonds in
DOTAP), are equally active in transfecting many types of
cells in vitro but exhibit a 10-fold difference in the level of
transgene expression in the animal lung after intravenous
administration.4 These results form the basis of the
present study in which an attempt has been made to
identify the structural characteristics of DOTMA that are
critical for its higher intravenous transfection activity. To
achieve this objective, a series of DOTMA and DOTAP
analogues were synthesized and evaluated for their in
vitro and in vivo transfection activity.
The structure of DOTMA and DOTAP could be looked
at as consisting of four different moieties: (1) a quaternary
ammonium head group, (2) a glycerol-based backbone,
(3) two linkage bonds, and (4) two hydrocarbon chains.
Different strategies were used to synthesize several new
analogues that resemble the overall structure of DOTMA
and DOTAP molecules.20,21 The structures of these new
cationic lipids are shown in Figure 1. Compared with the
structure of DOTMA, compounds 1a to 1d contain different backbones. Compounds 2a to 2d are DOTAP analogues with various backbones or aliphatic chains. In
compounds 3a to 3c, ether and ester bonds have been
used to bridge the backbone and the aliphatic chains.
The structural effect of these new cationic lipids on
their intravenous transfection activity was examined in
mice using a previously established procedure.3 Data in
Figure 2A show that the backbone that connects the cat-
Cationic lipid-mediated gene delivery in vivo
T Ren et al
765
Figure 1 Structures of cationic lipids. Cationic lipids 1a–1d, 2b–2d and
3a–3c were synthesized according to previously published procedures.20,21
Cationic lipid 2a was synthesized in the following steps: acetonidation of
1,2,4-butanetriol with a catalytic amount of p-toluenesulfonic acid
afforded the corresponding five membered acetonide. Bromination of the
primary alcohol in the acetonide, followed by acidic cleavage of ketal, liberated the corresponding diol. Diacylation of the diol furnished diester cationic lipid precursor. Finally, Menshutkin’s type quaternarization afforded
the diester cationic lipid 2a. All the synthetic new cationic lipids were
purified by column chromatography and characterized by 1H NMR
and 13C NMR.
ionic head group and the alkylchains plays an important
role in determining the overall in vivo transfection
activity. An increase in the number of carbons in the
backbone from m = 1 (DOTMA) to m = 2 (cationic lipid
1a) or to m = 4 (cationic lipid 1b) resulted in a 50-fold
decrease in luciferase activity in the lung. The level of
luciferase in other organs including the heart, liver,
spleen and kidney also decreased with increase in the
length of backbone.
Because of the lower transfection activity seen in cationic lipids 1a,1b that have a longer backbone, it was
decided to evaluate whether such a negative impact is
solely due to the length of the backbone or is due to the
relative location of the hydrophobic moiety to the cationic head group. To test this, compounds 1c,1d in which
one aliphatic chain moved from 3 position in 1a to 2 position in 1c, and from 5 position in 1b to 2 position in 1d
were synthesized. The resulting lipids 1c,1d contain the
quaternary amine head group and their neighboring
alkyl chain in a 1,2-relationship. Transfection activity of
cationic lipids 1c,1d (Figure 2B) showed the restoration
of the transfection activity due to change of position of
the hydrophobic moiety in the backbone region. An
identical level of luciferase gene expression was obtained
in all internal organs of animals transfected with lipid 1c
or 1d. The gene expression level was the same as that in
animals transfected with DOTMA. These results suggest
that the position of the aliphatic chain with respect to
the polar head group is critical for transfection activity.
Moreover, a lipid with a structure that has the quaternary
amine and its neighboring hydrocarbon chain in a 1,2fashion exhibits higher transfection activity. It is possible
that a close proximity of the hydrophobic chains with the
cationic head group is critical for the formation of
a DNA/lipid complex structure that is more active in
intravenous transfection.
The experiments summarized in Figure 2C were
designed to evaluate the transfection effect of backbone
length, and the relative position of one of the two aliphatic chains to the cationic head group in cationic lipids
Figure 2 Effect of variation in backbone and linkage bond of cationic lipids
on their in vivo transfection activity. Each CD-1 mouse (male, 18–20 g)
(Charlis River, Wilmington, MA, USA) received a tail vein injection of
200 ␮l of DNA and lipid mixture containing 25 ␮g of plasmid DNA complexed with 900 nmol of cationic lipid. pCMV-Luc plasmid DNA was
purified by the method of CsCl-ethidium bromide gradient centrifugation.34 DOTAP was purchased from Avanti Polar Lipids (Alabaster, AL,
USA). Lipid suspension was prepared by homogenizing the hydrated lipid
films in PBS using a Tissue Tearor (Biospec Products, Bartlesville, OK,
USA) according to a previously published procedure.3 The average diameter of the lipid particles was between 150 and 300 nm as measured by
light scattering using a submicron particle analyzer (Particle Sizing Systems, Santa Barbara, CA, USA). DNA/lipid complexes were prepared by
mixing equal volume of DNA solution (0.25 mg/ml) and lipid suspension
(9.0 ␮mol/ml). After gentle mixing, the DNA/lipid mixture was kept at
room temperature for 10 min before being injected into animals. For analysis of the level of gene expression in different organs including the lung,
heart, liver, spleen and kidney, animals were killed 8 h after injection.
Individual organs were dissected from the animal to which lysis buffer
(0.1 m Tris-HCl, 0.1% Triton X-100, 2 mm EDTA, pH 7.8) was added
in a ratio of 4–5 ␮l per mg of the tissue. Each tissue sample was homogenized for 20–30 s with the Tissue Tearor and the homogenate was centrifuged for 10 min at 12 000 g at 4°C. Protein concentration of the supernatant was determined using a Bio-Rad protein assay kit (Bio-Rad,
Hercules, CA, USA). The level of luciferase gene expression was evaluated
using 10 ␮l of supernatant and a luciferase assay kit (Promega, Madison,
WI, USA) in a luminometer (Autolumat LB953, EG & G, Berthhold,
Germany) with 10 s set for measurement. Luciferase activity was normalized to the amount of luciferase per mg of extracted proteins using a
standard curve in which luciferase protein (pg) was equal to 7.89 × 10−5
RLU + 0.093 (R2 = 0.999). Data represent mean ± s.d. (n = 3).
containing ester bonds. While lipid 2b showed a slightly
higher transfection activity than DOTAP, cationic lipid 2a
exhibited transfection activity similar to that of DOTAP.
These results suggest that the length of the backbone and
the relative position of the acyl chains to the cationic head
group of cationic lipids with ester linkage bond do not
significantly alter the transfection activity, indicating that
the positional effect seen in Figure 1A appears unique to
cationic lipids with two ether bonds.
One common feature in the more commonly used cationic lipids, which contain either alkyl or acyl chains as
a hydrophobic anchor, is that the two aliphatic chains are
identical.22,23 To explore whether such uniformity is critical for transfection activity, diester cationic lipids 2c,2d
in which one aliphatic chain is oleoyl and the other is
either a palmitoyl or a lauroyl chain were synthesized.
Data in Figure 3A show that uniformity of aliphatic
chains is pivotal with regard to the overall transfection
Gene Therapy
Cationic lipid-mediated gene delivery in vivo
T Ren et al
766
Figure 3 Effect of uniformity of aliphatic chains of cationic lipids on their
in vivo transfection activity. The experimental protocol was identical to
that described in Figure 1 legend. Data represent mean ± s.d. (n = 3).
activity. Cationic lipids 2c,2d with two different acyl
chains showed a lower transfection activity, compared to
that of lipid 2b, which possessed two identical aliphatic
chains. Similarly, cationic lipids 3b,3c with different aliphatic lengths and ether/ester linkage, showed lower
transfection activity in comparison with cationic lipid 3a,
which contains two identical hydrocarbon chains
(Figure 3B). These results support the previously demonstrated findings that the composition of the hydrocarbon
chains plays an important role in determining the
ultimate transfection activity.7,13
As far as transfection activity is concerned, the linkage
bond in cationic lipids 2b,2c and 3b,3c does not seem to
be important, since a similar level of transgene expression
was seen in animals transfected with cationic lipids containing either two ester bonds (Figure 3A) or ether/ester
bonds (Figure 3B).
Additional experiments were performed with cationic
lipid 1c to evaluate the extent to which cationic lipid to
DNA ratio affects the in vivo transfection activity. Various
buffer systems have been used by different investigators
in the preparation of DNA/lipid complexes.2,7,14 Therefore, DNA/lipid complexes were prepared in phosphatebuffered saline (PBS, pH 7.6), 5% glucose, distilled water,
or Hepes (pH 7.5) buffer. A standard intravenous transfection test was performed in mice using 25 ␮g of pCMVLuc plasmid DNA in the presence of different amounts
of cationic lipids. Data summarized in Figure 4 show that
cationic lipid to DNA ratio has an important effect on the
level of transgene product. For example, at the ratio of
36:1, the level of luciferase gene expression in the lung
was of similar magnitude regardless of the type of buffer
used. However, at a lower cationic lipid to DNA ratio,
an obvious difference in luciferase level in the lung was
seen in animals transfected with DNA/lipid complexes
prepared in different solutions. While a comparable level
of luciferase gene expression at cationic lipid to DNA
ratio of 12:1 to that of 36:1 was observed in DNA/lipid
complexes prepared in PBS for DOTMA and lipid 1c, a
Gene Therapy
Figure 4 Effect of cationic lipid to DNA ratio and different solutions on
in vivo transfection activity of cationic lipids. Lipid formulation and
DNA/lipid complexes were prepared in different solutions including phosphate-buffered saline (PBS), 5% glucose (Glu), distilled water (DW) and
Hepes buffer. Other experimental conditions were the same as those
described in Figure 1 legend. Data represent mean ± s.d. (n = 3).
much lower level of luciferase gene expression was seen
in animals transfected with complexes prepared in Hepes
buffer, 5% glucose solution or distilled water. These
results suggest that while an optimal transfection activity
can be obtained by using a higher cationic lipid to DNA
ratio, selection of an appropriate solution for preparation
of DNA/lipid complexes may become critical when a
small amount of cationic lipids has to be used.
Considering that the in vivo environment where gene
transfer occurs is different from that of a cell culture system, whether the structure–activity relationship seen in
vivo is comparable with that seen in vitro was evaluated.
For this purpose, transfection was performed on five different cell lines using cationic lipids that exhibited difference in in vivo transfection activity with intravenous
administration. Data summarized in Figure 5A show that
although different levels of gene product were obtained
in different cell lines, the major factor that determined
the level of gene expression was the cell line. Among the
cell types tested, 293 cells derived from human embryonic kidney showed a significantly higher level of gene
expression followed by BL-6 cells (murine melanoma). A
much lower level of gene expression was obtained in
HepG2 liver cells, NIH 3T3 and Hela cells. Interestingly,
no clear-cut pattern emerged regarding the structure–
activity relationship among the cationic lipids tested. Cationic lipids with different backbones, aliphatic chains and
linkage bonds exhibited similar transfection activity in a
given cell type. In addition, the transfection activity of
the new cationic lipids was almost identical to that of
DOTMA or DOTAP. The total protein recovery from the
transfected cells (Figure 5B) by these new lipids was
either similar to or higher than those of cells transfected
with either DOTMA or DOTAP, suggesting that none of
these new lipids has any significant toxic effect on
these cells.
Cationic lipid-mediated gene delivery in vivo
T Ren et al
Figure 5 Structural effect of cationic lipids on their transfection activity
in vitro. 293, BL-6, HeLa, HepG2 or NIH 3T3 cells (5 × 104 per well)
were seeded in 48-well plates 24 h before transfection. Each well received
freshly prepared DNA/lipid complexes (250 ␮l per well) containing 2 ␮g
of pCMV-Luc plasmid and 7.5 nmol of cationic lipids. Transfection proceeded for 5 h in serum-free medium before the transfection reagents were
removed and fresh medium was added. Forty-eight hours later, cells were
washed with PBS and lysed with lysis buffer (100 ␮l per well) for 15
min at room temperature. Cell lysates were collected and centrifuged in
a microcentriguge for 5 min. Ten ␮l of the supernatant were used for
luciferase assay. Protein concentration of the supernatant was determined
by protein assay using Bio-Rad protein assay reagents. (A) Luciferase gene
expression level; (B) Total protein recovery. Data represent mean ± s.d.
(n = 3).
Cationic lipids that have been developed to date
invariably consist of three structural components: (1) a
cationic head group, (2) backbone, and (3) hydrophobic
anchor.24 While many of the cationic lipids, including the
hydrocarbon chain-based13,18,19,25–29 and cholesterolbased,30–32 showed good transfection activities in vitro,
other studies have shown that hydrocarbon chain-based
cationic lipids with quaternary ammonium moiety as the
head group exhibit better intravenous transfection
activity.33 Results from our previous studies designed to
identify the factors that control the intravenous transfection activity of cationic lipids showed that DOTMA
appeared to be the most active lipid in transfecting lung
endothelial cells in mice via tail vein administration of
DNA/lipid complexes.3,4 The results summarized in this
report further support this claim.
While the evidence presented here clearly supports the
notion that the structure of cationic lipids plays a critical
role in determining the magnitude of their transfection
activity in vivo, the mechanisms by which the overall
transfection activity is regulated by the lipid structure
remains to be explored. Lack of a significant difference
in transfecting activity of these DOTMA analogues with
a cell line in vitro may suggest that the structure–activity
relationship observed in vivo is not due to the cells transfected in vivo, but rather due to the conditions where
transfection occurs. In general, upon entering the blood
circulation after its intravenous administration, cationic
lipid in the form of either free liposomes or as a complex
with DNA may interact with negatively charged blood
components (cellular and non-cellular). It is possible that
such an interaction would produce aggregates involving
cationic lipid, DNA and blood components. These aggregates, if large enough, are likely to be trapped in larger
blood vessels before reaching the capillary bed, resulting
in a limited exposure of DNA molecules to the lung
endothelial cells and a low level of transgene expression.
Similarly, the negatively charged blood components may
compete with DNA for binding to cationic liposomes,
such that DNA molecules are released from the
DNA/lipid complexes prematurely. This would result in
a transient exposure of DNA molecules to the lung endothelium, with minimal success of transfection. The best
cationic lipid structure for intravenous transfection, however, is the one with: (1) hydrocarbon chains adjacent to
the cationic head group, (2) ether bonds for bridging the
hydrocarbon chains to backbone, and (3) a pair of oleyl
chains as the hydrophobic anchor.
In summary, we have demonstrated in this study that
the structural characteristics of DOTMA for its higher
transfection activity include the close proximity of its
hydrocarbon chains to the cationic head group, its two
ether bonds, and paired oleyl chains as the hydrophobic
anchor. Although changes in these structures do not
affect the in vitro transfection activity, they do in fact
decrease intravenous transfection activity.
767
Acknowledgements
Assistance from Professor Balwant Dixit for critical reading of this manuscript is acknowledged. DOTMA was
kindly provided by Roche Bioscience. This work was
supported in part by a grant from NIH (CA 72925) and
research contract with Targeted Genetics Corporation.
References
1 Brigham KL et al. In vivo transfection of murine lungs with a
functioning prokaryotic gene using a liposome vehicle. Am J
Med Sci 1989; 298: 278–281.
2 Zhu N, Liggitt D, Liu R, Debs R. Systemic gene expression after
intravenous DNA delivery into adult mice. Science 1993; 261:
209–211.
3 Liu F, Qi H, Huang L, Liu D. Factors controlling the efficiency
of cationic lipid-mediated transfection in vivo via intravenous
administration. Gene Therapy 1997; 4: 517–523.
4 Song YK, Liu F, Chu SY, Liu D. Characterization of cationic liposome-mediated gene transfer in vivo by intravenous administration. Hum Gene Ther 1997; 8: 1585–1594.
5 Thierry A et al. Systemic gene therapy: biodistribution and longterm expression of a transgene in mice. Proc Natl Acad Sci USA
1995; 92: 9742–9746.
6 Liu Y et al. Cationic liposome-mediated intravenous gene delivery. J Bio Chem 1995; 270: 24864–24870.
Gene Therapy
Cationic lipid-mediated gene delivery in vivo
T Ren et al
768
7 Solodin I et al. A novel series of amphiphilic imidazolinium
compounds for in vitro and in vivo gene delivery. Biochemistry
1995; 34: 13537–13544.
8 Liu Y et al. Factors influencing the efficiency of cationic liposome-mediated intravenous gene delivery. Nat Biotech 1997; 15:
167–173.
9 Li S, Huang L. In vivo gene transfer via intravenous administration of cationic lipid-protamine–DNA (LPD) complexes. Gene
Therapy 1997; 4: 891–900.
10 Hong K, Zheng W, Baker A, Papahadjorpoulos D. Stabilization
of cationic liposome–plasmid DNA complexes by polyamine
and poly(ethyleneglycol) phospholipid conjugates for efficient
in vivo gene delivery. FEBS Lett 1997; 400: 233–237.
11 McLean JW et al. Organ-specific endothelial cell uptake of cationic liposome–DNA complexes in mice. Am J Physiol 1997; 273:
H387-H404.
12 Templeton NS et al. Improved DNA: liposome complexes for
increased systemic delivery and gene expression. Nat Biotech
1997; 15: 647–652.
13 Wang JK et al. Synthesis and characterization of long chain alkyl
acyl carnitine esters. Potentially biodegradable cationic lipids for
use in gene delivery. J Med Chem 1998; 41: 2207–2215.
14 Barron L, Mayer KB, Szoka FC Jr. Effects of complement
depletion on the pharmacokinetics and gene delivery mediated
by cationic lipids–DNA complex. Hum Gene Ther 1998; 9: 315–
323.
15 Mahato RI et al. Biodistribution and gene expression of
lipid/plasmid complexes after systemic administration. Hum
Gene Ther 1998; 9: 2083–2099.
16 Barron L, Gagne L, Szoka FC Jr. Lipoplex-mediated gene delivery to the lung occurs within 60 minutes of intravenous administration. Hum Gene Ther 1999; 10: 1683–1694.
17 Gao X, Huang L. Cationic liposome-mediated gene transfer.
Gene Therapy 1996; 3: 710–722.
18 Felgner PL et al. Lipofection: a highly efficient, lipid-mediated
DNA-transfection procedure. Proc Natl Acad Sci USA 1987; 84:
7413–7417.
19 Leventis R, Silvius JR. Interactions of mammalian cells with
lipid dispersion containing novel metabolizeable cationic
amphiphiles. Biochim Biophys Acta 1990; 1023: 123–132.
Gene Therapy
20 Ren T, Liu D. Synthesis of cationic lipids from 1,2,4-butanetriol.
Tetrahedron Lett 1999; 40: 209–212.
21 Ren T, Liu D. Synthesis of diether-linked cationic lipids for gene
delivery. Bioorg Med Chem Lett 1999; 9: 1247–1250.
22 Lee RJ, Huang L. Lipid vector system for gene transfer. Crit Rev
Therapeutic Drug Carrier Systems 1997; 14: 173–206.
23 Miller AD. Cationic liposomes for gene therapy. Angew Chem
Int Ed Engl 1998; 37: 1768–1785.
24 Byk G et al. Synthesis, activity, and structure–activity relationship studies of novel cationic lipids for DNA transfer. J Med
Chem 1998; 41: 224–235.
25 Behr JP et al. Efficient gene transfer into mammalian primary
endocrine cells with lipopolyamine-coated DNA. Proc Natl Acad
Sci USA 1989; 86: 6982–6986.
26 Felgner JH et al. Enhanced gene delivery and mechanism studies
with novel series of cationic lipid formulation. J Biol Chem 1994;
269: 2550–2561.
27 Wheeler CJ et al. A novel cationic lipid greatly enhances plasmid
DNA delivery and expression in mouse lung. Proc Natl Acad Sci
USA 1996; 93: 11454–11459.
28 Balasubramaniam RP et al. Structure and functional analysis of
cationic transfection lipids: the hydrophobic domain. Gene Therapy 1996; 3: 163–172.
29 Bennett MJ et al. Cationic lipid-mediated gene delivery to
murine lung: correlation of lipid hydration with in vivo transfection activity. J Med Chem 1997; 40: 4069–4078.
30 Gao X, Huang L. A novel cationic liposome reagent for efficient
transfection of mammalian cells. Biochem Biophys Res Commun
1991; 179: 280–285.
31 Lee ER et al. Detailed analysis of structures and formulations of
cationic lipids for efficient gene transfer to lung. Hum Gene Ther
1996; 7: 1701–1717.
32 Cooper RG et al. Polyamine analogues of 3␤-[N-(N′,N′dimethylaminoethane) carbamoyl]cholesterol (DC-Chol) as
agents for gene delivery. Chem Euro J 1998; 4: 137–151.
33 Huang L, Hung M, Wagner E (eds). Nonviral Vectors for Gene
Therapy. Academic Press: San Diego, 1999.
34 Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press: New York,
1989, pp 1.21–1.24.