788
T. Fichert et al. / Bioorg. Med. Chem. Lett. 10 (2000) 787±791
suspension in a white 96-well microtiterplate and by sub-
sequently measuring the enzymatic activity over a period
of 10 s in a microtiterplate luminometer. Luciferase
activity was expressed as RLU/mg of protein and com-
pared to that of DOTAP lipoplexes used as standard.
Discussion
Transfection pro®les
The transfection eciencies (TE) of all lipids, in depen-
dence on the lipid/DNA charge ratios, are presented in
Table 2. Using our lipids for transfection experiments
with COS-7 cells, apparently no cytotoxic side eects
could be observed. (This was concluded from the
amount of cellular protein, determined after transfec-
tion, which is closely related to the cytotoxicity of the
respective kind of lipid.) The ability of forming lipo-
somes (bilayer structures) seems to be a crucial pre-
requisite for high TE which is illustrated by the low TE
of the compounds 3 and 5. These two lipids do not form
liposomes and therefore show only marginal TE, an
observation which is consistent with previous results
(data not shown). For most lipids, the highest TE were
obtained at lipid/DNA charge ratios of about 3:1. Such
a typical transfection pro®le is depicted in Figure 3
(lipid 8). As an exception, the transfection pro®les of the
compounds 7 and 9 are characterized by a dierent
pro®le described in Figure 4 (lipid 7). Using these two
lipids no distinct maximum at the typical lipid/DNA
charge ratio could be observed. Instead, TE rose stea-
dily with increasing lipid/DNA charge ratio, leading to
a rather monotonous pro®le. We suppose that the lipo-
plexes containing lipids 7 and 9 are taken up into the
cell by a dierent mechanism, possibly by membrane
fusion and not by endocytosis.
Figure 1. Structures of the synthesized monocationic lipids.
Table 1.
Comp.
Lipid anchor
Spacer
Head group
5
8
9
6
10
11
12
136
14
15
3
Chol
Chol
Chol
Chol
Chol
Chol
Chol
Chol
Chol
Chol
DMG
DMG
Ac
Ac
Ac
C2
C2
C2
SO2
SN2
SO3
SN3
Ac
NMe2
NMe3
NMe2EtOH
NMe2
NMe3
NMe2EtOH
NMe2
NMe2
NMe2
NMe2
NMe2
NMe3
7
Ac
Lipoplex formation and cell transfection
Plasmid DNA (pCMVluc, containing a ®re¯y luciferase
gene driven by a CMV promoter) dissolved in HEPES
buer (pH 7.4) was added to eight dierent dilutions of
each liposomal dispersion, ®nally leading to eight dif-
ferent lipid/DNA charge ratios (from 1:1 to 15:1) at a
constant DNA concentration (125 ng/well, 96-well
plate). The DNA and the liposome dispersion were
gently mixed and allowed to stand for 1 h at rt to form
lipid/DNA complexes (lipoplexes). Twenty-four hours
prior to transfection, COS-7 cells, maintained in
EMEM/10% fetal calf serum/1% penicillin-streptomy-
cin at 37 ꢀC/5% CO2, were seeded in 96-well plates at
5000 cells per well. One hour before lipoplexes were
given to the cells, 50% of the cell culture medium was
removed from each well to be later replaced by an equal
volume (90 ml) of lipoplex dispersion. All experiments
were performed in triplicates, i.e., cells in three dierent
wells were transfected by the same lipoplex dispersion.
Cells were incubated for 4 h at 37 ꢀC, then the medium
was completely removed and replaced by fresh medium.
In¯uence of the lipid anchor
In this study cholesterol was chosen as a lipid anchor
because it represents a naturally occurring structure
which should be biodegradable and therefore non-toxic.
There are some hints that cholesterol is also responsible
for enhanced TE10 12 which is consistent with our results:
Compound 7 (containing the 1,2-dimyristyloxypropan-
3-ol anchor) and compound 8 (containing the choles-
terol anchor) have the acetyl spacer and the quaternary
ammonium head group in common. Comparing the
transfection results of these two compounds which dier
in their apolar subunits, the cholesterol derivative 8 shows
a higher TE than the DMG derivative 7 (see Figs. 3 and 4).
In¯uence of the spacer moiety
The TE was greatly in¯uenced by the spacer moiety due
to the inductive eect of the ester group on the amino
group and due to its polarity and length. The chemical
structure of the spacer in¯uences the protonation state
of the tertiary amino head group13 and therefore the
transfection properties of the lipid. From our results we
conclude that the acetyl spacer seems to decrease the pK
of the tertiary amine due to its strong inductive eect
Luciferase and protein assay
Forty-six hours post transfection, cells were washed and
lysed. Using equal volumes of the cell lysate, a reporter
gene assay and a BCA protein assay were performed.
Luciferase activity was determined by adding 80 ml of
luciferin reagent (25 mM glycylglycine, 5 mM ATP,
0.2 mM luciferin) to 20 ml of the homogenized cell