Lipopolycationic telomers for gene transfer: Synthesis and evaluation of their in vitro transfection efficiency

Article abstract of DOI:10.1021/jm9911579

We report on the synthesis of a series of lipopolyamine telomers I-14,n, I-18,n, and II-18,n and on their in vitro gene-transfer capability. Their structure consists of a polyamine polar moiety, including n primary amine functions (from 1 to 70), connected to a hydrophobic moiety, including two hydrocarbon C14 or C18 chains, through a mercaptopropanoyl or mercaptoglyceryl unit and an amide or ether function. They were obtained by telomerization of N-{2-[(BOC)-aminoethyl]}acrylamide with N,N-ditetradecyl- and N,N-dioctadecylpropanamide-3-thiol and rac-1,2-dioctadecyloxypropane-3- thiol, respectively, then BOC deprotection. For N/P ratios (N = number of telomer amine equivalents; P = number of DNA phosphates) from 0.8 to 10, these lipopolyamines condensed DNA, with or without the use of DOPE, forming lipopolyplexes or teloplexes of mean sizes less than 200 nm. Some trends, structure-activity and structure-toxicity relationships, were established toachieve both highest in vitro transfection levels and cell viability. Thus, DNA formulations based on telomers I-14,20 and I-18,20 and for N/P ratios lower than 5 led to the most efficient teloplex formulations for plasmid delivery to lung epithelial A549 cells.

Full text of DOI:10.1021/jm9911579

J . Med. Chem. 2000, 43, 1367-1379
1367
Lip op olyca tion ic Telom er s for Gen e Tr a n sfer : Syn th esis a n d Eva lu a tion of
Th eir in Vitr o Tr a n sfection Efficien cy
Ge´raldine Verderone,† Nathalie Van Craynest,† Otmane Boussif,‡ Catherine Santaella,†,§ Rainer Bischoff,‡,#
Hanno V. J . Kolbe,‡ and Pierre Vierling*,†
Laboratoire de Chimie Bio-Organique, ESA 6001 CNRS, Universite´ de Nice Sophia-Antipolis, Faculte´ des Sciences,
06108 Nice Ce´dex 2, France, and Transge`ne SA, 11 rue de Molsheim, 67082 Strasbourg, France
Received September 8, 1999
We report on the synthesis of a series of lipopolyamine telomers I-14,n , I-18,n , and II-18,n
and on their in vitro gene-transfer capability. Their structure consists of a polyamine polar
moiety, including n primary amine functions (from 1 to 70), connected to a hydrophobic moiety,
including two hydrocarbon C14 or C18 chains, through a mercaptopropanoyl or mercaptoglyceryl
unit and an amide or ether function. They were obtained by telomerization of N-{2-[(BOC)-
aminoethyl]}acrylamide with N,N-ditetradecyl- and N,N-dioctadecylpropanamide-3-thiol and
rac-1,2-dioctadecyloxypropane-3-thiol, respectively, then BOC deprotection. For N/ P ratios (N
) number of telomer amine equivalents; P ) number of DNA phosphates) from 0.8 to 10, these
lipopolyamines condensed DNA, with or without the use of DOPE, forming lipopolyplexes or
teloplexes of mean sizes less than 200 nm. Some trends, structure-activity and structure-
toxicity relationships, were established to achieve both highest in vitro transfection levels and
cell viability. Thus, DNA formulations based on telomers I-14,20 and I-18,20 and for N/ P
ratios lower than 5 led to the most efficient teloplex formulations for plasmid delivery to lung
epithelial A549 cells.
In tr od u ction
Although viral vectors are very efficient for gene
delivery, their use suffers from a number of disadvan-
tages, including safety and practical issues relating to
quality control, high-cost, and problematic large-scale
production. Retroviral vectors can only transduce divid-
ing cells; they cannot accommodate large-sized genes:
the retroviral genome is integrated into host cell DNA
and may thus cause genetic changes in the recipient cell,
and infectious viral particles could disseminate within
the organism or into the environment. Adenoviral
vectors, which do not integrate in the host cell DNA,
need, for long-term expression, to be administered
repeatedly and thus can induce a strong immune
response in treated patients.1
Gene therapy has been conceived as principally ap-
plicable to inherited deficiency diseases (cystic fibrosis,
dystrophies, hemophilias, etc.) where permanent cure
may be effected by introducing a functional gene.
Acquired diseases (cancer, AIDS, multiple sclerosis, etc.)
might also be treatable by transiently engineering host
cells to produce beneficial proteins. Another application
of gene therapy is DNA vaccination.
Successful gene therapy depends on the efficient
delivery of genetic material to cells of a living organism
and its effective expression within these cells. A func-
tional nucleic acid can be introduced into cells by a
variety of techniques resulting in either transient
expression of the gene of interest or permanent trans-
formation of the host cells resulting from incorporation
of the nucleic acid into the host genome.1 Viruses have
developed diverse and highly sophisticated mechanisms
to achieve this goal, including cell targeting, crossing
of the cellular membrane by fusion or endocytosis,
escape from endosomes and lysosomal degradation, and
finally delivery of their genome to the nucleus followed
by expression of the viral genome. In consequence, most
gene-delivery systems used to date and applied to
humans involve recombinant viral vectors, especially
adeno- and retroviral vectors.
Synthetic nonviral gene-transfer vectors, which are
free from the risks associated with viral vectors, will
be of considerable potential to the gene therapy field.
Over the past decade, a large variety of (poly)cationic
lipids, liposomes, and (lipo)polymers, eventually associ-
ated to molecular conjugates for improving cell target-
ing, cytoplasmic delivery, and/or nuclear transport, have
been used extensively to deliver genes to a large variety
of cell lines and tissues (see refs 2-5 and references
therein). These (poly)cationic systems are capable of
interacting with anionic DNA, condensing or compacting
DNA into small-sized complexes (e.g. lipoplexes or
polyplexes), neutralizing its negative charges, and thus
favoring its entry into the cell. Although the transfection
capabilities of such “artificial viruses” are far below
those of viral vectors, these systems present potential
advantages for the future with respect to low-cost and
large-scale production, safety, cell targeting, lower im-
munogenicity, and capacity to deliver large fragments
of DNA.
* Corresponding author. Tel: 33 (0)4 92 07 61 43. Fax: 33 (0)4 92
07 61 51. E-mail: vierling@unice.fr.
† Universite´ de Nice Sophia-Antipolis.
‡ Transge`ne SA.
§ Present address: Laboratoire d’Ecologie Microbienne de la Rhizo-
sphe`re, UMR CNRS-CEA 163, DSV/DEVM, CEN de Cadarache, 13108
Saint Paul lez Durance, France.
#
Present address: ASTRA-DRACO AB, Dept. of Bioanalytical
Chemistry, P.O. Box 34, S-221 00 Lund, Sweden.
10.1021/jm9911579 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/21/2000

1368 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7
Verderone et al.
Sch em e 1. Synthesis of Telogens L1-SH 3
compounds being particularly linked to the initial
taxogen/telogen molar ratio (Ro) makes it possible to
control and modulate the number of amine functions of
the telomers. Theoretically, if the transfer constant is
equal to unity, the aDPn should be equal to Ro.
The new cationic lipopolyamine telomer compounds
show a modular structure. They consist of a polar
moiety, including a variable number of primary amine
functions, connected to a hydrophobic moiety, including
two hydrocarbon chains of variable length (C18 or C14
chains), through various chemical units (mercaptopro-
panoyl, mercaptoglyceryl) and functions (amide, ether).
Their modular structure is aimed at modulating their
physicochemical and biological properties (e.g. chemical
stability, biodegradability, toxicity, and transfection
efficacy). A hydrophobic moiety composed of two long-
aliphatic chains has been introduced as several struc-
ture-function studies have shown a correlation of the
transfection capability with the hydrophobicity and/or
the physical state of the bilayers that theoretically the
lipid forms. For instance, cell transfection has been
observed with lipopolylysines but not with polylysines8
and with double-chain lipospermines but not with
single-chain lipospermines.9 A higher cell transfection
capability has been observed for some (poly)cationic
lipids and co-lipids which form, at the incubation
temperature, liquid-crystal rather than crystal bilay-
ers.10,11 As the use of co-lipids such as DOPE12,13 and
cholesterol14 (which when used alone do not enhance
gene transfer) was found in several cases to enhance
transfection activity, we report also on the transfection
efficacy of the telomers used in conjunction with DOPE.
F igu r e 1. Molecular structure of the lipopolycationic telomers
and of the pcTG90/DOPE reference gene-transfer lipids used
in this study.
Several studies have shown that the transfer ef-
ficiency of the complexed DNA into the cells, especially
in the case of in vivo transfer, can vary in function of
the interactions between the complexes and the cell
membranes (especially with anionic proteoglycans ex-
pressed at the cell surface6,7), the cell type involved, the
cationic components, the lipid composition of the cationic
components, the size of the complexes, and, more
particularly, the ratio of the positive to negative charges
of the different components of the complex. Currently,
little is also known concerning the intracellular traffic
after endocytosis of the DNA complexes, and the ongoing
research remains highly empirical. Consequently, there
is a need to develop new compounds, especially cationic
compounds, with characteristics and properties different
from those already described.
We report herein on the synthesis of new lipopolyamine
telomers (Figure 1) and on their capabilities of compact-
ing and transferring DNA into a cell as well as their
cellular compatibility. The term “lipopolyamine telomer”
designates products obtained from telomerizing (i) one
polymerizable monomer (termed taxogen M), for ex-
ample an acrylamide derivative comprising a (poly)-
amine part or one of its precursors, in the presence of
(ii) a transfer agent (termed telogen), for example
hydrophobic long-chain alkane thiols (RS-H), and of (iii)
an initiator, which could be a radical generator, for
example R,R′-azobis(isobutyronitrile) (AIBN). The li-
popolyamine telomers thus obtained are compounds of
formula RS-(M)n-H showing low degrees of polymeriza-
tion (1 e n < 200) and hence rather low molecular
weights, whereas compounds obtainable with classic
polymerization processes fit (M)n structure with high
polymerization degrees (n . 200) and hence high
molecular weights. Moreover, the average degree of
polymerization (aDPn) of the lipopolyamine telomer
Resu lts a n d Discu ssion
Telogen Syn th esis. Telomerization reactions are
more easily performed with transfer agents deriving
from primary thiols rather than from sterically hindered
secondary or tertiary thiols. Therefore we selected
telogens L1-SH 3 and L2-SH 7, the syntheses of which
are outlined in Schemes 1 and 2, respectively.
Two methods were tested for the synthesis of telogens
L1-SH. The coupling of the amine directly to unpro-
tected mercaptopropionic acid (step b1 in Scheme 1)
afforded in one step L1-SH 3a in ∼30% yield. The more
selective, but time-consuming, two-step procedure which
relies on the amidation of 3,3′-dithiodipropanoyl chloride
followed by the reduction of the disulfide (steps a1/a2
in Scheme 1) led to L1-SH 3b in an overall yield of
∼40%.

Lipopolycationic Telomers for Gene Transfer
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7 1369
Sch em e 2. Synthesis of Telogens L2-SH 7
respect to particle size measurements by quasi-elastic
light scattering (less than 250 nm), absence of “free”
plasmid upon gel electrophoresis, and reproducibility,
the dilution method gave the best results (data not
shown) and was therefore applied for the formulation
of the teloplexes. Table 2 collects the mean particle
diameters measured for the different teloplexes.
All the polyamine telomer compounds containing
6-60 primary amines are able, with or without the
addition of an equimolar amount of DOPE and for N/ P
ratios of 10, 5, and 2.5, to condense DNA into teloplexes
of a mean size of less than 200 nm. In most cases, the
preparations consist in a single population of particles.
Table 2 shows also that smaller (<100 nm) complexes
can be obtained frequently for the higher N/ P ratios.
For telomers bearing a number of amine groups below
6, one often observes a polydisperse population of large-
sized teloplexes that further tend to precipitate. The
number of primary amine groups seems indeed to affect
the size of the complexes: I-18,2 with its two amine
groups led, without DOPE, to teloplexes of much larger
mean size than those obtained with I-18,n (n ) 10, 20,
40, 50, or 60) or with II-18,n (n ) 6, 11, or 16).
Concerning the impact of the chemical unit connecting
the hydrophobic chains to the polar head on teloplex
size, one can observe that the mercaptoamidopropanoyl-
and mercaptoglyceryl-derived telomers behave very
similarly (see Table 2, entries I-18,10/II-18,n (n ) 6 or
11) and entries I-18,20/II-18,16).
No impact on teloplex size related to the length of the
hydrophobic chains could be demonstrated, telomers
I-18,n and I-14,n leading indeed to complexes of com-
parable size.
Telogen L2-SH (compound 7 in Scheme 2) was ob-
tained in 40% yield using a three-step procedure which
consists, starting from 3-mercaptopropane-1,2-diol, in
the selective S-benzylation followed by di-O-alkylation
by phase-transfer catalysis and then S-deprotection.
Telom er Syn th esis. All telomerization reactions
(Scheme 3) were performed under nitrogen by refluxing
in acetonitrile a mixture of taxogen A and the thiol
telogen 3 or 7 in fixed taxogen/telogen concentration and
ratio (Ro). When the temperature was stabilized, half
of the AIBN was added, the other half being added 1 h
after. At the end of the reaction, the solvent was
evaporated under reduced pressure and the residue
purified and fractionated by chromatography on a silica
column and gel permeation chromatography on a Sepha-
dex LH20 column. This allows to separate the N-BOC-
protected telomers from AIBN and residual disulfide
which has formed during telomerization. The BOC-
telomers thus obtained were then deprotected in excess
TFA at room temperature. The (poly)cationic telomers
were isolated as TFA salts by evaporation of excess TFA
in the presence of cyclohexane followed by lyophiliza-
tion. Quantitative deprotection was confirmed by 1H
NMR (absence of the signal corresponding to the methyl
BOC protons of the starting material). For each telogen
as well as Ro, the different telomers isolated and their
code names are reported in Table 1.
For each fraction collected, the aDPn values were
determined by 1H NMR on the BOC-protected telomers
by comparing the integration of the methylene NCH2-
CH2NHBOC protons (Hx) with that of the terminal
methyl chain protons (Hy): aDPn ()n) ) 3Hx/2Hy.
Electrospray ionization mass spectrometry (ESI-MS)
was further used to confirm the structure and mono-
dispersity of compounds I-18,1(BOC), I-18,2(BOC),
I-18,1, I-18,2, and II-18,1 and the polydispersity of
telomers I-18,n (BOC), I-14,n (BOC), II-18,n (BOC),
I-18,n , I-14,n , and II-18,n with n > 2.
No rationale concerning complex formation and sta-
bility (in terms of particle size evolution) can be deduced
from the use of DOPE as co-lipid with the telomers.
Indeed, and in several cases, the use of DOPE was found
either to have no impact on teloplex size or to lead to a
teloplex size and polydispersity increase and/or to
aggregation. In a few cases only, it led to smaller
teloplexes.
In Vitr o Tr a n sfection . The transfection potency of
the telomers, with and without DOPE, was assayed in
vitro on lung epithelial A549 cells, from human pulmo-
nary carcinoma. These assays were performed using
plasmid pTG11033 (pCMV-intronHMG-luciferase-SV40-
pA; 9572 bp) in the presence of 10% fetal calf serum for
24 h. All teloplex formulations that did not precipitate
after 15-18 h at 4 °C were tested. The transfection
efficiency of the teloplexes (expressed in fg of luciferase/
mg of protein) was evaluated for gradual N/ P ratios
(0.8, 1.25, 2.5, 5, and 10) and various amounts of DNA
(4, 2, 0.5, and 0.1 µg/well) as compared to naked DNA
and to the lipoplexes based on pcTG90/DOPE (1/1 molar
ratio). The cationic pcTG90 amphiphile (see structure
in Figure 1) is among the most efficient lipid for the
transfection of A549 cells, the highest level of luciferase
expression being observed for the lipoplexes formulated
together with DOPE and for a N/ P ratio of 5. [Under
the same conditions (plasmid and the A549 cells), but
with Transfectam alone or Transfectam/DOPE, lu-
ciferase expression levels between 105 and 106 or 106
and 5 × 106 fg of luciferase/mg of protein were mea-
sured, respectively (J . Gaucheron, C. Santaella, and P.
(Lip o)p olyp lex F or m a t ion a n d Ch a r a ct er iza -
tion . The capability of the polyamine telomer com-
pounds to condense DNA and to form lipopolyplexes or
teloplexes was analyzed with or without an equimolar
amount of the helper phospholipid DOPE. These studies
were performed with pTG11033 plasmid, which has
been used for our in vitro transfection assays (see
below). A screening procedure was applied for the
teloplex preparation which relies on three formulation
methods (dilution from DMSO/EtOH solution in buffer,
detergent dialysis, and extrusion) using N/ P ratios of
10, 5, 2.5, 1.25, and 0.8 (N ) number of amine functions
of the telomer; P ) number of DNA phosphates). With

1370 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7
Sch em e 3. Synthesis of Telomers I-14,n , I-18,n , and II-18,n
Verderone et al.
Ta ble 1. Telomerization of Telogens 3 and 7 with Taxogen A
and Code Names of the Isolated Telomers (after BOC
deprotection)a
Ta ble 2. Mean Sizes, Determined by Light Scattering
Spectroscopy, of the Teloplexes Formed by the Dilution Method
Applied to Plasmid pTG11033 and the Telomers I-m ,n and
II-18,n
telogen
Ro
aDPn (n)
telomer code name
teloplex
3b
10
2
10
20
40
50
60
10
20
70
6
I-18,2
mean diameter in nm (SD)
I-18,10
I-18,20
I-18,40
I-18,50
I-18,60
I-14,10
I-14,20
I-14,70
II-18,6
II-18,11
II-18,16
40
telomer
n
N/ P ratioa
alone
with DOPE
I-14,n 10
10
10
10
10
10
10
10
10
10
10
10
10
5
ntd
90 (15)
37 (100)
20
70
55 (20)
90 (15)
2640 (250)
55 (15)
60 (15)
65 (20)
65 (15)
70 (25)
90 (20)
50 (15)
50 (10)
ntd
75 (20)
85 (25)
3a
7
40
3
I-18,n
2
10
20
40
50
60
6
55 (20)/7150 (700) 75/25%b
55 (15)
70 (15)
11
16
ntd
55 (20)
100 (30)
100 (40)
70 (15)
a
II-18,n
Ro ) [taxogen A]/[telogen]; aDPn ) average degree of poly-
merization as determined by 1H NMR for the corresponding BOC-
protected telomers.
11
16
I-14,n 10
110 (40)
20
70
5
5
5
5
5
5
5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
60 (20)
90 (25)
60 (50)
75 (20)
60 (25)
60 (20)
120 (40)
95 (30)
2500 (250)
175 (20)
80 (15)
90 (25)
pc
85 (15)
65 (15)/700 (90) 75/25%b
Vierling, unpublished results). The Transfectam li-
pospermine is a highly efficient synthetic gene-transfer
agent widely documented in the literature; see refs 3
and 9.] Cells treated with naked DNA under equivalent
conditions showed expression levels of about 102 fg of
luciferase/mg of protein. The toxicity of the teloplexes
was also checked by determining the total protein
amount per well of the transfected cells relative to that
measured for untreated cells (for which the total protein
amount per well is in a 30-60 µg/well range).
The transfection and toxicity results are illustrated
in Figures 2-4 for the teloplexes formulated with
telomers I-14,n , I-18,n , and II-18,n , respectively. These
results indicate that the polyamine telomers do enable
transfection of the plasmid into the cells. All the
teloplexes tested so far showed indeed a higher ef-
ficiency to transfect A549 cells than naked DNA. For
the lowest amount of plasmid used (i.e. 0.1 µg/well), the
luciferase expression levels stood between 102 and 104
fg of luciferase/mg of protein. Furthermore, some for-
mulations based on I-18,2 telomer already peaked out
at 105-106 fg of luciferase/mg of protein for such a low
DNA amount (see Figure 3).
I-18,n 10
60 (20)/1100 (230) 85/15%b
20
50
60
55 (15)
ntd
55 (15)
80 (20)
pc
II-18,n 16
I-14,n 10
20
70
pc
125 (110)
85 (15)
80 (20)
115 (30)
I-18,n 10
20
40
50
150 (110) 80%b ntd
1300 (300) 20%b
60
2.5
220 (50)
200 (200)
2500 (960)
450 (680)
pc
70 (25)
I-14,n 10
1.25
1.25
1.25
0.8
0.8
0.8
pc
20
70
pc
pc
I-14,n 10
640 (100)/150 (40) 67/33%b
400 (80)/135 (35) 54/46%b
1400 (700)
20
70
pc
260 (70)
a
N ) amine equivalents; P ) plasmid phosphate equivalents.
Percent (%) of the overall population. c p ) precipitate. nt )
b
d
not tested.
The transfection efficiency of the teloplexes usually
increased on increasing the DNA amount from 0.1 to 2
µg/well, for which a plateau of luciferase expression
seemed to be reached. This plateau was most often
attained for a DNA amount of 0.5 µg/well. Concomi-
tantly, their cell toxicity also increased (see Figures
2-4). It remained, however, very low for DNA below
0.5 µg/well. For higher DNA amounts (e.g. 4 µg/well),
the transfection, even if higher, was in most cases
The transfection efficiency and the cell tolerance of
the teloplexes depend further on the amount of DNA,
on the N/ P ratio, on the series of telomers (structure I
or II, polymerization degree), and on the teloplex
formulation (DOPE added or not). Some of these pa-
rameters are linked and will be discussed together.

Lipopolycationic Telomers for Gene Transfer
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7 1371
F igu r e 2. Luciferase expression (bars) and cell viability (points) of the teloplexes made of pTG11033 and I-14,n telomers (A) or
I-14,n /DOPE (1/1 mol) lipids (B), as compared to the reference pcTG90/DOPE (1/1 mol; N/ P ) 5) in A549 cells. Means ( SEM are
given.
accompanied by a poor cell survival, except for the
lipoplexes based on the telomer II series (Figure 2).
Whatever the amount of plasmid used, luciferase
expression levels usually grew when the N/ P ratio
increased from 0.8 to 5 and then slumped. Simulta-
neously, cell tolerance decreased with the N/ P ratio
increase, slowly for DNA amounts below 2 µg/well and
dramatically when the plasmid amount exceeded 2 µg/
well. This transfection efficiency and toxicity increase
when raising the cationic charge of the teloplexes is
most probably related to an increase of the teloplex
cellular uptake which remains at a level compatible with
cell growth providing the N/ P ratio stayed below 5. For
N/ P ratios g 5, the teloplexes became too toxic for the
cells, likely as a result of a high level of cellular uptake
and/or of cell membrane disruption and/or as a result
of the presence of a high amount of “free” uncomplexed
telomer. As a consequence, reduced cell viability re-
sulted in a drop of luciferase expression. A transfection
optimum was reached for plasmid amounts between 0.5
and 2 µg/well for the I-14,n and I-18,n /DOPE teloplexes,
at N/ P ratios of 0.8-2.5, and for the I-14,n /DOPE and
I-18,n teloplexes, for N/ P ratios of 0.8-5.
Likewise, some trends were found for a moderate
toxicity on the A549 cell line. The amount of viable cells
increased while DNA amount dropped or when the N/ P
ratio declined. In most cases, the amount of DNA for
which 50% of cell growth inhibition can be observed lies
between 2 and 4 µg/well for N/ P e 2.5 and between 0.5
and 2 µg/well for N/ P g 5. An exception are the
teloplexes based on the telomer II series. These latter
formulations displayed indeed higher cell viabilities that
were barely affected by the amount of DNA or by the
N/ P ratio increase. However and likely related, the
luciferase expression levels remained low: among the
32 formulations based on telomers II tested, only eight
of them leveled out at 105-106 fg of luciferase/mg of
protein (see Figure 3 and discussion below).
When compared to the pcTG90/DOPE lipoplex (N/ P
) 5) reference which displays a high transfection
efficiency (106-107 fg of luciferase/mg of protein) already
for a DNA amount of 0.1 µg/well, only a few of the
formulations investigated here, i.e. I-18,2 (N/ P ) 0.8),
I-14,10 (1.25), I-14,20 (2.5), I-14,20/DOPE (2.5), and
I-18,20/DOPE (2.5), were found to be as efficient in
mediating DNA transfer and expression (see Figures 2

1372 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7
Verderone et al.
F igu r e 3. Luciferase expression (bars) and cell viability (points) of the teloplexes made of pTG11033 and I-18,n telomers (A) or
I-18,n /DOPE (1/1 mol) lipids (B), as compared to the reference pcTG90/DOPE (1/1 mol; N/ P ) 5) in A549 cells. Means ( SEM are
given.
and 3). However, and with respect to cell toxicity, it is
particularly noteworthy that these teloplexes, for DNA
amounts e 0.5 µg/well and over a range of N/ P ratios
lower than 5, allowed luciferase expression levels as
high as the reference but with lower cell toxicity. These
teloplexes seem to be appealing candidates for in vivo
administration.
Concerning the effects on A549 cell survival and
transfection (which cannot be analyzed without consid-
ering the effects on toxicity) resulting from the formula-
tion of the teloplexes with DOPE, no obvious trend could
be evidenced, these effects being rather at random.
Indeed, as compared to their respective teloplexes
formed without DOPE, the use of DOPE either had no
effect on the cell viability for nearly 50% of the cases or
increased the toxicity for the other 50% (data not
shown). (The increase in cell toxicity resulting from
adding DOPE was considered significant if the ratio of
protein amount per well measured for the formulation
with DOPE vs that measured for the corresponding one
without DOPE was lower than 0.8.) The transfection
efficiency upon DOPE addition, irrespective of the
amount of plasmid used, showed (i) an increase for about
15 of the 100 formulations tested (which also implies
that their cell viability has not been affected), (ii) no
effect on both transfection and cell viability for 35%, and
(iii) a decrease of either transfection and/or cell viability
for another 40% range (data not shown). (The effect on
transfection of DOPE was considered significant if the
ratio of luciferase amount measured for the formulation
with DOPE respectively to that measured for the
corresponding one without DOPE was out of the 0.2-5
range.) Finally, for less than 10% of the formulations
investigated, an increase of transfection could be evi-

Lipopolycationic Telomers for Gene Transfer
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7 1373
F igu r e 4. Luciferase expression (bars) and cell viability
(points) of the teloplexes made of pTG11033 and II-18,n
telomers or II-18,n /DOPE (1/1 mol) lipids, as compared to the
reference pcTG90/DOPE (1/1 mol; N/ P ) 5) in A549 cells.
Means ( SEM are given.
F igu r e 5. Variation of the II-18,n -based teloplexe’s trans-
fection efficiency (bars) and cell viability (points) of the number
n of amine functions of the II-18,n telomers.
I-14,n ′ homologue is higher than 5 and comparable if
this ratio is in the 0.2-5 range.) Almost one-third of
the I-14,20 teloplexes displayed a higher transfection
efficiency than their respective I-14,10 analogues, while
one-fifth of the I-14,10 formulations led to higher
transfection levels than the I-14,20 ones. By contrast,
a significant higher level was obtained for 60% or 80%
of the I-14,10- or I-14,20-based teloplexes with respect
to their 17 corresponding I-14,70 formulations, a com-
parable level of transfection being measured for the
complementary 40-20% ones, respectively. This is
mainly the case for low N/ P ratios. A similar analysis
of the I-18,n formulations showed that up to 85% and
40-50% of the 10 and 19 teloplexes formulated with
I-18,2 or I-18,20 telomers, respectively, gave a higher
level of luciferase expression than their corresponding
ones made with the I-18,40, I-18,50, or I-18,60 telomers,
a comparable transfection being measured for the
complementary 15% and 60-50% ones, respectively. To
ensure an optimum transfection level one should prefer-
ably use I-18,2-based teloplexes rather than I-18,20
ones. Indeed, 40% of the I-18,2 formulations were more
efficient than their corresponding I-18,20 ones while the
remaining 60% behave very similarly. Finally, no sig-
nificant difference in transfection was found for a
number of amine functions in the 40-60 range.
denced by adding DOPE, together with a consequent
toxicity increase. We could not detect any improvement
in transfection or in cell viability resulting from the
formulation of the teloplexes with DOPE and a given
telomer or a series of telomers, with respect to their
polymerization degree.
Str u ctu r e-Activity Rela tion sh ip s. Although a
variety of (poly)cationic lipids mediate DNA transfer to
cells, few systematic studies have been performed to
assess structure-activity relationships of these lipids,
partly because the synthesis of the materials is labor-
intensive and time-consuming.15-17 Solid-phase chem-
istries and combinatorial library approaches have fa-
cilitated the rapid generation of a wide array of
polyamine lipids4 and peptoid-phospholipid conju-
gates.18 The “telomer” design can access a diverse family
of structures that differ in the number of amine func-
tions and in the hydrophobic moiety (chain length and
linker), thus enabling the establishment of structure-
activity relationships.
An optimum transfection level was measured for a
number of amine functions in the 10-20 and 2-20
range for telomers I-14,n and I-18,n , respectively. For
a given N/ P ratio and DNA amount, a close analysis of
the 34 formulations based on I-14,10 or I-14,20 telomers
showed indeed that the levels of luciferase expression
obtained either with I-14,10 or with I-14,20 were in half
of the cases comparable. (The effect of the amine
function number was considered to significantly in-
crease transfection if the ratio of the luciferase amount
measured for the formulations with telomers I-14,n vs
that measured for the corresponding ones with the
For the telomer series II-18,n , a significant effect of
n (transfection increases with raising n from 6 to 11 and
to 16) could be evidenced only for low amounts of DNA
(see Figure 5). Where the effect of the polymerization
degree on the teloplex toxicity is concerned, it was also
found to be related to its N/ P ratio. For low N/ P ratios,
i.e. <2.5, increasing n from 10-20 to 40-70 units did

1374 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7
Verderone et al.
mention that no effect on cell viability related to the
chain length could be detected. These results indicate
that optimum transfection levels are most likely ensured
with telomers having C18 chains preferably to C14 ones.
The nature of the linker and the structure (telomers
of series I as compared to the corresponding molecules
of series II) was also found to have an impact on toxicity
and on transfection as well. As already mentioned above
and as illustrated in Figure 7, almost all the ether II-
18,n -based teloplexes are well-tolerated by A549 cells,
even for a N/ P ratio as high as 10 and for a 4 µg/well
amount of DNA. By contrast, this is not the case for
their most closely related amido I-18,n - or I-14,n -based
teloplexes which display an acceptable cell tolerance
only for a much lower (0.5 µg/well) amount of DNA.
These results make it difficult to compare the transfec-
tion efficiency of these two series. Nevertheless, our
results strongly suggest a superior transfection ef-
ficiency of the amido series I-based teloplexes. Indeed,
provided some cells are still alive, higher luciferase
expression levels (in a range of 106-107 fg of luciferase/
mg of protein) are preferably obtained with the te-
loplexes of the amido series I rather than with the ether
II-18,n -based telomers (Figure 7). These gave rise most
frequently to moderate levels of luciferase expression
(below 105 fg of luciferase/mg of protein), even for
optimum conditions (high N/ P ratio and high amount
of DNA, i.e. 4 µg/well). The higher transfection efficiency
of the telomers of series I is likely confirmed by the high
luciferase expression levels (∼107 fg of luciferase/mg of
protein) of the I-18,20 teloplexes which are obtained for
a lower N/ P ratio (2.5) and for a DNA amount of 2 µg/
well, for which a good cell viability is found (see Figure
3).
F igu r e 6. Impact on transfection efficiency of the hydrophobic
chain length as estimated by the ratio of luciferase amount
measured for the formulations with telomer I-14,n vs that
measured for the corresponding ones with the I-18,n homo-
logues. The impact is significant if the ratio value is not within
the gray zone (see text).
Con clu sion
The present work suggests that our new telomers
appear as potential synthetic vectors for gene delivery.
The variation of structure parameters allowed the
establishment of structure-activity and structure-
toxicity relationships and the optimization of the te-
loplex formulation, where both highest transfection
levels and concomitant cell viability were achieved.
Thus, a polymerization degree centered around 20
amine units and a N/ P ratio lower than 5 leads to the
most efficient formulations for plasmid delivery to A549
cells. Some of these telomers are currently being inves-
tigated for in vivo studies.
not impair the cell survival and in some cases improved
it. Teloplexes formulated with telomers of very small
n, such as n ) 2, combined with a low N/ P ratio seemed
rather more toxic than those made with telomers of
higher polymerization degree. Thus, for both an opti-
mum transfection and a low toxicity with the telomers
reported here, one should use teloplexes formulated with
a telomer having a polymerization degree centered
around 20 and having a N/ P ratio lower than 5.
Exp er im en ta l Section
Previous work has shown that a single fatty chain in
lipopolyamines prevented transfection activity9 and that
the optimal length of the double fatty chain might vary
for different products.4,9,19 Figure 6 displays the ratios
of luciferase level measured for the formulations with
telomer I-14,n vs that measured for the corresponding
I-18,n homologues. Assuming that the effect of the chain
length (C14 or C18) is significant if this ratio is not in
the 0.2-5 range, we found that telomers I-18,n which
possess C18 chains are more efficient in mediating
transfection than their I-14,n counterparts for 42% of
36 formulations investigated. For 47% of them, we found
that both series of telomers were as efficient, and only
14% of telomers I-14,n gave rise to a higher luciferase
expression than their I-18,n analogues. One should
Gen er a l Exp er im en ta l a n d An a lytica l Con d ition s. All
the reactions were performed in anhydrous solvents under dry
and oxygen-free nitrogen. The purifications by column chro-
matography were carried out using silica gel 60 (Merck, 70-
230 mesh) and chloroform (CHCl3), dichloromethane (CH2Cl2),
methanol (MeOH), diethyl ether (Et2O), or mixtures thereof
as indicated. Advancing of the polymerization reaction was
followed by thin-layer chromatography (TLC) on silica plates
F
254 (Merck). The following developing systems were used: UV
light; KMnO4; H2SO4/MeOH; Dragendorff reagents; DTNB
(5,5′-dithiobis(2-nitrobenzoic acid)); ninhydrin reagent (Sigma).
IR spectra were recorded using a Bruker FT-ITS 45 spec-
trometer. Solid compounds were analyzed using the KBr disk
method; liquids were analyzed as thin film between two KBr
blocks. 1H, 13C, and 19F NMR spectra were recorded at 200,
50.3, and 188.3 MHz, respectively, on a Bruker AC-200.

Lipopolycationic Telomers for Gene Transfer
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7 1375
F igu r e 7. Impact on transfection efficiency (bars) and cell viability (points) of the linker (II-18,n -based teloplexes as compared
to their corresponding I-m ,n -based ones with m ) 14 or 18). DNA: 4 µg/well (A), 2 µg/well (B), 0.5 µg/well (C), and 0.1 µg/well (D).
Chemical shifts were measured relative to CHCl3 (δ 7.27 ppm)
or CH3OD (δ 3.35 ppm) for 1H, relative to CDCl3 (δ 76.9 ppm)
for 13C and expressed indirectly in relation to TMS, and
relative to CCl3F as internal reference for 19F. The following
abbreviations are used to describe the signal multiplicities: s
(singlet), d (doublet), t (triplet), q (quadruplet), and m (mul-
tiplet). Chemical shifts are expressed in ppm and listed as
follows: shift in ppm (multiplicity, integration, coupling, and
attribution). Mass spectra (MS) were recorded on a Finningan
MAT TSQ 7000 equipped with an atmospheric pressure
ionization (API) source. Electrospray ionization mass spec-
trometry (ESI-MS) was used depending on the polar moiety
of the molecules to be studied. This method used in positive
mode can give either a M + H+ or a M + Na+ signal. Molecules
were also analyzed on a VG-ZAB-HF mass spectrometer in
FAB mode or on a Fisons BioQ quadrupol, VG BioTech in
positive ionization electrospray mode. Elemental analyses were
carried out by the Service Central de Microanalyses du CNRS.
Syn th esis of Telogen N,N-Diocta d ecylp r op a n a m id e-
3-th iol, 3b (Sch em e 1). The synthesis of telogen 3b was
carried out in two steps as shown in Scheme 1.
3,3′-Dith iobis(N,N-d iocta d ecylp r op a n a m id e), 2b. To a
solution of 4.0 g (7.2 mmol) dioctadecylammonium hydrochlo-
ride (compound 1b; Fluka) and 8 mL (58 mmol) triethylamine
in 50 mL anhydrous CHCl3 was added dropwise 0.88 g (3.6
mmol) 3,3′-dithiodipropanoyl chloride (obtained by reaction of
SOCl2 with 3,3′-dithiodipropionic acid (Fluka)) in 3 mL anhy-
drous CHCl3 at room temperature. After 2 days at room
temperature, the reaction mixture was washed with an aque-
ous solution of 0.1 N HCl. The organic phase was concentrated
and the concentrate was chromatographed on a silica column
(eluent: CH2Cl2). 2.1 g (1.75 mmol, 49%) 2b was obtained in

1376 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7
the form of a white solid. TLC (hexane/Et2O 1/1, KMnO4) Rf
Verderone et al.
) 6.6 Hz, CH2OMs). 13C NMR (CDCl3): δ 14.1 (CH3); 22.7 (CH2-
CH3); 25.4 (CH2CH2O); 29.0,29.2, 29.3, 29.4, 29.5, 29.6 and 29.7
((CH2)13); 31.9 (CH2CH2CH3); 37.4 (CH3S); 70.2 (CH2OMs). To
this solution were added 0.94 g (2.8 mmol) tetrabutylammo-
nium hydrogenosulfate (Aldrich) and 25 mL of an aqueous
solution of 9 N KOH. The reaction mixture was kept under
Et2O reflux for 96 h. The organic phase was washed with H2O,
then evaporated, and the residue was purified by chromatog-
raphy on a silica column (eluent: petroleum ether/Et2O, 95/
5). 10.5 g (15 mmol, 76%) compound 6 was obtained in the
form of a white solid. TLC (petroleum ether/Et2O, 95/5, H2-
SO4) Rf ) 0.8. 1H NMR (CDCl3): δ 0.82 (t, 6H, 3J ) 6.0 Hz,
CH3); 1.10-1.36 (m, 60H, (CH2)15); 1.39-1.62 (m, 4H, CH2-
CH2O); 2.40-2.68 (m, 4H, CH2SBn); 3.30-3.52 (m, 7H, CH2O
and CHO); 3.68 (s, 2H, SCH2Ph); 7.11-7.35 (m, 5H, C6H5).
13C NMR (CDCl3): δ 14.1 (CH3); 22.7 (CH2CH3); 26.2, 29.4;
29.5; 29.7; 29.8 and 30.1 ((CH2)14); 32.0 (CH2CH2CH3); 32.9
(CH2SBz); 37.1 (SCH2Ph);70.4; 71.6 and 71.8 (CH2O); 78.6
(CHO); 126.9 and 129.0 (C ortho and meta); 128.4 (C para);
138.6 (quarternary C).
1
) 0.4. IR (ν cm-1, KBr): 1639 (CO). H NMR (CDCl3): δ 0.83
(t, 12H,3J ) 6.0 Hz, CH3); 0.90-1.40 (m, 120 H, (CH2)15); 1.40-
1.65 (m, 8H, CH2CH2NCO); 2.60-2.80 (m, 4H, CH2CON);
2.80-3.05 (m, 4H, CH2S); 3.10-3.40 (m, 8H, CH2N). 13C NMR
(CDCl3): δ 14.2 (CH3); 22.8 (CH2CH3); 27.1, 27.2, 27.7, 29.5,
29.6, 29.7, and 29.9 ((CH2)14); 32.0 (CH2CH2CH3); 33.5 (CH2S);
34.9 (CH2CO); 46.5 and 48.5 (CH2N); 170.4 (s, CO). MS-ESI
(M + Na+) m/z ) 1240 (theoretical value for C78H156N2O2S2 +
Na+ ) 1239).
N,N-Diocta d ecylp r op a n a m id e-3-th iol, 3b. To a solution
of 2.04 g (1.67 mmol) 2b in 10 mL acetic acid at 60 °C was
added 700 mg (10.7 mmol) zinc powder in 4 portions during a
period of 3 h. The reaction mixture was stirred at 60 °C for 3
h, then filtered hot. CHCl3 was added and washing with H2O
was carried out until neutrality was reached. The organic
phase was dried over Na2SO4, then filtered and evaporated.
The obtained residue was chromatographed on a silica column
(eluent CHCl3). 1.67 g (2.73 mmol, 82%) telogen 3b was
obtained in the form of a white solid. TLC (hexane/Et2O 2/3,
1
DTNB) Rf ) 0.6. H NMR (CDCl3): δ 0.80 (t, 6H,3J ) 6.0 Hz,
r a c-1,2-Diocta d ecyloxyp r op a n e-3-th iol, 7. 2.3 g (3.3
mmol) compound 6 was solubilized in 35 mL anhydrous
tBuOH. 0.86 g (38 mmol) Na was added to this solution. The
reaction mixture was heated under reflux for 12 h. Then
several drops of H2O were added. After evaporation of the
solvent, the residue was taken up in hexane, then washed with
H2O. The organic phase was concentrated and the residue was
purified by chromatography on a silica column (eluent: CHCl3)
to yield 1.6 g (2.6 mmol, 78%) compound 7 in the form of a
white solid. TLC (CHCl3, DTNB) Rf ) 0.7. H NMR (CDCl3):
δ 0.80 (t, 6H, J ) 6.0 Hz, CH3); 1.00-1.37 (m, 60H, (CH2)15);
1.37-1.70 (m, 4H, CH2CH2O); 2.48-2.76 (m, 2H, CH2S); 3.27-
3.65 (m, 7H, CH2O and CH). 13C NMR (CDCl3): δ 14.0 (CH3);
22.6 (CH2CH3); 26.1 (CH2S); 29.3, 29.4, 29.6, 29.7, 30.0 and
31.9 ((CH2)15); 70.3, 71.0 and 71.6 (CH2O); 79.3 (CH).
CH3); 1.00-1.35 (m, 60 H, (CH2)15); 1.35-1.60 (m, 4H, CH2-
3
3
CH2N); 1.63 (t, 1H, J ) 8.0 Hz, SH); 2.55 (t, 2H, J ) 6.5 Hz
CH2CON); 2.65-2.85 (m, 2H, CH2S); 3.05-3.35 (m, 4H, CH2N).
13C NMR (CDCl3): δ 14.2 (CH3); 20.8 (CH2S); 22.8 (CH2CH3);
27.0, 27.2, 27.9, 29.2, 29.4, 29.5, and 29.8 ((CH2)14); 32.0 (CH2-
CH2CH3); 37.2 (CH2CO); 46.2 and 48.0 (CH2N); 170.2 (CO).
MS-ESI m/z ) 610 (calcd for C39H79NOS + H+ ) 610).
Syn th esis of Telogen N,N-Ditetr a d ecylp r op a n a m id e-
3-th iol, 3a . The synthesis of telogen 3a was carried out in
one step as described in Scheme 1. To a solution of 21 µL (0.24
mmol) 3-mercaptopropionic acid (Aldrich) in 2 mL anhydrous
CHCl3, kept at 0 °C, were added 51 mg (0.24 mmol) dicyclo-
hexylcarbodiimide (DCC; Aldrich), then 33 mg (0.24 mmol)
1-hydroxy-1H-benzotriazole (HOBt, Aldrich), and 100 mg (0.24
mmol) ditetradecylamine 1a in 8 mL anhydrous CHCl3. The
reaction mixture was kept under stirring at room temperature
for 12 h, then diluted with CHCl3 and washed with H2O. The
organic phase was dried over Na2SO4, filtered and evaporated.
The concentrate was chromatographed on a silica column
(eluent CHCl3). 37 mg (0.07 mmol, 31%) 3a was obtained in
the form of a white solid. TLC (CH2Cl2, DTNB) Rf ) 0.5. 1H
NMR (CDCl3): δ 0.81 (t, 6H,3J ) 6.0 Hz, CH3); 1.00-1.35 (m,
1
3
Syn th esis of Ta xogen N-{2-[(BOC)a m in oeth yl]}a cr yl-
a m id e, A. A solution of 1.7 g (18.7 mmol) acryloyl chloride
(Aldrich) in 5 mL anhydrous CHCl3 was added dropwise at 0
°C and during 30 min to 2.0 g (12.4 mmol) N-(BOC)ethylene-
diamine (Fluka) and 5.0 g (49,1 mmol) triethylamine in 35 mL
anhydrous CHCl3. The reaction mixture was left, under
stirring, for 90 min at room temperature, then concentrated
to dryness. After purification on a silica column (eluent:
CHCl3/MeOH 98/2), one obtained 2.0 g (9,2 mmol, 75%)
taxogen A as a white solid. TLC (CHCl3/MeOH, 98/2, KMnO4)
Rf ) 0.2. 1H NMR (CDCl3): δ 1.36 (m, 9H, CH3); 3.10-3.34
(m, 2H, CH2NH); 3.34-3.50 (m, 2H, NCH2); 5.55 (dd,1H, 3J )
3
44 H, (CH2)11); 1.35-1.60 (m, 4H, CH2CH2N); 1.65 (t, 1H, J
) 8.2 Hz, SH); 2.55 (t, 2H, 3J ) 6.5 Hz CH2CO); 2.65-2.85
(m, 2H, CH2S); 3.05-3.35 (m, 4H, CH2N). 13C NMR (CDCl3):
δ 14.2 (CH3); 20.5 (CH2S); 22.8 (CH2CH3); 27.0, 27.2, 27.9, 29.2,
29.4, 29.5, and 29.7 ((CH2)10); 32.0 (CH2CH2CH3); 37.2 (CH2-
CO); 46.2 and 48.0 (CH2NCO); 170.2 (CO).
2
3
3
9,6 Hz, J ) 2.2 Hz, CHdCH2); 6.05 (dd, 1H, J ) 9,6 Hz, J
) 17.0 Hz, CHdCH2); 6.20 (dd, 1H,3J ) 17.0 Hz, J ) 2.2 Hz,
2
CHdCH2).13C NMR (CDCl3): δ 28.3 (CH3); 40.1 and 40.7
(CH2N); 79.6 (C (BOC)); 126.1 (CH2dCH); 130.9 (CHdCH2);
157.0 (CO(BOC)); 166.3 (COCH). Anal. Calcd for C10H18N2O3
(M ) 214.2): C 56.05; H 8.46; N 13.07. Found: C 55.79; H
8.47; N 13.16.
Syn th esis of Telogen r a c-1,2-Diocta d ecyloxyp r op a n e-
3-th iol, 7 (Sch em e 2). Telogen 9 was obtained in three steps
from rac-3-mercaptopropane-1,2-diol, 4.
r a c-3-Ben zylth iop r op a n e-1,2-d iol, 5. 3.2 g (29 mmol) rac-
3-mercaptopropane-1,2-diol (Aldrich) was dissolved in a mix-
ture of 28 mL 2 N NaOH and 35 mL EtOH. To this solution,
kept at 0 °C, was added 5.6 g (44 mmol) benzyl chloride
dropwise. After complete reaction of the thiol (TLC), the pH
was ajusted to 5 with HCl, and the solvent was evaporated.
The residue was purified by distillation (bp: 150 °C, 0.4
mmHg) to yield 3.9 g (20 mmol, 68%) 5 in the form of a
colorless oil. TLC (CHCl3/MeOH 98/2, DTNB, UV) Rf ) 0.3.
Syn th esis of Telom er s. 1. Telom er s I-m ,n . Telom er s
I-m ,n (BOC): Gen er a l P r oced u r e for th e Telom er iza tion
Rea ction . A solution of 500 mg (0.82 mmol) telogen 3b and
1.68 g (7.85 mmol) taxogen A (i.e. a taxogen/telogen molar ratio
of 9.5) in 15 mL anhydrous acetonitrile was heated to 50 °C
before adding 40 mg (0.24 mmol) AIBN. The reaction medium
was stirred under reflux until the telogen 3b disappeared
(followed by TLC/DTNB) (∼24 h). The solution was concen-
trated and chromatographed on a silica column (eluent: CHCl3
to CHCl3/MeOH 98/2) and fractionated by gel permeation
chromatography on a Sephadex LH 20 column (Sigma) with
CHCl3 as eluent. In this way, 83 mg telomer I-18,1(BOC), 85
mg telomer I-18,2(BOC), 93 mg telomer I-18,3(BOC) of aDPn
) 3, and 1.64 g telomer I-18,10(BOC) of an aDPn ) 10, as
white solids, were isolated. The aDPn and structure were
determined by 1H NMR. The aDPn is equal to 3Hx/2Hy, Hx
being the integration of the signal corresponding to the taxogen
ethyldiamido methylenes (from to ppm) and Hy the integration
of the signal corresponding to the telogen methyl groups (t at
0.80 ppm). This protocol applied to telogen 3b and taxogen A
3
1H NMR (CDCl3): δ 2.59 (d, 2H, J ) 6.0 Hz, CH2SBn [Bn )
benzyl]); 3.45-3.78 (m, 2H, CH2O); 3.78-3.91 (m, 3H, CH and
SCH2Ph); 7.26-7.50 (m, 5H, C6H5). 13C NMR (CDCl3): δ 34.5
(CH2SBn); 36.5 (SCH2Ph); 65.3 (CH2O); 70.7 (CH); 127.2 and
128.9 (C ortho and para); 128.6 (C meta); 138.0 (quaternary
C).
r a c-1,2-Diocta d ecyloxy-3-ben zylth iop r op a n e, 6. 3.9 g
(20 mmol) compound 5 and 15 g (43 mmol) stearylmesylate
were solubilized in 30 mL Et2O (stearylmesylate was obtained
by reaction of mesyl chloride with 1-octadecanol in pyridine).
1
TLC (CH2Cl2, H2SO4, UV) Rf ) 0.6. H NMR (CDCl3): δ 0.85
3
(t, 3H, J ) 5.8 Hz, CH3); 1.15-1.55 (m, 30H, (CH2)15); 1.65-
3
1.85 (m, 2H, CH2CH2OMs); 2.97 (s, 3H, CH3S); 4.19 (t, 2H, J

Lipopolycationic Telomers for Gene Transfer
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7 1377
at a taxogen/telogen molar ratio of 40 led, after chromatogra-
phy and fractionation, to telomers I-18,4(BOC), I-18,5(BOC),
I-18,20(BOC), I-18,40(BOC), I-18,50(BOC), and I-18,60-
(BOC) of aDPn ) 4, 5, 20, 40, 50, and 60, respectively.
When applied to telogen 3a and taxogen A at a taxogen/
telogen molar ratio of 40, this protocol afforded telomers I-18,-
10(BOC), I-18,20(BOC), and I-18,70(BOC) of aDPn ) 10, 20,
and 70, respectively.
(9n+4)H, CH2CH2NCO, and CH3 (BOC)); 1.75-3.00 (m,
(3n+5)H, CHCO, CH2CON, CH2S, CH2CH2CO and CH2-
CHCO); 3.00-3.60 (m, (4+4n)H, CH2NCO and CH2NHCO);
5.15-5.65 (m, nH, NH (BOC)). MS-ESI shows the presence of
components of n ) 2-10: m/z ) (498 + 214.2n + 23) with n
) 2-10, in agreement with the masses calcd for [C31H63NOS
+ n(C10H18N2O3) + Na+] with n ) 2-10.
Telom er I-14,20(BOC): 1H NMR: identical to that of
telomer I-14,10(BOC) but with aDPn ) 20. MS-ESI shows
the presence of components of n ) 8-20: m/ 2z ) (249 +
Com p ou n d I-18,1(BOC): 1H NMR (CDCl3): δ 0.80 (t,
6H,3J ) 6.0 Hz, CH3); 1.00-1.30 (m, 60H, (CH2)15); 1.30-1.75
3
(m, 13H, CH2CH2NCO, and CH3 (BOC)); 2.43 (t, 2H, J ) 7.0
107.1n + 23) in agreement with the masses calcd for [C31H63
-
3
NOS + n(C10H18N2O3) + 2Na+]/2 with n ) 8-20.
Hz, CH2CO); 2.51 (t, 2H, J ) 7.0 Hz, CH2CO); 2.65-2.85 (m,
4H, CH2S); 3.05-3.35 (m, 8H, CH2NCO and CH2NHCO);
5.20-5.30 (m, 1H, NH (BOC)); 6.55-6.75 (m, 1H, NH). 13C
NMR (CDCl3): δ 14.1 (CH3); 22.7 (CH2CH3); 28.5 (CH3 (BOC));
27.0, 27.2, 27.3, 27.9, 28.3, 29.2, 29.4, 29.5 and 29.8 ((CH2)14);
32.0 (CH2CH2CH3); 33.4 (CH2S); 36.7 (CH2CO); 40.5 (CH2-
NHCO); 46.3 and 48.1 (CH2NCO); 79.5 (C(BOC)); 156.8 (NCO
(BOC)); 170.7 and 172.0 (CON). MS-ESI m/z ) 847 (calcd for
Telom er I-14,70(BOC): 1H NMR: identical to that of
telomer I-L1k (BOC) but with aDPn ) 70.
Telom er s I-m ,n : Gen er a l P r oced u r e for th e Dep r o-
tection of th e BOC-Telom er s. The deprotection of telomers
I-m ,n (BOC) was quantitatively achieved by dissolving and
keeping these telomers in a large excess of trifluoroacetic acid
(TFA; Aldrich) at room temperature for 1 h. The excess of TFA
was removed by evaporation in the presence of cyclohexane.
The telomers I-m ,n were obtained in the form of white solids.
C
49H97N3O4S + Na+ ) 847).
Com p ou n d I-18,2(BOC): 1H NMR (CDCl3): δ 0.80 (t,
6H,3J ) 6.0 Hz, CH3); 1.00-1.30 (m, 60H, (CH2)15); 1.30-1.75
(m, 22H, CH2CH2NCO, CH3 (BOC)); 1.75-1.95 (m, 2H, CHCH2-
CH2CONH); 1.95-2.30 (m, 2H, CHCH2CH2CONH); 2.30-2.85
(m, 7H, CH2CON, CHCONH, CH2S); 3.00-3.60 (m, 12H, CH2-
NCO and CH2NHCO); 5.35-5.60 (m, 2H, NH (BOC)); 6.75-
6.95 (m, 1H, NH); 6.95-7.10 (m, 1H, NH). MS-ESI m/z ) 1061
(calcd for C59H115N5O7S + Na+ ) 1061).
Com p ou n d I-18,1: 1H NMR (CDCl3/CD3OD): δ 0.85 (t,
6H,3J ) 6.0 Hz, CH3); 1.15-1.45 (m, 60H, (CH2)15); 1.45-1.60
(m, 4H, CH2CH2N); 2.51 and 2.57(t, t, 2H, 2H, 3J ) 7.0 Hz,
3
CH2CO); 2.70-2.90 (m, 4H, CH2S); 3.04 (t, 2H, J ) 5.5 Hz,
3
CH2NH3); 3.15-3.40 (m, 4H, CH2N); 3.45 (t, 2H, J ) 5.5 Hz,
CH2NH). 13C NMR (CDCl3/CD3OD): δ 14.2 (CH3); 22.8 (CH2-
CH3); 27.1, 27.2, 27.8, 29.5, 29.6, 29.8 ((CH2)14); 32.0 (CH2CH2-
CH3); 33.4 (CH2S); 34.8 (CH2CO); 37.5 (CH2NH); 39.9 (CH2NH3);
Telom er I-18,3(BOC): 1H NMR (CDCl3): δ 0.80 (t, 6H,3J
) 6.0 Hz, CH3); 1.00-1.30 (m, 60H, (CH2)15); 1.30-1.75 (m,
(4+9n)H, CH2CH2NCO, CH3 (BOC)); 1.75-2.90 (m, (3n+5)H,
CHCO, CH2CON, CH2S, CH2CH2CO and CH2CHCO); 3.00-
3.60 (m, (4+4n)H, CH2NCO and CH2NHCO); 5.15-5.65 (m,
nH, NH (BOC)); 6.45-7.20 (m, nH, NH). 13C NMR (CDCl3): δ
14.1 (CH3); 22.7 (CH2CH3); 28.5 (CH3 (BOC)); 27.0, 27.2, 27.4,
27.9, 28.1, 28.4, 29.2, 29.4 and 29.8 ((CH2)14 and CHCH2CH);
32.0 (CH2CH2CH3); 33.4 and 34.2 (CH2S); 35.0 (CH2CO); 39.9,
40.2, 40.5 and 40.9 (CH2NHCO); 46.0 (CHCONH); 46.3 and
48.1 (CH2NCO); 79.5 (C(BOC)); 156.8 (CO (BOC)); 171.0, 173.3
and 174.6 (CO).
1
46.3 and 48.1 (CH2N); 117.0 (q, J ) 300 Hz, CF3CO2-); 162.8
2
(q, J ) 46 Hz, CF3CO2-); 171.0 and 175.2 (CO).
Com p ou n d I-18,2: 1H NMR (CDCl3): δ 0.80 (t, 6H,3J )
6.0 Hz, CH3); 1.00-1.30 (m, 60H, (CH2)15); 1.30-1.65 (m, 4H,
CH2CH2NCO); 1.65-1.95 (m, 2H, CHCH2CH2CONH); 1.95-
2.30 (m, 2H, CHCH2CH2CONH); 2.30-2.85 (m, 7H, CH2CON,
CHCON, CH2S); 3.00-3.60 (m, 12H, CH2NH3, CH2NCO and
CH2NHCO). 13C NMR (CDCl3): δ 14.2 (CH3); 22.8 (CH2CH3);
27.1, 27.2, 27.8, 29.5, 29.6, 29.8 ((CH2)14); 32.0 (CH2CH2CH3);
33.4 (CH2S); 34.8 (CH2CO); 37.5 (CH2NH); 39.9 (CH2NH3); 46.0
Telom er I-18,4(BOC): 1H NMR: identical to that of
telomer I-18,3(BOC) but with aDPn ) 4. 13C NMR (CDCl3):
δ 14.1 (CH3); 22.7 (CH2CH3); 28.6 (CH3 (BOC)); 27.0, 27.2, 27.9,
29.2, 29.4 and 29.8 ((CH2)14 and CHCH2CH); 32.0 (CH2CH2-
CH3); 46.3 and 48.1 (CH2NCO); 79.6 (C (BOC)); 156.8 (CO
(BOC)). The resonances corresponding to the carbons CH2-
NHCO, CH2CO, CHCO and to the carbons NCO appear,
respectively, as large signals between 38 and 43 ppm and
between 170 and 177 ppm. The CH2S carbons are not detect-
able.
1
(CHCO); 46.3 and 48.1 (CH2NCO); 117.0 (q, J CF ) 300 Hz,
2
CF3CO2-); 162.8 (q, J CF ) 46 Hz, CF3CO2-); 171.0, 175.2 and
177.0 (CON). MS-ESI m/z ) 839, in agreement with the calcd
mass for M[)C39H79NOS + 2(C5H10N2O)] + H+.
Telom er s I-18,n , for n g 3: 1H NMR (CD3OD): δ 0.80 (t,
6H,3J ) 6.0 Hz, CH3); 1.00-1.35 (m, 60H, (CH2)15); 1.35-2.60
(m, (9+3n)H, CH2CH2N, CH2CON, CHCO, CH2S, CH2CH2CO
and CH2CHCO); 2.70-3.60 (m, (4n+4)H, CH2NH3+, CH2NCO
and CH2NHCO) with, respectively, an aDPn ) 3 (telomer
I-18,3), 4 (telomer I-18,4), 5 (telomer I-18,5), 10 (telomer I-18,-
10), 20 (telomer I-18,20), 40 (telomer I-18,40), 50 (telom er
I-18,50), and 60 (telomer I-18,60). 13C NMR (CD3OD): δ 14.3
(CH3); 23.3 (CH2CH3); 29.0, 29.6, 29.9 and 30.3 (CHCH2CH
and (CH2)14); 32.5 (CH2CH2CH3); 37.7 (CH2NCO); 40.0 (CH2-
Telom er I-18,5(BOC): 1H NMR: identical to that of
telomer I-18,3(BOC) but with aDPn ) 5. 13C NMR: identical
to that of telomer I-18,4(BOC).
Telom er I-18,10(BOC): 1H NMR: identical to that of
telomer I-18,3(BOC) but with aDPn ) 10. 13C NMR: identical
to that of telomer I-18,4(BOC). MS-ESI shows the presence
of compounds with n ) 3-15: m/z ) (610 + 214.2n + 23) with
n ) 3-15 in agreement with the masses calcd for [C39H79NOS
+ n(C10H18N2O3) + Na+] with n ) 3-15.
2
NH3+); 117.0 (q, 1J ) 300 Hz, CF3CO2-); 162.8 (q, J CF ) 46
Hz, CF3CO2-); 176.2 (NCO). The resonances corresponding to
carbons CHCO and NHCO appear as broad signals between
41 and 45 ppm and between 176.5 and 179 ppm, respectively.
The resonances corresponding to carbons CH2S and CH2CO
are not detectable. MS-ESI of telomer I-18,10 shows the
presence of compounds with n ) 3-15: m/z ) (610 + 114.1n
+ 1) with n ) 3-10, in agreement with the calcd masses for
[C39H79NOS + n(C5H10N2O) + H+] with n ) 3-10; m/ 2z )
(305 + 57n + 1) with n ) 3-15, in agreement with the calcd
masses for [C39H79NOS + n(C5H10N2O) + 2H+]/2 with n )
3-15; m/ 3z ) (203 + 38n + 1) with n ) 5-13 in agreement
with the calcd masses for [C39H79NOS + n(C5H10N2O) + 3H+]/
3, with n ) 3-13.
Telom er I-18,20(BOC): 1H NMR: identical to that of
telomer I-18,3(BOC) but with aDPn ) 20. 13C NMR: identical
to that of telomer I-18,4(BOC). MS-ESI shows the presence
of compounds with n ) 4-33: m/z ) (610 + 214.2n + 23) with
n ) 4-15, in aggreement with the masses calcd for [C39H79
-
NOS + n(C10H18N2O3) + Na+] with n ) 4-15; m/ 2z ) (305 +
107.1n + 23) with n ) 9-33 in agreement with the masses
calcd for [C39H79NOS + n(C10H18N2O3) + 2Na+]/2 with n )
9-33.
Telom er s I-18,40(BOC), I-18,50(BOC), a n d I-18,60-
(BOC): 1H NMR: identical to that of telomer I-18,3(BOC) but
with aDPn ) 40, 50, and 60, respectively. 13C NMR: identical
to that of telomer I-18,4(BOC).
Telom er I-14,10(BOC): 1H NMR (CDCl3): δ 0.80 (t, 6H,3J
) 6.0 Hz, CH3); 1.00-1.30 (m, 44H, (CH2)11); 1.30-1.75 (m,
Telom er s I-14,n , w ith n ) 10, 20, 70: 1H NMR (CD3OD):
δ 0.80 (t, 6H,3J ) 6.0 Hz, CH3); 1.00-1.35 (m, 44H, (CH2)11);
1.35-2.60 (m, (9+3n)H, CH2CH2N, CH2CON, CHCO, CH2S,
CH2CH2CO and CH2CHCO); 2.70-3.60 (m, (4n+4)H, CH2-
NH3+, CH2NCO and CH2NHCO) with respectively an aDP of

1378 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7
Verderone et al.
10 (telomer I-14,10), 20 (telomer I-14,20), and 70 (telomer
I-14,70). 13C NMR: identical to those of telomers I-18,n with
n g 3.
163.4 (q, 2J CF ) 34 Hz, CF3CO2-); 174.8, 175.1 and 177.1 (NCd
O). 19F NMR (CDCl3, CD3OD): δ -76.6 (CF3CO2-).
Telom er II-18,6: 1H and 19F NMR (CDCl3, CD3OD): iden-
tical to those of telomer II-18,3, but with n ) 6. 13C NMR
(CDCl3, CD3OD): δ 14.8 (CH3); 23.7 (CH2CH3); 27.1, 30.4, 30.5,
30.7 and 31.1 ((CH2)14 and CH2CHCO); 33.0 (CH2CH2CH3);
34.0 (CH2S); 35.0 (CH2CO); 38.1 (CH2N); 40.7 (CH2NH3); 71.5,
2. Telom er s II-18,n . Telom er s II-18,n (BOC). The telom-
erization protocol described above for the synthesis of telomers
I-m ,n (BOC) was applied to telogen 7 and taxogen A, at a
taxogen/telogen molar ratio of 3 and 5. It led to two series of
telomers in the form of white solids after purification by
chromatography on a silica column (eluent: CHCl3) and with
a yield of 60%. The fractionation on Sephadex LH-20 of the
first series led to compound II-18,1(BOC) and telomers II-
18,3(BOC) and II-18,11(BOC); the fractionation of the second
series allowed to isolate, in addition to the first three telomers,
telomers II-18,6(BOC) and II-18,16(BOC). The aDP values
2
72.8 and 72.9 (CH2O); 79.6 (CHO); 163.0 (q, J CF ) 36 Hz,
CF3CO2-); 176.2 and 177.9 (NCdO). The CH(CdO) carbons
appear as broad resonance signals between 41 and 46 ppm.
MS-FAB and MS-ESI show the presence of n ) 4-9 oligo-
mers: m/z ) 1069.5; 1183.5; 1297.6; 1411.6; 1525.6; 1641.2 in
agreement with M[()C39+5nH80+11nN2nO2+nS) + H+] for n )
4-9, respectively.
1
were determined by H NMR, as detailed above for telomers
Telom er II-18,11: 1H and 19F NMR (CDCl3, CD3OD):
identical to those of telomer II-18,3, but with n ) 11. 13C NMR
(CDCl3, CD3OD): δ 14.8 (CH3); 23.7 (CH2CH3); 27.1, 30.3, 30.6
and 30.7 ((CH2)14 and CH2CHCO); 32.3 (CH2CH2CH3); 33.2
(CH2S); 163.0 (q, 2J CF ) 36 Hz, CF3CO2-); 175.0 (NCdO). The
resonances corresponding to carbons CH2N, CH2CO and CHCO
appear as broad signals between 38 and 43 ppm; the reso-
nances corresponding to carbons CH2O and CHO are not
detectable. MS-FAB and MS-ESI show the presence of oligo-
mers of n ) 4-15: m/z ) 1069.9; 1183.9; 1298.0; 1412.1;
1526.1; 1640.2; 1754.3; 1868.4; 1982.4; 2097.5; 2210.5 in
agreement with M()C39+5nH80+11nN2nO2+nS) for n ) 4-14,
respectively; m/ 2z ) 763.8; 821.0; 878.0; 935.0; 992.0; 1049.3;
1106; 1163 in agreement with [M()C39+5nH80+11nN2nO2+nS) +
2H+]/2, for n ) 8-15, respectively.
I-m ,n .
Com p ou n d II-18,1(BOC): 1H NMR (CDCl3): δ 0.80 (t,
6H,3J ) 6.0 Hz, CH3); 1.02-1.32 (m, 60H, (CH2)15); 1.37 (s,
9H, CH3 (BOC)); 1.44-1.55 (m, 4H, CH2CH2O); 2.39 (t, 2H, 3J
) 6.9 Hz, CH2CO); 2.64 (d, 2H, 3J ) 4.8 Hz, CH2S); 2.78 (t,
2H, 3J ) 6.2 Hz, SCH2CH2CO); 3.12-3.60 (m, 11H, CH2O,
CHO and CH2N); 5.16 and 6.60 (m, 2H, NH). 13C NMR
(CDCl3): δ 14.0 (CH3); 22.7 (CH2CH3); 28.4 (CH3 (BOC)); 26.1,
29.3, 29.5, 29.6, 29.7 and 30.0 ((CH2)14); 31.9 (CH2CH2CH3);
33.9 (CH2S); 36.7 (CH2CO); 40.3 and 40.5 (CH2N); 70.5, 71.5
and 71.7 (CH2O); 78.5 (CHO); 79.5 (O-C (BOC)); 156.7 (CdO
(BOC)); 171.9 (CdO).
Telom er II-18,3(BOC): 1H NMR (CDCl3): δ 0.80 (t, 6H,3J
) 6.0 Hz, CH3); 0.98-1.30 (m, 60H, (CH2)15); 1.30-1.60 (m,
(9n+4)H, CH3 (BOC) and CH2CH2O); 1.60-3.00 (m, (3n+3)H,
CH2CO, CHCO, CH2CHCO, CH2CH2CO and CH2S); 3.05-3.55
(m, (4n+7)H, CH2O, CHO and CH2N) with n ) 3. 13C NMR
(CDCl3): δ 14.1 (CH3); 22.6 (CH2CH3); 28.4 (CH3 (BOC)); 26.1,
29.3, 29.5, 29.6, 29.7 and 30.1 ((CH2)13 and CH2CHCO); 31.9
(CH2CH2CH3); 33.8 and 34.5 (CH2S); 35.6 (CH2CO); 35.8 and
36.2 (CHCO); 70.4, 71.6 and 71.7 (CH2O); 78.5 (CHO); 79.6
(O-C (BOC)); 156.7 (CdO (BOC)); 172.0, 173.1 and 174.4 (CO).
The CH2N carbons appear as a broad resonance between 39
and 41 ppm.
II-18,n (BOC), w ith n ) 6, 11, 16: 1H NMR: identical to
that of II-18,3(BOC) but with n ) 6, 11, and 16, respectively.
13C NMR (CDCl3): δ 14.0 (CH3); 22.6 (CH2CH3); 28.4 (CH3
(BOC)); 26.0, 29.3, 29.4, 29.6 and 29.7 ((CH2)14 and CH2CHCO);
31.8 (CH2CH2CH3); 33.6 and 34.4 (CH2S); 70.4, 71.7 and 71.9
(CH2O); 78.5 (CHO); 79.6 (O-C (BOC)); 156.6 (CdO (BOC));
the CH2(CdO), CHCdO, CH2N, and NHCdO carbons appear
as very broad resonance signals between 38 and 44 ppm and
between 173 and 176 ppm.
Telom er II-18,16: 1H and 19F NMR (CDCl3, CD3OD):
identical to those of telomer II-18,3, but with n ) 16. 13C NMR
(CDCl3, CD3OD): δ 14.8 (CH3); 23.7 (CH2CH3); 27.1, 30.3, 30.6
and 30.7 ((CH2)14 and CH2CHCO); 32.3 (CH2CH2CH3); 33.2
2
(CH2S); 163.0 (q, J CF ) 36 Hz, CF3CO2-). The resonances
corresponding to carbons CH2N, CH2CO, CHCO, and NCO,
respectively, appear as broad signals between 38 and 43 ppm
and between 176 and 179 ppm, respectively; the resonances
corresponding to carbons CH2O and CHO are not detectable.
MS-FAB and MS-ESI show the presence of oligomers of n )
6-17: m/z ) 1298.0; 1412.1; 1526.2; 1640.3; 1755.3; 1869.4;
1983.5; 2097.5; 2211.6; 2325.4; 2439; 2553; 2667; 2782 in
agreement with M()C39+5nH80+11nN2nO2+nS) for n ) 6-17,
respectively.
P r epar ation of Com plexes Com posed of th e P olyam in e
Telom er Com p ou n d s a n d P la sm id p TG11033. The plas-
mid pTG11033 was produced by Transgene SA. The endotoxin
content of the plasmid preparation was checked using a
limulus amebocyte lysat kit (Biogenic, France). This value was
below 1 endotoxin unit/mg of plasmid, hence below the 5 e.u./
mg of DNA recommended for in vivo protocols. The quantities
of compound used were calculated according to the desired
DNA concentration of 0.1 mg/mL (for in vitro tests), the charge
ratio, the molar weight and the number of potential positive
charges in the selected compound. The polyamine telomer/
DNA complex is formulated by adding a desired volume of
telomer preparation at a concentration of 10 mg/mL (in 20 mM
HEPES buffer, pH 7.5, containing 10% DMSO and 10% EtOH)
to the desired volume of DNA solution to reach the desired
DNA concentration (for example 0.1 mg/mL). Thus for the
preparation of the polyamine telomer I-18,40/DNA complex
at charge ratio 5 and 0.1 mg/mL DNA, 40 µL of a polyamine
telomer solution (10 mg/mL in CHCl3) was transferred to a
borosilicate glass tube (16 × 100 mm). The solvent was
evaporated in Rotavap evaporation system (45 °C, 30 pm, 0.2
bar, 40 min). 4 µL of DMSO, 4 µL of EtOH and 32 µL of HEPES
20 mM, pH 7.5, were added to the film obtained. The
preparation was heated at 50 °C, while stirring (30 rpm), for
5 min. After cooling to room temperature, 37 µL of this
preparation was added to 963 µL DNA solution [100 µL DNA
(1 mg/mL) diluted with 863 µL HEPES buffer]. This prepara-
tion can be used for further experiments.
Telom er s II-18,n . The removal of the BOC protection
groups from telomers II-18,n (BOC) was quantitatively achieved
by dissolving and keeping these telomers in a large excess of
TFA at room temperature for 1 h. The excess TFA was
removed by coevaporation in the presence of cyclohexane.
Telomers II-18,n were obtained in the form of white solids.
Com p ou n d II-18,1: IR (ν cm-1, KBr): 1875 (CdO). 1H
NMR (CDCl3): δ 0.80 (t, 6H,3J ) 6.0 Hz, CH3); 1.00-1.36 (m,
60H, (CH2)15); 1.37-1.70 (m, 4H, CH2CH2O); 2.43 (t, 2H, 3J )
3
6.1 Hz, CH2CO); 2.62 (d, 2H, J ) 5.0 Hz, CH2S); 2.74 (t, 2H,
3J ) 6.2 Hz, SCH2CH2CO); 3.00-3.25 (m, 2H, CH2NH3); 3.30-
3.65 (m, 9H, CH2O, CHO and CH2N). 13C NMR (CDCl3): δ
14.1 (CH3); 22.7 (CH2CH3); 26.1, 29.3, 29.5, 29.7 and 30.0
((CH2)14); 31.9 (CH2CH2CH3); 33.8 (CH2S); 36.1 (CH2CO); 37.5
(CH2N); 39.4 (CH2NH3); 70.6, 71.6 and 71.8 (CH2O); 78.3
(CHO); 171.9 (CdO). 19F NMR (CDCl3): δ -76.0 (CF3CO2-).
Telom er II-18,3: 1H NMR (CDCl3, CD3OD): δ 0.84 (t,
6H,3J ) 5.8 Hz, CH3); 1.00-1.45 (m, 60H, (CH2)15); 1.45-1.75
(m, 4H, CH2CH2O); 1.75-3.25 (m, (5n+3)H, CH2CH2CO, CH2-
CHCO, CH2CO, CHCO, CH2NH3 and CH2S); 3.25-3.70 (m,
(2n+7)H, CH2O, CHO and CH2N) with n ) 3. 13C NMR (CDCl3.
CD3OD): δ 15.0 (CH3); 23.7 (CH2CH3); 27.2, 29.0, 29.1, 30.4,
30.6, 30.7, 30.8 and 30.9 ((CH2)14 and CH2CHCO); 33.0 (CH2-
CH2CH3); 33.7 (CH2S); 38.1 (CH2CO); 40.7 (CH2N); 41.0 (CH2-
NH3); 43.4 (CHCO); 70.5, 70.6 and 71.8 (CH2O); 78.3 (CHO);
When the composition further comprises DOPE, the desired
volume of 1 mg/mL chloroform solution (to obtain mole/mole

Lipopolycationic Telomers for Gene Transfer
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 7 1379
(2) Felgner, P. L.; Gadek, T. R.; Holm, M.; Roman, R.; Chan, H. W.;
Wenz, M.; Northrop, J . P.; Ringold, G. M.; Danielsen, M.
Lipofection: a highly efficient, lipid-mediated DNA-transfection
procedure. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 7413-7417.
(3) Remy, J . S.; Abdallah, B.; Zanta, M. A.; Boussif, O. Behr, J . P.;
Demeinex, B. Gene transfer with lipospermines and polyethyl-
enimines. Adv. Drug Deliv. Rev. 1998, 30, 85-95.
(4) Byk, G.; Dubertret, C.; Escriou, V.; Frederic, M.; J aslin, G.;
Rangara, R.; Pitard, B.; Crouzet, J .; Wils, P.; Schwartz, B.;
Scherman, D. Synthesis, activity, and structure-activity rela-
tionship studies of novel cationic lipids for DNA transfer. J . Med.
Chem. 1998, 41, 224-235.
(5) Miller, A. D. Cationic liposomes for gene therapy. Angew. Chem.,
Int. Ed. 1998, 37, 1768-1785.
(6) Mislick, K. A.; Baldeschwielr, J . D. Evidence for the role of
proteoglycans in cation-mediated gene transfer. Proc. Natl. Acad.
Sci. U.S.A. 1996, 93, 12349-12354.
(7) Mounkes, L. C.; Zhong, W.; Cipres-Paladin, G.; Heath, T. D.;
Debs, R. J . Proteoglycans mediate cationic liposome-DNA com-
plexe based gene delivery in vitro and in vivo. J . Biol Chem.
1998, 273, 26164-26170.
(8) Zhou, X. H.; Klibanov, A. L.; Huang, L. Lipophilic polylysines
mediate efficient DNA transfection in mammalian cells. Biochim.
Biophys. Acta 1991, 1065, 8-14.
(9) Remy, J . S.; Sirlin, C.; Vierling, P.; Behr, J . P. Gene transfer
with a series of lipophilic DNA-binding molecules. Bioconjugate
Chem. 1994, 5, 647-654.
(10) Felgner, J . H.; Kumar, R.; Sridhar, C. N.; Wheeler, C. J .; Tsai,
Y. J ., Border, R., Ramsey, P., Martin M.; Felgner, P. L. Enhanced
gene delivery and mechanism studies with a novel series of
cationic lipid formulations. J . Biol. Chem. 1994, 269, 2550-2561.
(11) Gao, X.; Huang, L. A novel cationic liposome reagent for efficient
transfection of mammalian cells. Biochem. Biophys. Res. Com-
mun. 1991, 179, 280-285.
(12) Felgner, P. L.; Holm, M.; Chan, H. Cationic liposome mediated
transfection. Proc. West. Pharmacol. Soc. 1989, 32, 115-121.
(13) Gao, X.; Huang, L. Cationic liposomes and polymers for gene
transfer. J . Liposome Res. 1993, 3, 17-30.
(14) Crook, K.; Stevenson, B. J .; Dubouchet, M.; Porteous, D. J .
Inclusion of cholesterol in DOTAP transfection complexes in-
creases the delivery of DNA to cells in vitro in the presence of
serum. Gene Ther. 1998, 5, 137-143.
(15) Lee, E. R.; Marshall, J .; Siegel, C. S.; J ian, C.; Yew, N. S.;
Nichols, M. R.; Nietupsky, J . B.; Ziegler, R. J .; Lane, M. B.;
Wang, K. X.; Wan, N. C.; Scheule, R. K.; Harris, D. J .; Smith,
A. E.; Cheng, S. H. Detailed analysis of structures and formula-
tions of cationic lipids for efficient gene transfer to the lung.
Hum. Gene Ther. 1996, 7, 1701-1717.
(16) Balasubramaniam, R. P.; Malone, R. W. Structural and func-
tional analysis of cationic transfection lipids: the hydrophobic
domain. Gene Ther. 1996, 3, 163-172.
cationic telomer/DOPE) was added to the volume of polyamine
telomer solution (10 mg/mL in CHCl3) and then the mixture
was transferred to a borosilicate glass tube. The following steps
were as described above.
Mea su r em en t of th e Size of th e Com p lexes F or m ed .
The average sizes were measured by photon correlation
spectroscopy using a Coulter N4Plus particle size analyzer.
The analyses were performed at 25 °C, after equilibrating the
sample at 25 °C for 20 min. One aliquot of the sample was
aspirated and discharged several times prior to pipetting. The
sample was diluted with 20 mM HEPES in the measurement
tube and homogenized. These analysis were thus carried out
on complexes having a 10 µg/mL concentration of DNA. After
waiting 180 s, the light diffraction at 90° was measured for
180 s. The range was 3 nm to 10 µm, using 31 bins. The size
measurement is valid if the sample gives between 50 000 and
1 000 000 counts/s. The formulations and analyses were
reproduced twice.
Aga r ose Gel Electr op h or esis. Each sample (0.2 µg of
plasmid) was analyzed by electrophoresis for about 90 min
under 75 V/cm, through a 1% agarose gel in TAE 1× (Tris-
acetate-EDTA) buffer. The DNA was visualized in TAE buffer
containing 12 µL/200 mL ethidium bromide (Sigma).
In Vitr o Tr a n sfection of A549 Cells. 24 h before trans-
fection, A549 cells (epithelial cells derived from human
pulmonary carcinoma) were grown in Dulbeco-modified Eagle
culture medium (DMEM), containing 10% fetal calf serum,
FCS (Gibco BRL), in 96-well plates (2 × 104 cells/well), in a
wet (37 °C) and 5% CO2/95% air atmosphere. Volume of DNA/
polyamine telomer complex (40, 20, 5, and 1 µL, respectively)
was diluted to 100 µL in DMEM or DMEM supplemented with
10% FCS in order to obtain various amounts of DNA (4, 2,
0.5, and 0.1 µg, respectively) in the preparation. The culture
medium was removed and replaced by 100 µL of DMEM
supplemented with 10% FCS and containing the desired
amount of DNA. After 2 and 24 h, 50 µL and 100 µL of DMEM
supplemented with 10% FCS were added, respectively.
48 h after transfection, the culture medium was discarded
and the cells were washed with 100 µL of PBS and then lysed
with 50 µL of lysis buffer (Promega). The lysates were frozen
at -80 °C awaiting analysis of luciferase activity. This
measurement was done for 1 min on 20 µL of the lysis mixture
in a Berthold LB96P luminometer in dynamic mode, using the
“luciferase” determination system (Promega) in 96-well plates.
The total protein concentration per well was determined
using conventional techniques (BCA test, Pierce). For cells
grown in the absence of lipoplexes, a well contains around 30-
60 µg of proteins. Each formulation was tested twice.
(17) Legendre, J . Y.; Trzeciak, A.; Bur, D.; Deuschle, U.; Supersaxo,
A. N-acyl-(R,γ-diaminobutyric acid)n hydracide as an efficient
gene transfer vector in mammalian cells in culture. Pharm. Res.
1997, 14, 619-624.
(18) Huang, C. Y.; Uno, T.; Murphy, J . E.; Lee, S.; Hamer, J .D.
Escobedo, J . A.; Cohen, F. E.; Radhakrishnan, R.; Dwarki, V.;
Zuckermann, R. N. Lipitoids- novel cationic lipids for cellular
delivery of plasmid DNA in vitro. Chem. Biol. 1998, 5, 345-
354.
Su p p or tin g In for m a tion Ava ila ble: Four figures show-
ing respectively the effect of DOPE on transfection efficiency
and cell viability of the teloplexes and the impact of the
number of amine functions on transfection efficiency and cell
viability for the I-14,n - and I-18,n -based teloplexes. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(19) Lasic, D. D.; Templeton, N. S. Liposomes in Gene Therapy. Adv.
Drug Deliv. Rev. 1996, 20, 221-266.
Refer en ces
(1) Wivel, N. A.; Wilson, J . M. Methods of gene delivery. Hematol.
Oncol. Clin. North Am. 1998, 12, 483-501.
J M9911579