Gene Therapy (2002) 9, 736–739
 2002 Nature Publishing Group All rights reserved 0969-7128/02 $25.00
www.nature.com/gt
Visualization of the transgene distribution according to
the administration route allows prediction of the
transfection efficacy and validation of the results
obtained
P Delepine1,2,*, T Montier1,*, C Guillaume1, L Vaysse5, A Le Pape4 and C Ferec1,2,3
1
INSERM EMI-U 01-15, UBO, Brest, France; 2CHU Morvan, Brest, France; 3EFS-Bretagne, Brest, France; 4INSERM EMI-U
00-10, Faculté de Médecine and Biopharm Consulting, Tours, France; and 5Université de Bordeaux 2, Bordeaux, France
Gene transfer to the lung can be achieved via a systemic,
that targets the endothelium, or local, that targets the epithelium, delivery route. In the present study, we followed the
distribution of a plasmid after transfection using some of our
phosphonolipids, which have previously shown their
efficiency in transfecting mouse lungs. The plasmid was
radiolabeled and varying combinations of plasmid/
phosphonolipid were administered by intravenous injection,
or by endotracheal spray. The distribution of radioactive
labeling was observed over a time course using a ␥-camera.
These images were then correlated with the results for
luciferase expression levels in the lungs. In each case, lungs
were well targeted. However, whereas an intravenous injection reaches all of the lung immediately, progressive dif-
fusion occurs when the plasmid/phosphonolipid is administered via an aerosol. Elimination of the radioactivity
associated with plasmid occurs via the urinary tract after
intravenous injections, and via the feces using the aerosol
delivery approach. The radioactivity detected in the lungs
correlated strongly with transgene expression. Thus, such
an imaging technique is a powerful strategy to predict the
formulation that will generate the best transfection efficiency.
This study reveals that scintigraphic imaging permits both
validation of the administration method and the results
obtained for each animal, thereby reducing the statistical
variability of in vivo experiments.
Gene Therapy (2002) 9, 736–739. DOI: 10.1038/sj/gt/3301742
Keywords: lipoplexes; biodistribution; transfection efficiency; intravenous; endotracheal spray
Gene transfer targeted to the lung is a promising
approach for the treatment of numerous lung diseases,
such as cancers or cystic fibrosis. The ideal delivery strategy for such a therapy would be one that allows the gene
of interest to specifically target the lung. When using
viral vectors, local administration methods are employed,
including aerosolisation,1,2 nasal instillation,3 and lobar
instillation.4 Alternatively, nonviral vectors are delivered
via systemic5,6 or local7–9 routes.
Whatever the route chosen, the problem remains that
gene transfer needs to be located specifically in the targeted organ, ie the lung. This can be confirmed by visualizing the biodistribution of the administered material.
Some groups have studied the distribution of lipoplexes,
usually looking for the lipoplexes in individual tissues by
radioactively labeling the complexes.10–12 Alternatively,
the cell type transfected in a particular organ (usually the
lung) is analyzed by immunohistochemical techniques.13
However, whole body distribution studies looking at the
whole mouse using imaging are rare, and are usually performed on excised organs from dead animals.14 This
Correspondence: C Ferec, Inserm EMI-U 01-15, 46, Rue Felix Le Dantec,
29275 Brest, France
*The first two authors contributed equally to this work
report deals with the strategy we took to follow our lipoplexes by visualization. This technique has been
described for an adenovirus lung deposition study15 and
for observing gene induction.16 However, to the best of
our knowledge, the approach has not been used for a
kinetic study of lipoplex distribution in living rodents.
With the aim of developing synthetic gene transfection
reagents for cystic fibrosis gene therapy, a few years ago
our group developed an original family of cationic lipids,
called phosphonolipids. These have been shown to be
efficient for gene transfer in vitro, as well as in vivo, particularly in gene transfer to mouse lung tissue.5,17,18 Biodistribution studies using the cationic phosphonolipids/
DNA lipoplexes have been carried out after intravenous
administration. These studies show that the lipoplexes
are rapidly dissociated and the phosphonolipids
degraded.12 The aim of the present study was first, to
assess the biodistribution for each administration route,
ie intravenous versus endotracheal spray. Second, to validate the protocols and the devices we used to achieve
mouse lung-directed gene transfer, and third, to evaluate
the interest and the possible applications of this scintigraphic imaging for animal studies.
The biodistribution study was performed by radiolabeling the plasmid DNA with Tc99m. The lipoplexes formed
were intravenously injected in the first instance. As pre-
Visualization of transgene distribution
P Delepine et al
viously described,6 the lipoplexes accumulate very rapidly in the lung. The lipoplex aggregates in the lungs are
progressively degraded, and the plasmid DNA that has
not been internalized is eliminated via the urinary tract
(Figure 1a). It was noted that the whole lung was reached
using this route for lipoplex administration. The significance of the recorded signal cannot be determined
beyond the 6 h time-point due to some exchange of 99m
Tc from the Tc99m-plasmid chelate with plasma proteins.
When administered via an endotracheal spray, the lipoplexes reach only the air volume inside the lungs with
the signal much more localized (Figure 1b). In parallel
with a progressive diffusion to the upper lungs, the signal intensity slowly decreases because of an elimination
of the radioactivity through the digestive tract due to
mucociliary clearance and deglutition.
The relationship between plasmid present in the lungs
and expression of the transgene demonstrates that the
more plasmid that persists in the lung, the more efficient
the transfection. Indeed, the formulation that allows the
plasmid DNA to remain in the lungs for the longest times
are the ones in which the most luciferase expression is
observed, irrespective of the route used for administration (Figures 2a and 3). The results shown here explain
why a formulation that can be efficient when administered via one route is totally inefficient via another.
Indeed, formulation A, which exhibits the highest
expression when sprayed into the trachea (Figure 3), is
less efficient when injected intravenously (Figure 2). This
is a consequence of very rapid elimination of the plasmid
due to the instability of this formulation in the blood-
stream. In contrast, this formulation allows the greatest
persistence of the plasmid in the lung when administered
as a spray.
Through these experiments, we have demonstrated the
efficiency of our phosphonolipids for gene transfer
directed to the lung. Moreover, we have demonstrated
the application of the microsprayer as a means to achieve
aerosolisation in which the quantity of material that
reaches the mouse lung can be accurately determined.
This device would facilitate the establishment of a
relationship between the aerosolized material and the
transfection level obtained. Therefore it would allow the
comparison of the efficiency of various nebulized vectors,
independently of the nebulization technique used.
With regard to the imaging technique used, firstly, it
permits observation of the plasmid biodistribution. This
is of particular interest for formulation studies, so that
targeting to a specific organ can be improved. Indeed,
we have shown that the efficacy of a given preparation
depends on the administration route chosen. Secondly,
quantification and comparison of the remaining signal
intensity allows us to predict the relative efficacy of various preparations. Moreover, imaging the preparation in
the animal body allows validation of each administration.
Animals with an unsatisfactory distribution can be
eliminated from the analysis.
In conclusion, the scintigraphic imaging is of great
interest for biodistribution studies, and will allow significant progress to be made in choosing the most valuable lipoplex formulations, particularly when used in
combination with reproducible devices, such as the
microsprayer.
737
Figure 1 Biodistribution study. Plasmid DNA (pCMVLuc) was radiolabeled with technetium 99m (99mTc). The specific activity obtained was 5 ␮Ci/␮g.
The GLB43 cationic phosphonolipid powder used in this experiment was dissolved in chloroform in a glass vial. Chloroform was then evaporated under
vacuum to give a dry lipid film. An appropriate amount of sterile saline solution (NaCl 0.9% w/v) was then added to the lipid film. The vials were
subsequently stored at 4°C overnight. The resulting solution was sonicated for 10 min in a bath sonicator (Prolabo, Paris, France) before use. Cationic
lipid and plasmid DNA were mixed and incubated for 30 min before administration. Each mouse received 50 ␮g of the radiolabeled plasmid formulation.
The radioactivity was evaluated by means of a specially designed high resolution imaging device (gamma imager, Biospace, France) (a) Biodistribution
of the plasmid DNA after intravenous injection. Mice were anesthetized by ether inhalation before administration of 200 ␮l of the lipoplex solution via
the tail vein. Images were recorded just after injection, and then at 30 min, 3 h and 6 h after injection. (b) Biodistribution of the plasmid after endotracheal
administration using a microsprayer. Mice were anesthetized by fluothane inhalation. The tip of the microsprayer, adapted for use on mice (PennCentury, PA, USA), was inserted into the trachea close to the carena. Lipoplex solution (150 ␮l) was administered. Images were recorded just after
injection and at 30 min, 1 h, 6 h and 20 h.
Gene Therapy
Visualization of transgene distribution
P Delepine et al
738
Figure 2 Correlation between luciferase expression in the lung and plasmid DNA remaining in the lung after intravenous injection. Plasmid DNA
was radiolabeled as described in Figure 1 and was combined with the cationic phosphonolipids GLB 43 or GLB 73. Those two lipids were prepared as
described in Figure 1. Four preparations called A, E, F and G were also obtained. Images were recorded at 0, 30 min, 3 h and 6 h. The record presented
in this figure was obtained 3 h after injection. Mice were killed 24 h after injection. The lungs were harvested and crushed in a mortar cooled with
liquid nitrogen. Lysis buffer (500 ␮l) was added to each sample, and stirred for 30 min before sample centrifugation (10 min, 10 000 g). A chemiluminescent luciferase assay (Promega, Madison, WI, USA) was performed on each of the supernatants. Luciferase enzymatic activity in each of the supernatants
was measured using a microtiter plate luminometer (Dynatech Laboratories, Guyancourt, France). The total protein concentration of each supernatant
was evaluated using a Coomassie plus protein assay reagent kit (Pierce, Rockford, IL, USA). The results are expressed in RLU (relative light unit) per
milligram of protein.
Figure 3 Correlation between luciferase expression in the lungs and persistence of plasmid DNA after endotracheal nebulization. Plasmid DNA was
radiolabeled as described in Figure 1 and was combined with the cationic phosphonolipids GLB 43 or GLB 73. These two lipids were prepared as described
in Figure 1. Four preparations called A, B, C and D were also obtained. Mice were treated as described in Figure 1 and the four formulations were
administered by means of the microsprayer. Images were recorded at 0, 30 min, 1 h, 6 h and 20 h. The image presented in Figure 2 was obtained at
the 6 h time-point. Mice were killed 72 h after administration, and the lungs were treated as described in Figure 2. Luciferase expression results are
expressed in RLU per milligram of total protein.
Acknowledgements
This work was supported by the French associations
‘Vaincre la Mucoviscidose’ and ‘Association pour la
recherche sur le cancer’ (ARC) and by the conseil general
de Bretagne.
Gene Therapy
References
1 Bellon G et al. Aerosol administration of a recombinant adenovirus expressing CFTR to cystic fibrosis patients: a phase I
clinical trial. Hum Gene Ther 1997; 8: 15–25.
2 Perricone MA et al. Aerosol and lobar administration of a recom-
Visualization of transgene distribution
P Delepine et al
3
4
5
6
7
8
9
binant adenovirus to individuals with cystic fibrosis. II. Transfection efficiency in airway epithelium. Hum Gene Ther 2001; 12:
1383–1394.
Boucher RC et al. Gene therapy for cystic fibrosis using E1deleted adenovirus: a phase I trial in the nasal cavity. The University of North Carolina at Chapel Hill. Hum Gene Ther 1994;
5: 615–639.
Joseph PM et al. Aerosol and lobar administration of a recombinant adenovirus to individuals with cystic fibrosis. I. Methods,
safety, and clinical implications. Hum Gene Ther 2001; 12:
1369–1382.
Floch V et al. Systemic administration of cationic lipids/DNA
complexes and the relationship between formulation and lung
transfection efficiency. Biochim Biophys Acta 2000; 1464: 95–103.
Barron LG, Gagne L, Szoka Jr FC. Lipoplex-mediated gene
delivery to the lung occurs within 60 minutes of intravenous
administration. Hum Gene Ther 1999; 10: 1683–1694.
Guillaume-Gable C et al. Cationic phosphonolipids as nonviral
gene transfer agents in the lungs of mice. Hum Gene Ther 1998;
9: 2309–2319.
Alton EWFW et al. Cationic lipid-mediated CFTR gene transfer
to the lungs and nose of patients with cystic fibrosis: a doubleblind placebo-controlled trial. Lancet 1999; 353: 947–954.
Chadwick SL et al. Safety of a single aerosol administration of
escalating doses of the cationic lipid GL-67/DOPE/DMPE-PEG
5000 formulation to the lungs of normal volunteers. Gene Therapy 1997; 4: 937–942.
10 Ishiwata H et al. Characteristics and biodistribution of cationic
liposomes and their DNA complexes. J Control Rel 2000; 65:
139–148.
11 Thierry AR et al. Systemic gene therapy: biodistribution and
long-term expression of a transgene in mice. Proc Natl Acad Sci
USA 1995; 92: 9742–9746.
12 Delépine P et al. Phosphonolipids: a class of nonviral vectors
rapidly eliminated and efficient in mice lung-directed gene
transfer. Hum Gene Ther (submitted).
13 Uyechi LS et al. Mechanism of lipoplex gene delivery in mouse
lung: binding and internalization of fluorescent lipid and DNA
components. Gene Therapy 2001; 8: 828–836.
14 Osaka G et al. Pharmacokinetics, tissue distribution, and
expression efficiency of plasmid [33P]DNA following intravenous administration of DNA/cationic lipid complexes in
mice: use of a novel radionuclide approach. J Pharm Sci 1996;
85: 612–618.
15 Lerondel S et al. Radioisotopic imaging allows optimization of
adenovirus lung deposition for cystic fibrosis gene therapy.
Hum Gene Ther 2001; 12: 1–11.
16 Sun S et al. Quantitative imaging of gene induction in living
animals. Gene Therapy 2001; 8: 1572–1579.
17 Delépine P et al. Cationic phosphonolipids as nonviral vectors:
in vitro and in vivo applications. J Pharm Sci 2000; 89: 629–638.
18 Guillaume C et al. Characterization and optimization of aerosolized cationic lipid-DNA complexes. Biochem Biophys Res Com
2001; 286: 464–471.
739
Gene Therapy