| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;73:432-437
© 2002 The Society of Thoracic Surgeons
a Division of General Thoracic Surgery, University Hospital, Berne, Switzerland
* Address reprint requests to Dr Schmid, Division of General Thoracic Surgery, University Hospital, CH-3010 Berne, Switzerland
e-mail: ralph.schmid{at}insel.ch
Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29–31, 2001.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Methods. The F344 rats underwent instillation of 300 µg DNA (pCIluc, luciferase) to the left lung. Naked DNA, linear polyethylenimine, branched polyethylenimine, and lipid GL-67 (in either atelectatic or ventilated lungs) were assessed (n = 5 per group). After 24 hours, left lung PaO2 (mm Hg) and luciferase activity (RLU/mg) were measured. The median (range) was given, and the analysis of variance was applied, followed by the planned comparison on log-transformed data.
Results. In atelectatic lungs, lipid GL-67 was best (927 [330 to 4112] RLU/mg; p < 0.001 versus other groups of atelectatic lung; p < 0.001 versus all other groups), but highest luciferase activity in all groups was measured in ventilated lungs using linear polyethylenimine (1,240 [922 to 2519] RLU/mg; p < 0.001 versus other groups of ventilated lung; p < 0.001 versus all other groups). In comparison with naked DNA, all nonviral vector systems significantly impaired PaO2 24 hours after airway transfection (p < 0.001; naked DNA versus all other groups). Regardless of transfection technique, PaO2 was worst in lungs transfected by linear polyethylenimine.
Conclusions. Highest transfection was achieved with GL-67 in atelectatic lungs and with linear polyethylenimine in ventilated lungs. All gene delivery systems impaired gas exchange of the transduced lung in comparison with naked DNA.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Intravenous application of cationic lipids is limited by interaction of nonviral vectors with serum proteins [3]. Conversely, adeovirus based vectors have a high efficiency in vitro and in vivo, but are limited by inflamma-tory host response and formation of antibodies. Indeed, recent studies demonstrated prolonged adenoviral mediated transgene expression and less inflammation by combined application of immunosuppressive agents generally used after organ transplantation [4], but the efficiency of gene transfer to the target organ still has to be improved.
In this study we investigated transduction efficiency of three nonviral vectors in comparison with naked DNA to atelectatic and ventilated donor lungs in a small animal model and evaluated the effect of this procedure on oxygenation capacity of the transfected lung.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Gene transfer vectors
Branched 25 KDa polyethylenimine (brPEI) (Sigma-Aldrich, Schnelldorf, Germany) was prepared as a 0.1 mol/L aqueous solution as previously described [5].
Linear 22 KDa polyethylenimine (liPEI) (Exgene 500, Euromedex, Souffelweyersheim, France) was purchased as a 1 mol/L aqueous stock solution. The cationic lipid GL-67 (Genzyme, Framingham, MA) was obtained as a dried lipid film and reconstituted as a 2 mmol/L aqueous solution immediately before use.
Preparation of transfection complexes
The GL-67/DNA lipoplexes were prepared in purified sterile water, and the complexes between plasmid DNA and polymers were prepared in 5% dextrose. The appropriate quantity of liposome/polymer was diluted to a volume of 250 µL and 300 µg of plasmid DNA was separately diluted to the same volume. The diluted liposome and polymer solution was then added to the DNA solution followed by mixing with a pipette tip. After mixing, complexes were incubated for either 10 minutes (polyethylenimine) or 30 minutes (GL-67) before administration to rats. For complex preparation, the following amounts of vector were used per 300 µg DNA: 4.5 µL liPEI; 0.227 µmol GL-67; 46 µL brPEI.
Animals
Inbred male Fischer F344 rats weighing in the range of 230 to 260 g were used. All animals received humane care in compliance with the European Convention of Animal Care. The study protocol was approved by the local Animal Study Committee.
Study groups
Rats were divided into four groups with 5 animals per group undergoing atelectasis technique and 5 per group undergoing isolated ventilation. The groups were naked DNA, liPEI/DNA, brPEI/DNA, and GL-67/DNA.
Administration of complexed plasmid DNA
After induction of anesthesia by inhalation of halothane in a glass chamber, rats were intubated by a 14-gauge catheter to the trachea. The animals were ventilated by a Harvard rodent ventilator (Harvard Apparatus, South Natick, MA) at a tidal volume of 8 mL/kg, breathing frequency of 100, FIO2 of 1.0 and halothane 2% to 2.5%. A small anterior right thoracotomy was performed and lungs were transfected at body temperature. In the isolated ventilation groups, the right main bronchus was occluded by a microvascular clip. A solution of 500 µL, containing 300 µg plasmid DNA alone or complexed gene delivery vector, was instilled followed by left lung ventilation for 10 minutes. In the corresponding subgroup that underwent resorption atelectasis of the left lung, the solution was instilled and then a clip was placed on the left lung, and ventilation was maintained for 30 minutes on the right lung.
After this time, the microvascular clip was removed, a chest tube was inserted, and the suture was closed. The chest tube was removed from the animal when spontaneous breathing began; then the rat was returned to its cage.
Twenty-four hours after DNA administration, the animal was anesthetized by intraperitoneal administration of pentobarbital (50 mg/kg) and heparinized (500 IU/kg). Ventilation was carried out with an FIO2 of 1.0, a frequency of 100 per minute, and a tidal volume of 8 mL/kg body weight by a tracheotomy. For functional assessment of the transfected left lung, the right hilum was dissected, and the right pulmonary artery and right main bronchus were occluded by microvascular clips. Five minutes after occlusion, a steady state was reached and an arterial blood gas sample was drawn from the thoracic aorta. After removal of the clips, lungs were perfused with 20 mL saline through the pulmonary vein, and the lobes were divided into upper, middle and lower portions and were assessed independently. Samples were snap frozen in liquid nitrogen and stored at -80°C until analysis.
Luciferase assay
For luciferase assay, lung tissue was weighed and homogenized (Polytron, Littau, Switzerland) in ice-cold reporter lysis buffer (Promega, Zurich, Switzerland) for 1 minute, freeze thawed three times, and centrifuged at 14,000 g for 15 minutes. The luciferase activity in 20 µL of lysate was measured on a luminometer (Mediators PhL, Mediators Diagnostika, Vienna, Austria) programmed to inject 100 µL of luciferase assay buffer (Promega). Integrated light units were collected for more than 10 seconds, and all results were normalized to protein content in each sample (Dc protein assay kit, Bio-Rad, Reinach, Switzerland) using bovine serum albumin as a standard. Data are expressed as relative light units per mg protein (RLU/mg).
Statistics
The STATISTICA 5.1 software (StatSoft, Tulsa, OK) was used. For each animal, three measurements of luciferase activity were carried out for either left or right lung. The median of these values was used. Data were not normally distributed; therefore, analysis of variance with planned comparison (contrast vector analysis) was applied on log-transformed data. Arterial oxygen content was analyzed by analysis of variance followed by planned comparison. A p value of less than 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Ventilated lungs
Highest luciferase activity in all groups was measured in ventilated lungs using liPEI (1240 [922–2519] RLU/mg; p < 0.001 versus other groups ventilated lung; p < 0.001 versus all other groups; p < 0.001 versus liPEI atelectasis lung) (Fig 1). Transfection was much less in all other vector systems (brPEI 532 [409–635] RLU/mg; p = 0.039 versus liPEI; GL-67 390 [56–1280] RLU/mg; p < 0.001 versus liPEI; naked DNA 85 [23–162] RLU/mg; p < 0.001 versus liPEI).
|
Oxygenation
A significant statistical difference between groups (p < 0.001) and between atelectatic and ventilated lungs (p = 0.0016) was observed, but interaction was significant, too (p < 0.001). Therefore, planned comparison with contrast analysis was applied.
In comparison to naked DNA, all nonviral vector systems significantly impaired PaO2 24 hours after airway transfection (p < 0.001; naked DNA versus all other groups) (Fig 2). Regardless of transfection technique, PaO2 was worst in lungs transfected by liPEI (atelectatic lung 94 [55–109] mm Hg; p = p < 0.001 versus naked DNA atelectatic lung; ventilated lung 116 [82–253] mm Hg; p < 0.001 versus naked DNA ventilated lung).
|
Distribution of gene delivery
No preference for gene distribution between upper, middle, or lower portion of the lung was observed (p = 0.76 for all groups; n = 120).
Regardless which delivery technique and vector system was used, luciferase activity in the right lung was minimal with a median between 0.0 RLU/mg and 9.05 RLU/mg, indicating gene delivery to the left lung only.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
To date viral and nonviral gene delivery systems may be used to transduce the donor organ before transplantation. Adenovirus, which seems to be the most effective vector for the lung in clinical trials, may raise some safety concerns in the setting of organ transplantation.
Simon and colleagues [6] evaluated the toxicity of adenovirus-mediated transfer of the cystic fibrosis transmembrane conductance regulator gene to baboons by endoscopic instillation. At higher adenoviral levels, a perivascular lymphocytic and histiocytic infiltrate was observed. The intensity of inflammation increased between 4 and 21 days leading to diffuse alveolar wall damage and edema. In macaques, administration of 1010 plaque-forming units of a recombinant adenovirus coding the cystic fibrosis transmembrane conductance regulator-gene led to a progressive increase in CD8+ lymphocytes in bronchoalveolar fluid [7]. Histologic examination demonstrated peribronchial and perivascular cuffing by inflammatory cells and accumulation of neutrophils and macrophages in the alveolar space. Administration of only 4 x 106 plaque-forming units was not associated with an immune response, but cystic fibrosis transmembrane conductance regulator expression was low. McCray and coworkers [8] found an unexpected reactive hyperplasia and squamous metaplasia in the epithelia of the transfected trachea of fetal lambs. Within 1 week after transfection, the adenovirus-treated fetuses had massive inflammatory cell infiltrates develop, and transgene expression faded coincident with serologic evidence of anti-adenoviral antibodies and inflammation. After repeated administration of adenovirus to rat lungs, the induction of antiviral immune response significantly reduced the transgene expressison level [9]. A negative correlation between expression and serum neutralizing anti-adenoviral antibodies was observed.
In contrast, triple immunosuppressive drugs (cyclosporine, azathioprine, and methylprednisolone) increase transfection efficiency and allow retransfection after 5 weeks [4]. But preformed antibodies might limit this approach, as for example, in a cystic fibrosis population, IgG antibodies against wild-type adenovirus are observed in up to 96% of the population, which most likely limits the efficacy of adenoviral vectors in this group of lung transplant recipients [10].
The nonviral vectors used in this study were chosen based on promising reports regarding their efficiency. Lipid GL-67/DOPE is a cholesterol-based cationic lipid, formulated with the fusogenic helper lipid, dioleoylphosphotidylethanolamine. This liposome preparation was reported to be the least toxic and most efficient in a series of compounds developed for pulmonary gene transfer to mice lung [11]. Polyethylenimine, first described in 1995 [5] for the transfection of mammalian cells, was later shown to be effective for transfection of various tissues including the lungs [12], liver, and brain [13]. It has been stated that cytotoxicity in vitro was low in concentrations needed for optimal transfections, and that the optimal PEI cation and anion balance for in vitro transfection is only slightly on the cationic side, which should be advantageous for in vivo delivery [5]. Polyethylenimine has the highest cationic charge density of any known organic polymeric macromolecules and its superior transfection activity is believed to originate from its ability to act as a "proton sponge" under acidic conditions. Essentially the protonation profile for PEI increases from 20% to 45% between pH 7 and pH 5 (conditions which would be encountered within the endosome). This buffering is believed to prevent the degradation of internalized plasmid DNA. In addition to its endosomal buffering capabilities, PEI has recently been shown to facilitate transfer of plasmid DNA across the nuclear membrane, thereby overcoming one of the fundamental barriers to successful gene delivery. This is an advantage over the use of liposomes that are retained within endosomal compartments for considerable periods of time after cellular uptake [14].
The main theoretical advantage of nonviral vectors in comparison with viral vectors is that they do not cause host response with subsequent inflammation. However, Freimark and coworkers [15] demonstrated induction of cytokines and cellular influx in mice lungs after airway delivery of a positively charged cholesterol and plasmid complex. Tumor necrosis factor-alpha is rapidly induced by the complex or lipid alone, as assessed by bronchoalveolar lavage. The induction of the innate immune response led to an acute cellular influx, which is mainly due to the lipid and only to a much lesser extent due to the plasmid itself.
In contrast, even the repeated administration of linear PEI for up to 21 consecutive days to animals did not lead to toxic side effects or elicit an immune response against liPEI [16]. Another study evaluated the pulmonary toxicity of GL-67 [17]. A dose-dependent inflammation mainly by neutrophils in association with elevated levels of pro-inflammatory cytokines was observed. The per se lesser toxicity of nonviral vectors may be outweighed by the high amount of complexes needed for relevant transduction, and this fact may explain the reduced oxygenation capacity of lungs treated with either polymers or lipids in this study.
Adenovirus-mediated gene transfer of ß galactosidase in a rat lung transplantation model demonstrated uneven distribution of ß galactosidase expression, which ranged from only a few cells staining per slide to up to 75% [18]. In our study, variability of expression between upper, middle, and lower portions of the lung was not significant; however, transduction efficiency was inferior in comparison with studies using adenoviral vectors. This lower transduction efficiency with nonviral vectors may at least in part be explained by the surfactant layer of the lung. The gene expression of naked DNA delivered endobronchially in mice is completely abrogated in the presence of synthetic surfactant, suggesting that surfactant acts as a barrier to transfection of the airways [19]. Using two different cationic lipid formulations and the luciferase reporter gene system for transfection of respiratory cell lines, Duncan and colleagues [20] demonstrated inhibition of transfection by pulmonary surfactant lipids and proteins. The higher transduction efficiency of GL-67 in atelectatic lungs, in which the presence of surfactant is reduced, in comparison with ventilated lungs might be explained by this mechanism. In contrast, liPEI demonstrated highest transduction efficiency in the ventilated lung, suggesting that transduction with liPEI is less influenced by the amount of surfactant present during transfection.
Clinical relevance of this study regarding lung transplantation may be limited by the fact that transfection has been carried out in perfused lungs at body temperature, and most transfection studies on experimental lung transplantation used ex vivo transfection through the vascular bed at low temperatures. Conversely, the kind of transfection presented in this article demonstrates another approach that makes transfection of the donor lung already on the ventilator, and therefore before multi-organ harvest feasible. However, future studies have to confirm that results of airway transfection in a transplantation setting after cold ischemia are at least comparable with hitherto observed transfection rates.
All nonviral vectors reduced gas exchange of the transduced lung 24 hours after gene delivery. This effect seems to be an inverse correlation with transduction efficiency and it is very relevant for PEI vectors. Toxicity of GL-67 is less pronounced in our hands. Taking all factors into account, GL-67 used in atelectatic lungs seems to be the best compromise between efficiency and toxicity for nonviral gene transfer to rat lungs.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
DR STAMMBERGER: This study was planned as a screening study to evaluate the best delivery mechanism for transfection of lung tissue with nonviral vectors, therefore, we did not assess oxygenation capacity in atelectatic lungs for a longer period. At least one explanation for the effect we observed, might be that, as it has been shown with artificial surfactant, cationic lipids and polymers mix and interact with surfactant, thereby destroying the surfactant layer.
DR MARK I. BLOCK (San Francisco, CA): Did you look at any other organs, liver, heart, to see if your delivery system got out of the lung and into the peripheral circulation?
DR STAMMBERGER: No, we did not look at other organ systems because several groups showed that expression in other organs is very low using gene delivery through the airways. We looked at upper, middle, and lower parts of the right lung to confirm that targeted gene delivery only to the left lung is feasible. In addition, luciferase activity in the trachea was assessed, which was very low in comparison with the activity measured in the right lung.
DR BLOCK: And then a global question. One of the problems down the road for this technology will be particularly when you introduce genes into patients that do not have them, such as alpha-1 antitrypsin deficiency; expression of those genes will generate an immune response because the gene product will be seen as a foreign antigen. So how do you perceive this technology proceeding in the face of these immunologic concerns? My question is not so much the vector you use, but the gene that you deliver. For example, if you deliver alpha-1 antitrypsin to a patient with alpha-1 antitrypsin deficiency, their immune system will see alpha-1 antitrypsin as a foreign protein and mount a response against that protein. So it is not really the vector I am so much concerned about, but when you are delivering new genes into patients who have not seen them before. How are you going to deal with that problem?
DR STAMMBERGER: We agree that the immune response against either the vector or the expressed protein may be a major issue; however, it has been demonstrated that most of the immune response is against the viral surface proteins of viral vectors. The immune response against the functional protein has to be minimized by using adequate genes.
DR ZWISCHENBERGER: You have described three different gene transfection techniques, all of which show promise. How much is enough? How much gene transfection do you need to accomplish before you can deliver a gene of significance to impact on the clinical state you are trying to change?
DR STAMMBERGER: I think that this depends on the underlying disease. For example, in cystic fibrosis it has been shown that only very low levels of protein expression are needed, but, on the other hand, repeated administration is needed for the lifetime of the patient. So an ideal vector for this disease may have comparable low in vivo efficiency, but should have minimal immunogenic side effects. On the other hand, in ischemia and reperfusion following lung transplantation, high levels of expression for a short period are needed, and immunogenic side effects may be less important because the patient is under immunosuppression.
DR DAO M. NGUYEN (Bethesda, MD): Luciferase is a very sensitive reporter system. You can have one cell transduce and have a high level of expression of the luciferase activity. Have you looked at another reporter system, such as beta-gal, to see what is the percentage of the pneumocytes transduced with your system? Probably that is a better way of reporting the activities of the delivery system that you want to study.
DR STAMMBERGER: Thank you for your comment. We checked the expression with beta-gal. The percentage of transduced cells is still quite low, but other groups showed a biological response (eg, reduced rejection of allografts) with a quite low transduction rate.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
