The purpose of this study was to investigate a new method of in vivo gene transfer to the lung parenchyma by the percutaneous approach. The plasmid that contains the gene for firefly luciferase driven by a cytomegalovirus (CMV) promoter (pCMVL) in combination with cationic lipids was percutaneously injected into the lung parenchyma. Luciferase activities were localized to the lobes of the lung where the plasmids with cationic lipids were injected. Percutaneous injection of the plasmid containing the human endothelin-1 (hET-1) gene driven by a CMV promoter (pRc/CMVhET-1) in combination with cationic lipids into the lungs caused pulmonary fibrosis localized to the injection site in the peripheral lungs. We concluded that percutaneous in vivo gene transfer to the lungs is a unique and important approach to introduce exogenous gene expression in the limited area of the lung parenchyma. This method of gene transfer will be applicable for human gene therapy for targeted areas of peripheral lung and will also be useful to assess the function of the proteins expressed by a gene in the local area of the lungs.
- cytomegalovirus promoter
- firefly luciferase
- human endothelin-1
- pulmonary fibrosis
exogenous gene transfer to the lung parenchyma provides an important strategy of gene therapy for the treatment of various heritable and acquired lung diseases (2, 5, 7). Treatment of airway diseases such as cystic fibrosis requires delivery of an exogenous gene to the airway epithelium. In contrast, gene therapy for alveolar diseases such as α1-antitrypsin deficiency, pulmonary fibrosis, and lung cancer requires delivery of therapeutic genes to the cells of alveoli.
Several methods of gene transfer into the lungs have been reported. Intravenous gene transfer has been reported as a strategy to transfer the exogenous gene to the lungs (16, 17). Compared with the systemic administration, direct intrapulmonary gene transfer is a relatively less invasive procedure and provides a high local concentration of exogenous genes. A transtracheal approach for gene delivery to the lungs with naked plasmids or plasmids with cationic lipid (1, 13, 27) and viral vectors (6, 24) has been reported as successful. However, this approach seems to be difficult in the transfer of the exogenous gene to the lung parenchyma due to the presence of mucociliary clearance of the airways (1,6, 9, 13, 24, 27).
The purpose of this study was to confirm the method of direct gene transfer targeted to the peripheral lungs by the percutaneous approach in vivo. Using a plasmid that includes firefly luciferase or human endothelin (hET)-1 driven by a cytomegalovirus (CMV) promoter in combination with a cationic lipid, we demonstrate potent luciferase activity and specific pathological change induced by hET-1 in the targeted lung area where plasmids were injected percutaneously.
MATERIALS AND METHODS
A549 cells, a cell line derived from a human lung adenocarcinoma (HSRBB Cell Bank, Osaka, Japan), were cultured in RPMI 1640 medium containing 10% heat-inactivated fetal calf serum, 25 U/ml of penicillin, and 25 μg/ml of streptomycin. The cells were cultured in a humidified incubator containing 5% CO2 at 37°C. All studies were carried out when these cells were 70–80% confluent.
We used a luciferase expression plasmid, pCMVL, as previously described (21). The hET-1 expression plasmid (pRc/CMVhET-1) was constructed by insertion of hET-1 cDNA into the pRc/CMV plasmid (Invitrogen, San Diego, CA). cDNA of hET-1 was a 672-bp cDNA segment constructed with a polymerase chain reaction of a human aorta cDNA library (Clontech, Palo Alto, CA) and primers specific for hET-1 cDNA (hET-1 sense, 5′-GGCCTCTAGACAGAATGGATTATTTGCTCA-3′ and hET-1 antisense, 5′-GATCTCTAGACGAAGTCTGTCACCAATGTG-3′) (14). The plasmids were propagated in Escherichia coli DH5α (TOYOBO, Osaka, Japan) grown overnight at 37°C in Luria-Bertani medium containing 100 U/ml of penicillin. Plasmid DNA was isolated with a plasmid extraction kit (Promega, Madison, WI).
In vitro transfection of plasmids.
Plasmid DNA (1 mg/ml) was mixed with synthesized cationic lipids (Tfx-50, Promega) (4) in a polystyrene tube (charge ratio of cationic lipids to DNA was 3:1), and the plasmids and cationic lipids were allowed to form a plasmid-liposome complex for 15 min at room temperature. The cultured cells were washed with basal medium twice and exposed to the plasmid-liposome complex with culture medium for 24 h.
In vivo gene transfer to rat lungs.
For in vivo studies, a luciferase expression plasmid was used as a gene transfer marker. Male Sprague-Dawley rats (5 wk, 120–130 g) purchased from Japan SLC (Hamamatsu, Japan) were used. All experiments using rats in this study were approved by the Institutional Animal Care and Use Committee. The rats were anesthetized with an intraperitoneal injection of pentobarbital sodium (5 mg/100 g body wt). The plasmid-liposome complex was percutaneously injected into the right upper lobe with a 27-gauge needle. Plasmid or cationic lipid alone was also injected as a negative control. In the dose-response experiments, 10, 50, or 100 μg of pCMVL-liposome complex were percutaneously injected into the right lungs. To evaluate the differences between gene transfer by transtracheal instillation and by percutaneous injection, another group was given 100 μg of pCMVL-liposome complex via the trachea. In the time-course experiments, 50 μg of pCMVL-liposome complex were percutaneously injected into the right upper site of the lungs. The rats were killed on days 2, 7, and 14, and the lungs were harvested for measuring luciferase activity and examining pathological changes as described inLung histology.
To evaluate the functional expression of the exogenous gene transfected by a percutaneous needle injection, we used a hET-1 expression plasmid, pRc/CMVhET-1. A complex of pRc/CMVhET-1 and cationic lipid was injected via either a percutaneous or an intratracheal route. The rats were killed on days 2, 7, 14, and28, and the excised lungs were evaluated for pathological changes.
Luciferase activity was evaluated in A549 cells 48 h after transfection. The cells were washed twice with PBS and incubated with 100 μl of lysate solution (Promega) for 15 min at room temperature. The cell lysates were recovered in Eppendorf tubes and centrifuged (15,000 rpm) for 2 min at 4°C. The supernatant was used for luciferase assay.
To evaluate luciferase activity in rat lungs in vivo, the rats were killed under deep anesthesia, and the lungs were excised after the pulmonary arteries were infused with ice-cold PBS. The lung was divided into each lobe (right: upper, middle, lower, and caudal; left: upper and lower), homogenized separately in 500 μl of lysate solution, and centrifuged (15,000 rpm) at 4°C for 2 min, and the supernatant was recovered. Luciferase activity in a mixture of 20 μl of lysate and 100 μl of luciferase assay substrate (Promega) was measured with a luminometer (luminescensor JNR, Atto, Tokyo, Japan) (8). Data for luciferase activity are expressed as relative light units (RLU) per milligram of protein. The total protein concentration in the cell lysates was evaluated by the Bradford method (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard.
ET-1 level in cultured cells.
Twenty-four and forty-eight hours after pRc/CMVhET-1 transfection, the supernatant of A549 cells was evaluated for hET-1 levels with an enzyme-linked immunosorbent assay (ELISA; IBL, Fujioka, Japan). Briefly, a microtiter plate was coated with a rabbit monoclonal antibody specific for human ETs. Experimental samples or standard samples of recombinant hET-1 were added to individual wells and incubated overnight at 4°C. After seven washes to remove unbound proteins, a monoclonal antibody specific for hET-1 conjugated to horseradish peroxidase was added to the wells and incubated at 37°C for 1 h. After another nine washes, a substrate solution containing hydrogen peroxide and tetramethylbenzidine was added and incubated for 30 min in the dark. The reaction was stopped by adding 1 N sulfuric acid. The color generated was determined by measuring the optical density at 450 nm with a spectrophotometric microtiter plate reader (model 450, Bio-Rad). The standard curve was lined with a log-log scale and subjected to a regression analysis.
ET-1 level in lung tissues.
After pRc/CMVhET-1 transfection, the rats were killed under deep anesthesia, and the lungs were harvested for measurement of ET-1 level on days 2 and 7. Briefly, the lungs were homogenized with a manual homogenizer in 1 M acetic acid (Wako Pure Chemical Industries, Osaka, Japan) containing 20 mM hydrochloric acid (Wako Pure Chemical Industries) and conditioned to 100 mg lung wet weight/ml. Homogenates were immediately boiled for 10 min and centrifuged (10,000 rpm) at 4°C for 10 min. ET-1 in 500 μl of supernatant was adsorbed to a SepPak C18 cartridge (Millipore, Milford, MA) and eluted with 60% acetonitrile (Wako Pure Chemical Industries) containing 0.1% trifluoroacetic acid in water. The samples were frozen, evaporated, and powdered by vacuum centrifugation (EYELA CPD-30, Tokyo Rikakikai, Tokyo, Japan). The samples were reconstituted with PBS and evaluated for ET-1 level with an ELISA as described inET-1 level in cultured cells.
On days 2, 7, 14, and 28after plasmid, cationic lipid, or plasmids-liposome complex transfer, the rats were intubated with a 16-gauge needle via a tracheotomy under deep anesthesia. After blood was drawn by direct puncture of the right ventricle, the lungs were fixed with 4% paraformaldehyde in PBS (pH 7.4) for 25 min at a constant pressure of 23 cmH2O. The fixed lungs were embedded in paraffin, and sagittal sections of each lobe were cut at 3-μm thickness. Each section was stained with hematoxylin and eosin or elastica-Masson-Goldner stain for histological evaluation.
The significance of differences in luciferase activity was evaluated by one-way ANOVA with Bonferroni-Dunn correction for multiple comparison after testing with StatView on a Macintosh personal computer. Differences were considered significant at P < 0.05.
Cationic lipid enhanced the efficiency of gene transfer in A549 cells as expressed by an increase in the luciferase activity. A lower level of luciferase activity was shown in A549 cells when pCMVL was transfected to the cells without cationic lipids. Luciferase activity in A549 cells was elevated with the dose of plasmid with cationic lipids in a concentration-dependent manner (data not shown).
In a preliminary study, we injected trypan blue dye into the lung via a percutaneous route to establish the technique for targeting the injection. We confirmed proper targeting and no complications such as lung collapse or life-threatening events in all animals in this study.
There was no histological change in the rat lungs in which pCMVL alone, cationic lipid alone, and pCMVL with cationic lipid were percutaneously injected or intratracheally instilled on days 2 and7 (data not shown).
Two days after in vivo transfection, significant levels of luciferase activity were detectable in the right upper and middle lobes where we directly injected the pCMVL-liposome complex via a percutaneous route. In contrast, there was no luciferase activity in the caudal lobes and left lungs in the same rat. We could not detect any significant luciferase activity in the rat lungs when cationic lipid alone or pCMVL without cationic lipid was injected (Fig.1). There was a larger variation in luciferase activity between the right upper and middle lobes of the rat lungs where the pCMVL-liposome complex was injected.
In Fig. 2, left, we show luciferase activity in all lobes of the rat lungs in which 100 μg of pCMVL with cationic liposome were intratracheally administered. In contrast, luciferase activity was demonstrated only in the right upper and middle lobes where the pCMVL-liposome complex was injected percutaneously.
In the dose-dependent experiment, luciferase activity was increased in accordance with the dosage of plasmid (data not shown). It is important to note that even if we injected the higher dose of plasmid-liposome complex in vivo, we could detect little luciferase activity in any lobe except the right upper and middle lobes in the rat lungs. Because the injections were performed percutaneously, it is possible that some rats received the plasmid in the upper lobe while others received it in the middle lobe. The data for individual rats are consistent with this hypothesis (Fig. 2, solid lines).
In time-course experiments, luciferase activity was detectable only onday 2, not on day 7 or 14 when we used 50 μg of pCMVL-liposome for percutaneous transfection (data not shown).
A549 cells spontaneously secreted hET-1 protein. The addition of cationic lipid alone to cultured cells did not affect the secretion of hET-1 from A549 cells. Transfection of pRc/CMVhET-1 with cationic lipid to A549 cells produced significantly higher levels of hET-1 (Fig.3). Thus we confirmed the functional property of pRc/CMVhET-1 in vitro.
ET-1 protein level in lung tissues was evaluated by ET-1-specific ELISA (Table 1). Lung tissue from the lobe in which the plasmid-liposome complex was transfected contained much higher levels of ET-1. In contrast, lung tissue from the untreated lobe of the same rat contained lower levels of ET-1 similar to those in control rats. Injection of pRc/CMVhET-1 or cationic lipid alone did not affect the level of ET-1 in the lung tissue.
The distribution of lung lesions was evaluated in a gross view of a section of whole rat lungs 7 days after transfection (Fig.4). There was no abnormality in rat lungs in which pCMVL (data not shown), pRc/ CMVhET-1, or cationic lipid alone was percutaneously injected (Fig. 4, Aand B, respectively). The severe fibrotic change was demonstrated as dense staining in the localized area of the right lung in which pRc/CMVhET-1 with cationic lipid was injected (Fig.4 C, arrowhead). In contrast, focal dense staining areas were distributed over both lungs in which pRc/ CMVhET-1 with cationic lipid was intratracheally administered (Fig.4 D, arrowheads).
At higher magnification, there was no pathological change in the lungs in which pRc/CMVhET-1 without cationic lipid was percutaneously injected (Fig. 5, A andD). In rat lungs with percutaneous injection of pRc/CMVhET-1 with cationic lipid, there was a marked pathological lesion in the right upper lobe of rat lungs (Fig. 5, B and E). Massive alveolitis was evident, and pulmonary fibrosis spread from the alveolar lesion to the visceral pleura (Fig. 5 E). Intratracheal administration of pRc/CMVhET-1 with cationic lipid caused a rather patchy distribution of fibrotic lesions in bilateral lungs (Fig. 5, C and F). Alveolitis was mainly observed around the peribronchial sites, and fibrotic changes spread in association with the bronchial tree, with bronchial wall thickening and epithelial hyperplasia (Fig. 5 F).
A chronological change in the histopathological findings was observed in rat lungs injected with pRc/ CMVhET-1 plus cationic lipid percutaneously (Fig. 6). Two days after injection of pRc/CMVhET-1 with cationic lipid into the lungs, the exudation of inflammatory cells into the alveolar space and alveolar septal thickening were observed (Fig. 6 A). The formation of connective tissue plaques in the parenchyma and extended collagen deposition were observed 7 days after transfection (Fig. 6,B and C). The fibrotic changes remained in the injected lobe on day 28 (Fig. 6 D).
In this study, exogenous genes (pCMVL and pRc/CMVhET-1) were introduced into the lung parenchyma in vivo by percutaneous direct gene transfer in a plasmid-liposome complex. We demonstrated that genes transferred by direct injection were expressed in the localized region of the injected site. Although attempts of percutaneous gene transfer to the lungs have already been reported by Nabel et al. (20) and Waddill et al. (26), their targets for gene transfer were metastatic tumors, not lung parenchyma per se. Thus we think that this is the first report demonstrating successful gene transfer to the lung parenchyma by a percutaneous approach.
Cationic liposome has been used to enhance the delivery and expression of plasmid DNA in vitro and in vivo (1, 13, 18, 27). A variety of studies using both viral vectors and cationic lipid as the principal applications of gene therapy for cystic fibrosis have been reported (1, 6, 13, 24, 27). Unlike viral vector systems such as adenoviruses or adeno-associated viruses, plasmid-based gene therapies with liposomes appear not to give rise to immune responses (7, 12, 22, 23). On this background, we used plasmids and a synthetic cationic lipid (4, 10, 25) that have been applied for efficient gene transfer in vitro and in vivo.
Our method of gene transfer to the lungs has several substantial advantages over the method of gene transfer with the transtracheal route. First, direct gene transfer to the lungs can provide a higher concentration of plasmid-liposome complex in the targeted region of the lung parenchyma. When the plasmid-liposome complex was administered via the transtracheal route, it was hindered by the mucociliary clearance of the airway and a minimal amount of plasmid-liposome complex reached the alveolar area (1, 6, 9, 13, 24, 27). Thus we were able to get the maximal effect of a transgene with a minimal dose of plasmid-liposome complex with percutaneous approaches.
Second, liposome-mediated ET-1 gene delivery by a percutaneous route expressed a specific pathological change in lungs. ET-1, a potent vasoconstrictive factor, has various effects on cells, e.g., DNA synthesis, hypertrophy, expression of the protooncogene, and proliferation with other growth factors (3). Liposome-mediated transtracheal instillation of a hET-1 expression plasmid caused pulmonary fibrosis that was patchily distributed and localized along the small airways. This pathological change was similar to the lung pathology reported by Takeda et al. (25). They demonstrated experimental bronchiolitis obliterans induced by in vivo hemagglutinating virus of Japan-liposome-mediated ET-1 gene transfer via a transtracheal route. In contrast, percutaneous gene transfer of pRc/ CMVhET-1 resulted in fibrosing alveolitis without bronchiolitis obliterans in our study. Independent of the mechanism of pulmonary fibrosis caused by ET-1 gene transfer, we clarified the importance of hET-1 in the development of pulmonary fibrosis by using a percutaneous gene transfer into the lungs. Thus our study demonstrated the importance of methods or routes of gene transfer into the lung, not only what kind of genes are expressed in the lung but also where genes are dominantly expressed within the lung. Several lines of evidence suggest that ET-1 is one of the potent mediators in the development of pulmonary fibrosis (11, 15, 19), although the precise mechanisms of how ET-1 causes development of pulmonary fibrosis remain to be elucidated. Overexpression of ET-1 in lung tissue may cause proliferation of lung fibroblasts. Luciferase studies demonstrated that gene expression lasted for at least 2 days. In contrast, ET studies showed that ET-1 levels in lung tissue were elevated for 7 days after transfection as demonstrated by ELISA. We speculate that ET-1 expressed by a transgene may induce endogenous ET-1, and it could be driven by a paracrine/autocrine-like mechanism (3) and accelerate the fibrotic change in the lung.
Finally, in a clinical aspect of gene therapy, this method, under the guidance of computed tomography, could be applicable to localized lung disorders such as lung cancer, intractable infectious lesions including fungus and tuberculosis, or localized pulmonary fibrosis. The helical computed tomography scan is able to clarify targeted lesions during the injection in real time. We can easily regulate the location and/or depth of the injection on the order of every centimeter. In addition, our method of gene delivery can be performed repeatedly with adequate volumes of injected plasmids, and the therapeutic effect at each time period can be evaluated.
In conclusion, percutaneous in vivo gene transfer to the lungs with liposomes is one of the important methods for analyzing the functions of exogenous gene products in the alveolar area of the lungs on a molecular basis. This method is also applicable as gene therapy for various lung parenchymal diseases.
We thank Dr. Kunihiko Yoshimura for the generous gift of luciferase plasmid and Eiji Tsuchida for technical assistance.
Address for reprint requests and other correspondence: H. Tomoike, The First Dept. of Internal Medicine, Yamagata Univ. School of Medicine, 2-2-2 Iida-Nishi, Yamagata City 990-9585, Japan (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society