Gene Therapy (1997) 4, 162–171 1997 Stockton Press All rights reserved 0969-7128/97 $12.00
High efficiency reporter gene transfection of vascular tissue in vitro and in vivo using a cationic lipid–DNA complex M-C Keogh1, D Chen1, F Lupu1, N Shaper1, JF Schmitt1, VV Kakkar1 and NR Lemoine1,2 1
Thrombosis Research Institute, Emmanuel Kaye Building, London, and 2Imperial Cancer Research Fund, Molecular Oncology Unit, Hammersmith Hospital, London, UK
Efficient transfection conditions for a number of human, rat and rabbit primary cells and established lines of vascular origin have been determined using a complex of a commercially available cationic lipid transfection agent (Tfx50) and luciferase reporter plasmid constructs. The optimised conditions have also been successfully applied to rabbit carotid arteries in vivo and a series of human arteries in vitro. The most critical factors influencing the efficiency of gene transfection with this protocol are: DNA concentration; ratio of lipid reagent to DNA; transfection time and the presence or absence of serum. Immunohistochemical analysis shows that a high percentage of
cells (approximately 30–80% dependent on lineage) were transfected under optimal conditions with minimal toxicity effects. Similar analyses performed on undamaged rabbit carotid vessels transfected in vivo and human arteries transfected in vitro show high-efficiency transfer and strong expression of the luciferase vector as demonstrated by reporter gene expression. The optimisation of gene transfer into vascular cells with this cationic lipid complex will be valuable for molecular studies of genes implicated in cardiovascular diseases and as a possible method of gene delivery with therapeutic intent.
Keywords: gene therapy; cationic liposomes; luciferase; Tfx-50; transfection; vascular tissues
Introduction The use of percutaneous transluminal coronary angioplasty (PTCA) has increased greatly over the past 10 years. However, the long-term luminal restenosis rate has remained constant at 30 to 50%, despite drug regimens which have been used in an attempt to control it.1,2 The exact cause of restenosis remains controversial, although the migration and overproliferation of vascular smooth muscle cells (VSMCs) and simultaneous production of extracellular matrix have been implicated.3–5 This process is proposed to be due to an exaggerated healing response and vascular remodelling by the vessel following PTCAinduced damage. In support of this hypothesis, examination of restenosed vessels has shown high-level expression of proliferating cell nuclear antigen (PCNA) and other growth markers in smooth muscle cells.6,7 A number of strategies for gene therapy have been suggested which may prevent the progressive proliferation of VSMCs and alleviate the problem of restenosis.8,9 The antisense approach involves the use of synthetic oligonucleotides or expression constructs designed to produce nucleic acids complementary in sequence to target growth-associated genes in order to block their translation by a variety of mechanisms.10–13 Alternatively, entire genes have been inserted into cells, either being Correspondence: NR Lemoine, ICRF Molecular Oncology Unit, Hammersmith Hospital, London W12 0NN, UK The contribution of the first two authors is considered equal Received 19 August 1996; accepted 5 November 1996
expressed in a modified form to compete with the wildtype,14 or as a cytotoxic gene product to interfere with cellular metabolism.15 However, many of these strategies are hampered by the low transfection efficiencies of present protocols. Many protocols involving cationic liposomes for the delivery of DNA,16–19 RNA20,21 or antisense oligonucleotides11,12,22 into cells and tissues have been described in vitro and in vivo.23 Indeed, the protocols have progressed to the point where a number of human gene therapy trials have been devised utilising cationic liposomes to facilitate gene delivery,24,25 although none yet target cardiovascular disease. Cationic liposomes have a number of potential advantages over viral gene delivery systems. These include the ability to use a range of gene constructs from simple plasmids to chromosomal fragments; the potential for cellspecific targeting (based on the narrow efficiency windows for different cell lineages); reduced safety concerns relative to those associated with viral genomes; easier transfection protocols and less propensity to induce an immune response in vivo.23,26–31 The mechanism by which cationic liposomes mediate their activity relies on neutralizing the negative charge of the DNA intended for transfection. This results in a condensation reaction and the formation of stable complexes with a net positive charge which then associate with the negatively charged surface of the cell.32,33 This complex then either fuses with, or causes a transient destabilisation of, the cell membrane. The net result of this is macromolecule delivery to the cytoplasm while
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
avoiding fusion with lysosomes and subsequent degradation.34 The Tfx-50 reagent utilised in these experiments is a mixture of a synthetic cationic lipid molecule (N,N,N′,N′-tetramethyl-N, N′-bis(2-hydroxyethyl)2,2,-dioleoyloxy-1,4-butanediammonium iodide) and ldioleoylphosphatidylethanolamine (DOPE).35 The use of DOPE is common to many cationic liposome formulations and its role is to facilitate the membrane fusion or destabilisation step in macromolecule delivery. 34,36 Studies on the mechanism of Tfx-50-mediated gene transfection indicate that it is not mediated by acidic endocytotic vesicles.37 Hence it is classed as a ‘cytofectin’, which describes the ability of a compound to facilitate macromolecule entry into living cells while avoiding the lysosomal pathway.22,34 Although the application of physical gene transfection techniques has begun to show promise, the efficiency of introduction of foreign DNA probably falls short of that required for effective gene therapy for vascular disease. However, this could depend to a large extent on whether the gene product is cell-localised or secreted to mediate its effects on the surrounding population.38,39 Additional difficulties have been encountered because each cell type from each species has a characteristic set of optimal transfection conditions which complicates the translation of results from model systems to clinical applications and vice versa. To determine the optimal transfection conditions for the introduction of foreign genes into vascular cells, the present studies were carried out for a number of vascular lineages from different species using the new liposome compound Tfx-50.35
Results Gene transfection in vitro – optimization of transfection conditions in cultured cells The efficiency of various transfection conditions for rabbit, rat and human arterial smooth muscle cells (SMCs), human umbilical vein endothelial cells (HUVECs) and the human VSMC (ATCC CRL-1999) cell line have been determined using Tfx-50 and the pGL3 luciferase vectors. The conditions used for the HepG2 cell line were as previously described.35 The transfection optimisation for three representative populations (human aortic smooth muscle cells, HUVECs and the HepG2 cell line) are shown in Figure 1. These demonstrate that: (1) the transfection profile for each population varies; and (2) a narrow window for efficient transfection exists – conditions outside those optimal can result in a dramatic reduction of luciferase activity. Unlike the HepG2 cell line which was successfully transfected in the presence of serum, we observed that luciferase activity in the other lineages was dramatically reduced under these conditions. For intrapopulation comparisons the luciferase enzymatic activity derived from each following a representative transfection under relevant optimal conditions is shown (Table 1). Transfection efficiency in these experiments is determined by enzymatic analyses of luciferase activity. Time course: transfections In order to determine the time course of DNA uptake under the optimised conditions, a series of transfection analyses were performed. In these experiments the transfection solution (Tfx-50:pGL3 control luciferase) was
applied to the cells for the appropriate period, following which it was removed and the cells cultured in complete medium for 48 h before harvesting and enzymatic analyses. The results (Figure 2) show that luciferase activity is detectable even when the transfection time is as short as 5 min and this activity increases with extension of transfection time. A plateau of luciferase activity is reached for each lineage such that increasing the time within the parameters studied results in no increase in transfection. Parallel experiments utilising a 2-h exposure of each population to naked DNA at the same concentration produced no detectable luciferase expression indicating that it is the presence of the Tfx-50 in the transfection mix which is mediating the transfection as opposed to the vector being taken up via an alternative mechanism.
Toxicity of Tfx-50 or Tfx-50–plasmid DNA complexes The optimised conditions determined for a panel of populations were applied to determine whether the Tfx-50–DNA complexes have a toxic effect. Comparisons were made in each case to nontransfected and Tfx50 only controls. Relative toxicity was then determined 48 h later by comparing the proliferation level of each population assayed following a 4-h pulse labelling with tritiated thymidine. There is no significant difference in the level of thymidine incorporation between transfected and nontransfected human and rabbit smooth muscle cells (95% confidence interval). However HUVECs and rat SMCs do exhibit reductions in incorporation (approximately 10 and 25% respectively) indicating that the transfection complexes have a significant toxic effect on these populations. Certain lines do exhibit a major toxic response to Tfx-50–DNA complexes; it proved impossible to optimise transfection in the L6 rat skeletal muscle myoblast cell line due to rapid cell death following exposure to the complexes (data not shown). Immunohistochemical analyses Luciferase activity analyses of a transfected population do not indicate whether all the cells or merely a subpopulation have been successfully transfected. For this reason immunohistochemical analyses were performed. Populations were transfected with the pGL3 control luciferase vector utilising optimal or suboptimal conditions (reduction in amount of vector DNA, all other variables constant) as previously determined by enzymatic analyses. The populations were then assayed for luciferase expression 48 h after transfection. Under the tested conditions .85% of rabbit VSMCs express the luciferase protein at 48 h, while use of the higher concentrations of vector DNA (to optimal) results in a greater level of luciferase expression per cell (Figure 3a–c). Control rabbit VSMCs cells transfected with the promoterless pGL3 basic luciferase vector under the same conditions show no luciferase expression (Figure 3d). Similar analyses performed on rat and human VSMCs and the HepG2 cell line show that a lower percentage of cells (>30%) in these populations express luciferase (Table 2) when transfected under optimal conditions. Transfection analyses on human arteries in vitro The previous experiments have shown that following optimisation, Tfx-50 mediated transfection efficiently facilitates gene delivery into cell monolayers. However if
163
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
164
Figure 1 Optimisation of Tfx-50–DNA transfection conditions. Cells were transfected in triplicate utilising the 32 (16 for HepG2) conditions shown. Conditions are numbered 1–32 and expressed as follows: mg DNA, ratio of Tfx-50 to DNA, transfection time (in h), presence (+) or absence (−) of serum. Data points are expressed as the mean of triplicates of luciferase enzymatic activity with the standard error of the mean.
Table 1 Optimal transfection conditions for each population Cells
HUVEC Rab VSMC Rat VSMC Hum VSMC HepG2 HuSMC line
DNA per well (mg)
Serum (±)
Charge ratio Tfx-50–DNA
Incubation time (h)
Average value (light units ± s.e.m.)
0.5 0.5 0.5 0.5 0.5 0.5
− − − − + −
4:1 4:1 4:1 4:1 4:1 4:1
2 1 1 1 2 2
2209 ± 100 4082 ± 254 1929 ± 126 1361 ± 381 4880 ± 584 1431 ± 36
the protocol is to be successfully applied to a gene therapy situation it is essential that it should be applicable to intact tissues. A series of human vessels were obtained from transplant donors and transfected as described utilising conditions determined as optimal from the relevant isolated cells (human VSMCs, Table 1). Both normal and atherosclerotic vessel segments (as determined
by grossly visible atherosclerotic plaques) were assayed. Analyses on vessel segments fixed 48 h after transfection show strong luciferase expression in the cells of the intimal layer with scattered staining in cells of the medial layer (Figure 4 a, c and d). The luciferase expression in these cells is localised in the cytoplasm, with the nonfluorescent nucleus being clearly outlined (indicated). Non-
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
165 Table 2 Transfection efficiency in each population Cells
Figure 2 Tfx-50–DNA complex transfection time course. Cells were transfected with the optimal complex for the appropriate time period as described. Data for each time-point (untransfected, 5, 10, 20, 30, 45, 60 and 120 min) are expressed as the mean of triplicates of luciferase enzymatic activity with the standard error of the mean.
transfected vessel sections analysed using the same protocol show no detectable staining (Figure 4b).
Transfection analyses on rabbit common carotid arteries in vivo To determine the transfection efficiency and extent of penetrance of the Tfx-50–DNA complexes into the rabbit common carotid artery in an in vivo situation, experiments were performed as described. Immunocytochem-
pGL3 control per well (mg)
% Luc (+) cells
Rat VSMC
1.0 0.5 0.25
33.14 29.1 22.4
Hum SMC
1.0 0.5 0.25
40.5 35 15
HepG2
1.0 0.5 0.25
19 29.9 27.7
ical analyses of the transfected arteries show the majority to exhibit luciferase expression localised to the intimal layer (Figure 5a) while in some cases strong luciferase expression which penetrates through the internal elastic lamina into the media is seen (Figure 5b). Nontransfected vessels analysed using the same protocol show no detectable staining (Figure 5c). It is possible that these differences are due to the integrity of the endothelial layer with its tight junctions such that those vessels which exhibit deep penetrance have had the endothelial layer damaged as a consequence of the transfection procedure utilised. This is supported by the colocalisation of von Willebrand factor (vWf) (indicating the presence of endothelial cells) and luciferase expression in a vessel which exhibits intimal luciferase expression (Figure 6, inset).
Duration of gene expression The peak of gene expression following transient transfection is generally acknowledged to occur at 48–96 h and
Figure 3 Immunocytochemical detection of luciferase expression in rabbit VSMC monolayers. Cells were transfected under optimal conditions using the pGL3 control luciferase or pGL3 basic luciferase vectors and assayed as described. (a) 0.5 mg pGL3 control luciferase; (b) 0.25 mg pGL3 control luciferase; (c) 0.125 mg pGL3 control luciferase; (d) 0.125 mg pGL3 basic luciferase (negative control). Magnification: (a–d) × 300.
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
166
Figure 4 Immunocytochemical analyses of human arterial segments following Tfx-50 mediated transfection in vitro. Vessel sections were transfected in vitro as described. (a, b) Healthy arterial segments were transfected with the transfection complexes under optimal conditions as described. Luciferase expression is detectable by immunofluorescent staining and localised mainly to the immediate intimal layer and some underlying smooth muscle cells (a, arrowed). Transfection with the negative control pGL3 basic luciferase vector (b) shows no detectable luciferase expression. (c, d) Atherosclerotic arterial segments transfected with the pGL3 control luciferase vector show a heterogeneous distribution of labelling through several intimal smooth muscle cell layers. Note the elongated shape of the arrowed cells characteristic of VSMCs of the contractile phenotype. Magnification (a–d) × 40.
reduce thereafter as the episomal gene constructs are degraded. Experiments were performed to determine the level of luciferase gene expression after this time utilising the Tfx-50 protocol. Human and rabbit vessel sections were analysed for luciferase expression by immunohistochemistry 5 and 10 days after transfection as described (Figures 6 and 7). Both the human and rabbit arteries show strong luciferase expression at day 5 (Figures 6a and 7a). At day 10 however a dramatic reduction in luciferase expression is readily apparent in rabbit carotid arteries (Figure 6b) relative to the earlier time-points (Figure 6a), while human vessels analysed for luciferase expression 5 or 10 days after transfection are indistinguishable (Figure 7a and c).
Discussion Recent studies indicate that the use of cationic liposomes to facilitate the functional delivery of reporter genes such as luciferase, b-galactosidase or chloramphenicol acetyl
transferase (CAT) into vascular cells in vitro,39,40 or into the vascular wall in vivo18,41,42 is possible. In this investigation we have verified that using a new commercially available liposome, Tfx-50, it is possible to transfer the luciferase gene to various vascular cells both in vitro and in vivo at higher efficiencies than previously reported for a complex of this type.32,40,43 The DNA concentration, the ratio of liposome to DNA (which affects the net charge of the complexes formed), the transfection time and the effect of serum are the most critical factors in each transfection, and some of the conditions vary depending on the lineage tested. Previous in vivo transfection analyses utilising liposomes have been performed to transfer reporter genes into the vessel walls of intact animals, but with disappointing results. The reasons for the lower transfection efficiencies than those recorded in in vitro analyses are not clear, but may be related to a number of factors: (1) the optimal transfection conditions for vascular cells in vitro may not correspond to those in vivo or the optimal
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
plexes with serum proteins such as lipases, or a blocking effect by the glycocalyx layer which covers the luminal surface of intact vessels. The efficiency of luciferase expression following the transfection of intact vessels is encouraging but raises a number of questions: why do certain cells deep within the vessel express high levels of luciferase while their neighbours are apparently nontransfected? The majority of positive cells have a classical contractile phenotype, being elongated in shape. It is possible that the efficiency of transfection is clonal in nature and this is currently being investigated. Previous studies utilising liposomemediated transfection of isolated vascular smooth muscle cells has shown differences in DNA uptake dependent on whether the cells came from healthy tissue, atherosclerotic plaque or restenotic lesions,40 which further supports this possibility. The in vivo experiments described herein involve a relatively nontraumatic transfection protocol (isolate and dwell). It may be possible to increase the transfection efficiency still further by incorporation of other approaches such as: (1) catheter-mediated gene delivery using a double balloon system; (2) the exploitation of systems which have been used successfully to improve cationic liposome-mediated gene transfection, such as complexing with polylysine conjugates, inactivated Sendai virus or adenovirus particles, AAV-based plasmids or cholesterol.32,44–46 Using the protocol we describe, transfection of intact vessels is not only possible but efficient. In summary: (1) gene transfection occurs within 5 min and gene expression is present in 30–85% of cells in an optimally transfected cell monolayer in vitro; (2) the transfection complexes appear to traverse the endothelial and internal elastic lamina in undamaged vessels; (3) toxicity is not excessive in vitro not in vivo; (4) luciferase expression is still easily detectable 10 days after transfection. Taken together these results suggest that these optimised conditions for gene transfer into vascular cells with Tfx-50 will be valuable for molecular analyses requiring transfection and gene therapy approaches to treat cardiovascular diseases such as restenosis.
Materials and methods
Figure 5 Immunocytochemical analyses of rabbit carotid arteries following Tfx-50 mediated transfection in vivo. Vessel sections were transfected in vivo as described. (a) Vessels showing luciferase expression localised to the immediate endothelial and intimal smooth muscle cell layers. (b) Deep penetrance of luciferase expression as occasionally seen. (c) Nontransfected vessels (negative controls) show no detectable staining for luciferase expression. Magnification: (a, b) × 130; (c) × 80.
conditions obtained from different types of tissues may not be similar to those of animal arteries; (2) the transfection of liposome–DNA mixtures into cell monolayers in serum-free medium in vitro reflects a different condition to exposure to DNA injected into the lumen of target arteries during gene transfer in vivo. This could be due to inhibitory in vivo interactions of the transfection com-
Materials Tfx-50 and the luciferase assay system were purchased from Promega (Southampton). Directly conjugated polyclonal rabbit anti-luciferase*FITC and polyclonal rabbit anti-luciferase antibodies were purchased from Europa Research Products (Cambridge, UK). Goat anti-rabbit IgG*FITC was obtained from Sigma Chemicals (Dorset, UK). Monoclonal mouse anti-a-actin was from Boehringer Mannheim (East Sussex, UK) and polyclonal antivon Willebrand factor (vWf) was from Dako (High Wycombe, UK). The pGL3 luciferase control vector (SV40 promoter and enhancer) and pGL3 luciferase basic vector (promoterless and enhancerless) were purchased from Promega, transformed into E. coli (JM109) and purified on Qiagen-500 (Qiagen, Surrey, UK) columns before transfection. Heat-inactivated FCS was purchased from Harlan Seralabs (Crawley Down, UK). Medium and supplements were purchased from Gibco Life Technologies (Paisley, UK). Tritiated thymidine (3 H-TdR) was
167
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
168
Figure 6 Demonstration of luciferase expression in rabbit carotid arteries. Rabbit carotid arteries were transfected in vivo as described and assayed for luciferase expression 5 or 10 days later. (a) Luciferase expression is detectable through a number of layers, with the number of positive cells decreasing towards the medial layer. (b) Luciferase expression after 10 days is very low, present in a very small number of scattered cells. Inset: double staining for luciferase expression and the endothelial cell marker von Willebrand factor (vWf) demonstrating that the presence of an intact endothelial layer could prevent deep penetrance of the Tfx-50–DNA transfection complexes. Magnification: (a) × 125; (b) × 105; inserts, × 55.
purchased from Amersham Life Science (Amersham, UK). Human arteries were kindly provided by King’s College Hospital Liver Transplant Unit, London, UK, from organ transplant donors with the consent of relatives.
Cell culture and transfection Smooth muscle cells from rabbit and rat aorta, or human arteries (aorta, inferior mesenteric or splenic), and endothelial cells from human umbilical veins (HUVECs) were routinely grown as follows: rabbit SMCs: M199, 10% FCS, 1% nonessential amino acids (NEAA), 5% CO2 , passage 2 (p2) only; rat SMCs: DMEM, 10% FCS, 10% CO2 , p8–
p15; human SMCs: DMEM, 10% FCS, 10% CO2, p2–p8; HUVECs, M199, 10% FCS, 5% CO2, 20 mg/ml ECGF, 80 mg/ml heparin, fungizone, p2–p8. The HepG2 cell line was grown in DMEM, 10% FCS, 5% CO2. The cultures were maintained at 37°C in a humidified incubator with CO2 as described. All cultures also contained penicillin, streptomycin and l-glutamine as standard. For these studies, cell cultures were maintained in the exponential growth phase and used between the passages described. Cells for transfection were plated at 5 × 104 cells per well in 12- or 24-well plates (dependent on cell size) 2 days before transfection. On the day of transfection the medium was aspirated and the cells were washed twice
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
169
Figure 7 Demonstration of luciferase expression in human artery segments. Arterial segments were transfected in vitro as described and assayed for luciferase expression 5 or 10 days later. The level of luciferase expression was indistinguishable after 5 (a) and 10 (c) days. (b) Double staining for luciferase and VSMC a-actin expression demonstrates the colocalisation of the two antigens, demonstrating that the positively staining cells are VSMCs. Luciferase expression in medial VSMCs was occasionally detected (d). Magnification: (a, c) × 130; (b) × 90; (d) × 145.
in serum-free DMEM or M199. To optimize the Tfx-50mediated transfection conditions, triplicate cultures were incubated as appropriate according to the conditions shown (Table 1). After the appropriate length of time, 1.0 to 1.5 ml of 10% DMEM or M199 was added per well and cells were cultured for 48 h before assaying for luciferase activity.
Assay of luciferase activity Luciferase activity was measured as described in the manufacturer’s instructions (Promega). Briefly, cells were washed twice in PBS (Ca2+ and Mg2+ free), and then harvested in 100–150 ml of cell lysis buffer (for 24- or 12-well plates respectively). The resulting cell extracts were then subjected to three rounds of freeze–thaw (liquid N2 for 3
min, 3 min at 37°C) and centrifuged at 10 000 g in a benchtop microfuge for 2 min. Supernatants were then stored at −20°C until the luciferase assays were performed. For enzymatic analyses, 40 ml of the supernatant was mixed in 100 ml of luciferase assay buffer and the light units produced immediately measured on a luminometer (Turner Designs luminometer model 20e). Data (in light units) are the mean of triplicates with the standard error of the mean also shown.
Transfection of rabbit carotid arteries in vivo Twelve 16- to 18-week-old male rabbits (brown half-lop or New Zealand White (NZW), mean weight 3080 g, range 2860–3499 g) were anaesthetized with halothane
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
170
(2%) (fluranisone 10 mg/ml and fentanyl citrate 0.315 mg/ml) in nitrous oxide/oxygen via nose cone or endotracheal tube after induction with 10 mg methohexitone sodium intravenously (i.v.) via the ear vein. Heparin (300 U/kg) was administered i.v. The right common carotid artery was surgically exposed and dissected free of the vagus nerve. The experimental segment (approximately 2 cm in length) was then isolated by means of soft, nontraumatic microvascular clamps and flushed with serum-free M199. The transfectant solution (2.5 mg/ml pGL3 control luciferase vector, Tfx-50 ratio 4:1) was directly instilled into the isolated vessel segment with a 30 gauge needle, and allowed to remain for 30 min (the ‘dwell’ technique). At the end of this period, the clamps were removed and free flow of blood to the vessel was restored which washed the remaining transfection complexes into the circulation. The wound was closed with 4.0 subcutaneous vicryl (J & J, Ascot, UK) sutures, dusted with 2% chlortetracycline hydrochloride antibiotic powder and the rabbits were allowed to recover from anaesthesia and returned to their cages. In addition to the above, the 5 and 10 day survival rabbits were given 0.5 ml of a solution of trimethoprim 40 mg/ml, sulphadiazine 200 mg/ml subcutaneous antibiotic. Free access to food and water was allowed and analgesia given as required. After the appropriate period of time (2, 5 or 10 days), the rabbits were premedicated with 0.25 mg intramuscular (i.m.) fluanisone 10 mg/ml and 0.3 ml i.v. heparin (5000 IU/ml) and killed by standard i.v. barbiturate overdose (8 ml pentabarbitone sodium, 60 mg/ml), and the experimental segments of vessel were removed for incubation and examination by histochemical analyses. The negative control for each rabbit was the untreated left common carotid vessel. Arteries were sectioned into three or four serial segments each approximately 3 mm in length. The segments were embedded in OCT embedding medium (Bayer, London, UK) and frozen immediately in isopentane cooled with liquid N2. For immunocytochemical staining, 5 mm cryosections were cut, airdried for 1 h at room temperature, rinsed three times with PBS for 5 min each and fixed for 30 min in methanol at −20°C.
Transfection of human arteries in vitro Human vessels (aorta, inferior mesenteric and splenic arteries) were provided by Kings College Hospital within 12 h of removal from transplant donors. Following dissection and the removal of connective tissue the vessel was divided into flat sections (approximately 9 × 3 mm) which were rinsed briefly in serum-free DMEM, placed in transfection solution (2.5 mg/ml pGL3 control luciferase vector, Tfx-50 ratio 4:1) sufficient to cover the segments for 30 min to 2 h and finally overlaid with complete DMEM. After the appropriate incubation period (2, 5 or 10 days at 37°C, 5% CO2 with medium changes every 2 to 3 days), the segments were harvested and bisected. Segments were washed three times with PBS, fixed in Zamboni’s fixative (0.1 m phosphate buffer pH 7.2, 1% paraformaldehyde (PFA), 15% picric acid) for 90 min at room temperature, rinsed twice in PBS-sucrose (0.43 m sucrose in PBS 0.01% sodium azide) for 10 min, embedded in OCT embedding medium and frozen immediately in isopentane cooled with liquid N2. Cryosections for immunocytochemical staining were cut from the blocks and dried for 1 h at room temperature.
Immunocytochemical studies For immunocytochemical studies of cell lines and primary cultures 5 × 104 cells were grown on coverslips in 12- or 24-well plates and transfected as above. Forty-eight hours after transfection, the coverslips overlaid with a cell monolayer were rinsed briefly in PBS, fixed for 10 min in methanol at −20°C, air dried at room temperature for 5 min and stored desiccated at −20°C. Immediately before staining, coverslips were rinsed briefly in PBS and cells permeabilised by immersion for 10 min in 0.1% saponin. Coverslips were then drained thoroughly and placed in a humid chamber. Sixty microlitres of the primary antibody solution (affinity-purified rabbit anti-luciferase, 1:50 in PBS, 0.5% BSA, 0.01% sodium azide) was added to each coverslip and incubated overnight at 4°C. The excess antibody solution was then removed and the coverslips dipped briefly in PBS and rinsed in three changes of PBS for 5 min with gentle agitation. Sixty microlitres of the second antibody (goat anti rabbit IgG*FITC, 1:100) was added per coverslip and then incubated for 1 h at room temperature followed by washing as above. Each coverslip was then inverted on to a small drop Vectashield antifading mountant on a clean glass slide and observed using a laser confocal microscope (BioRad MRC 600, Hemel Hempstead, UK). For histochemical studies of rabbit and human vessels, the sections were rinsed briefly three times in PBS, permeabilised for 30 min in 0.2% triton-X100, blocked for 1 h with 5% horse serum in PBS and drained thoroughly. Fifty microlitres of directly conjugated anti-luciferase* FITC antibody (1:50) was added and the slides were incubated in darkness overnight at 4°C. The slides were then washed three times in PBS and mounted in a 1:1 mixture of PBS:glycerol. Analyses were performed by confocal microscopy as described. Proliferation assays Proliferation was measured by tritiated thymidine (methyl-3 H-TdR) incorporation as assayed by liquid scintillation spectroscopy in a betaplate counter. Cells for proliferation assays were plated and transfected as described above, with the difference that the transfection assays were performed in sextuplet rather than triplicate. Each well was pulsed with 2 mCi/ml (methyl-3H-TdR) 40–48 h after treatment. Four hours after the pulse, the cells were rinsed twice with PBS, and washed three times with 5% trichloroacetic acid (TCA). Following this the wells were rinsed with 70% ethanol, air dried and stored at −20°C until analysis. For scintillation counting, the cell monolayers were suspended by incubation in 400 ml 0.25 m NaOH at 37°C for 1 h with gentle shaking. Samples were then transferred to scintillation vials, 4 ml of Optiphase Hisafe scintillation cocktail (Wallac, Milton Keynes, UK) was added and the vials were assayed. Results are expressed as mean counts per minute (c.p.m.) of the sextuplet samples with the standard error of the mean.
Acknowledgements These studies were supported by funding from the Garfield Weston Foundation. We would like to thank Mrs Sally Mill for technical assistance and the staff of the Liver Transplant Unit, Kings College Hospital, Denmark Hill for supplying us with human vessels.
Cationic liposome-mediated transfection of vascular tissues M-C Keogh et al
References 1 Kadel C et al. Long-term follow up after PTCA in patients with single-vessel disease. Am Heart J 1992; 124: 1159–1169. 2 CABRI Trial Participants. First-year results of CABRI (Coronary angioplasty versus bypass revascularisation investigation). Lancet 1995; 346: 1179–1183. 3 Haudenschild CC. Pathobiology of restenosis after angioplasty. Am J Med 1993; 94: 4A–40S. 4 Casscells W, Engler D, Willerson JT. Mechanisms of restenosis. Texas Heart Inst J 1994; 21: 68–77. 5 O’Brien ER, Schwartz SM. Update on the biology and clinical study of restenosis. Trends Cardiovasc Med 1994; 4: 169–178. 6 Sawa H et al. Increased intramural expression of plasminogen activator inhibitor type I after balloon injury: a potential progenitor of restenosis. J Am Coll Cardiol 1994; 24: 1742–1748. 7 Mattsson E, Clowes AW. Current concepts in restenosis following balloon angioplasty. Trends Cardiovasc Med 1995; 5: 200–204. 8 Mazur W et al. Coronary restenosis and gene therapy. Texas Heart Inst J 1994; 21: 104–111. 9 Gibbons GH, Dzau VJ. Molecular therapies for vascular diseases. Science 1996; 272: 689–693. 10 Dzau VJ, Pratt RE. Antisense technology to block autocrine growth factors. J Vasc Surg 1992; 15: 934–935. 11 Bennett RM et al. Inhibition of vascular smooth muscle cell proliferation in vitro and in vivo by c-myc antisense oligodeoxynucleotides. J Clin Invest 1994; 93: 820–828. 12 Wagner R. Gene inhibition using antisense oligodeoxynucleotides. Nature 1994; 372: 333–335. 13 Simons M et al. Antisense c-myb oligonucleotides inhibit intimal arterial smooth muscle cells accumulation in vivo. Nature 1992; 359: 67–70. 14 Chang MW et al. Cytostatic gene therapy for vascular proliferative disorders with a constitutively active form of the retinoblastoma gene product. Science 1995; 267: 518–522. 15 Ohno T et al. Gene therapy for vascular smooth muscle cell proliferation after arterial injury. Science 1994; 265: 781–784. 16 Felgner PL et al. Lipofection: a highly efficient lipid-mediated DNA-transfection procedure. Proc Natl Acad Sci USA 1987; 84: 7413–7417. 17 Nabel EG, Facc GP, Nabel GJ. Gene transfer into vascular cells. J Am Coll Cardiol 1991; 17: 189B–194B. 18 Stewart MJ et al. Gene transfer in vivo with DNA–liposome complexes: safety and acute toxicity in mice. Hum Gene Ther 1992; 3: 267–275. 19 Ghoumari AM et al. Gene transfer in hepatocarcinoma cell lines: in vitro optimization of a virus-free system. Gene Therapy 1996; 3: 483–490. 20 Malone RW, Felgner PL, Verma IM. Cationic-liposome mediated RNA transfection. Proc Natl Acad Sci USA 1989; 86: 6077–6081. 21 Lu D et al. Optimization of methods to achieve mRNA-mediated transfection of tumour cells in vitro and in vivo employing cationic liposome vectors. Cancer Gene Ther 1994; 1: 245–252. 22 Lewis JG et al. A serum-resistant cytofectin for cellular delivery of antisense oligodeoxynucleotides and plasmid DNA. Proc Natl Acad Sci USA 1996; 93: 3176–3181. 23 Nabel EG. Gene therapy for cardiovascular disease. Circulation 1995; 91: 541–548. 24 Caplen N et al. Liposome-mediated CFTR gene transfer to the nasal epithelium of patients with cystic fibrosis. Nature Med 1995; 1: 39–46. 25 Nabel GJ et al. Direct gene transfer with DNA–liposome complexes in melanoma: expression, biological activity, and lack of
26
27
28
29
30 31
32
33 34
35 36
37
38
39
40 41 42
43
44
45
46
toxicity in humans. Proc Natl Acad Sci USA 1993; 90: 11307– 11311. Yang Y et al. Cellular immunity to viral antigens limits E1deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 1994; 91: 4407–4411. Zabner J et al. Safety and efficacy of repetitive adenovirusmediated transfer of CFTR cDNA to airway epithelia of primates and cotton rats. Nat Genet 1994; 6: 75–83. Simon RH et al. Adenovirus-mediated transfer of the CFTR gene to lung of nonhuman primates: toxicity study. Hum Gene Ther 1993; 4: 771–780. Knowles MR et al. A controlled study of adenovirus-vectormediated gene transfer in the nasal epithelium of patients with cystic fibrosis. New Engl J Med 1995; 333: 823–831. Hug P, Sleight RS. Liposomes for the transfection of eukaryotic cells. Biochim Biophys Acta 1991; 1097: 1–17. Schulick AH et al. In vivo gene transfer into injured carotid arteries: optimization and evaluation of acute toxicity. Circulation 1995; 91: 2407–2414. Morishita R et al. Novel and effective gene transfer technique for study of vascular renin angiotensin system. J Clin Invest 1993; 91: 2580–2585. Felgner PL, Holm M, Chan H. Cationic liposome mediated transfection. Proc West Pharmacol Soc 1989; 32: 115–121. Felgner JH et al. Enhanced gene delivery and mechanism studies with a novel series of cationic lipid formulations. J Biol Chem 1994; 269: 2550–2561. Promega. Tfx-50 reagent for the transfection of eucaryotic cells Technical Bulletin 216, 1995. Duzgunes N et al. Fusion of liposomes containing a novel cationic lipid, N-(2,3(dioleoyloxy)propyl)-N,N,N-trimethylammonium: induction by multivalent anions and asymetric fusion with acidic phospholipid vesicles. Biochemistry 1989; 28: 9179–9184. Labroille G, Bonnefille S, Belloc F. Tfx-50 reagent increases the uptake of oligonucleotides by leukaemic cells. Promega Notes 1996; 56: 8–13. Losordo DW et al. Use of the rabbit ear artery to serially assess foreign protein secretion after site-specific arterial gene transfer in vivo. Circulation 1994; 89: 785–792. Takeshita S et al. Increased gene expression after liposomemediated arterial gene transfer associated with intimal smooth muscle cell proliferation. In vitro and in vivo findings in a rabbit model of vascular injury. J Clin Invest 1994; 93: 652–661. Pickering JG et al. Liposome-mediated gene transfer into human vascular smooth muscle cells. Circulation 1994; 89: 13–21. Liu Y et al. Cationic liposome-mediated intravenous gene delivery. J Biol Chem 1995; 270: 24864–24870. Chapman GD et al. Gene transfer into coronary arteries of intact animals with a percutaneous balloon catheter. Circul Res 1992; 71: 27–33. Dalesandro J et al. Gene therapy for donor hearts: ex vivo liposome-mediated transfection. J Thorac Cardiovasc Surg 1996; 111: 416–421. Sawa Y et al. Efficiency of in vivo gene transfection into transplanted rat heart by coronary infusion of HJV liposome. Circulation 1995; 92: 11479–11482. Vieweg J et al. Efficient gene transfer with adeno-associated virus-based plasmids complexed to cationic liposomes for gene therapy of human prostate cancer. Cancer Res 1995; 55: 2366– 2372. Bennett MJ et al. Cholesterol enhances cationic liposomemediated DNA transfection of human respiratory epithelial cells. Bioscience Rep 1995; 15: 47–53.
171
