Cell delivery, intracellular trafficking and expression of an integrin-mediated gene transfer vector in tracheal epithelial cells

Abstract

The mechanism of cell entry and intracellular fate of a gene transfer vector composed of a receptor-targeting, DNA-condensing peptide, RGD-oligolysine, a luciferase encoding plasmid DNA (pDNA) and a cationic liposome was examined. We demonstrate by confocal microscopy, electron microscopy and subcellular fractionation that the major mechanism of entry of the vector is endocytic. The vector complex rapidly (5 min) internalizes into early endosomes, then late endosomes and lysosomes. Entry involves, at least in part, clathrin-coated pit-mediated endocytosis since different conditions or drugs known to influence this pathway modify both uptake of pDNA and its expression. The observed increase in expression with addition of a lip some correlated with an increase in the rate of transfer of the pDNA to lysosomes, a decrease in intracellular recycling and exocytosis of the pDNA and an increase in the amount of pDNA in the nuclear fraction. Trafficking within the cell involved endosome fusion and the acid environment of the endosomes–lysosomes was beneficial for expression. After 30 min both the peptide and pDNA localized to the nucleus and the amount of intact pDNA in the nuclear fraction was highest with liposome and peptide. A better understanding of the cellular mechanisms by which vectors transfer to and traffic in cells should help design improved vectors.

Introduction

For gene therapy to be successful the vector system must be efficient in delivering the transgene to the cell and in expressing the gene product. In addition, expression should be prolonged and the vector non-toxic (for review see Ref. 1).

Viral vectors are still considered to be the most efficient gene transfer systems but the efficiency of liposome-based nonviral vectors is continuously improving.2 In addition, such nonviral systems provoke less host inflammatory and immune responses and are relatively easy to prepare in large quantities. Receptor-mediated molecular conjugates are also used to deliver plasmid DNA (pDNA) and consist mainly of a receptor-binding ligand conjugated to a DNA-binding moiety, usually high molecular weight oligo-L-lysine (for review see Ref. 3).

Knowledge of the mechanisms of entry of the adenovirus into cells by interaction with integrins and the coxsackievirus and adenovirus receptor (CAR) inspired us4 and more recently others5 to develop peptide-based gene transfer vectors which mimic the adenovirus. We have demonstrated the principle of integrin-mediated gene delivery using a bifunctional peptide comprised of an integrin receptor-targeting domain containing an arginine–glycine–aspartic acid tri-peptide motif and a DNA-binding moiety consisting of a stretch of 16 lysine residues, Arg–Gly–Asp/oligolysine ([K]16RGD).467 We showed that this peptide condensed pDNA (approximately 10 nm) rendering it resistant to nuclease degradation and delivers pDNA in a specific manner to integrin-expressing airway and intestinal cells with an efficiency 10- to 100-fold that of naked pDNA and approaching that of the commercial liposome.67 The efficiency could be further improved 30-fold by the addition of a liposome, lipofectamine to tracheal cells6 and a 10- to 100-fold increase was also observed with lipofectin together with similar RGD-containing peptides in ECV304 endothelial cells.8

Improvements in the efficiency of expression of nonviral vectors have essentially been made by testing different vector compositions. The development of future gene transfer vectors will probably benefit from an improved understanding of the interaction of such vectors with the extracellular and intracellular environment. However, only a few studies, which have led to contradictory conclusions, have addressed the mode of entry and intracellular fate of mainly cationic liposome constructs.9101112 We have previously shown that the pDNA:[K]16RGD vector complex in the absence or presence of liposomes enters cells and colocalizes with horseradish peroxidase, a marker of fluid-phase endocytosis, suggesting that entry occurs by an endocytic pathway.6

Several routes of endocytic entry of extracellular material into cells exist: the clathrin-dependent pathway, macropinocytosis, the caveolar pathway, a clathrin- and caveolin-independent pathway and phagocytosis. The former four pathways concern small (<0.2 μm diameter) vesicles transporting fluid and macromolecules while the latter involve uptake of larger particles (>0.5 μm diameter) (for review see Refs 13 and 14). The initial step of the endocytic pathway involves internalization of membrane proteins including receptors and ligands into early endosomes. These proteins are either rapidly recycled back to the plasma membrane via recycling vesicles or are transported to late endosomes and then lysosomes where they are degraded.

With the aim of understanding better the mechanism by which the [K]16RGD vector system, and possibly other systems, lead to efficient expression we investigated further the cell delivery and intracellular trafficking of the pDNA:[K]16RGD:lipofectamine vector system.

Results

Trafficking of the RGD-oligolysine and plasmid DNA components of the lipofectamine containing complexes from the plasma membrane to the nucleus

To examine the intracellular trafficking of the pDNA:[K]16RGD:lipofectamine complexes we labeled the [K]16RGD peptide with FITC (green) and the pGL3 luciferase encoding plasmid DNA (pDNA) with the carbocyanine dye Cy5 (pseudo blue). We had previously verified that fluorescent labeling did not modify the level of luciferase expression (data not shown) and that it remained bound.6

At 1 min the complex is seen to interact initially with the cell surface and the label forms a halo around the cell with predominant colocalization of peptide and pDNA which gives turquoise on merging of the green and blue (Figure 1a). At 5 min both the peptide and pDNA start to appear in the cytoplasm just below the plasma membrane with some dissociation of the two components indicated by a few separate green and blue spots. This increases at 15 min where both colors are visible as punctate spots throughout the cell cytoplasm characteristic of location in vesicles. After 30 min the majority of the peptide was associated with the nucleus giving a yellow composite color with the red nuclear DNA staining. At this time-point a substantial amount of the pDNA was still seen in the cytoplasm. However, at 2 h both the peptide and the pDNA were predominantly associated with the nucleus. The peptide appeared to show a partial preference for the nucleoli and the nuclear membrane while the pDNA remained predominantly in concentrated regions around nucleoli and sometimes associated with the periphery of the nucleus giving a white merging color of blue, green and red. For the three later times the nucleus was stained with propidium iodide (red). At 4 h the pDNA label was slightly more diffuse. The aggregated aspect of the label may suggest that the pDNA localizes to distinct nuclear territories as shown for certain chromosomes.15 Separation of the three signals, green (FITC-[K]16RGD), blue (Cy5-pDNA) and red (propidium iodide-labeled cellular DNA and RNA) of the 2-h image (Figure 1a) shows colocalization of the peptide with nucleic acid material inside and outside the nucleus (Figure 1b). Since we did not treat the cells with RNase the red staining material around the nucleus is probably RNA. The transgene DNA appears to aggregate in either the nucleus or cytoplasm in partial association with the peptide. In the cytoplasm it can be seen to also stain red with propidium iodide.

Figure 1
figure1

Intracellular trafficking of fluorescent peptide and pDNA in the presence of lipofectamine. Tracheal cells were incubated with Cy5-pDNA:FITC-peptide:lipofectamine at 37°C for different times and examined by confocal microscopy. (a) 1, 5, 15 and 30 min and 2 and 4 h. The FITC-labeled [K]16RGD peptide is shown in green and the Cy5-labeled pDNA is shown in pseudo blue. For the short times (1, 5 and 15 min) the cells were viable while for the longer times (30 min, 2 and 4 h) the cells were fixed and the nucleus stained with propidium iodide (red). Merged images of the FITC and Cy5 fluorophores are shown for the early times while merged images with also propidium iodide are shown for the longer times. (b) Shows the individual fluorescent labels at the 2-h time-period. The bar represents 10 μm.

When stained with propidium iodide these cells showed visible nucleoli which suggested that uptake did not seem to be related to cell cycle-dependent formation of the nucleolus.

These observations suggest vesicular trafficking of the complex within the cell and the uptake into the nucleus of both the pDNA and [K]16RGD components.

Multilamellar liposomes are visible in clathrin-coated pits and endosome- and lysosome-like vesicles

Tracheal epithelial cells were incubated in the presence of vector complex (pDNA:[K]16RGD:lipofectamine) for different times and then examined by transmission electron microscopy. Single or several piled together multilamellar lipid particles each of about 100 nm in diameter could be seen to attach to the cell surface (Figure 2a and f). No similar structures were seen in control cells not incubated with complex. Several coated-pits containing electron dense material with a liposome-like multilamellar structure were visible at 5 min (Figure 2b and c). The liposomes appeared collapsed and sometimes spread over the edge of the pits on the cell membrane (Figure 2d and e). Since the enclosed volume of clathrin-coated pits does not normally contain electron dense material and the material retained some multilamellar appearance we conclude that it is liposome lipids which condensed or dissociated as a result of the proximity to the hydrophobic environment of the cell membrane. At 1 h multilamellar structures could be seen to occupy partially the internal space of large (200–300 nm diameter) vesicles (Figure 2f, arrow) a size equivalent to that of multivesicular bodies. This result thereby confirms entry of the liposomes by endocytosis. At 4 h an accumulation and compaction of multilamellar structures which took on either a fingerprint- or stack-like appearance were visible (Figure 2g and h). No similar structures were observed in cells not incubated with complex. The size of these multilamellar-containing vesicles and the density of the multilamellar material at 4 h suggests that fusion of endosomes had occurred. No equivalent concentrations of multilamellar lipid could be observed in the nucleus, which suggests either that the lipid component did not passage into the nucleus or that it passages in a degraded or disaggregated form. These results demonstrate that the lipid component of the vector also enters cells by an endocytic mechanism which, at least in part, involves clathrin-coated pits.

Figure 2
figure2

Vesicle location of lipofectamine. Transmission electron micrographs of tracheal epithelial cells incubated for different times with vector complex. (a) A single multilamellar liposome attached to the cell surface at 1 min; (b–e) condensed liposome material in clathrin-coated pits (arrow) and a liposome plastered on the edge of the pit on the cell membrane (arrow head) after 5 min; (f) a stack of multilamellar liposomes attached to the cell surface and multilamellar lipid material partially filling the internal space of vesicles (arrow) after incubation for 1 h; (g) finger print- and stack-like multilamellar lipid in vesicles after 4 h; (h) highly concentrated aggregates of multilamellar lipid in vesicles after 4 h. The bars represents 200 nm.

Cell entry of the pDNA containing RGD-oligolysine liposomes is at least partially receptor mediated

In a previous study we showed by confocal microscopy that in the presence or absence of lipofectamine the pDNA:[K]16RGD complex colocalized in cells with the fluid-phase marker horseradish peroxidase.6 This suggested that the complex enters cells by an endocytic pathway. To investigate further the precise mechanism of cell entry of the pDNA:[K]16RGD:lipofectamine complex we examined the effect of depletion of K+, exposure to hypertonic medium and different drugs, all of which are known to modify endocytosis, on internalization and expression of the complex. The uptake of 35S-labeled pDNA and subsequent luciferase expression of the transgene were investigated (Figure 3a and b, respectively). Uptake was determined by subtracting the binding at 4°C from the cell association (binding and internalization) at 37°C. At 4°C classical endocytosis does not occur but ligand can bind to its receptor. An example of this is the adenovirus which associates with RGD binding integrins at 4°C but is not internalized.16 We showed previously that internalization was close to maximal after incubation for 2 h with complex and that uptake (37–4°C) is considerably higher in the presence of lipofectamine.6

Figure 3
figure3

Implication of endocytosis in internalization and expression of the luciferase reporter gene. Effect on cell uptake of 35S-labeled pDNA (a) and luciferase expression (b) of conditions which modify clathrin-coated pits and intracellular trafficking. (a) Control cells, K+-depleted cells, cells exposed to hypertonic medium (0.33 M sucrose) or cells preincubated for 30 min at 37°C with wortmannin (100 nM) or LY294002 (30 μM) or with cytochalasin B (5 μg/ml, 45 min) were examined. 35S-labeled pDNA complexed to the [K]16RGD peptide together with lipofectamine was added to cells maintained either at 4°C (cell binding) (hatched bars) or 37°C (binding and internalization) (open bars) for 2 h. Uptake is the 37°C–4°C value. (b) Luciferase activity determinations of cells incubated with the complex containing lipofectamine; assays were performed 48 h after the 4-h transfection period. The results are the mean ± s.e.m. of three experiments, normalized to the control, each performed in triplicate.

Cell depletion of K+ has been shown to inhibit receptor-mediated endocytosis by removing clathrin from the plasma membrane.17 Uptake in the presence (Figure 3a) and absence of lipofectamine (data not shown) was considerably decreased as was expression (Figure 3b) in K+-depleted cells. To verify that K+ depletion of the tracheal cells used in this study does indeed modify receptor-mediated (clathrin-dependent) endocytosis we incubated depleted and non-depleted cells in the presence of FITC-labeled transferrrin, a marker of receptor-mediated endocytosis, and examined the cell fluorescence on a fluorescence flow cytometer. K+ depletion resulted in an approximately 50% decrease in the mean fluorescence intensity after subtraction of the cellular autofluorescence (data not shown). This result indicates that K+ depletion does indeed bring about a decrease in receptor-mediated endocytosis in the cells used in this study.

Treatment of cells with hypertonic media (45 min preincubation and present during transfection) also inhibits receptor-mediated endocytosis by rendering clathrin unavailable for assembly into normal coated pits.18 Indeed, incubation of our cells in hypertonic medium composed of either 0.33 or 0.45 M sucrose resulted in a 90% decrease in the endocytosis of FITC-labeled transferrin which enters cells by a clathrin-coated pit mechanism. In addition, hypertonic medium decreased significantly both uptake (Figure 3a) and expression (Figure 3b) of the vector complex; 0.33 M and 0.45 M sucrose gave similar results. Inhibition was more complete with hypertonic medium than with K+ depletion which probably reflects the 90% and 50% inhibition, respectively, of the endocytosis of transferrin. In addition, flow cytometric analysis of these treated cells did not show cell damage. Controls for expression performed either in K+-free buffer supplemented with 4 mM KCl or in isotonic sucrose gave the same values as in OptiMEM.

Since hypertonic medium could be envisaged to disrupt the vector complex and thereby explain reduced internalization and expression we also tested preincubation of cells in hypertonic medium and subsequent exposure of rinsed cells to the vector complex in OptiMEM for a transfection of 1 h. A significant reduction in both internalization and expression was still observed compared with the control (data not shown). Thus, the inhibition could not be explained solely by a disruption of the vector complex.

Drugs which affect vesicle trafficking also influence expression

As phosphatidyl inositol-3-OH kinase (PI 3-Kinase) has been shown to play a role in endosome–endosome and endosome–lysosome fusion19 we investigated the effect of two inhibitors of PI 3-kinase, wortmannin and LY294002 ([2-(4-morpholinyl)-8-phenyl-4H-1-benzo- pyran-4-one]) on internalization and expression of the vector. These two chemically unrelated inhibitors rapidly and irreversibly impair the enzyme by different mechanisms.20 To verify that these two agents inhibit PI 3-K activity in our cell system we assayed for this enzyme and found a 50% decrease in activity with both inhibitors (data not shown). Wortmannin and LY294002 do not modify significantly uptake of vector (Figure 3a), however, they result in a decrease in expression. This result suggests that partial inhibition of the fusion of endosomes was responsible for the decrease in expression and thereby again points to an endocytic mechanism for entry and trafficking of the vector complex.

The cytoskeleton of the cell is also known to be involved in controlling vesicle movement in cells, therefore we examined the drug cytochalasin B which disrupts the assembly of the F-actin microfilament network and blocks non-coated pit-mediated endocytosis and phagocytosis but does not affect receptor-mediated endocytosis.2122 We observed that cytochalasin B increased slightly but consistently the uptake (Figure 3a). This result suggests that assembly of microfilaments may not be essential for internalization of complex and that its entry does not involve non-coated pit-mediated endocytosis or phagocytosis. Expression of the transgene in the presence of cytochalasin B is also increased (Figure 3b) which possibly reflects the slight increase in internalization.

The vehicle dimethyl sulphoxide used to solubilize the different drugs did not modify significantly the internalization or expression levels compared with the control.

Absence of phagocytosis of RGD-oligolysine:pDNA:lipofectamine complexes

To examine further the possibility that the complex enters cells by phagocytosis we incubated cells with fluorescent complex in the presence of a marker of phagocytosis. Merged images of the three signals in the presence of the phagocytic marker did not show colocalization (Figure 4), which may suggest that phagocytosis is not the major entry mechanism at least for these epithelial cells which are probably poorly phagocytic. This result together with the electron microscopy data and results obtained with drugs that modify different endocytic pathways suggest that the major entry mechanism is receptor-mediated endocytosis.

Figure 4
figure4

Absence of colocalization of a marker of phagocytosis with fluorescently labeled complex. Tracheal cells were incubated with Cy5-pDNA:FITC-[K]16RGD:lipofectamine at 37°C for 30 min and then for a further 15 min in the presence of Nile red-labeled fluorospheres (1 μm), a marker of phagocytosis. The Cy5-pDNA is shown in pseudo blue, FITC-[K]16RGD in green and the Nile red fluorospheres in red. The bar represents 5 μm.

Release of lipo–oligoplexes (pDNA:[K]16RGD:lipofectamine) from vesicles may require an acid pH

We reported previously that chloroquine, a weak base which neutralizes the acid pH of endocytic vesicles, brings about a significant increase in expression of the pDNA:[K]16RGD complex, in the absence of lipofectamine, in several cell lines.7 However, in contrast, in the presence of lipofectamine, chloroquine at concentrations used to modify the pH brought about a substantial decrease in the level of expression (Figure 5). This result may suggest that a possible beneficial inhibition of lysosomal enzymes and thus diminished degradation of the vector complex by raising the lysosomal pH is outweighed by the requirement of an acid pH for release of the liposome vector from endocytic vesicles. Certain viruses are also known to be dependent on an acid environment for release from endosomes23 and this may also be the case for the complex containing lipofectamine. This result also further confirms the presence of the vector complex in intracellular vesicles.

Figure 5
figure5

Beneficial effect of an acid endosomal–lysosomal pH on expression of the luciferase reporter gene. Chloroquine was added at the indicated concentrations to cells together with the vector complex (pDNA:[K]16RGD:lipofectamine) in OptiMEM and incubated for 4 h. The cells were washed and the luciferase activity was determined 48 h later.

Endosomal location of vector complexes and traffic to lysosomes in the presence of lipofectamine

To determine further the subcellular localization of vector we performed subcellular fractionation on Percoll gradients of cells incubated with 35S-labeled pDNA with or without the [K]16RGD peptide and/or lipofectamine. However, we first characterized the gradient of homogenized cells and observed that the upper gradient fractions contained early and late endosomes at a density of about 1.03 and 1.04, respectively, while the lower fractions contained lysosomes at a density of about 1.10. These conclusions were based on: (1) the detection of the activity of horseradish peroxidase internalized by cells for either 5 (early endosomes, EE) or 30 min (late endosomes, LE) in the upper fractions of the Percoll gradient; (2) β-hexosaminidase activity, a marker of lysosomes (Lys), predominantly in the bottom fractions; (3) detection of the lysosomal-associated membrane protein-1, a marker of lysosomes in the bottom fractions; and (4) electron microscopy of the upper fractions showing the presence of endosomes and absence of lysosomes. In addition, plasma membranes and cytosolic proteins floated to the upper most fractions as demonstrated by the activity of: (1) the plasma membrane marker protein alkaline phosphatase which peaked in the top 2.5 ml of the gradient and (2) the cytosolic marker enzyme lactate dehydrogenase which was found to float to the top 2 ml of the gradient.

Cells incubated for 4 h with either pDNA alone, pDNA:lipofectamine, pDNA:[K]16RGD or pDNA:[K]16RGD: lipofectamine were rinsed and then incubated in medium for a further chase period of either 0, 4 or 12 h. Subcellular fractionation of the post-nuclear supernatant (PNS) of the cells in a Percoll gradient was then carried out as described above. At the zero chase time the radioactivity of pDNA alone floated predominantly to upper fractions and the amount of radioactivity in the overall fractions declined progressively (Figure 6a upper left). When [K]16RGD was present the radioactivity at the zero chase time was slightly displaced to denser fractions corresponding to that of endosomes and also declined progressively with time (Figure 6a upper right).

Figure 6
figure6

Early endosomal (EE), late endosomal (LE) and lysosomal (Lys) location of 35S-labeled pDNA. (a) Cells were incubated with the different complexes as indicated for 4 h and then rinsed and incubated for a further 0 (□), 4 () or 12 () h chase period before homogenization and subcellular fractionation on a 25% Percoll gradient. The same amount of post-nuclear supernatant was loaded on each gradient. The results are representative of two separate experiments. (b) The different vector complexes were loaded on a Percoll gradient and centrifuged either in homogenization buffer (left) or in an homogenate of cells (right): pDNA alone, ▪; pDNA: lipofectamine (L), •; pDNA:[K]16RGD, ; pDNA:[K]16RGD:lipofectamine (L), ×. The results are representative of two separate experiments.

In the presence of lipofectamine the pDNA first appears in early endosomal fractions and then progressively shifts to late endosomal fractions and lysosomal fractions (Figure 6a lower left). The amount of pDNA in the late endosomal fractions is similar for the 4- and 12-h chase periods while the amount in the lysosomal fractions increases with time. This is possibly the result of progressive internalization of cell surface-bound pDNA which was not removed by rinsing. In the presence of lipofectamine and peptide the pDNA first appears in endosomal fractions and then shifts to the lysosomal fractions (Figure 6a lower right). The amount of pDNA in the endosomal fractions increases slightly at the 4-h period and then declines slightly at 12 h. The amount in the lysosomes is also higher at 4 h than at 12 h which is in contrast to that for the pDNA:lipofectamine complex. This finding suggests that either the pDNA in the lysosomes in the presence of peptide transfers more rapidly or is more rapidly degraded or that the complex exits the lysosomes.

To check that the labeled pDNA was actually in vesicles separated by the Percoll gradient and not free pDNA we examined the behavior of the different vector complexes on their own on a Percoll gradient when centrifuged in either just homogenization buffer (Figure 6b left) or in an homogenate of cells (Figure 6b right). About the same amount of labeled pDNA as in the PNS was loaded on to the gradient. The pDNA floated to the top 2 ml of the gradient as observed previously when plasmid was incubated with cells. A slight shift to denser fractions was seen when the pDNA was added to a cell homogenate. The pDNA:lipofectamine complex was present predominantly in the upper 2 ml fractions but also throughout the gradient particularly when centrifuged in buffer. The pDNA:[K]16RGD:lipofectamine complex floated in the upper 2.5 ml of the gradient as did the pDNA:[K]16RGD complex. These results suggest that some of the radioactivity in the endosomal fractions in Figure 6 may be free complex but does not exclude the possibility that a proportion of the radioactivity is in endosomes. The results demonstrate that in the presence of lipofectamine the complex is present in lysosomes which confirms that some of the radioactivity in the upper fractions is in endosomes since endosomes are the vesicles that either mature into or pass on their contents to lysosomes.13 In the absence of lipofectamine little radioactivity was found in the lysosomal fractions. In addition, the displacement of the radioactivity to more dense fractions was not the result of partial degradation of the pDNA. Lysosomal fractions showed the same proportion of degraded and intact pDNA after electrophoresis on an agarose gel as did the endosomal fractions (data not shown).

These results confirm that the mechanism of entry of lipoplexes (pDNA:lipofectamine) and lipo–oligoplexes (pDNA:[K]16RGD:lipofectamine) into cells involves endocytosis.

An increased amount of pDNA is cell-associated and nuclear located in the presence of [K]16RGD and lipofectamine

Cells were incubated for a 4-h transfection period with complex and then rinsed and chased for 0, 4 or 12 h in fresh medium. The radioactivity, thus ng of pDNA, recovered in this medium increased with the chase-period for the different complexes (Table 1). This pDNA may represent either released membrane-bound pDNA or exocytosed previously internalized pDNA. In the presence of lipofectamine less pDNA was observed in the medium in accordance with increased cell association. This result may suggest that lipofectamine not only improves the vector-to-cell contact but also reduced egress of the vector by the cell. The percentage of cell-associated radioactivity increased during the different chase periods in the absence of liposome but remained relatively stable in its presence. This result suggests that the cell association was complete during the transfection period when lipofectamine was present but not in its absence. The overall amount of pDNA in the PNS is lower without than with lipofectamine. Indeed, when the pDNA is complexed with both [K]16RGD and lipofectamine the overall amount of pDNA in the PNS is slightly higher at the 4- and 12-h periods compared with lipofectamine:pDNA. Since the amount of pDNA in the PNS was also higher in the presence of liposome these results suggest that the pDNA enters to a greater extent in the presence of liposome than in the absence and at a faster rate. This confirms our previous findings, which suggested that the same number of cells were transfected with or without lipofectamine but that more complex entered per cell when lipofectamine was present.6 In addition, the amount of radioactivity in the nuclear fractions was considerably higher in the presence of lipofectamine and in particular with [K]16RGD (Table 1). While a slight decrease with time was noted for pDNA:[K]16RGD a steady increase with time was seen for the pDNA:[K]16RGD:lipofectamine complex; values remained stable for the pDNA:lipofectamine complex. These results suggest that [K]16RGD not only facilitates cell entry but also nuclear transport and this is reflected in the ratio of the radioactivity in the nucleus over that in the PNS, which indicates partitioning between the cytoplasm and the nucleus (Table 1). This ratio is higher for pDNA:[K]16RGD:lipofectamine compared with pDNA:lipofectamine and higher for pDNA:[K]16RGD compared with pDNA alone. This suggests preferential partitioning of the pDNA into the nucleus in the presence of [K]16RGD and lipofectamine. The ratio also remains relatively stable for pDNA:[K]16RGD:lipofectamine for the different chase periods compared with pDNA:lipofectamine, which suggests that nuclear passage had already occurred during the transfection period and thus occurs more rapidly in the presence of [K]16RGD. This result confirms that of confocal microscopy which shows nuclear localization after 2–4 h for both the pDNA and peptide.

Table 1 Extracellular and intracellular distribution of pDNA of different complex combinations after different chase periods

The quality of the nuclear preparations was verified by: (1) examining the morphology of the nuclei by phase contrast microscopy and (2) determining the activity of the cytoplasmic and plasma membrane enzymes, respectively, lactate dehydrogenase and alkaline phosphatase. Intact nuclei were visible and levels of enzyme activity, below 3 and 10% respectively for lactate dehydrogenase and alkaline phosphatase, of whole cells were detected indicating only very low levels of contamination.

In conclusion, the results shown in Table 1 indicate that there exists a correlation between the amount of pDNA in the PNS and associated with the nucleus and the expression level. The pDNA:[K]16RGD:lipofectamine complex shows: (1) the highest amount of pDNA in both the PNS and nuclear fraction; (2) the highest ratio of the amount of pDNA in the nuclear fraction compared to the PNS; and (3) the highest expression level (see below).

Early expression of the internalized plasmid DNA in the presence of [K]16RGD and lipofectamine

We examined the expression of the pGL3 luciferase reporter gene at the different chase periods and found that expression was detectable at the 4- and 12 h-chase periods only in the presence of lipofectamine (Figure 7). Expression was detectable for pDNA and [K]16RGD:pDNA after 24 and 48 h and was increased for the latter complex, as reported previously.67 These results show that lipofectamine promotes early expression probably by protecting the pDNA from degradation and preventing recycling to the plasma membrane and the peptide enhances the efficacy possibly by promoting internalization and transfer to the nucleus.

Figure 7
figure7

Early expression of the luciferase reporter gene in the presence of lipofectamine. Expression of the different vector complexes at the different chase periods after transfection. The luciferase activity of the pDNA was examined at 4 and 12 h after the 4-h transfection period. Transfection was performed with either pDNA alone or pDNA complexed to the peptide [K]16RGD in the absence or presence of lipofectamine.

Intact pDNA complexed with lipofectamine is present in nuclear fractions after the different chase periods

To investigate further the possibility that the peptide and lipofectamine protect the pDNA we extracted the pDNA from the nuclear fractions at the different chase periods and examined it by autoradiography. In the absence of lipofectamine only a slight signal was visible while pDNA complexed in the presence of lipofectamine with or without peptide was detectable (Figure 8). Although the pDNA was substantially degraded, bands corresponding to the supercoiled and coiled plasmid were detectable. This result further confirms the increase in the amount of DNA in the nuclear fraction and explains the efficacy of the lipofectamine-containing vector.

Figure 8
figure8

Intactness of 35S-labeled pDNA in the nuclear fraction at different chase periods. DNA isolated from the nuclear fractions of cells incubated for 4 h in the presence of vector containing lipofectamine with or without peptide was electrophoresed on an agarose gel, blotted on to a nylon membrane and exposed to film: pGL3, plasmid DNA; [K]16RGD, integin-binding peptide; L, lipofectamine.

Discussion

Several authors have demonstrated that lipoplexes, pDNA–liposome complexes enter cells by endocytosis by both the coated pit (receptor-mediated) and non-coated pit pathways10122425 or by phagocytosis involving clathrin-coated pits9 but little information is available concerning the mechanism of entry and intracellular trafficking of ‘lipo–oligoplexes’; cationic oligomer–DNA–liposome composite gene delivery vectors.

We recently demonstrated the utility of an RGD–oligolysine peptide ([K]16RGD) in gene transfer.67 Such oligoplexes, cationic oligopeptide–pDNA complexes, composed of a receptor targeting moiety would be expected to be internalized by a receptor-mediated mechanism. In addition, since the ligands of integrins have been shown to be internalized by coated pit endocytosis26 and since certain RGD-recognizing integrins and other integrins have been shown to be present in endosomes of a variety of cells2728 we postulated that the pDNA:[K]16RGD vector system enters cells by endocytosis involving a coated pit mechanism. However, since we found that the efficacy of this system could be enhanced significantly by the addition of a liposome,6 where the peptide and liposome led to a significant synergistic effect, we were interested in understanding better the function of these two components in DNA transfer.

By definition receptor-mediated vectors (oligoplexes) would be expected to involve a clathrin-dependent pathway unless the presence of the pDNA modified the charge interactions or the size of the complex was prohibitive. Clathrin-coated vesicles range in size from 80 to 160 nm while [K]16RGD-condensed plasmid ranges in size from 1 to 10 nm.7 In addition, the association of the pDNA:[K]16RGD complex with a cationic liposome (20–150 nm) might be expected to modify both the mechanism of entry and intracellular trafficking.

To investigate the mechanism of cell entry of this lipo–oligoplex we performed confocal microscopy and electron microscopy both of which indicated that entry occurred predominantly by endocytosis. The labeled pDNA and peptide showed within cells: (1) a punctate pattern which is characteristic of presence in an endosomal compartment; (2) label which partially colocalized with labeled horseradish peroxidase a marker of fluid-phase endocytosis; and (3) label which did not colocalize with labeled beads which enter by phagocytosis. In addition, multilamellar lipid structures were observed within multivesicular bodies as previously observed for liposomes composed of either lipopolyamine11 or DOTAP/DOPE.29 Lipofectamine a mixture of DOSPA/DOPE was used in the present study.

To investigate further the endocytic mechanism of entry we examined cell depletion of K+ which inhibits receptor-mediated endocytosis by removing clathrin from the membrane.17 We found that K+ depletion had a considerable inhibitory effect on both entry and expression of RGD–oligolysine-associated pDNA in the presence and absence of a cationic liposome. A similar inhibition of entry due to K+ depletion of a doubly labeled liposome–pDNA complex has been reported.9 Exposure to hypertonic medium, an additional condition which inhibits clathrin-coated pit endocytosis,18 also resulted in inhibition of entry and thus expression of the vector complex. These results point to a similar clathrin-dependent entry mechanism for lipoplexes and lipo– oligoplexes. However, since the clathrin coat may be involved in phagocytosis30 and since phagocytosis has been suggested to be the mechanism for cell entry of lipoplexes9 we examined the effect of the drug cytochalasin B which inhibits phagocytosis by disrupting the assembly of the F-actin microfilament network beneath the plasma membrane; cytochalasin also blocks non-coated pit-mediated endocytosis.21 We observed that cytochalasin B increased slightly but consistently the uptake and expression of pDNA associated with [K]16RGD in both the presence and absence of liposome. Zhou and Huang12 also showed an increase in expression of lipoplexes and suggested that since cytochalasin also increases the diffusion coefficient in the cytoplasm of the cell this facilitated translocation from the cytoplasm to the nucleus leading to increased transfection. However, Watanabe et al,25 showed a dose-dependent decrease in β-galactosidase activity and Nicolau and Sene24 and Matsui et al,9 showed a decrease in uptake of labeled lipoplexes with cytochalasin B. Such differences may result from the different cell systems and liposomes studied.

The inhibition of the activity of PI 3-K, an enzyme thought to be involved in vesicle fusion and delivery of recycling endosomes to the plasma membrane,20 resulted in a partial inhibition of expression of the vector. This result further confirms trafficking of the vector complex in intracellular vesicles and the requirement for fusion of endosomes for optimal expression.

The low pH of late endosomes and lysosomes appears to facilitate endosome fusion and/or release of the vector into the cytoplasm since chloroquine, a weak base which increases the pH of intracellular vesicles,31 reduced the efficacy of the vector complex. A number of studies have shown increased expression, in the presence of chloroquine, of oligoplexes732333435 and lipoplexes.1236 Several suggestions have been proposed to explain the influence that chloroquine has on expression of vectors, these include: (1) decreasing the activity of lysosomal enzymes by raising the pH of the lysosomes and thus protecting the pDNA from degradation; (2) inhibiting the fusion of endosomes with lysosomes; (3) binding of chloroquine to the pDNA; and (4) facilitating escape from endosomes resulting from osmotic swelling. Nonetheless, several studies have also shown decreased activity in the presence of chloroquine but all in the presence of liposomes including lipofectamine3738 or adenovirus.39 Certain virus are known to be dependent on an acid environment for release from endosomes23 and this may also be the case for [K]16RGD:lipofectamine where the oligoplex itself may facilitate release from endosomes. In addition, if the pDNA in the complex within lysosomes is protected from enzymatic degradation by a liposome the presence of this complex in lysosomes may not be deleterious for subsequent expression and on the contrary may help release into the cytoplasm due to the acid environment. In addition, our subcellular fractionation results suggest that: (1) [K]16RGD increases uptake of the pDNA and that this increase is further enhanced with a liposome; (2) liposome prevents recycling of the complex to the plasma membrane and subsequent exocytosis; and (3) the presence of [K]16RGD together with a liposome increases the rate of transfer to lysosomes as well as its release into the cytoplasm. The latter result is in contrast to a study which indicated that a cationic liposome delays transfer of plasmid DNA into lysosomes in the liver;40 however, the authors used a lipoplex and not a lipo–oligoplex. Our findings may suggest that if there is no transfer to lysosomes there is a greater risk that the vector be recycled back to the plasma membrane. Despite exposure of the complex to lysosomal enzymes and thus degradation in the lysosomes escape may be facilitated by the lower pH of lysosomes. The neutral helper lipid DOPE, which is one of the components of lipofectamine, has been suggested to destabilize lipid bilayers12 and thereby escape from vesicles. However, the accumulation of lipid in vesicles seen by electron microscopy may suggest that the lipid has difficulty getting out and thus the presence of an endosomal/lysosomal disrupting agent may help release.

Translocation of proteins into the nucleus occurs through the nuclear pore complex which functions as a molecular sieve whereby molecules smaller than 40–45 kDa diffuse in and out while larger proteins require a nuclear localization signal (NLS). Thus the [K]16RGD (4.3 kDa) peptide when on its own would be able to transfer passively but when complexed to the pDNA or lipofectamine it would require an NLS. There is no general consensus sequence for NLSs but the best characterized sequences often involve a short sequence rich in basic amino acid residues such as lysine and arginine residues separated by a spacer. The peptide sequence used in this study was N-[K]16GGCRGDMFGCA which contains both lysine and arginine residues and thus may act as an NLS for translocation of pDNA to the nucleus. The k-nearest neighbor algorithm of the program PSORT II which predicts the subcellular localization of proteins from their amino acid sequence predicts nuclear localization (74%). This would explain the facile passage to the nucleus of the peptide and its intense signal, as demonstrated by confocal microscopy. In addition, the higher amount of pDNA in the nucleus in the presence of [K]16RGD indicates that the peptide is helping the pDNA enter the nucleus. This hypothesis is presently the subject of further investigation.

The nucleotide binding capacity of the peptide may also explain its intense nuclear signal and this may result in competition of the transgene with the cellular nuclear material and thereby lead to a dissociation of the transgene from the peptide. This possibility is also under investigation.

A recent publication also employing a peptide-based gene transfer strategy inspired by the mechanism of transfer of the adenovirus showed nuclear targeting of an adenovirus fiber peptide containing a polylysine sequence.5

In conclusion, our results demonstrate that the pDNA:[K]16RGD:lipofectamine complex enters cells at least in part by a clathrin-coated pit-dependent endocytic pathway and that endosomes then act as transport vesicles which fuse and became packed with lipo–oligoplexes. The acid environment of intracellular vesicles then facilitates release into the cytoplasm from where the complex, probably devoid of the liposome component, penetrates into the nucleus. Further investigation into endosomal escape and nuclear transport of this vector complex is in progress and should improve our knowledge concerning the prerequisites for an efficient gene transfer system.

Materials and methods

Cell culture

A fetal human tracheal epithelial cell line 56FHTE8o obtained from a non-cystic fibrosis fetus was kindly provided by Dr D Gruenert.41 Cells were grown in a 1:1 mixture of Dulbecco's minimum essential medium (DMEM) and Ham's F-12 nutrient medium (DMEM/F-12) and supplemented with fetal calf serum (10% v/v) (Boehringer Mannheim, Meylan, France). Cells were maintained at 37°C in 5% CO2–95% air and the media changed every second day.

Peptide and pDNA preparation and labeling

The peptides N-[K]16GGCRGDMFGCA ([K]16RGD) was synthesized, labeled (FITC) and cyclized as previously described.7 An American firefly (Phoptinus pyralis) luciferase DNA plasmid (5256 bp) (pGL3) under the control of an SV40 promoter and enhancer was used as a reporter gene. Plasmid DNA was purified and labeled with Cy5 by a modified nick-translation technique as previously described.642

Complex formation

The [K]16RGD peptide (FITC labeled or non-labeled) was mixed with the pDNA at a ratio of five to one (w:w), respectively, and incubated for 15 min at room temperature in Hepes-buffered saline. Peptide at this ratio condenses the DNA as shown by electron microscopy and renders it resistant to DNase.7 Lipofectamine (2 mg/ml) a 3:1 (w:w) mixture of a polycationic lipid (DOSPA) and a neutral lipid (DOPE) was then added (final ratio DNA: peptide:lipofectamine; 1:5:18 (w:w:w)) and the mixture incubated for a further 15 min.

Transfection

Cells were seeded on 12-well cell culture plates at a density of 4 × 105 cells per well and incubated at 37°C for about 12 h until 80% confluence. Transfections were performed on cells exposed to different conditions and drugs as follows: (1) For K+-depleted cells the cells were incubated for 45 min in K+-free buffer (140 mM NaCl, 20 mM Hepes pH 7.4, 1 mM CaCl2 and 1 mM MgCl2). Wells were rinsed twice with 500 μl of K+-free buffer, once with 500 μl K+-free hypotonic buffer (1:1 dilution in deionized water of K+-free buffer) and twice with 500 μl K+ free-buffer all at 37°C. After a further incubation for 15 min in K+-free buffer the complex, suspended in K+-free buffer, was added to the cells. Controls were performed in OptiMEM and in K+-free buffer supplemented with 4 mM KCl. (2) Cells were preincubated in hypertonic media (0.33 or 0.45 M sucrose) for 45 min at 37°C and transfection was performed in hypertonic media. Controls were performed in OptiMEM and in isotonic sucrose (0.25 mM). (3) Cells were also preincubated for 30 min at 37°C with wortmannin (100 nM) or LY294002 (30 μM) in OptiMEM and maintained during transfection. (4) Cells were also pre-incubated with or without cytochalasin B (5 μg/ml) for 45 min at 37°C in OptiMEM. Dimethyl sulphoxide was used to solubilize the above drugs and the final concentration was 1%. The cells were then washed twice with PBS before addition of the vector complex. (5) Chloroquine was added in OptiMEM at the indicated concentrations together with the vector complex during the transfection period.

The components of the vector complex (pDNA:[K]16 RGD:lipofectamine, 0.08 μg:0.4 μg:1.4 μg, per well) alone or together were added to the cells in a final volume of 300 μl of Opti-MEM (except for K+ depletion or exposure to hypertonic medium, see above). Cells were then incubated at 37°C for 4 h after which the components of the complex were removed. The cells were then washed in PBS, the normal supplemented media added and incubated for a further 48-h post-transfection period. Forty-eight hours was found to give optimal luciferase activity.

To verify that K+ depletion and hypertonic medium but not cytochalasin B inhibited receptor-mediated endocytosis in our cell system, treated and control cells were incubated in the presence of FITC-labeled transferrin (Molecular Probes, Eugene, OR, USA) (50 μg/ml) for 5 min at 37°C. The cells were then rinsed twice with citrate buffer (pH 4.6), to remove cell surface-bound transferrin,43 and twice with PBS. After trypsinization the cell fluorescence was examined by flow cytometry on a Cytofluorograph IIs (Ortho Diagnostic System, Westwood, MA, USA).

Luciferase activity was assayed as previously described6 and the protein concentration of cell lysates was determined using the Bradford method.44

The PI 3-K activity of cells treated with or without either wortmannin (100 nM) or LY294002 (30 μM) was measured as described.45

Cell uptake of 35S-labeled DNA

Cells at approximately 80% confluency (4 × 105 cells per well) in 12-well plates were treated with or without drugs or depleted of intracellular K+ as indicated above and incubated with 35S-labeled DNA:[K]16RGD complexes (0.02 μg pDNA:0.1 μg [K]16RGD per well) with or without lipofectamine (0.36 μg per well) for 2 h at 4 or 37°C. The cells were rinsed with PBS, harvested with 250 μl trypsin/EDTA per well for 10 min at 37°C or for 1 h at 4°C before being blocked with 250 μl of 1% BSA in PBS and recovered by centrifugation. The cell pellet (10 min, 1100 g) was rinsed once in PBS by centrifugation before lysis in 1 N NaOH. The protein concentration was determined by the method of Bradford and the radioactivity measured by scintillation counting with Ecolite (ICN, Orsay, France) in an LKB Rackbeta 1209 counter (EG & G Instruments, Evry, France).

Percoll density gradient fractionation of cell homogenates

Subcellular fractionation was performed essentially as described previously.4647 For characterization of the Percoll gradient fraction flasks (75 cm2) of cells were incubated at 37°C with horseradish peroxidase (HRP, 1.8 mg/ml) prepared in an internalization medium (minimum Eagle's medium (MEM), 10 mM Hepes, 5 mM D-glucose, pH 7.4) with gentle shaking for either 5 min (to label early endosomes) or 30 min (to label late endosomes). Cells were rinsed twice with cold PBS++ (137 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, 6.5 mM Na2HPO4, 1 mM CaCl2, 1 mM MgCl2) containing 5 mg/ml of bovine serum albumin (BSA) and twice with PBS++. The cells were then scrapped into 2.5 ml of PBS and centrifuged (5 min, 500g). The pellet was resuspended in 5 ml of homogenization buffer (250 mM sucrose, 3 mM imidazole, pH 7.4) and centrifuged for 10 min at 800 g. The pellet was gently resuspended in 1 ml of homogenization buffer containing protease inhibitors (pepstatine A, 1 mg/ml and leupeptine, 1 mg/ml), homogenized using a dounce (30 strokes); nuclei were still intact as determined by microscopy. The suspension was centrifuged for 15 min at 800 g to obtain a post-nuclear supernatant (PNS) and a nuclear pellet. A 1-ml cushion of 1 M sucrose was layered with a syringe and needle under a 25% Percoll (Pharmacia, Orsay, France) suspension in homogenization buffer in a centrifuge tube and the PNS was loaded on top. The gradient was centrifuged at 4°C for 90 min at 20000 g in a Beckman 90 Ti fixed angle rotor (Beckman, Gagny, France). Fractions (approximately 0.3 ml) were collected from the bottom of the tube using a peristaltic pump. The density of the different fractions was determined by running parallel tubes containing colored density marker beads (Pharmacia). The protein concentration in each fraction was determined by the Bradford method44 after precipitation of the Percoll with a final 0.2 N NaOH.

Cells were also transfected for 4 h with different combinations of vector using 1 μg of DNA per flask as described above, rinsed twice and chased for 0, 4 and 12 h in medium-containing serum. Subcellular fractionation was then performed as described above.

Purification of the nuclear fraction

The nuclear pellet obtained above was washed three times by centrifugation (750g, 10 min) in a 10 mM Tris HCl, pH 7.4 buffer, containing 10 mM NaCl, 3 mM MgCl2, 1 mM EDTA, 1 mM EGTA, 0.1% Nonidet P40 and the pellet resuspended in a 50 mM Tris HCl, pH 7.4 lysis buffer containing 100 mM NaCl, 3 mM MgCl2, 1 mM EDTA, 1 mM EGTA, 0.5% deoxycholate, 0.1% SDS, 0.2% Nonidet P40 and centrifuged (12000 g, 15 min). The supernatant was assayed for radioactivity, measured by scintillation counting with Ecolite (ICN, Orsay, France) in a LKB Rackbeta 1209 counter. The amount of nuclear DNA in the supernatant was determined by fluorimetry (excitation 346 nm, emission 460 nm) using the Hoechst dye.48 The morphology of the nuclei was assessed by phase contrast microscopy using a Leitz DMIL (Leica, Rueil Malmaison, France) inverted microscope. The cytoplasmic marker lactate dehydrogenase was assayed with a Synchron CX4 CE (Beckman) analyzer using pyruvate and NADH as substrate. The plasma membrane marker alkaline phosphatase was assayed with a kit from Sigma (St Quentin Fallavier, France) using p-nitrophenyl phosphate as substrate.

The 35S-labeled plasmid DNA was isolated from the nuclear fraction using a Qiagen (Courtaboeuf, France) DNA isolation QIAamp blood kit (small scale), electrophoresed on a 0.8% agarose gel and blotted on to a nylon Hybond N+ membrane (Amersham, Les Ulis, France) before autoradiography.

Organelle marker detection

The activity of the internalized fluid phase marker HRP was detected with the substrate 4-amino antipyridine. 100 μl of each fraction was added to 2 ml of 2.5 mM 4-amino antipyridine, 1.7 mM H2O2 in 0.2 M phosphate buffer, pH 7.0 and the mixture incubated at room temperature until color development and the supernatant (centrifugation for 30 min at 25200 g) read at 510 nm in a spectrophotometer. The activity of the lysosomal marker β-hexosaminidase was determined with the fluorescent substrate 4-methylumbelliferyl-β-D-galactoside. 100 μl of each fraction was added to a mixture of 1.2 mM 4-methylumbelliferyl-β-D-galactoside, 0.1% Triton X-100, 0.1 M sodium acetate, pH 4.4 in a final volume of 500 μl and incubated at 37°C for 30 min. The reaction was stopped by the addition of 1.5 ml of 0.5 M glycine in a 0.5 M carbonate, pH 10.0 buffer. The fluorescence was measured with a Perkin-Elmer LS 50B fluorimeter (excitation, 365 nm; emission, 460 nm) (Paris, France). The results of the activities of the enzyme markers are expressed as a percentage of the total activity.

The lysosomal-associated membrane protein (LAMP-1) was detected in fractions by immunodot blotting. Fractions (50 μl) were deposited on a nitrocellulose membrane equilibrated in TBS (0.15 M NaCl, 10 mM Tris HCl pH 7.5) and mounted in a dot-blot Minifold apparatus (Schleicher and Schuell, Ecquevilly, France). The membrane was incubated at 37°C for 1 h in 0.5% powdered milk in TBS then for 1.5 h at room temperature in a solution containing an anti-LAMP-1 (Pharmingen, San Diego, CA, USA) antibody (1/5 000 dilution), 0.5 M NaCl, 20 mM Tris HCl, pH 7.5, 0.3% powdered milk and 0.05% Tween 20. The membrane was first rinsed with 0.5 M NaCl, 20 mM Tris HCl, pH 7.5, containing 0.05% Tween 20 then with 0.5 M NaCl, 20 mM Tris HCl, pH 7.5 before incubation with a secondary peroxidase-coupled antibody (1/10000 dilution) in 0.5 M NaCl, 20 mM Tris HCl, pH 7.5, 0.3% powdered milk and 0.05% Tween 20. The membrane was washed three times for 5 min each in TBS then in distilled water for 15 min before chemiluminescense detection (Amersham). Lactate dehydrogenase and alkaline phosphatase activities were assayed as described above.

Confocal laser scanning microscopy

Cells (1.5 × 104 cells per well) were plated on Labtek 8-well plates (Gibco BRL, Cergy Pontoise, France) and the following day Cy5-labeled pDNA:fluorescein isothiocyanate (FITC)-labeled [K]16RGD complex, at the above indicated optimal ratio (0.3 μg pDNA per well), was added together with or without lipofectamine in OptiMEM for the indicated times; the same batch of labeled pDNA and peptide were used. Controls were performed previously to verify that the fluorescent labels remained bound.6 The cells were washed twice in PBS and confocal microscopy performed as previously described.6

Electron microscopy

Endosomes were prepared by ultracentrifugation on a Percoll gradient as described above. The fractions corresponding to endosomes were pooled and the Percoll removed by successive dilution and ultracentrifugation. The resulting pellet was fixed in 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer pH 7.4 at 4°C for 30 min, centrifuged for 15 min at 25000 g and resuspended in phosphate buffer. Osmium tetroxide was added to a final 2% and the preparation incubated for 30 min at room temperature. The pellet was then dehydrated in a graded series of ethanol (70, 95 and 100%), embedded in Epon, sectioned and stained first with uranyl acetate then with lead citrate.

The fixation, post-fixation and staining of whole cells incubated for different times in the presence of the vector complex (pDNA:[K]16RGD:lipofectamine; 1:5:18) was performed as described above. The endosome preparation and whole cells were viewed with a JOEL 1010 electron microscope.

Statistical analysis

Data are presented as the mean ± s.e.m. of triplicate determinations and are representative of results obtained in three independent experiments that produced similar relative results. Statistical analysis was performed using the nonparametric rank test of Wilcoxon. Probability values <0.05 were considered statistically significant.

References

  1. 1

    Coutelle C . Gene therapy for cystic fibrosis. Strategies, problems and perspectives. In: Strauss M, Barranger JA (eds) Concepts in Gene Therapy Walter de Gruyter & Co: Berlin, New York 1997 313–343

    Google Scholar 

  2. 2

    Lee ER et al. Detailed analysis of structures and formulations of cationic lipids for efficient gene transfer to the lung Hum Gene Ther 1996 7: 1701–1717

    CAS  Article  Google Scholar 

  3. 3

    Lasic DD, Templeton NS . Liposomes in gene therapy Adv Drug Del Rev 1996 20: 221–266

    CAS  Article  Google Scholar 

  4. 4

    Hart SL et al. Gene delivery and expression mediated by an integrin-binding peptide Gene Therapy 1995 2: 552–554

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5

    Zhang F et al. A transfecting peptide derived from adenovirus fiber protein Gene Therapy 1999 6: 171–181

    CAS  Article  Google Scholar 

  6. 6

    Colin M et al. Liposomes enhance delivery and expression of an RGD-oligolysine gene transfer vector in human tracheal cells Gene Therapy 1998 5: 1488–1498

    CAS  Article  Google Scholar 

  7. 7

    Harbottle RP et al. An Arg–Gly–Asp/oligolysine peptide: a prototype construct for integrin mediated gene delivery Hum Gene Ther 1998 9: 1037–1047

    CAS  Article  Google Scholar 

  8. 8

    Hart SL et al. Lipid-mediated enhancement of transfection by a nonviral integrin-targeting vector Hum Gene Ther 1998 9: 575–585

    CAS  Article  Google Scholar 

  9. 9

    Matsui H, Johnson LG, Randell SH, Boucher RC . Loss of binding and entry of liposome–DNA complexes decreases transfection efficiency in differentiated airway epithelial cells J Biol Chem 1997 272: 1117–1126

    CAS  Article  Google Scholar 

  10. 10

    Zabner J et al. Cellular and molecular barriers to gene transfer by a cationic lipid J Biol Chem 1995 270: 18997–19007

    CAS  Article  Google Scholar 

  11. 11

    Labat-Moleur F et al. An electron microscopy study into the mechanism of gene transfer with polyamines Gene Therapy 1996 3: 1010–1017

    CAS  PubMed  PubMed Central  Google Scholar 

  12. 12

    Zhou X, Huang L . DNA transfection mediated by cationic liposomes containing lipopolylysine: characterization and mechanism of action Biochim Biophys Acta 1994 1189: 195–203

    CAS  Article  Google Scholar 

  13. 13

    Griffiths G, Gruenberg J . The arguments for pre-existing early and late endosomes Trends Cell Biol 1991 1: 5–9

    CAS  Article  Google Scholar 

  14. 14

    Riezman H, Woodman PG, van Meer G, Marsh M . Molecular mechanisms of endocytosis Cell 1997 91: 731–738

    CAS  Article  Google Scholar 

  15. 15

    Misteli T, Spector DL . The cellular organization of gene expression Curr Opin Cell Biol 1998 10: 323–331

    CAS  Article  Google Scholar 

  16. 16

    Wickham TJ, Mathias P, Cheresh, DA, Nemerow GR . Integrins αvβ3 and αvβ5 promote adenovirus internalization but not virus attachment Cell 1993 73: 309–319

    CAS  Article  Google Scholar 

  17. 17

    Larkin JM, Donzell WC, Anderson RGW . Potassium-dependent assembly of coated pits: new coated pits form as planar clathrin lattices J Cell Biol 1986 103: 2619–2627

    CAS  Article  Google Scholar 

  18. 18

    Heuser JE, Anderson RGW . Hypertonic media inhibit receptor-mediated endocytosis by blocking clathrin-coated pit formation J Cell Biol 1989 108: 389–400

    CAS  Article  Google Scholar 

  19. 19

    Li G et al. Evidence for phosphatidylinositol 3-kinase as a regulator of endocytosis via activation of Rab5 Proc Natl Acad Sci USA 1995 92: 10207–10211

    CAS  Article  Google Scholar 

  20. 20

    Kurashima K et al. Endosomal recycling of the Na+/K+ exchanger NHE3 isoform is regulated by the phosphatidylinositol 3-kinase pathway J Biol Chem 1998 273: 20828–20836

    CAS  Article  Google Scholar 

  21. 21

    Salisbury JL, Keller GA . Structural investigations on the role of microfilaments in ligand translocation Meth Enzymol 1983 98: 368–375

    CAS  Article  Google Scholar 

  22. 22

    Brown EJ . Phagocytosis BioEssays 1995 17: 109–117

    CAS  Article  Google Scholar 

  23. 23

    Leopold PL et al. Fluorescent virons: dynamic tracking of the pathway of adenoviral gene transfer vectors in living cells Hum Gene Ther 1998 9: 367–378

    CAS  Article  Google Scholar 

  24. 24

    Nicolau C, Sene C . Liposome-mediated DNA transfer in eukaryotic cells. Dependence of the transfer efficiency upon the type of liposomes used and the host cell cycle stage Biochim Biophys Acta 1982 721: 185–190

    CAS  Article  Google Scholar 

  25. 25

    Watanabe Y et al. Highly efficient transfection into primary cultured mouse hepatocytes by use of cation-liposomes: an application for immunization J Biochem 1994 116: 1220–1226

    CAS  Article  Google Scholar 

  26. 26

    Memmo LM, McKeown-Longo P . The αvβ5 integrin functions as an endocytic receptor for vitronectin J Cell Sci 1998 111: 425–433

    CAS  Google Scholar 

  27. 27

    Timar J et al. Immunomorphological characterization and effects of 12-(S)-HETE on a dynamic intracellular pool of the αIIbβ3-integrin in melanoma cells J Cell Sci 1995 108: 2175–2186

    CAS  PubMed  Google Scholar 

  28. 28

    Pol A, Ortega D, Enrich C . Identification and distribution of proteins in isolated endosomal fractions of rat liver: involvement in endocytosis, recycling and transcytosis Biochem J 1997 323: 435–443

    CAS  Article  Google Scholar 

  29. 29

    Friend DS, Papahadjopoulos D, Debs RJ . Endocytosis and intracellular processing accompanying transfection mediated by cationic liposomes Biochim Biophys Acta 1996 1278: 41–50

    Article  Google Scholar 

  30. 30

    Isberg RR, Tran Van Hhieu G . The mechanism of phagocytic uptake promoted by invasin–integrin interaction Trends Cell Biol 1995 5: 120–124

    CAS  Article  Google Scholar 

  31. 31

    Maxfield FR . Weak bases and ionophores rapidly and reversibly raise the pH of endocytic vesicles in cultured mouse fibroblasts J Cell Biol 1982 95: 676–681

    CAS  Article  Google Scholar 

  32. 32

    Wagner E et al. Influenza virus hemagglutinin HA-2 N-terminal fusogenic peptides augment gene transfer by transferrin–polylysine–DNA complexes: toward a synthetic virus-like gene-transfer vehicle Proc Natl Acad Sci USA 1992 89: 7934–7938

    CAS  Article  Google Scholar 

  33. 33

    Midoux P et al. Specific gene transfer mediated by lactosylated poly-L-lysine into hepatoma cells Nucleic Acids Res 1993 21: 871–878

    CAS  Article  Google Scholar 

  34. 34

    Erbacher P, Roche AC, Monsigny M, Midoux P . Putative role of chloroquine in gene transfer into a human hepatoma cell line by DNA/lactosylated polylysine complexes Exp Cell Res 1996 225: 186–194

    CAS  Article  Google Scholar 

  35. 35

    Fominaya J, Uherek C, Wels W . A chimeric fusion protein containing transforming growth factor-α mediates gene transfer via binding to the EGF receptor Gene Therapy 1998 5: 521–530

    CAS  Article  Google Scholar 

  36. 36

    Legendre J-Y, Szoka FC Jr . Delivery of plasmid DNA into mammalian cell lines using pH-sensitive liposomes: comparison with cationic liposomes Pharm Res 1992 9: 1235–1242

    CAS  Article  Google Scholar 

  37. 37

    Farhood H, Serbina N, Huang L . The role of dioleoyl phosphatidylethanolamine in cationic liposome mediated gene transfer Biochim Biophys Acta 1995 1235: 289–295

    Article  Google Scholar 

  38. 38

    Meunier-Durmont C et al. Mechanism of adenovirus improvement of cationic liposome-mediated gene transfer Biochem Biophys Acta 1997 1330: 8–16

    Article  Google Scholar 

  39. 39

    Fasbender A et al. Complexes of adenovirus with polycationic polymers and cationic lipids increase the efficiency of gene transfer in vitro and in vivo J Biol Chem 1997 272: 6479–6489

    CAS  Article  Google Scholar 

  40. 40

    Wattiaux R et al. Cationic lipids delay the transfer of plasmid DNA to lysosomes Biochem Biophys Res Commun 1996 227: 448–454

    CAS  Article  Google Scholar 

  41. 41

    Gruenert DC et al. Characterization of human tracheal epithelial cells transformed by an origin-defective simian virus 40 Proc Natl Acad Sci USA 1998 85: 5951–5955

    Article  Google Scholar 

  42. 42

    Wheeler VC, Coutelle C . Nondegradative in vitro labelling of plasmid DNA Anal Biochem 1994 225: 374–376

    Article  Google Scholar 

  43. 43

    Ghosh RN, Gelman, DL, Maxfield FR . Quantification of low density lipoprotein and transferrin endocytic sorting in HEp2 cells using confocal microscopy J Cell Sci 1994 107: 2177–2189

    CAS  PubMed  Google Scholar 

  44. 44

    Bradford M . A rapid and sensitive method for the quantification of microgram quantities of protein utilizing protein-dye binding Anal Biochem 1976 72: 248–254

    CAS  Article  Google Scholar 

  45. 45

    Whitman M et al. Association of phosphatidylinositol kinase activity with polyoma middle-T competent for transformation Nature 1985 315: 239–242

    CAS  Article  Google Scholar 

  46. 46

    Galloway CJ et al. Acidification of macrophage and fibroblast endocytic vesicles in vitro Proc Natl Acad Sci USA 1983 80: 3334–3338

    CAS  Article  Google Scholar 

  47. 47

    Green SA, Zimmer K-P, Griffiths G, Mellman I . Kinetics of intracellular transport and sorting of lysosomal membrane and plasma membrane proteins J Cell Biol 1987 105: 1227–1240

    CAS  Article  Google Scholar 

  48. 48

    Cesarone CF, Bolognesi C, Santi L . Improved microfluorometric DNA determination in biological material using 33258 Hoechst Anal Biochem 1979 100: 188–197

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This work was supported by grants from the ‘Institut National de la Santé et de la Recherche Médicale’ (INSERM) and from the ‘Association Française de Lutte contre la Mucoviscidose’. We thank Dr E Coudrier of CNRS UMR 144 at the ‘Institut Curie’ for advice on subcellular fractionation. Our thanks to the members of the INSERM units 482 and 489 at the St-Antoine and Tenon Hospitals, respectively for access to their electron microscope. We also thank Dr M Caron, Dr E Lasnier, Ms M Auclair and Ms L Grivot of INSERM U402. CC is funded by the Muller Bequest/CF-Trust, the MRC and the March of Dimes Birth Defects Foundation. RPH and AK are supported by the March of Dimes Birth Defects Foundation (19-FY97-0632).

Author information

Affiliations

Authors

Corresponding author

Correspondence to M C Brahimi-Horn.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Colin, M., Maurice, M., Trugnan, G. et al. Cell delivery, intracellular trafficking and expression of an integrin-mediated gene transfer vector in tracheal epithelial cells. Gene Ther 7, 139–152 (2000). https://doi.org/10.1038/sj.gt.3301056

Download citation

Keywords

  • integrin
  • intracellular traffic
  • gene transfer
  • liposome
  • receptor-mediated endocytosis

Further reading

Search