HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 36, 33875-33880, September 7, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/36/33875    most recent M011553200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ruponen, M. Articles by Urtti, A. Search for Related Content PubMed PubMed Citation Articles by Ruponen, M. Articles by Urtti, A. Social Bookmarking                      What's this?

Extracellular Glycosaminoglycans Modify Cellular Trafficking of Lipoplexes and Polyplexes^*

Marika Ruponen^§, Seppo Rönkkö, Paavo Honkakoski, Jukka Pelkonen^¶, Markku Tammi^**, and Arto Urtti

From the Departments of  Pharmaceutics, ^¶ Clinical Microbiology, and ^** Anatomy, University of Kuopio and the  Department of Pediatrics, Kuopio University Hospital, FIN-70211 Kuopio, Finland

Received for publication, December 21, 2000, and in revised form, June 4, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

It has been shown that extracellular glycosaminoglycans (GAGs) limit the gene transfer by cationic lipids and polymers. The purpose of this study was to clarify how interactions with anionic GAGs (hyaluronic acid and heparan sulfate) modify the cellular uptake and distribution of lipoplexes and polyplexes. Experiments on cellular DNA uptake and GFP reporter gene expression showed that decreased gene expression can rarely be explained by lower cellular uptake. In most cases, the cellular uptake is not changed by GAG binding to the lipoplexes or polyplexes. Reporter gene expression is decreased or blocked by heparan sulfate, but it is increased by hyaluronic acid; this suggests that intracellular factors are involved. Confocal microscopy experiments demonstrated that extracellular heparan sulfate and hyaluronic acid are taken into cells both with free and DNA-associated carriers. We conclude that extracellular GAGs may alter both the cellular uptake and the intracellular behavior of the DNA complexes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Gene therapy holds great promise in the treatment of genetic and acquired diseases. The majority of gene transfer protocols utilizes viral delivery systems. Although many promising results have been achieved, the safety concerns and the difficulty of production on a large scale limit the usefulness of the recombinant viral vectors. This has prompted further development of viral vectors but also the search for efficient, nonimmunogenic, and easy-to-prepare nonviral vector systems.

In nonviral gene therapy, the disease is treated with exogenously given plasmid DNA that is transcribed and translated to produce the therapeutic protein(s). However, the delivery of naked plasmid DNA to the target cells is not efficient because of the large size and multiple negative charges of the DNA molecule (1). Therefore, increasing attention has focused on the development of nonviral carriers such as cationic lipids (2-4) or cationic polymers (5-7). These carriers complex with DNA to form condensed structures (usual mean diameter, 40-200 nm) that are termed either lipoplexes or polyplexes (8-10). The complexes have a positively charged surface, which facilitates the delivery of DNA into the target cells due to electrostatic binding to the cell surface (7). Although numerous gene delivery vehicles work in cultured cells, they are not efficient enough in vivo. Sometimes gene delivery by the complexes is less than it is with naked DNA (11-13). The factors that control gene transfer to the cells are still poorly understood (14). A good understanding of the mechanisms of gene delivery and identification of the limiting factors should help in the development of efficient delivery systems for DNA.

In recent years, extracellular glycosaminoglycans (GAGs)¹ have been found to be one of the main biological factors that affect gene delivery (15-17). GAGs are linear, negatively charged polysaccharides. They are the major components in the extracellular matrix of many tissues (for example, vascular walls and connective tissues), but they are also found inside the cells and on the cell surface (18). Recently, GAGs have been shown to have a dual role in gene delivery. Cell membrane-associated GAGs have been shown to mediate the cellular entry of DNA complexes both in vitro (15) and in vivo (16), suggesting that GAGs may act as (central) receptors for gene delivery complexes. On the other hand, the interactions between the extracellular or secreted GAGs and various complexes decrease the gene transfer depending on the structures and charge densities of the carriers and GAGs (15, 17, 19). Previously, we have shown that the GAG-mediated inhibition of transgene expression is rarely associated with DNA release or relaxation of the complexes by GAGs (17). This indicates that other mechanisms of GAG-mediated inhibition must exist.

In this study, we show that the inhibitory effects of GAGs on gene transfer cannot be explained, in most cases, by the decreased cellular uptake of the complexes. Furthermore, GAGs may bind to the cationic carrier or to the surface of positively charged complexes, and in some cases, GAG may eventually replace DNA in the complex resulting in the uptake of GAG into the cells instead of DNA. Finally, by confocal microscopy we found that extracellular GAGs are taken into cells by the cationic carriers and the cationic complexes and that they may alter the intracellular behavior of the complexes.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Polycations-- Polyethyleneimine with a mean molecular weight of 25 kDa was obtained from Aldrich and was used as a 10 mM aqueous stock solution (6). Poly-L-lysine hydrobromide (PLL; mean molecular weight of 200,000), obtained from Sigma, was dissolved in water (3 mg/ml).

Liposomes-- 1,2-Dioleyl-3-phosphatidylethanolamine (DOPE) and N-(1-(2,3-dioleoyloxy)propyl)-N,N,N-trimethyl ammonium methylsulfate (DOTAP) were purchased from Avanti Polar Lipids. Cationic liposomes composed of DOTAP or DOTAP/DOPE at a molar ratio of 1:1 were prepared by evaporating a chloroform solution of lipids, resuspending the lipids in water at a 3.2 mM concentration, and sonicating the liposomal suspension under argon until a translucent lipid solution was obtained.

Plasmids-- The reporter gene plasmid that encodes -galactosidase under the control of cytomegalovirus (CMV) promoter (20) was a gift from Dr. F. C. Szoka, Jr. (University of California, San Francisco). The green fluorescent protein (GFP) S65T mutant was excised from pTR5-DC/GFP plasmid (21) as a BamHI fragment that was inserted into the BamHI site of CMV-driven pCR3 plasmid (Invitrogen). The plasmids were amplified in Escherichia coli and purified on ion exchange columns (Qiagen). Plasmid integrity was confirmed by agarose gel electrophoresis. The concentrations of DNA were determined by absorbance at 260 nm. Rhodamine-labeled CMV-driven -galactosidase plasmid (pGeneGrip) was obtained from Gene Therapy Systems (San Diego, CA).

Plasmid Labeling-- Ethidium monoazide (EMA, Molecular Probes) forms covalent bonds with DNA bases during photoactivation. Ethidium monoazide-labeled DNA (EMA-DNA) was prepared according to the procedure of Zabner et al. (14) with minor modifications. Briefly, EMA in water (5 µg/ml) was added to CMV-GFP plasmid in water (200 µg/ml), and the mixture was incubated for 10 min at room temperature. The solution was then exposed to UV light at 312 nm for 2 min. Gel filtration on NAP-10 columns (Amersham Pharmacia Biotech) was used to purify the labeled DNA from free EMA. To remove intercalated but not covalently bound EMA, cesium chloride was added to a concentration of 1.1 g/ml and plasmid was extracted with CsCl-saturated isopropanol. CsCl was removed by dialysis against Tris-EDTA buffer, and the labeled EMA-DNA plasmid was recovered with ethanol precipitation.

Fluorescein-labeled Hyaluronic Acid-- Hyaluronic acid (from bovine trachea, purity more than 95%, mean molecular weight 1.4 million Da; CarboMer, Westborough, MA) was labeled with fluorescein amine (Fluka), as previously described by de Balder and Wik (22), and lyophilized. Labeled hyaluronic acid was used as a 3.3 mg/ml aqueous solution. The molar ratio of fluorescein molecules/hyaluronic unit of disaccharide was obtained from the absorbance of the fluorescein-labeled hyaluronic acid at 490 nm. The molar ratio was about 1:10.

Characterization of Fluorescein-labeled Hyaluronic Acid-- Gel filtration of the fluorescein-labeled hyaluronic acid on 1.0 × 30 cm columns of Sephacryl S-400 and S-1000 resins (Amersham Pharmacia Biotech) eluted at 0.4 ml/min with 100 mM ammonium hydrogen carbonate showed that the molecular size of the hyaluronic acid ranged from 100,000 to a few million Da according to Amersham Pharmacia Biotech's calibration curves. These values corresponded to the average 1.4 million stated by the manufacturer. Fluorescein-labeled hyaluronic acid (100 µg) was incubated overnight at 37 °C in 50 units of testicular hyaluronoglucosaminidase (Sigma, type IV-S, EC 3.2.1.35) in 0.1 M sodium chloride, 0.1 M sodium acetate buffer, pH 6.0. The digest was analyzed by gel filtration on a 1.0 × 30 cm Superdex Peptide column (Amersham Pharmacia Biotech), eluted with 100 mM ammonium hydrogen carbonate at 0.5 ml/min, and monitored for fluorescence at 488/530 nm excitation/emission. After digestion the fluorescence shifted completely to an elution position that corresponded to the size of underivatized oligosaccharides of 12-18 monosaccharide units. This indicates that the hyaluronic acid did not contain protein contaminants and that the fluorescein-labeled hyaluronic acid remained susceptible to hyaluronoglucosaminidase, suggesting retained biological activity.

Fluorescein-labeled Heparan Sulfate-- Fluorescein was coupled to heparan sulfate using the procedure of Nagasawa and Uchiyama (23). Heparan sulfate (from porcine intestinal mucosa, purity 90%, mean molecular weight of 14,000; Sigma) was dissolved in water (2 mg/ml), and fluorescein isothiocyanate isomer I (Sigma) was dissolved in 0.5 M sodium carbonate, pH 8.5 (1 mg/ml). Solutions were mixed (1 mg heparan sulfate with 0.15 mg fluorescein) and stirred for 16 h at 37 °C. After incubation, free label was removed from the labeled heparan sulfate on a NAP-10 column at 4 °C with 0.1 M sodium acetate, pH 6.8, as elution buffer. Labeled heparan sulfate was precipitated with absolute ice-cold ethanol (20 ml) and with saturated sodium chloride (250 µl). The yellow precipitate was washed with 70% ethanol. After lyophilization, the final yield was determined, and precipitate was dissolved in water as a 3.3 mg/ml solution. The molar ratio of label molecule to disaccharide unit was ~1:50 as determined by spectrophotometry.

Characterization of the Fluorescein-labeled Heparan Sulfate-- Aliquots of the fluorescein-labeled heparan sulfate (80-100 µg) were digested with testicular hyaluronoglucosaminidase as described above and with 10 units of heparitin-sulfate lyase III (Sigma, EC 4.2.2.8) in 5 mM calcium acetate buffer, pH 7.0, overnight at 37 °C. Part of the heparitin-sulfate lyase III-digested fluorescein-labeled heparan sulfate was further treated by nitrous acid according to Lindahl et al. (24). Analysis of native and testicular hyaluronoglucosaminidase-digested fluorescein-labeled heparan sulfate (100 µg) on the Sephacryl S-400 column showed an elution position well in agreement with the average size of 14,000 Da announced by the manufacturer for the underivatized material and no change by hyaluronoglucosaminidase treatment, indicating no significant contamination by hyaluronic acid or chondroitin sulfate. About 85% of the heparitin-sulfate lyase III-digested fluorescein-labeled heparan sulfate moved from its original position in the excluded volume of the Superdex Peptide column, indicating its susceptibility to the enzyme specific for the monosulfated and nonsulfated regions of this polymer. Nitrous acid treatment, attacking the N-sulfated regions of heparan sulfate, shifted about 70% of the fluorescein-labeled heparan sulfate from the excluded volume of the Superdex peptide column. After a combination of heparitin-sulfate lyase III and nitrous acid digestion, the fluorescence was completely shifted from its original position in the excluded volume, which indicated the purity of the fluorescein-labeled heparan sulfate. Aliquots of the heparitin-sulfate lyase III digest were also subjected to reducing end labeling with 2-aminoacridone and polyacrylamine gel electrophoresis (25), which revealed two major bands with migration positions close to but not identical with unsulfated and monosulfated disaccharides of chondroitin sulfate. Furthermore, fluorescein-labeled heparan sulfate and hyaluronic acid were applied to a 1-ml Hi-Trap^TM Q column (Amersham Pharmacia Biotech) eluted at 1 ml/min with a linear 0.1-1.2 M sodium chloride gradient in 0.1% CHAPS and 50 mM Tris, pH 7.4. Fluorescein-labeled heparan sulfate eluted as a single peak in the gradient separate from the hyaluronic acid.

Cell Culture-- The RAA SMC cell line (smooth muscle cells from rabbit aortic media) was a kind gift from Dr. Seppo Ylä-Herttuala (University of Kuopio, Finland). The cells were grown in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% heat-inactivated fetal bovine serum and penicillin (100 units/ml)-streptomycin (100 µg/ml) (all from Life Technologies, Inc.) at 37 °C in 7% CO₂. The cells were subcultured twice a week.

Transfection Protocol-- The cells were seeded onto 6-well plates at a density of 120 000 cells/well in 2 ml of growth medium. After the overnight incubation prior to transfection, the medium was replaced with serum-free medium. Complexes were prepared by adding a solution of DNA (3 µg/well) in 50 mM MES, 50 mM HEPES, 75 mM NaCl buffer, pH 7.2, to an equal volume of carrier in buffer at charge ratio ± 4. After a15-min incubation, GAGs were added to the complexes at 3-fold the anionic charge excess. Solutions containing DNA, carriers, and GAGs were added to the cells for 5 h. The cells were fixed before analysis by flow cytometry. First, the cells were washed twice with phosphate-buffered saline and once with 1 M sodium chloride solution to remove the complexes attached to the plasma membrane. Then cells were detached from the bottom of the wells with 1 ml of trypsin (0.05 g/liter)-EDTA (0.02 g/liter) (Life Technologies, Inc.). After harvesting, the cells were fixed by incubating them for 5-10 min in 1% paraformaldehyde. After incubation, the cells were washed twice with 1% paraformaldehyde and stored at +4 °C prior to analysis by flow cytometry.

In cellular uptake experiments (EMA-DNA uptake and fluorescein-GAG uptake), the cells were fixed and analyzed immediately after 5 h of exposure to complexes. In the GFP experiments, the cells were washed with phosphate-buffered saline after exposure, and 1.5 ml of growth medium was added to cells. The cells were analyzed for GFP expression 24 h after removal of the complexes.

Flow Cytometry Analysis-- Cellular uptake of both DNA and GAG and expression of GFP were analyzed by flow cytometry (FACScan, Becton Dickinson) with an argon ion laser (488 nm) as the excitation source. Fluorescence of GFP and fluorescein-labeled GAGs was collected at 525 nm (FL 1), and red fluorescence of EMA was collected at 670 nm (FL 3). For each sample, 10,000 events were collected. Cells cultured under normal culture conditions (control cells) were visualized on a forward angle light scatter (FSC) versus a 90° light scatter (SSC) display, and living cells were defined by gating the major population of the cells; only the cells within this gate were analyzed.

Cellular Uptake of EMA-DNA-- EMA-DNA was used as marker for intracellular delivery of DNA. The number of positive events was analyzed from FL 3 histogram (Fig. 1, A-D). The gate of positive events for each carrier was adjusted according to the negative control (transfections were made with unlabeled DNA-carrier complexes). The percentage of positive cells was calculated as the number of positive events in the FL 3 histogram divided by the total number of events in the live cell gate.

GFP Expression-- The number of GFP positive cells was analyzed from the FL 1 versus FL 3 dot plot (Fig. 1, E-H). The positive events were separated from the autofluorescence by setting a gate, G1 (Fig. 1, E-H).

Cellular Uptake of Fluorescein-GAGs-- A plasmid expressing -galactosidase was used in these experiments. The number of fluorescein-GAG positive cells was analyzed in a similar manner as for cells expressing GFP.

Confocal Microscopy-- For the confocal microscopy experiments, the cells were cultured on 8-well-chambered cover glasses (Lab-Tek II, Nunc). Transfections were done as described, above but the amount of DNA was 1 µg/well, and the images were taken at 24 h post-transfection from living cells by confocal microscopy on a UltraVIEW confocal imaging system (PerkinElmer Life Sciences) with an Eclipse TE300 microscope (Nikon) using a 40× (NA 0.6) oil immersion objective (Plan Fluor, Nicon). Fluorescein-labeled GAGs were imaged using the 488 nm excitation line of krypton/argon laser, and green fluorescence was detected at 515-545 nm. Rhodamine-labeled DNA was detected at 590-610 nm after excitation at 568 nm. Exposure times were between 0.2 and 0.4 s, and confocal images were collected with a cooled digital CCD (charge-coupled device) camera (PerkinElmer Life Sciences). Serial images of fluorescein and rhodamine fluorescence at ~0.5 µm Z-intervals were recorded separately and then co-localized. Images were processed and analyzed using the confocal assistant software program (UltraVIEW) and Photoshop (Adobe Systems Inc.).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Autofluorescence of Cells in Flow Cytometry Analysis-- After exposing the cells to carrier-DNA complexes, we noticed that the autofluorescence of the cells was increased and the histogram plot was shifted to the right (Fig. 1, A-D). The increase in the mean fluorescence intensity at the FL 3 channel varied from about 2.1-fold (DOTAP) to 1.2-fold (PLL, DOTAP/DOPE). A similar increase in the autofluorescence was also detected at the FL 1 channel (Fig. 1, E-H). Therefore, it was very important to have a negative control for each carrier (i.e. transfections done with unlabeled DNA) to determine the autofluorescence for each treatment separately instead of the autofluorescence of untransfected cells.

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Analysis of internalized EMA-labeled DNA (A-D) and GFP-expression (E-H) by flow cytometry. The cellular uptake of EMA-labeled DNA by various carriers (A, PEI; B, PLL; C, DOTAP; D, DOTAP/DOPE) was measured immediately after 5 h transfection. Dotted lines show the autofluorescence of the RAA SMC cells without any treatment. Solid lines (with gray area) show the influence of complexes on autofluorescence of the cells. Dark solid lines represent the cell population of EMA positive cells. The number of EMA positive cells was analyzed by setting a gate according to negative control (solid lines with gray area). GFP expression of the complexes (E, PEI; F, PLL; G, DOTAP; H, DOTAP/DOPE) was analyzed 24 h post-transfection from the FL 1 versus FL 3 dot plots. The positive events were separated from the autofluorescence by setting a gate (G1).

Cellular Uptake of DNA and GFP Expression-- EMA-labeled DNA was used to measure the cellular uptake of DNA, the results of which are shown as a percentage of EMA positive cells. The cellular uptake of the polyplexes (PEI, PLL) was more than 50%, whereas 17-25% of cells took up lipoplexes (DOTAP, DOTAP/DOPE) (Table I). Thus, it seems that polyplexes were taken up by the cells more efficiently than the lipoplexes. The percentage of GFP positive cells was low, maximally only about 2% (Table I and Fig. 1, E-H). The number of GFP positive cells (2.1%) after transfection with PEI polyplexes was about 20-fold higher than with PLL and DOTAP/DOPE complexes (Table I and Fig. 1, E-H). The mean intensity of GFP positive cells was about two times higher with PEI and DOTAP complexes than with PLL and DOTAP/DOPE complexes (Fig. 1, E-H). The rank order of both GFP positive cells and the mean fluorescence intensities matches with our earlier report using -galactosidase as the marker gene (17).

These results also indicate that the cellular uptake of the DNA does not correlate with the gene expression. PLL polyplexes were taken into the cells two times more efficiently than lipoplexes, but the levels of GFP expression ranked oppositely (Table I).

Each carrier was studied at charge ratios of ±2 and ±4, but because the conclusions were similar for both cases only the data for a charge ratio of ±4 is presented. In addition, we studied the cellular uptake of anionic DNA complexes (charge ratio ± 0.25). In the case of PLL and DOTAP, the cellular uptake was below the detection limit. Only 2 and 4% of cells took up anionic PEI and DOTAP/DOPE complexes, respectively. Furthermore, the cellular uptake of DNA and GFP expression by several other carriers (different molecular weights of PEI and PLL, DOTAP/cholesterol and dioctadecylamidoglycylspermine (DOGS)) were also tested, but only a few carriers that represented different classes of behavior were chosen for further experiments.

Influence of GAGs on Cellular Uptake and Expression of DNA Complexes-- Next, we studied the effects of extracellular hyaluronic acid and heparan sulfate on cellular DNA uptake and gene expression. The cellular uptake of the PEI polyplexes was decreased to one-third by hyaluronic acid and completely blocked by heparan sulfate (Table I). Both GAGs increased the internalization of PLL polyplexes from 50 to ~75%, whereas GAGs had only a moderate influence on the uptake of the lipoplexes (Table I).

The effects of extracellular GAGs on GFP expression were clear. In the case of PEI, hyaluronic acid clearly decreased and heparan sulfate totally blocked GFP expression (Table I). With PLL polyplexes and both lipoplexes, hyaluronic acid actually increased the number of GFP positive cells by about 1.6-2.3-fold, whereas heparan sulfate reduced the GFP expression in all cases (Table I).

These results indicate that the decrease or total block in gene expression, caused by extracellular GAGs, could not be explained merely by a decrease in the cellular uptake of the complexes.

The experiments were repeated with human retinal pigment epithelial cell line (D407) with similar results. This suggests that the effects of GAGs on the delivery of carrier-DNA complexes were not cell line specific.

The Cellular Uptake of GAGs with Free Carrier and Carrier-DNA Complexes-- The internalization of fluorescein-labeled GAGs (3-fold anionic charge excess) with complexes was tested by flow cytometry at various charge ratios (carrier-DNA) ranging from an excess of positive charges (±4) to an excess of negative charges (±0.5) (Fig. 2). The experiments were also carried out without DNA, using corresponding amounts of cationic carriers to obtain information about the cellular delivery of GAGs by the free carriers. The results are shown as a percentage of positive cells.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   The cellular uptake of hyaluronic acid (A and B) and heparan sulfate (C and D). Fluorescein-labeled GAGs were incubated with complexes (filled symbols) or with free carrier (open symbols) at various charge ratios (±4-±0.5). RAA SMC cells were exposed to complexes for 5 h, and the internalized GAGs were analyzed by flow cytometry. The results are shown as a percentage of fluorescein-GAG positive cells (mean of triplicates ± S.D.).

The cellular uptake of fluorescein-labeled hyaluronic acid and heparan sulfate as free molecules was below the detection limit in flow cytometry. However, at 20× higher concentrations of fluorescein-labeled GAGs, the cellular uptake was enhanced to 10-20% depending on the GAG.

The cellular uptake of hyaluronic acid was very similar for all carrier-DNA complexes (Fig. 2A). Hyaluronic acid was delivered efficiently into the cells at charge ratios higher than unity (Fig. 2, A and B) when the complexes had positive -potential and free carrier molecules were also present. At lower charge ratios, DNA binds all carrier molecules and the surface of the complexes becomes negatively charged. These negatively charged complexes did not transfer hyaluronic acid into the cells, whereas hyaluronic acid was efficiently taken into the cells by corresponding amounts of free carrier (Fig. 2, A and B). This suggests that the cellular uptake of hyaluronic acid was mediated for the most part by the free carrier and/or by complexes bearing a net positive charge.

The cellular uptake of heparan sulfate (Fig. 2, C and D) by PLL or DOTAP/DOPE complexes was similar compared with the cellular uptake of hyaluronic acid. In other words, the heparan sulfate was transferred by the free carrier and/or positively charged complexes (Fig. 2, C and D). The cellular uptake of heparan sulfate by DOTAP and PEI complexes, however, was quite different. In these cases, the heparan sulfate was taken into the cells as well with complexes as with free carrier, also at a charge ratio below unity (Fig. 2, C and D). This suggests that heparan sulfate is not only taken up by free carrier but also with these negatively charged complexes.

Intracellular Distributions of DNA and GAGs-- Flow cytometry studies could not distinguish whether GAGs were internalized into the cells by the free carrier or by the DNA complexes. At charge ratios above unity, both free carrier and DNA complexes are present, and GAGs might interact with either species. We carried out confocal microscopy experiments to address this problem.

Experiments with rhodamine-labeled DNA and unlabeled carriers showed that DNA was located mainly near the plasma membrane, around the nucleus, and also inside the nucleus in few cells (Fig. 3, A-D). The amount of labeled DNA in the nuclei was dependent on the carrier, being most prevalent with PEI (Fig. 3C), which coincides to the GFP expression results. Labeled DNA was usually found distributed in the cells within discrete vesicles, which are probably endosomes and/or lysosomes. This is in line with earlier reports suggesting that intracellular factors like endosomal escape, diffusion in cytoplasm, and nuclear uptake limit gene delivery and transgene expression in the cells (14, 26, 27).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Intracellular distribution of rhodamine-DNA (A-D), fluorescein-hyaluronic acid (E), and fluorescein-heparan sulfate (F). RAA SMC cells were exposed to lipoplexes (A, DOTAP; B, DOTAP/DOPE) or polyplexes (C, PEI; D, PLL) with rhodamine-DNA. Free cationic carriers at amounts similar with the carrier-DNA complexes at a charge ratio of ±4 were complexed with fluorescein-hyaluronic acid (E) or fluorescein-heparan sulfate (F). In panels E and F, the carrier was DOTAP/DOPE. RAA SMC cells were exposed to the complexes for 5 h, and confocal microscopy images were taken from living cells 24 h after complex removal. N indicates the nucleus.

Studies with fluorescein-labeled GAGs and unlabeled carriers showed that the cellular uptake of GAGs was clearly increased by the carriers, e.g. DOTAP/DOPE (Fig. 3, E and F). In the absence of carriers the cellular uptake of GAGs was below detection limit (data not shown). GAGs could be found in vesicles outside the nucleus, and the nuclear uptake of labeled GAGs was insignificant regardless of the carrier.

Double labeling studies were also done with rhodamine-labeled DNA and fluorescein-labeled GAGs (Fig. 4). These experiments demonstrated that both of the GAGs were localized not only with free carrier (green spots) but also with positively charged complexes (yellow spots).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Effects of hyaluronic acid and heparan sulfate on intracellular distribution of complexes. 3-Fold charge excess of fluorescein-labeled hyaluronic acid (A-D) or heparan sulfate (E-H) was added to lipoplexes (A and E, DOTAP/DOPE; B and F, DOTAP) or polyplexes (C and G, PEI; D and H, PLL) at charge ratio ± 4 with rhodamine-labeled DNA. Images were taken 24 h post-transfection from living cells by confocal microscopy. Red fluorescence represents labeled DNA without GAG, and green fluorescence represents labeled GAG without DNA. GAG associated with DNA is seen as yellow. N indicates the nucleus.

Fig. 4 also shows how differently carrier-DNA complexes behave after contacting GAGs. In the case of PEI polyplexes, mainly green fluorescence could be detected inside the cells (Fig. 4, C and G). Hyaluronic acid and heparan sulfate were taken into the cells primarily with free PEI, and heparan sulfate probably displaced DNA from the carrier (Fig. 4G). Both GAGs were internalized into the cells with PLL polyplexes, and heparan sulfate in particular was strongly associated with DNA (yellow color; Fig. 4, D and H). The association of both GAGs with DOTAP/DOPE lipoplexes was similar (Fig. 4, A and E). In the case of DOTAP lipoplexes, the colocalization with DNA was weaker with heparan sulfate (Fig. 4F) than with hyaluronic acid (Fig. 4B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Negatively charged extracellular glycosaminoglycans are one of the major biological barriers for gene transfer mediated by both cationic lipids and polymers (15, 17, 19). The objective of this study was to understand why GAGs, especially heparan sulfate, inhibit the transfection by lipoplexes and polyplexes. Previously we showed (17) that DNA is only rarely released prematurely from the complexes by the strong electrostatic attraction between the cationic carrier and anionic heparan sulfate, as in the case of PEI. Since transfection was inhibited in many other cases as well (17), other mechanisms for inhibition of gene transfer must exist. Interactions with GAGs might alter the composition of the complexes and thereby affect cellular uptake or intracellular trafficking of the complexes. Our present results, however, show that GAGs do not have a significant effect on cellular uptake of various complexes. Thus, the inhibition of gene transfer is, in most cases, caused by the changes in the intracellular distribution of the lipoplexes and polyplexes.

In this study, we show that the influence of GAGs on the complexes depends on the properties of both the GAG and the carrier. Sulfated GAGs have stronger interactions with carrier-DNA complexes than hyaluronic acid because of the higher negative charge density of sulfated GAGs compared with hyaluronic acid. Our results show that hyaluronic acid can bind to the surface of the positively charged complexes but does not interact with negatively charged complexes. Heparan sulfate, on the other hand, can interact with DOTAP and PEI complexes even if the surface of the complexes is negatively charged. This indicates that weakly charged hyaluronic acid cannot compete with DNA for cationic binding sites of carrier, whereas heparan sulfate can.

Previously we have shown that based on the effects of GAGs on gene transfer, the delivery systems can be divided into the following three groups: (i) carriers with endosomal buffering capacity, (ii) liposomes with fusogenic DOPE, and (iii) others (17). The first group includes the systems that are most sensitive to the effects of GAGs, whereas liposomes with fusogenic lipid (DOPE) are the most resistant.

PEI polyplexes, with endosomal buffering capacity, are the most sensitive, whereas non-buffering PLL polyplexes are more resistant to interactions with GAGs. Heparan sulfate is known to bind PEI and release DNA from the complexes (17). Therefore, both the cellular uptake and gene expression are blocked by heparan sulfate (Table I). Hyaluronic acid, on the other hand, decreased the cellular uptake of PEI-containing polyplexes, which partially explains the decreased gene expression. On the basis of our earlier studies (17), PLL forms very stable complexes with DNA probably because of the high charge density and flexibility of PLL molecules. In fact, the stability of PLL polyplexes, as demonstrated by their resistance to charge excess of heparin sulfate, may be the reason for the low transfection efficiency by PLL polyplexes (Table I). DNA may not be released from these polyplexes in the cells either. The different behaviors of PEI and PLL is probably because PLL contains only primary amino groups, whereas PEI contains primary, secondary, and tertiary amino groups (28). Also the different molecular structures of the linear PLL versus the branched PEI may influence the electrostatic complexation by these carriers (28).

Our results show that lipoplexes with DOTAP are more sensitive to interactions with GAGs, especially heparan sulfate, than lipoplexes with DOTAP/DOPE. Recently Xu et al. (29) showed that DOTAP and DOTAP/DOPE lipoplexes have similar colloidal properties, but the lipoplexes differed morphologically from each other. DOTAP forms with DNA multilamellar structures in which DNA is entrapped between the lamellae (29-31). DOTAP/DOPE, on the other hand, forms many extensively elongated tubular hexagonal phase lipid structures around the DNA chains (29, 31, 32). It is possible that multilamellar structures of DOTAP allow heparan sulfate to interact and disturb these lipoplexes, whereas the tubular structures of DOTAP/DOPE lipoplexes might be more stable against heparan sulfate.

Recently it has been shown that cationic lipoplexes are endocytosed into the cells after binding to the negatively charged heparan sulfate proteoglycans located in the cell surface (15, 16). The surface charge of the complexes is crucial for the cell uptake of the complexes because negatively charged complexes (charge ratio ± 0.25) were not internalized significantly into the cells in our study, although the size of these complexes was comparable with those at charge ratio ± 4 (29). In our experiments, we used a 3-fold negative charge excess of extracellular GAGs compared with carrier. This amount of GAGs should cover the complexes and change the surface charge of the complexes negative. Interestingly, in most cases the extracellular GAGs did not alter the cellular uptake of DNA (Table I). This finding suggests that the cellular uptake route of complexes associated with GAGs may be different than in complexes without GAGs. Although GAGs did not affect the cellular uptake of complexes, they had a major impact on gene expression of complexes. Hyaluronic acid increased GFP expression of the complexes, except in complexes with PEI, whereas heparan sulfate significantly reduced the gene expression in all cases. This suggests that the complexes covered by hyaluronic acid and heparan sulfate may have different route of entry or distribution inside the cells, causing differences in gene expression by the complexes.

Both heparin (33, 34) and hyaluronic acid (35) have been shown to be internalized into smooth muscle cells. Heparins are taken into the smooth muscle cells via receptor-mediated endocytosis and are degraded rapidly in endosomes (33) without further accumulation into the nucleus (34). Exogenous hyaluronic acid is taken efficiently into proliferating smooth muscle cells by endocytosis (35), presumably via CD44 receptors (36). In our studies, however, the cellular uptake of heparan sulfate and hyaluronic acid was much less than the cellular uptake of GAGs with either cationic carrier or cationic complexes. We observed that the cationic carriers increase the cellular uptake of GAGs, which has also been shown by others (19). In contrast to the data of Belting and Petersson (19), we could not see a significant accumulation of GAGs into the nucleus. This difference might be explained by our use of a different cell line. Additionally, Belting's group (19) studied the cellular uptake of secreted macromolecules from conditioned medium that contained large amounts of anionic macromolecules. These will compete against each other for the binding sites of the cationic carrier.

In summary, extracellular GAGs do not significantly affect the cellular uptake of most carrier-DNA complexes, although the surface properties of the complexes are changed. Instead, extracellular GAGs have an influence on intracellular behavior of the complexes, thus significantly affecting the gene expression mediated by the complexes. Furthermore, the extracellular GAGs are internalized into the cells with free carrier, with cationic carrier-DNA complexes, and in some cases with carrier released from DNA. In this study we have demonstrated the influence of GAG interactions on cellular trafficking of lipoplexes and polyplexes. In addition to these factors, internalization of GAGs might have some unknown effects on translation and transcription. It is important to understand the role of molecules that interfere with gene delivery. Such information provides new strategies for the development of safe and efficient delivery vehicles for gene therapy.

	ACKNOWLEDGEMENTS

-Galactosidase-expressing plasmid and pTR5-DC/GFP were generous gifts from Dr. F. C. Szoka, Jr. (San Francisco) and Dr. Dick D. Mosser (Montreal), respectively. We are grateful to Mika Reinisalo for cloning the CMV-GFP-expressing plasmid. We thank Lea Pirskanen and Kaarina Pitkänen for assistance.

	FOOTNOTES

^* The work was supported in part by grants from the Graduate School in Pharmaceutical Research (to M. R.), Technology Development Center of Finland (to A. U.), Academy of Finland (to A. U., P. H., and M. T.), and Finnish Cultural Foundation of Northern Savo (to M. R.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Dept. of Pharmaceutics, University of Kuopio, P. O. Box 1627, FIN-70211 Kuopio, Finland. Tel.: 358-17-162491; Fax. 358-17-162456; E-mail: Marika.Ruponen@uku.fi.

Published, JBC Papers in Press, June 4, 2001, DOI 10.1074/jbc.M011553200

	ABBREVIATIONS

The abbreviations used are: GAG, glycosaminoglycan; lipoplex, cationic lipid-nucleic acid complex; polyplex, cationic polymer-nucleic acid complex; GFP, green fluorescent protein; PEI, polyethylenimine; PLL, poly-L-lysine; DOTAP, N-(1-(2, 3-dioleoyloxy)propyl)-N,N,N-trimethyl ammonium methylsulfate; DOPE, 1,2-dioleyl-3-phosphatidylethanolamine; CMV, cytomegalovirus immediate-early gene promoter; EMA, ethidium monoazide; RAA SMC, smooth muscle cells from rabbit aortic media; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; MES, 4-morpholineethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. de la Fuente, B. Seijo, and M. J. Alonso Novel Hyaluronic Acid-Chitosan Nanoparticles for Ocular Gene Therapy Invest. Ophthalmol. Vis. Sci., May 1, 2008; 49(5): 2016 - 2024. [Abstract] [Full Text] [PDF]

				L. Peeters, N. N. Sanders, K. Braeckmans, K. Boussery, J. Van de Voorde, S. C. De Smedt, and J. Demeester Vitreous: A Barrier to Nonviral Ocular Gene Therapy Invest. Ophthalmol. Vis. Sci., October 1, 2005; 46(10): 3553 - 3561. [Abstract] [Full Text] [PDF]

				M. Boustta, L. Leclercq, and M. Vert Affinity Chromatography as a Tool to Analyze Polyanion-Polycation Complexes: the Case of Poly(Llysine Citramide)-Poly(L-lysine) Systems Journal of Bioactive and Compatible Polymers, May 1, 2004; 19(3): 155 - 171. [Abstract] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/36/33875    most recent M011553200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ruponen, M. Articles by Urtti, A. Search for Related Content PubMed PubMed Citation Articles by Ruponen, M. Articles by Urtti, A. Social Bookmarking                      What's this?