October 2005
Volume 46, Issue 10
Free
Biochemistry and Molecular Biology  |   October 2005
Vitreous: A Barrier to Nonviral Ocular Gene Therapy
Author Affiliations
  • Liesbeth Peeters
    From the Laboratory of General Biochemistry and Physical Pharmacy and the
  • Niek N. Sanders
    From the Laboratory of General Biochemistry and Physical Pharmacy and the
  • Kevin Braeckmans
    From the Laboratory of General Biochemistry and Physical Pharmacy and the
  • Koen Boussery
    Department of Physiology and Pathophysiology, Ghent University, Ghent, Belgium.
  • Johan Van de Voorde
    Department of Physiology and Pathophysiology, Ghent University, Ghent, Belgium.
  • Stefaan C. De Smedt
    From the Laboratory of General Biochemistry and Physical Pharmacy and the
  • Joseph Demeester
    From the Laboratory of General Biochemistry and Physical Pharmacy and the
Investigative Ophthalmology & Visual Science October 2005, Vol.46, 3553-3561. doi:10.1167/iovs.05-0165
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Liesbeth Peeters, Niek N. Sanders, Kevin Braeckmans, Koen Boussery, Johan Van de Voorde, Stefaan C. De Smedt, Joseph Demeester; Vitreous: A Barrier to Nonviral Ocular Gene Therapy. Invest. Ophthalmol. Vis. Sci. 2005;46(10):3553-3561. doi: 10.1167/iovs.05-0165.

      Download citation file:


      © 2015 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements

purpose. Intravitreal injection of therapeutic DNA, complexed to nonviral carriers such as cationic liposomes, may be promising in the treatment of many severe retinal eye diseases. However, after intravitreal injection, such DNA/cationic liposome complexes—called lipoplexes (LPXs)—which are typically hundreds of nanometers in size, must first diffuse through the vitreous before they can reach the retina. The aim of this study was to elucidate whether vitreous is a barrier for the LPXs and to find strategies to overcome this barrier.

methods. Fluorescent polystyrene nanospheres and LPXs were mixed with vitreous, and their mobility was monitored by fluorescence recovery after photobleaching (FRAP), a microscopy-based technique. The stability of LPXs and naked plasmid DNA in vitreous was studied by gel electrophoresis.

results. We showed that polystyrene nanospheres, in our first experiments used as a model for the LPXs, do not diffuse freely into the vitreous but adhere to fibrillar structures in the vitreous, most likely to collagen fibers. Making the surfaces of the polystyrene nanospheres hydrophilic by attaching hydrophilic polyethylene glycol (PEG) chains at their surfaces circumvented the binding to fibrillar structures in the vitreous. FRAP revealed that “pegylated” polystyrene nanospheres, as long as they are smaller than 500 nm, are indeed mobile in the vitreous. It was further demonstrated that LPXs severely aggregate in vitreous and strongly bind to biopolymers in the vitreous, which immobilizes them completely. However, as observed for the polystyrene nanospheres, coating of the LPXs with PEG averted their aggregation in the vitreous and their binding to fibrillar structures.

conclusions. Modifying the surfaces of LPXs with hydrophilic PEG chains prevents them from aggregating in vitreous. In this way, LPXs are obtained that can freely move in vitreous, an absolute criterion for reaching the retina after intravitreal injection.

During the past decennium, the genetic basis of many severe ocular diseases has been studied in detail, 1 2 3 paving the way for the treatment of these diseases through gene therapy. Target cells for ocular gene therapy are often located in the neuroretina or the retinal pigment epithelium (RPE). For example, in retinitis pigmentosa (RP), the photoreceptors are the major target cells, 4 whereas the RPE cells should be transfected in proliferative vitreoretinopathy. 5  
The most convenient way to deliver therapeutic DNA or small interfering (si) RNA (siRNA) to these cells would be topical application. However, it is well known that high-molecular-weight drugs have difficulty reaching the retina and even the vitreous after topical application, mainly because of limited diffusion through the sclera. 6 Moreover, the retina and the vitreous are inaccessible by systemic administration because the blood-retina barrier is, like the blood-brain barrier, very tight. 7 Therefore, in most ocular gene therapy trials, genes have been administered to the retinal cells through subretinal or intravitreal injection. 8 9 10 11 12 13 Good results have been obtained after subretinal injection of viral vectors containing the DNA of interest in animal models. 13 14 However, subretinal injection is invasive and is not the first choice for treating ocular diseases, which are in general not life threatening. 15 Moreover, the use of viral vectors has some disadvantages. For one thing, it is difficult to guarantee the safety of their use. 15 In addition, viral vectors are difficult to produce on a large scale, and the size of the therapeutic DNA that can be incorporated is limited. 16 17 Therefore, intravitreal injection of nonviral gene complexes, which is less invasive than subretinal injection, may be a clinically acceptable alternative. This technique, however, requires that the nonviral gene complexes must diffuse through the vitreous before they can reach the retina and the RPE cells. 
Vitreous is a greatly hydrated (98% water) transparent biogel consisting mainly of collagen (∼300 μg/mL), hyaluronan (65–400 μg/mL), and proteoglycans containing chondroitin sulfate and heparan sulfate. 18 Besides these components, vitreous contains noncollagenous proteins and serum components and low numbers of hyalocytes. 18 19 The gel-like behavior of vitreous arises from an ordered, 3-dimensional network of collagen fibrils bridged by proteoglycans filaments. Hyaluronan is known to bind the chondroitin sulfate part of proteoglycans and to fill up the interfibrillar spaces. 18 19  
Considering its structure and composition, vitreous may decrease nonviral gene transfer to the retina and the RPE cells in different ways. First, binding of negatively charged glycoaminoglycans (GAGs), such as chondroitin sulfate, heparan sulfate, and hyaluronan, to the (mostly) cationic nonviral gene complexes might release the plasmid (p) DNA from the complexes or change the charge and size of the complexes, thereby affecting their cellular uptake or intracellular behavior. 20 21 22 Second, vitreous may act as a molecular sieve that limits the transport of the injected nonviral gene complexes. 23 Finally, gene complexes may stick to, for example, collagen fibrils in the vitreous. 
Aside from the recent work of Pitkänen et al., 23 no information about the influence of vitreous on nonviral gene delivery to the retina is available. Although Pitkänen et al. 23 showed for the first time in vitro that nonviral gene delivery to RPE cells is substantially limited by the vitreous, the exact mechanism(s) by which this occurs is still not completely known. Knowledge about how vitreous limits nonviral gene transfer may help in the development of new, more promising nonviral carriers for DNA and siRNA delivery to the retina. Therefore, the aim of this work was to elucidate how vitreous impedes nonviral gene transfer. In a first approach, the mobility of FITC-dextran and fluorescent polystyrene nanospheres, as models for the more complicated nonviral gene complexes, was studied in vitreous using fluorescence recovery after photobleaching. In additional experiments, the mobility of polyethylene glycol (PEG)–coated polystyrene nanospheres, lipoplexes (LPXs), and PEG-coated LPXs in vitreous was also determined. Finally, we studied whether naked pDNA in vitreous degrades and whether vitreous releases pDNA from LPXs. 
Materials and Methods
Vitreous
Vitreous gel came from bovine eyes obtained from the local abattoir. Bovine eyes were enucleated within half an hour after the animals were slaughtered and were transported to the laboratory in ice-cold Krebs-Ringer bicarbonate solution. Fresh bovine eyes were cleaned of extraocular tissue and opened with a scalpel. Subsequently, the vitreous was carefully collected without affecting the retina. 
FITC-Dextran and Fluorescent Nanospheres
FITC-dextran with a molecular weight of 2 × 106 (FD2000), 4.64 × 105 (FD464), and 1.67 × 105 (FD167) g/mol were obtained from Sigma-Aldrich (Bornem, Belgium). Fluorescent (yellow-green) polystyrene nanospheres (NS) of different sizes, bearing carboxyl groups on their surfaces, were purchased from Molecular Probes (Eugene, OR). The average hydrodynamic diameter of the FITC-dextran and the nanospheres (NS) was determined by dynamic light scattering (DLS; Autosizer 4700; Malvern, Worcestershire, UK). For this purpose the FITC-dextran and nanospheres were dissolved/dispersed in HEPES buffer (20 mM HEPES, pH 7.4). For the nanospheres the zeta potential (ζ) was also measured by determining electrophoretic mobility (Malvern ζ-sizer 2000; Malvern). The hydrodynamic diameter and zeta potential (mean ± SD) of the FITC-dextran and nanospheres are summarized in Table 1 . Nanospheres with diameter of 37 nm, 120 nm, 182 nm, and 540 nm will be referred to as NS37, NS120, NS182, and NS540, respectively. 
PEG Coating of the Fluorescent Polystyrene Nanospheres
Fluorescent polystyrene nanospheres were coated with increasing amounts of a block-copolymer of PEG and polypropylene oxide (Pluronic F-127; Sigma-Aldrich), using the following procedure: 150 μL sonicated (for 10 minutes) polystyrene nanospheres containing 7 × 1014 NS37, 2 × 1013 NS120, and 6 × 1012 NS540 nanospheres/mL, respectively, were mixed with 850 μL distilled water and 2 mL solution (Pluronic F-127; Sigma-Aldrich) in distilled water. The concentrations of solutions ranged from 0 to 10 mg/mL, leading to final concentrations in the nanosphere dispersions between 0 and 6.7 mg/mL. After vortexing, the nanospheres were incubated for 1 hour at room temperature. Subsequently, 500 μL nanosphere suspension was transferred to a centrifugal filter device (MWCO 100,000) (Microcon YM-100; Millipore, Bedford, MA) and centrifuged for 12 minutes at 14,000g. After centrifugation, the concentrate with the coated nanospheres was collected by placing the sample reservoir upside down in a new vial and spinning it again for 3 minutes at 1000g. The volume of the collected nanospheres was adjusted to 500 μL with HEPES buffer, and the size and the zeta potential of the coated nanospheres was determined as described (see Table 1 ). 
Preparation and Purification of Plasmid DNA
The pDNA used in this study consisted of 5803 base pairs and contained as a reporter gene secretory alkaline phosphatase (SEAP) under the control of a simian virus 40 promoter. The pDNA was amplified in Escherichia coli and purified as previously described. 24 The pDNA concentration was set at 1.0 mg/mL in HEPES buffer assuming that the absorption of 50 μg DNA/mL at 260 nm equals 1. The pDNA showed a high purity because the 260:280 nm absorption ratio was between 1.8 and 2.0. 
Preparation of Cationic Liposomes
The phospholipids DOTAP (N-(1-(2,3-dioleoyloxy)propyl)-N,N,N-trimethylammonium chloride, DOPE (dioleoylfosfatidylethanolamine), DSPE-PEG (distearoyl phosphatidylethanolamine polyethylene glycol) and FITC-DOPE (fluorescein isothiocyanate dioleoylfosfatidylethanolamine) were purchased from Avanti Polar Lipids (Alabaster, AL). 
Cationic liposomes containing DOTAP and DOPE in a 1:1 molar ratio with 0.1 mol% FITC-DOPE and 0 to 16.7 mol% DSPE-PEG were prepared as previously described. 24 The hydrodynamic diameter and the zeta potential of the cationic liposomes were measured as described for the fluorescent nanospheres. The average hydrodynamic diameters of the DOTAP/DOPE liposomes were 134 ± 2 nm, 126 ± 1 nm, 131 ± 1 nm, 136 ± 1 nm, 134 ± 1 nm, and 98 ± 1 nm (mean ± SD) for liposomes containing 0 mol%, 1.9 mol%, 3.8 mol%, 5.6 mol%, 9.1 mol%, and 16.7 mol% DSPE-PEG, respectively. The zeta potential of the DOTAP/DOPE liposomes was 50 ± 2 mV, 31 ± 3 mV, 28 ± 1 mV, 18 ± 2 mV, 15 ± 1 mV, and 14 ± 2 mV for liposomes with 0 to 16.7 mol% DSPE-PEG, respectively. 
Preparation of Lipoplexes
Lipoplexes (with different percentages of DSPE-PEG) were prepared at a ± charge ratio of 4, as described previously, 24 with the ± charge ratio defined as the ratio of the number of the positive charges (originating from DOTAP) to the number of the negative charges (originating from the pDNA). Briefly, pDNA was first diluted (0.41 mg/mL) and subsequently added to an equal volume of (pegylated) cationic liposomes (5 mM DOTAP). Immediately after the addition of pDNA to the liposomes, HEPES buffer was added until the final concentration of pDNA was 0.126 mg/mL. This mixture was then vortexed and incubated at room temperature for at least 30 minutes. Average hydrodynamic diameters and zeta potentials of the LPXs were measured and are reported in Table 1
Agarose Gel Electrophoresis Experiments
To study the stability of naked pDNA in vitreous, 50 μL pDNA solution containing 2 μg pDNA was incubated for 3 hours at 37°C with 50 μL diluted vitreous. This diluted vitreous was a fivefold dilution of vitreous with HEPES buffer. After incubation, 5 μL loading buffer (50% [wt/vol] sucrose in HEPES buffer) was added to 50 μL pDNA/vitreous mixtures. Subsequently, 25 μL resultant mix was loaded on a 1% (wt/vol) agarose gel prepared in electrophoresis buffer (10.8 g/L Tris base, 5.5 g/L boric acid, and 0.85 g/L ethylenediaminetetraacetic acid). The gel was run at a voltage of 100 V for 1 hour. After this time span, the gel was incubated into a bath of ethidium bromide (0.5 μg/mL water). After 30 minutes, the gel was illuminated with ultraviolet light and photographed. 
To know whether vitreous releases pDNA from the cationic (pegylated) LPXs, 35 μL LPXs was incubated with 65 μL fivefold-diluted vitreous at 37°C for 3 hours. After incubation, 5 μL loading buffer was added to 50 μL LPX/vitreous mixture. Subsequently, 25 μL resultant mix was loaded on a 1% (wt/vol) agarose gel prepared in electrophoresis buffer. Electrophoresis, staining, and visualization procedures were as described. 
Diffusion Experiments in Vitreous by Photobleaching
Fluorescence recovery after photobleaching (FRAP) experiments were performed to determine the mobility of FITC-dextran, fluorescent polystyrene nanospheres, and fluorescent LPXs in vitreous. FRAP experiments were performed using an inverted confocal scanning laser microscope (CSLM; model MRV 1024 UV; Bio-Rad, Hemel Hempstead, UK) equipped with a 10× lens (CFI Plan Apochromat; Nikon, Badhoevedorp, The Netherlands). 
Basically, as illustrated and explained in Figure 1 , in a FRAP experiment, the fluorescent molecules or fluorescent particles in a small area (30 μm) of the (vitreous) sample are irreversibly photobleached by an intense laser beam focused in the sample by the objective lens of the CSLM. After this short “bleaching step,” the fluorescence in the area where the bleaching occurred is measured in time at low laser power. Diffusion of fluorescent molecules (surrounding the bleached zone) into the bleached area leads to a recovery of fluorescence in the photobleached area. The diffusion coefficient (D) and the fraction of mobile molecules (k) can be derived from the rate and amount of recovery of fluorescence in the bleached zone by fitting the data points to appropriate equations. 25 The settings used in the FRAP experiments and the way D and k were extracted from the fluorescence recovery curve have been described previously by our group. 26  
For FRAP experiments on FITC-dextran, fluorescent polystyrene nanospheres, and fluorescent LPXs in HEPES buffer, the solutions/dispersions were applied between a microscope slide and a coverslip. A 120-μm–thick double adhesive spacer (Secure-Seal Spacer; Molecular Probes, Leiden, The Netherlands) was placed between the slide and the coverslip. For FRAP on FITC-dextran, fluorescent polystyrene nanospheres, and fluorescent LPXs in vitreous, approximately 400 μL vitreous gel (from the central zone of the vitreous sample) was carefully removed with a small scalpel and immediately transferred into a well of an 8-well plate with a glass bottom suitable for fluorescence microscopy (Nalge Nunc International, Naperville, IL). Subsequently, 40 μL FITC-dextran solution, fluorescent polystyrene nanospheres, fluorescent liposomes, or fluorescent LPXs was injected into the middle of the vitreous sample. Samples were gently stirred, approximately 10 times, by hand with a plastic rod, and the fluorescent molecules/nanoparticles were allowed to spread throughout the sample for 30 to 60 minutes at room temperature before taking FRAP measurements. In addition, before photolysis experiments were performed, the distribution of the fluorescent molecules/particles in the vitreous samples was visualized with the use of a CSLM. FRAP measurements were taken in regions of the vitreous “far” from the regions where the fluorescent molecules/particles were injected into the vitreous. The values for D and k were the mean of 3 independent measurements. For each measurement, another vitreous sample was used. 
The following concentrations were used in the FRAP experiments: 2 mg/mL for FD2000, 1.5 mg/mL for FD464, 4 mg/mL for FD167, 4 × 1013 nanospheres/mL for NS37, 3 × 1012 nanospheres/mL for NS120, 2 × 1011 nanospheres/mL for NS182, and 7 × 109 nanospheres/mL for NS540. Concentrations of LPXs in the FRAP experiments was set at 136 μM DOTAP. These concentrations of the different types of fluorescent molecules belonged to the concentration range wherein a linear relation existed between fluorescence and concentration of fluorescent molecules/particles. Only under these conditions can accurate FRAP measurements be performed. 26  
Results
Mobility of FITC-Dextran in Vitreous
Before determining the diffusion coefficient of the FITC-dextran in vitreous, the spatial distribution of FITC-dextran in the vitreous was visualized by fluorescence microscopy (CSLM). Homogeneous spreading of FD2000 (see Fig. 3 ) and the other FITC-dextrans (data not shown) throughout the vitreous was observed, suggesting that FITC-dextran does not drastically bind to the biopolymers in the vitreous and that it can move freely in the vitreous network. FRAP measurements confirmed this. Indeed, as summarized in Figure 2A , the mobile fraction (k) in vitreous for all FITC-dextran was 1.0, indicating that all the FITC-dextrans were mobile in the vitreous. Consequently, we calculated the ratio of the diffusion coefficient in vitreum (D vit) to the diffusion coefficient in HEPES buffer (D HB). This ratio (D vit/D HB), which theoretically ranges between 0 and 1, indicates to which extent the biopolymer network in the vitreous decreases the mobility of the fluorescent molecules/particles by a viscous drag or by a steric hindrance. Figure 2Areveals that the FITC-dextrans in vitreous are not drastically hindered because their diffusion in vitreous was slowed down by only a factor of 1.2 to 1.5 compared with their diffusion in HEPES buffer. Figure 2Afurther clearly shows that the larger the FITC-dextran, the lower D vit/D HB,, which should be explained by stronger steric retardation by the biopolymer network in the vitreous. 
Mobility of Polystyrene Nanopheres in Vitreous
In contrast to the mixing of FITC-dextran with vitreous, mixing fluorescent polystyrene nanospheres with vitreous did not result in homogeneous spreading of the nanospheres in the vitreous; rather, the fluorescent nanospheres seemed to stick to large fibril-like structures in the vitreous (Fig. 3A) . We also performed FRAP experiments on these samples, but the fluorescence did not recover at all in the bleached area (k = 0), confirming that the polystyrene nanospheres, regardless of their size, were completely immobilized in vitreous. 
Mobility of Coated Polystyrene Nanospheres in Vitreous
It is clear from Figure 3and photolysis experiments that the polystyrene nanospheres strongly bind to the biopolymers in the vitreous. Given that both the polystyrene nanospheres and the biopolymers in the vitreum are negatively charged, we speculate that the nature of this binding is not electrostatic but rather hydrophobic. In addition, because collagen mainly consists of the hydrophobic amino acids glycine, proline, and hydroxyproline, we hypothesize that the fibrils to which the polystyrene nanospheres bind are collagen fibrils. To overcome the (hydrophobic) binding of the polystyrene nanospheres to the biopolymers in the vitreous, we proposed to attach hydrophilic PEG chains to the surfaces of the nanospheres by coating them with a block-copolymer of PEG and polypropylene oxide (Pluronic F-127; Sigma-Aldrich). Different concentrations of coating were tested: at a concentration of 6.7 mg/mL, the nanospheres had reached their minimal zeta potential, which, depending on the nanosphere size, ranged between −8 mV and −17 mV (Table 1) . This concentration (6.7 mg/mL) was, therefore, used in all our experiments for coating the nanospheres. We also observed that coating the nanospheres with this concentration caused a moderate increase in the size of the nanospheres (Table 1) , as expected. 
After mixing the coated nanospheres with vitreous, homogeneous spreading was observed for all nanospheres except for the largest (575 nm; data not shown). Figure 1shows the typical outcome of a FRAP experiment on coated nanospheres in vitreous. Nanoparticles seem to be mobile in vitreous, as the fluorescence completely recovers (k values did not significantly differ from 1.0), clearly different from the behavior of noncoated polystyrene nanospheres, which were completely immobile in vitreous. Results of the FRAP measurements on PEG-coated polystyrene nanospheres are depicted in Figure 2B . Compared with their diffusion in buffer, the NS37-, NS120-, and NS182-coated nanospheres diffused, respectively, 1.4, 2.1, and 3.4 times slower in vitreous. As observed for FITC-dextran (Fig. 2A) , the larger the coated nanospheres, the lower D vit/D HB, indicating that the (steric) hindrance of the diffusion by the vitreous network is stronger for the larger coated nanospheres than for the smaller ones. 
Stability of Lipoplexes in Vitreum
After the preparation of the fluorescent LPXs, their size and zeta potential were measured. Table 1shows that the higher the amount of pegylated lipids (DSPE-PEG) in LPXs, the smaller they are and the lower their zeta potential. 
Before performing FRAP experiments on LPXs in vitreous, we first checked whether the LPXs—complexes between negatively charged pDNA and cationic lipids—remain intact in vitreous. Indeed, dissembling the fluorescent LPXs into liposomes and free pDNA would lead to an incorrect interpretation of the FRAP results. Therefore, the dissociation of LPXs with increasing mol% DSPE-PEG in vitreous and HEPES buffer (control) was determined through agarose gel electrophoresis. None of the LPXs released pDNA after 3 hours of incubation with HEPES buffer (data not shown) or vitreous (Fig. 4) . Moreover, all the fluorescent liposomes and pDNA clearly colocalized in the slots of the agarose gel, indicating that no free fluorescently labeled lipids were present. To further ensure that the absence of released pDNA on the gel in Figure 4was not caused by degradation of the released pDNA by nucleases in the vitreous (this would result in small DNA fragments that would easily migrate through the agarose gel and possibly run out of the gel), naked pDNA was incubated with HEPES buffer or vitreous for 3 hours at 37°C. No degradation of pDNA was seen in HEPES buffer. Figure 5shows that even though the supercoiled form of pDNA gradually changed to the circular form in vitreous as a function of incubation time, after 3 hours of incubation with vitreous free pDNA did not disappear from the agarose gel. Results of the agarose gel experiments indicated that LPXs in vitreous remain stable and do not disassemble. 
Mobility of Lipoplexes in Vitreous
Fluorescent LPXs containing an increasing amount of pegylated lipids (DSPE-PEG) were mixed with vitreous and visualized with the use of a CSLM before FRAP experiments. LPXs without DSPE-PEG clearly aggregated in vitreous as large fluorescent clusters were observed (Fig. 6A) . Incorporation of small amounts of DSPE-PEG (1.9% and 3.8%) in the LPXs partially protected them from aggregation in the vitreous (Figs. 6B 6C) . However, the lowly pegylated LPXs still showed the tendency to bind to the fibrillar network in the vitreous. Further increasing the degree of pegylation of the LPXs (to 5.6% and 9.1% DSPE-PEG) completely prevented aggregation in the vitreous, but binding of the LPXs to the biopolymers in the vitreous still occurred (Figs. 6D 6E) . We discovered that further increasing the amount of DSPE-PEG (to 16.7 mol%) prevented aggregation of the LPXs in the vitreous and binding of LPXs to the fibrils in the vitreous. Indeed, homogeneous spreading of nonaggregated LPXs was observed when they contained 16.7 mol% DSPE-PEG (Fig. 6F) . FRAP measurements on LPXs in vitreous could only be performed on the latter LPXs. FRAP experiments on these 16.7 mol% DSPE-PEG containing LPXs revealed that they were indeed mobile in vitreous: D vit/D HB equaled 0.53 ± 0.02 and was similar to the D vit/D HB value obtained for the coated NS132 nanospheres, which were of sizes comparable to those of the LPXs containing 16.7 mol% DSPE-PEG. The k value of these highly pegylated LPXs did not significantly differ from 1, confirming that these LPXs were not immobilized by the vitreous. In addition to FRAP measurements on the pegylated LPXs in vitreous, we also performed measurements on pegylated liposomes that differed in the amount of DSPE-PEG. Similar observations were made as for the pegylated LPXs (data not shown). 
Discussion
The delivery of therapeutic DNA or siRNA to the retina may be a promising treatment of many severe retinal eye diseases, such as RP and proliferative vitreoretinopathy. 4 5 Subretinal injection of viral gene carriers has led to a correction of the genetic defect in animal models. 13 14 Unfortunately, viral gene carriers have many disadvantages, 15 16 17 and repeated subretinal injection is not without any risk. Therefore, intravitreal injection of nonviral gene complexes may form a safer alternative for treating severe eye diseases. However, taking into account the size (20–500 nm) and the cationic surfaces of currently studied nonviral carriers, vitreous may decrease their efficiency in various ways. First, as proposed by Urtti et al., 21 22 the negatively charged GAGs present in the vitreous might bind to the gene complexes, which can subsequently lead to dissociation of the gene complexes or alteration of their cellular uptake and intracellular processing. Second, the mobility of the gene complexes in the vitreous can be restricted in two ways: they might become stuck to the biopolymers in the vitreous or be sterically hindered by the 3-dimensional biopolymer network in the vitreous. The aim of this study was to contribute to a better understanding of the vitreous barrier for nonviral gene complexes with the hope of obtaining insight for improving nonviral ocular gene therapy. 
We studied the diffusion of FITC-dextran and fluorescent polystyrene nanospheres in vitreous using FRAP. FITC-dextrans up to 64 nm were able to diffuse through the vitreous, almost as quickly as through water (Fig. 2A) , indicating that the interfibrillar spaces in the vitreous are large enough to allow more or less free passage of FITC-dextran. Similar observations have been reported in studies dealing with the mobility of FITC-dextran and proteins in colon and cervical mucus. 27 28 29 30  
Mixing of polystyrene nanospheres (37–540 nm) with vitreous showed that these polystyrene nanospheres stuck to, most likely, collagen fibers within the vitreous (Fig. 3) . This sticking probably resulted from hydrophobic interactions. Olmsted et al. 27 reported similar observations in cervical mucus. After mixing polystyrene nanospheres within the cervical mucus, the nanospheres stuck tightly to the mucins, which collapsed the mucus gel into thick cables of aggregated mucin strands. 
It is well known that coating nanoparticles with PEG can prevent aspecific adsorption of proteins to the surfaces of the nanoparticles, which increases their circulation time after intravenous administration. 31 32 33 In this study, pegylation of the nanospheres prevented them from sticking to the collagen fibers and rendered them mobile in the vitreous. As observed for FITC-dextran, the larger the coated nanospheres, the stronger the retardation by the vitreous because of a stronger steric hindrance by the vitreous network. Based on the observation that coated nanospheres measuring 575 nm in diameter did not spread homogeneously throughout the vitreous after mixing but remained localized near the injection site, we can conclude that the maximum particle size that allows diffusion through the vitreous is situated between 220 nm and 575 nm. 
In our study both the nonpegylated and the pegylated LPXs incubated in vitreous did not dissociate (Fig. 4) , which is in disagreement with previously reported data. Indeed, Ruponen et al. 20 showed that solutions of chondroitin sulfate B/C, heparan sulfate, and hyaluronic acid induce some DNA release from nonpegylated DOTAP/DOPE LPXs. This discrepancy may be explained by differences in interactions between LPXs and GAGs, respectively, in a solution (as in the study of Ruponen et al. 20 ) and in real vitreous gel (as in our experiments). 
The aggregation of (nonpegylated) cationic LPXs, which we clearly observed in vitreous, may explain the results of Pitkänen et al. 23 They found that the cellular uptake of cationic DOTAP LPXs was much more affected by a thin layer of vitreous than the cellular uptake of neutral FITC-dextran and was of a size comparable to that of the LPXs (before aggregation). The aggregation may be explained by the fact that negatively charged polymers such as GAGs are known to bind to cationic LPXs: this neutralizes the positive zeta potential of the LPXs, favoring their aggregation. 34 35 36  
Moderate pegylation of cationic LPXs seems to avoid aggregation of the LPXs in the vitreous. Moreover, when high mol% DSPE-PEG (16.7%) is incorporated into the LPXs, the interactions with the fibrillar structures in the vitreous can be circumvented and the LPXs rendered mobile (Fig. 6F)
The observation that nanoparticles of comparable size, namely DOTAP/DOPE LPXs with 16.7 mol% DSPE-PEG (±106 nm) and NS120 (±132 nm) and FD2000 (±64 nm) and the coated NS37 spheres (±61 nm), are equally slowed by the vitreous indicates that nanoparticles of the same size, and with a neutral and hydrophilic surface, are equally slowed by the vitreous network. 
Incubation of pDNA in the vitreous showed that a part of the supercoiled pDNA converted into open circular pDNA. This is in agreement with the results of Pitkänen et al., 23 who also found a conversion of the supercoiled pDNA to circular pDNA. This observation indicates moderate DNase or topoisomerase activity in the vitreous gel. 
As outlined in the introduction, this study is part of a research effort aimed at understanding the barrier in the vitreous toward gene complexes injected intravitreally. A major question is whether intact (nonaggregated, nondisassembled) gene complexes can reach the retina after intravitreal injection. The data on mobility, aggregation, and stability of LPXs in vitreous presented in this study will certainly contribute to answering this question. However, we would like to emphasize some important points. First, the in vivo transport of drugs in vitreous does not occur only by diffusion but may also occur by convection in the vitreous. 37 This should be taken into account when determining whether nanoparticles will be able to reach the retina after intravitreal injection. For low-molecular-weight drugs convection plays a minor role, but for high-molecular-weight (slower diffusing) drugs, convection becomes more important. 38 In addition, the contribution of convection to drug transport in the vitreous is more important in larger animal species (such as humans) in which convection accounts for roughly 30% of the total drug transport. 37 Second, in this study we used bovine vitreous. The structural components of human and bovine vitreous are the same, but the exact organization may differ. Nevertheless, it is highly probable that the human vitreous forms a diffusive barrier for nanoparticles similar to that of bovine gel. 23 Third, we must consider that manipulation of the vitreous can lead to alteration of the network structure, which may influence the macromolecular transport. Indeed, manipulation of the vitreous may lead to the aggregation of collagen and, hence, to the redistribution of the macromolecules. On the other hand, human vitreous gel undergoes progressive liquefaction with age, 18 39 and the conditions for permeation also vary in vivo. Fourth, in this research we focused on transport in the central parts of the vitreous samples. One must be aware that the cortical vitreous zone, which contains a more densely arranged network of collagen, and the inner limiting lamina may be important additional barriers to the diffusion of gene complexes into the retina. 
Table 1.
 
Diameter and Zeta Potential (ζ) of the FITC-Dextrans, Polystyrene Nanospheres, and Lipoplexes
Table 1.
 
Diameter and Zeta Potential (ζ) of the FITC-Dextrans, Polystyrene Nanospheres, and Lipoplexes
Nanoparticle Size (nm) ζ (mV)
FD167 18 ± 1
FD464 30 ± 2
FD2000 64 ± 7
NS37 37 ± 2 −50 ± 9
NS120 120 ± 2 −42 ± 2
NS182 182 ± 9 −47 ± 4
NS540 540 ± 10 −54 ± 1
Coated NS37 61 ± 6 −17 ± 10
Coated NS120 132 ± 2 −14 ± 1
Coated NS182 220 ± 1 −8 ± 1
Coated NS540 575 ± 10 −15 ± 1
LPX 0% PEG 233 ± 12 39 ± 1
LPX 1.9% PEG 252 ± 3 15 ± 1
LPX 3.8% PEG 200 ± 1 15 ± 1
LPX 5.6% PEG 252 ± 3 16 ± 1
LPX 9.1% PEG 128 ± 2 8 ± 4
LPX 16.7% PEG 106 ± 5 0 ± 1
Figure 1.
 
(A) FRAP experiment performed on vitreous containing PEG-coated polystyrene nanospheres (NS37). The first image, the prebleach image, shows the sample before bleaching. Consequently, in a small area (typically 30 μm in diameter) of the vitreous, the polystyrene nanospheres are bleached by an intense laser beam (image at t = 0). Next, the laser intensity is switched back to a low intensity to record the fluorescence recovery in the bleached area over multiple images at regular time intervals. Details on the experimental procedure are described by Braeckmans et al.26(B) The curve shows how the fluorescence in the bleached area recovers in time (F indicates the fluorescence in the bleached zone at time t, and F0 indicates the fluorescence in this zone before bleaching). The fluorescence recovery is caused by the diffusion of bleached NS37 nanospheres out of the bleached area and the diffusion of fluorescent NS37 nanospheres from the surrounding unbleached zone into the bleached area. Experimental data are indicated by dots on the curve. A best fit of the model (solid line) finally yields the diffusion coefficient (D) and the mobile fraction (k).
Figure 1.
 
(A) FRAP experiment performed on vitreous containing PEG-coated polystyrene nanospheres (NS37). The first image, the prebleach image, shows the sample before bleaching. Consequently, in a small area (typically 30 μm in diameter) of the vitreous, the polystyrene nanospheres are bleached by an intense laser beam (image at t = 0). Next, the laser intensity is switched back to a low intensity to record the fluorescence recovery in the bleached area over multiple images at regular time intervals. Details on the experimental procedure are described by Braeckmans et al.26(B) The curve shows how the fluorescence in the bleached area recovers in time (F indicates the fluorescence in the bleached zone at time t, and F0 indicates the fluorescence in this zone before bleaching). The fluorescence recovery is caused by the diffusion of bleached NS37 nanospheres out of the bleached area and the diffusion of fluorescent NS37 nanospheres from the surrounding unbleached zone into the bleached area. Experimental data are indicated by dots on the curve. A best fit of the model (solid line) finally yields the diffusion coefficient (D) and the mobile fraction (k).
Figure 2.
 
(A) Dvit/DHB and k values for FITC-dextran in vitreous. (B) Dvit/DHB and k values for the coated polystyrene nanospheres in vitreous. Each value is the mean of 3 independent measurements. For each measurement another vitreous sample was used.
Figure 2.
 
(A) Dvit/DHB and k values for FITC-dextran in vitreous. (B) Dvit/DHB and k values for the coated polystyrene nanospheres in vitreous. Each value is the mean of 3 independent measurements. For each measurement another vitreous sample was used.
Figure 3.
 
Fluorescence microscopy image (CSLM) of FD2000 FITC-dextran (A), NS37 (B), NS120 (C), and NS540 (D) polystyrene nanospheres in vitreous.
Figure 3.
 
Fluorescence microscopy image (CSLM) of FD2000 FITC-dextran (A), NS37 (B), NS120 (C), and NS540 (D) polystyrene nanospheres in vitreous.
Figure 4.
 
Agarose gel after electrophoresis on fluorescent lipoplexes (with increasing amounts of DSPE-PEG) mixed with vitreous for 3 hours. (A) Agarose gel before staining with ethidium bromide showing the localization of the fluorescent lipids on UV illumination. (B) Agarose gel after staining with ethidium bromide showing the localization of pDNA. The amount of pDNA, either free or complexed, loaded in each slot equaled 0.91 μg.
Figure 4.
 
Agarose gel after electrophoresis on fluorescent lipoplexes (with increasing amounts of DSPE-PEG) mixed with vitreous for 3 hours. (A) Agarose gel before staining with ethidium bromide showing the localization of the fluorescent lipids on UV illumination. (B) Agarose gel after staining with ethidium bromide showing the localization of pDNA. The amount of pDNA, either free or complexed, loaded in each slot equaled 0.91 μg.
Figure 5.
 
Agarose gel showing free pDNA after 0, 1, 2, and 3 hours of incubation in HEPES buffer or vitreous at 37°C. The amount of pDNA in each slot was 0.45 μg.
Figure 5.
 
Agarose gel showing free pDNA after 0, 1, 2, and 3 hours of incubation in HEPES buffer or vitreous at 37°C. The amount of pDNA in each slot was 0.45 μg.
Figure 6.
 
Fluorescence microscopy images (CSLM) of lipoplexes (with increasing mol% DSPE-PEG) in vitreous.
Figure 6.
 
Fluorescence microscopy images (CSLM) of lipoplexes (with increasing mol% DSPE-PEG) in vitreous.
 
The authors thank Bart Leroy and Christophe De Laey for their advice. 
McLaughlinME, SandbergMA, BersonEL, DryjaTP. Recessive mutations in the gene encoding the beta-subunit of rod phosphodiesterase in patients with retinitis pigmentosa. Nat Genet. 1993;4:130–134. [CrossRef] [PubMed]
FarrarGJ, KennaP, JordanSA, et al. A three-base-pair deletion in the peripherin-RDS gene in one form of retinitis pigmentosa. Nature. 1991;354:478–480. [CrossRef] [PubMed]
KajiwaraK, HahnLB, MukaiS, TravisGH, BersonEL, DryjaTP. Mutations in the human retinal degeneration slow gene in autosomal dominant retinitis pigmentosa. Nature. 1991;354:480–483. [CrossRef] [PubMed]
DelyferMN, LeveillardT, Mohand-SaidS, HicksD, PicaudS, SahelJA. Inherited retinal degenerations: therapeutic prospects. Biol Cell. 2004;96:261–269. [CrossRef] [PubMed]
YasukawaT, OguraY, TabataY, KimuraH, WiedemannP, HondaY. Drug delivery systems for vitreoretinal diseases. Prog Retinal Eye Res. 2004;23:253–281. [CrossRef]
WeinrebRN. Enhancement of scleral macromolecular permeability with prostaglandins. Trans Am Ophthalmol Soc. 2001;99:319–343. [PubMed]
DuvvuriS, MajumdarS, MitraAK. Drug delivery to the retina: challenges and opportunities. Expert Opin Biol Ther. 2003;3:45–56. [CrossRef] [PubMed]
MiyoshiH, TakahashiM, GageFH, VermaIM. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA. 1997;94:10319–10323. [CrossRef] [PubMed]
TakahashiM, MiyoshiK, VermaIM, GageFH. Rescue from photoreceptor degeneration in the rd mouse by human immunodeficiency virus vector-mediated gene transfer. J Virol. 1999;73:7812–7816. [PubMed]
SarraGM, StephensC, de-AlwisM, et al. Gene replacement therapy in the retinal degeneration slow (rds) mouse: the effect on retinal degeneration following partial transduction of the retina. Hum Mol Genet. 2001;10:2353–2361. [CrossRef] [PubMed]
HarveyAR, KamphuisW, EggersR, et al. Intravitreal injection of adeno-associated viral vectors results in the transduction of different types of retinal neurons in neonatal and adult rats: a comparison with lentiviral vectors. Mol Cell Neurosci. 2002;21:141–157. [CrossRef] [PubMed]
SarraGM, StephensC, SchlichtenbredeFC, et al. Kinetics of transgene expression in mouse retina following sub-retinal injection of recombinant adeno-associated virus. Vision Res. 1901;42:541–549.
BennettJ, MaguireAM, CideciyanAV, et al. Stable transgene expression in rod photoreceptors after recombinant adeno-associated virus-mediated gene transfer to monkey retina. Proc Natl Acad Sci USA. 1999;96:9920–9925. [CrossRef] [PubMed]
AclandGM, AguirreGD, RayJ, et al. Gene therapy restores vision in a canine model of childhood blindness. Nat Genet. 2001;28:92–95. [PubMed]
SakamotoT, IkedaY, YonemitsuY. Gene targeting to the retina. Adv Drug Delivery Rev. 2001;52:93–102. [CrossRef]
De SmedtSC, DemeesterJ, HenninkWE. Cationic polymer based gene delivery systems. Pharm Res. 2000;17:113–126. [CrossRef] [PubMed]
GaoX, HuangL. Cationic liposome-mediated gene transfer. Gene Ther. 1995;2:710–722. [PubMed]
BishopP. The biochemical structure of mammalian vitreous. Eye. 1996;10:664–670. [CrossRef] [PubMed]
ScottJE. The chemical morphology of the vitreous. Eye. 1992;6:553–555. [CrossRef] [PubMed]
RuponenM, YlaHS, UrttiA. Interactions of polymeric and liposomal gene delivery systems with extracellular glycosaminoglycans: physicochemical and transfection studies. Biochim Biophys Acta Biomembr. 1999;1415:331–341. [CrossRef]
RuponenM, HonkakoskiP, RonkkoS, PelkonenJ, TammiM, UrttiA. Extracellular and intracellular barriers in non-viral gene delivery. J Control Release. 2003;93:213–217. [CrossRef] [PubMed]
RuponenM, RonkkoS, HonkakoskiP, PelkonenJ, TammiM, UrttiA. Extracellular glycosaminoglycans modify cellular trafficking of lipoplexes and polyplexes. J Biol Chem. 2001;276:33875–33880. [CrossRef] [PubMed]
PitkanenL, RuponenM, NieminenJ, UrttiA. Vitreous is a barrier in nonviral gene transfer by cationic lipids and polymers. Pharm Res. 2003;20:576–583. [CrossRef] [PubMed]
SandersNN, Van RompaeyE, De SmedtSC, DemeesterJ. Structural alterations of gene complexes by cystic fibrosis sputum. Am J Respir Crit Care Med. 2001;164:486–493. [CrossRef] [PubMed]
MeyvisTKL, De-SmedtSC, Van-OostveldtP, DemeesterJ. Fluorescence recovery after photobleaching: a versatile tool for mobility and interaction measurements in pharmaceutical research. Pharm Res. 1999;16:1153–1162. [CrossRef] [PubMed]
BraeckmansK, PeetersL, SandersNN, De SmedtSC, DemeesterJ. Three-dimensional fluorescence recovery after photobleaching with the confocal scanning laser microscope. Biophys J. 2003;85:2240–2252. [CrossRef] [PubMed]
OlmstedSS, PadgettJL, YudinAI, WhaleyKJ, MoenchTR, ConeRA. Diffusion of macromolecules and virus-like particles in human cervical mucus. Biophys J. 2001;81:1930–1937. [CrossRef] [PubMed]
SandersNN, De-SmedtSC, DemeesterJ. The physical properties of biogels and their permeability for macromolecular drugs and colloidal drug carriers. J Pharm Sci. 2000;89:835–849. [CrossRef] [PubMed]
SaltzmanWM, RadomskyML, WhaleyKJ, ConeRA. Antibody diffusion in human cervical mucus. Biophys J. 1994;66:508–515. [CrossRef] [PubMed]
SzentkutiL. Light microscopical observations on luminally administered dyes, dextran, nanospheres and microspheres in the pre-epithelial mucus gel layer of the rat distal colon. J Control Rel. 1997;46:233–242. [CrossRef]
BlumeG, CevcG. Liposomes for the sustained drug release in vivo. Biochim Biophys Acta. 1990;1029:91–97. [CrossRef] [PubMed]
SeniorJ, DelgadoC, FisherD, TilcockC, GregoriadisG. Influence of surface hydrophilicity of liposomes on their interaction with plasma protein and clearance from the circulation: studies with poly(ethylene glycol)-coated vesicles. Biochim Biophys Acta. 1991;1062:77–82. [CrossRef] [PubMed]
DunnSE, BrindleyA, DavisSS, DaviesMC, IllumL. Polystyrene-poly (ethylene glycol) (PS-PEG2000) particles as model systems for site specific drug delivery, 2: the effect of PEG surface density on the in vitro cell interaction and in vivo biodistribution. Pharm Res. 1994;11:1016–1022. [CrossRef] [PubMed]
MislickKA, BaldeschwielerJD. Evidence for the role of proteoglycans in cation-mediated gene transfer. Proc Natl Acad Sci USA. 1996;93:12349–12354. [CrossRef] [PubMed]
XuY, SzokaF-CJ. Mechanism of DNA release from cationic liposome/DNA complexes used in cell transfection. Biochemistry. 1996;35:5616–5623. [CrossRef] [PubMed]
ZelphatiO, SzokaF-CJ. Mechanism of oligonucleotide release from cationic liposomes. Proc Natl Acad Sci USA. 1996;93:11493–11498. [CrossRef] [PubMed]
XuJ, HeysJJ, BarocasVH, RandolphTW. Permeability and diffusion in vitreous humor: implications for drug delivery. Pharm Res. 2000;17:664–669. [PubMed]
StayMS, XuJ, RandolphTW, BarocasVH. Computer simulation of convective and diffusive transport of controlled-release drugs in the vitreous humor. Pharm Res. 2003;20:96–102. [CrossRef] [PubMed]
BalazsEA, DenlingerJL. Aging and Human Visual Function. 2005;AR Liss New York.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×