Surface-Tethered DNA Complexes for Enhanced Gene Delivery

Enabling customization of non-viral gene delivery systems for individual cell types by surface-induced mineralization. Bingbing Sun , Kenny K. Tran , ...
0 downloads 0 Views 467KB Size
Download PDF
Bioconjugate Chem. 2002, 13, 621−629

621

Surface-Tethered DNA Complexes for Enhanced Gene Delivery Tatiana Segura† and Lonnie D. Shea†,‡,* Departments of Chemical Engineering and Biomedical Engineering, Northwestern University, 2145 Sheridan Road E156, Evanston, Illinois 60208-3120. Received November 2, 2001; Revised Manuscript Received February 6, 2002

Overcoming the barriers to efficient gene transfer is a fundamental goal of biotechnology. A versatile approach to enhance the delivery of nonviral DNA involves complexation with cationic polymers, which can be designed to overcome the barriers to effective gene transfer. More recently, DNA release from a polymer substrate or scaffold has been shown to enhance gene transfer, likely by increasing DNA concentrations in the cell microenvironment. We propose a novel approach that combines these two strategies in which cationic polymer/DNA complexes are tethered to a substrate that supports cell adhesion. The cationic polymers package the DNA for efficient internalization and the surface tethering functions to maintain elevated concentrations in the cell microenvironment for cells adhered to the substrate. The cationic polymer polylysine (degree of polymerization equal to 19 or 150) was modified with biotin groups, which was confirmed by mass spectrometry and biochemical analysis. Complex formation of DNA with biotinylated-polylysine, or mixtures of biotinylated and nonbiotinylated polylysines, was confirmed by gel electrophoresis. Plasmid DNA encoding for the reporter gene β-galactosidase was complexed with different mixtures of biotinylated and nonbiotinylated polylysine and incubated on neutravidin (nonglycosylated avidin)-coated surfaces. DNA surface densities ranging from 0.1 to 4.3 µg/cm2 were observed and found to be a function of the number of biotin groups, the molecular weight of the polylysine, and the amount of DNA. HEK293T or NIH/3T3 cells were then seeded onto the DNA-modified surfaces, and transfection was quantified at 48 and 96 h. Transfection by the DNA surfaces was observed with both cell lines, and expression levels up to 100 fold greater than bulk delivery of the complexes was obtained. Transfection was found to be a function of the surface DNA quantities and the number of tethers on the complex. Transfected cells were observed only in the region in which DNA complexes were tethered, suggesting that the location of transfected cells can be specifically controlled. Surface tethering of DNA represents a promising approach to enhancing gene transfer and spatially controlling gene delivery, which may have applications to a multitude of fields ranging from tissue engineering to functional genomics.

INTRODUCTION

Developing systems capable of controlled and efficient gene transfer is a fundamental goal of biotechnology, with applications ranging from basic science to clinical medicine. In the area of basic science, increasing or decreasing the expression level of a gene within a cell has the power to reveal or confirm the roles of specific components of signaling pathways and can lead to a mechanistic understanding of cell behavior, disease pathogenesis, and drug action (1, 2). Assays based on cellular responses to altered gene expression allow gene function to be examined within a biological context that is more representative of the physiological situation (3, 4). The growing collection of gene sequences and cloned cDNAs has led to the development of microarrays of cell clusters expressing gene products as a high-throughput approach to screen for a phenotype of interest (5). Alternatively, therapeutic strategies using gene delivery are under development for inherited genetic disorders, cancer, and tissue regeneration (6, 7). The successful application of gene transfer for basic science and clinical medicine requires the ability to manipulate the expres* Address correspondence to Lonnie D. Shea, Department of Chemical Engineering, Northwestern University, 2145 Sheridan Rd./E156, Evanston, IL 60208-3120. Phone: 847-491-7043. Fax: 847-491-3728. E-mail: [email protected]. † Department of Chemical Engineering. ‡ Department of Biomedical Engineering.

sion of target genes in the desired cell population. Numerous techniques are under development to overcome the barriers to gene transfer, which includes processes such as cellular internalization, endosomal escape, and nuclear trafficking. Cationic polymers provide a versatile approach for gene transfer, as the polymers can be designed or modified to overcome the barriers to gene transfer (8, 9). Complexation with cationic polymers functions to condense DNA, to produce a complex with a less-negative surface charge, to enhance cellular internalization of DNA, and to protect the DNA from degradation. Although many types of cationic polymers have been explored (10-13), polymers based on poly(L-lysine) (PLL) (14-16), poly(ethylenimine) (PEI) (17-19), and poly(amidoamine) (PAMAM) (20, 21) are among the most commonly utilized. Numerous studies have examined the relationship between the physical properties of the cationic polymer (e.g., molecular weight) with the ability to condense DNA and to efficiently transfer the DNA to cells (17, 22). Complexes are formed at nitrogen:phosphate ratios ranging from 0.5:1 to 20:1 and typically produces complexes with sizes ranging from 50 to 150 nm (23-25). An advantage of the cationic polymers is that functional domains can be covalently incorporated to design a synthetic vector capable of overcoming the barriers to gene transfer. Fusogenic peptides and other functional groups have been attached to enhance endosomal escape of the DNA complexes (26,

10.1021/bc015575f CCC: $22.00 © 2002 American Chemical Society Published on Web 03/15/2002

622 Bioconjugate Chem., Vol. 13, No. 3, 2002

27). Alternatively, ligands that bind to specific cell surface receptors have been attached to increase binding of the DNA complexes to the cell surface and can enhance cellular internalization (28-31). Recently, the polymeric systems that were developed to deliver biologically active proteins have been adapted to deliver nonviral DNA. Polymeric scaffolds have been fabricated from a variety of materials, both natural (e.g., collagen) and synthetic (e.g., poly(lactide-co-glycolide)), which function as a support for cell adhesion and migration (7). These scaffolds act to increase the local concentration of DNA within the cellular microenvironment either by providing a sustained release of DNA (32) or by maintaining the DNA locally (33, 34). Enhanced gene transfer by these polymer scaffolds in vivo likely results from an increased concentration at the cell surface, which has been shown to enhance gene transfer in vitro (35). We present a novel approach toward enhancing DNA delivery that combines DNA complexation with polymeric delivery. DNA is complexed with a cationic polymer and subsequently tethered to a substrate the supports cell adhesion. The cationic polymer polylysine is covalently modified with biotin groups, complexed with DNA, and the resulting complexes are bound to a neutravidincoated surface. The quantity and location of DNA on the surface can be controlled and cells cultured on the surface are transfected by the surface-associated DNA. EXPERIMENTAL PROCEDURES

Materials. Plasmid DNA encoding for β-galactosidase (pNGVL1-β-gal) was purified from bacteria culture using Quiagen (Santa Clara, CA) reagents and stored in TrisEDTA buffer solution (10 mM Tris, 1 mM EDTA, pH ) 7.4). Fluorescein tagged β-galactosidase vector (Fl-β-gal) was purchased from Gene Therapy Systems (San Diego, CA). Two polylysine (PLL) peptides were used for DNA complexation: Cys-Trp-Lys19 (K19, BioPeptide, San Diego, CA) and Lys150 (K150, average molecular weight of 20000, Sigma, St Louis, MO). Avidin and biotin reagents for peptide modification and surface tethering were purchased from Pierce (Rockford, IL). All other reagents were obtained from Fisher Scientific (Fairlawn, NJ) unless otherwise noted. Synthesis of Biotinylated Polylysine. The peptide K19 was modified with a biotin group through the terminal cysteine residue by reaction of the sulfhydryl group with the iodoacetyl group of the biotinylation reagent, EZ-link-PEO-iodoacetyl-biotin. The peptide K19 (10 mg) was dissolved in 850 µL of buffer (50 mM Tris, 5 mM EDTA, pH ) 8.3) that was previously bubbled with nitrogen gas. The EZ-link-PEO-iodoacetyl-biotin (3.8 mg) was also dissolved in 150 µL of buffer (0.1 M sodium phosphate, 5 mM EDTA, pH ) 6.0). The biotin solution was added dropwise to the peptide solution, mixed gently, covered with aluminum foil, and incubated for 90 min. The starting peptide solution and the reaction mixture were analyzed by HPLC to determine if the reaction had gone to completion. The starting peptide solution and the reaction mixture were resolved by injecting 20 µg through a C18 RP-HPLC column eluted with water (0.1% trifluoroacetic acid (TFA)) and an acetonitrile gradient (0.1% TFA, 0 to 95% over 50 min at 60 °C) while detecting the absorbance at 260 nm. For purification, sephadex (G15) was equilibrated in deionized water for 30 min prior to packing in a glass column (2 cm diameter × 12 cm height). The reaction mixture was passed though the column using deionized water. Twenty fractions were collected, and the presence of the tryptophan side chain

Segura and Shea

was examined by measuring the absorbance at 260 nm (Beckman Instruments Inc., Fullerton, CA). The fractions with the greatest absorbance at 260 nm were lyophilized (Labconco Corp., Kansas City, MO) and analyzed by mass spectrometry. The purified biotinylated peptide (K19-B) was stored as a powder at -20 °C. The peptide K150 was biotinylated using succinimide ester (NHS)/amine chemistry. K150 (10 mg) was dissolved in 1 mL of phosphate buffered saline (PBS, pH ) 7) and EZ-link-Sulfo-NHS-LC-biotin (2.8 mg) was added directly to the solution, mixed gentlyb and incubated for 2 h at 4 °C. The reaction mixture was purified using dialysis cassettes immersed in deionized water. The dialyzed product was further purified using a monomeric avidin column to separate the biotinylated components from nonbiotinylated species. The biotinylated product was eluted with 10 mL of a 10 mM biotin solution and dialyzed to remove the unconjugated biotin. The purified biotinylated peptide (K150-B) was then lyophilized and stored as a powder at -20 °C. The degree of biotinylation of K150-B was determined by quantifying the mole ratio of biotin to K150 using 2-(4′-hydroxyazobenzene)benzoic acid (HABA). The absorbance at 500 nm of a HABA/ avidin solution in PBS was recorded before and after the addition of the K150-B and used to calculate the molar ratio of biotin to K150. Complex Formation and Surface Tethering. The ability of the biotinylated and nonbiotinylated polylysine (K150, K150-B, K19, K19-B) to condense DNA was assessed by gel electrophoresis. Biotinylated and nonbiotinylated peptides were mixed and added in a stepwise manner (1 µL of 1 mg/mL) to a DNA solution (200 µL of 20 µg/mL). After each addition step, the solution was vortexed for 4 s and incubated for 10 min and a sample (10 µL) removed. Upon complete addition of peptide, trypsin was added to digest the polylysine. Gel electrophoresis was performed to assess the extent of complex formation for the samples and the trypsin-digested DNA solution. DNA/PLL complexes were incubated on surfaces to specifically tether the complexes through the biotinneutravidin binding. DNA (90 µL of 44.4 µg/mL) was complexed at a charge ratio (() of 5.5:1 with the four peptides (K150, K150-B, K19, or K19-B) individually or with mixtures of biotinylated and nonbiotinylated peptides. The number of tethers on each complex is varied by mixing biotinylated and nonbiotinylated PLLs prior to complexation with DNA. Complexes were incubated after mixing for 30 min at room temperature and then allowed to bind to prewashed neutravidin coated surfaces for 2 h. The unbound complexes were then removed from the wells and washed with tris-buffered saline (TBS). The surface quantities of DNA were determined by incubating with trypsin (100 µL) at 37 °C for 2 h to degrade the PLL and release the DNA into solution. The quantity of DNA was measured with a fluorometer (Turner Designs TD360) using the Hoechst dye. Tethered DNA complexes were visualized by fluorescence microscopy using the Fltagged β-galactosidase vector before and after washing with TBS. Cell Culture and Transfection. Transfection of cells on the DNA-modified surfaces was examined using a β-galactosidase plasmid and two cell lines (HEK293T and NIH/3T3). Polyethyleneimine (PEI, 22 kDa, MBI-Fermentas, Hanover, MD, 0.5 µL of 10 µM) was added to DNA-modified surfaces and incubated for 5 min. Cells were then plated and cultured on the surfaces for 48-96 h and then lysed for assay of the β-galactosidase enzyme activities (Promega, Madison, WI) and protein levels (BioRad, Hercules, CA). Alternatively, cells were

Surface-Tethered DNA Complexes

Bioconjugate Chem., Vol. 13, No. 3, 2002 623

stained with X-gal to determine the location of transfected cells. HEK293T and NIH/3T3 were cultured at 37 °C and 5% CO2 in Dubelcco’s modified eagle medium (DMEM, Life Technologies, Gaithersburg, MD) supplemented with 10% heat inactivated FBS and 1% penicillin/ streptomycin. Control experiments to characterize the effectiveness of surface-mediated delivery were performed by using bulk delivery of DNA complexed with the nonbiotinylated K150 or K19. For bulk delivery of K150/DNA complexes, the quantity of DNA delivered was determined based on the amount of surface associated DNA obtained when DNA was complexed only with K150-B. For bulk delivery of K19/DNA, the quantity of DNA delivered was determined based on the amount of surface-associated DNA obtained when DNA was complexed with a mixture of K150-B/K19, which had a predominance of K19 over K150-B. Upon formation of the K150/DNA and K19/DNA complexes, PEI was added at the same molar ratio of DNA:PEI that was used for the substrate-mediated delivery. For the bulk delivery experiments, HEK293T and NIH/3T3 cells were plated at 1 day prior to transfection and cultured in complete media. RESULTS

PLL Biotinylation and DNA Complexation. Polylysine was covalently modified with biotin residues either through an N-terminal cysteine side chain (K19) or through an amine (K150). HPLC analysis of the initial peptide (K19 ) and the peptide reaction mixture (K19-B) demonstrated that the reaction proceeded to completion (Figure 1a), as evidenced by the increase in molecular weight in the reaction mixture and the absence of the lower molecular weight K19 peak. Mass spectrometry (Figure 1b) suggests that the approach for modifying the peptide K19 results in the attachment of a single biotin group. A good correspondence was found between the theoretically expected molecular weight for K19 modified with a single biotin (3158 Da) and the experimentally obtained molecular weight (3157.66 ( 0.77 Da). Alternatively, the chemistry employed for modification of K150 allows for multiple biotin residues to be attached per PLL. The K150-B synthesis resulted in a 3.1:1 molar ratio of biotin to K150 by using a 10:1 molar ratio of biotinylation reagent to K150 in the reaction mixture. Biotinylated and nonbiotinylated peptides were subsequently analyzed for their ability to complex with DNA using gel electrophoresis. The nonbiotinylated peptides (K19, K150) completely eliminate the electrophoretic mobility at charge ratios of 3.1 and 1.2, respectively (Figure 2a,c). However, the presence of the biotin group on the peptide affected the charge ratio at which the mobility is eliminated, which is 4.6 and 4.9 for K19-B and K150-B, respectively (Figure 2b,d). Mixtures of K150-B with K19 were also examined by gel electrophoresis for DNA complexation at a charge ratio of 5.5:1. The inhibition of mobility was observed for all combinations of the two peptides (Figure 2e-g). The results of Figure 2 illustrate that the biotinylated peptides, nonbiotinylated peptides, and mixtures of biotinylated and nonbiotinylatd peptides are capable of electrostatically neutralizing plasmid DNA and eliminating its electrophoretic mobility. Surface Tethering of DNA/PLL Complexes. DNA complexed with biotinylated PLL was examined to determine if the biotin groups on the PLL were available for tethering to a surface using the affinity of biotin for neutravidin (nonglycosylated avidin). Surface-associated DNA was visualized for DNA complexed with K150-B.

Figure 1. HPLC and mass spectrometry confirm the formation of biotyinylated K19. (a) HPLC of the K19 starting material 1 and the reaction mixture 2. The shift in molecular weight between 1 and 2 and absence of the low molecular weight peak in 2 confirmed that the reaction went to completion. (b) Mass spectrometry of the purified reaction mixture.

Fluorescence images taken after the initial incubation on the surface and before the wash demonstrates the presence of complexes across the entire surface (Figure 3a). Thorough washing of the surfaces resulted in a reduction of the quantity of surface-associated DNA (Figure 3b). All subsequent studies used surfaces that were thoroughly washed to ensure binding specificity of the complexes. The quantity of surface associated DNA was subsequently measured as a function of the PLL peptides and the number of biotin groups per complex. Nonbiotinylated peptides used for DNA condensation resulted in low surface densities ( 0.05) and 1705 biotin groups and its bulk control (p > 0.05) were found. The distribution of transfected cells throughout the cell population also differed between the delivery mechanisms. For bulk delivery, transfected cells were seen throughout the cell population (not shown); however, surface-mediated delivery resulted in cells that were transfected in clusters (Figure 6). Additionally, the location of transfected cells on the surface was consistent with the location of surfaceassociated DNA seen with the fluorescently tagged plasmid (Figure 3b).

Surface-Tethered DNA Complexes

Figure 4. Density of surface-associated DNA. Complexes were formed by mixing DNA with (a) K150-B or (b) K19-B with either K19 or K150 at a charge ratio of 5.5. The moles of biotin per mole DNA (i.e., average number of biotin groups per complex), which varied from 28 to 1705 for K150-B and 217 to 4342 for K19-B, are listed beneath each bar. The symbol * indicates statistical significance (p < 0.05) in the surface densities relative to complexes without biotin.

The expression levels of protein at 96 h by cells cultured on DNA-modified surfaces increased relative to the that observed at 48 h and, for all biotinylated DNA complexes, was greater than that obtained by bulk delivery. Maximal expression levels by HEK293T cells was obtained for the complexes containing 28 biotin groups and was statistically significant from all other conditions tested (p < 0.05) (Figure 7a). The expression level decreased as the average number of biotin groups on the complex increased (p < 0.05). The complexes formed with K150-B (1705 biotin groups) had the lowest transfection level of the surface associated delivery; however, the expression level was significantly greater than the bulk control (p < 0.01). The expression levels for the NIH/3T3 cells were less dependent on the number of biotin groups, yet the decreasing expression levels for increasing numbers of biotin groups was again observed. For all conditions tested with surface associated delivery of DNA complexes tethered to the surface with biotin groups, an increased level of transfection was observed relative to the delivery of nonbiotinylated DNA complexes and the bulk delivery of DNA complexes (p < 0.01).

Bioconjugate Chem., Vol. 13, No. 3, 2002 625

Figure 5. Expression levels by surface mediated transfection after 48 h of incubation. The expression levels for (a) HEK293T and (b) NIH/3T3 cells were determined using an assay for β-galactosidase activity. Bulk delivery of complexes resembling the limiting conditions for surface-associated DNA, 1705 and 28 biotin groups, respectively, were used as a control. For bulk delivery of K150/DNA complexes, the quantity of DNA delivered was determined based on the amount of surface-associated delivery for K150-B that has 1705 biotin groups. For bulk delivery of K19/DNA complexes, the quantity of DNA delivered was determined based on the amount of surface-associated DNA for K150-B/K19 that has 28 biotin groups, which has a predominance of K19 over K150. The symbol * indicates statistical significance in transfection levels (p < 0.05) relative to both bulk delivery conditions. DISCUSSION

This report presents a novel approach to DNA delivery that combines DNA condensation with polymeric delivery by tethering DNA complexes to a surface that supports cell adhesion in order to enhance gene transfer by increasing the concentration of DNA in the cellular microenvironment. In the first step of the process, cationic peptides are covalently modified with a functional group. A mixture of modified and unmodified cationic peptides are then complexed with nonviral DNA, and the resulting complexes are tethered to a substrate, which can support cell adhesion, through the functional group on the cationic peptide. Surface densities of DNA ranging from 0.1 to 4.3 µg DNA/cm2 were obtained with coupling efficiencies up to 35%. The density of surfaceassociated DNA was determined to be a function of the

626 Bioconjugate Chem., Vol. 13, No. 3, 2002

Segura and Shea

Figure 6. Photomicrograph of cells transfected by surfaceassociated delivery. (a) HEK293T and (b) NIH/3T3 were stained with X-gal after incubation on surfaces for 48 h.

number of tethers, the molecular weight of the cationic peptide, and the initial amount of DNA. Cells cultured on these surfaces express the transgene, and the level of transgene expression is dependent upon the exposure time of cells to the DNA and the surface tethering. At 96 h of culture, surface-mediated delivery resulted in significantly greater expression of the transgene than that obtained by bulk delivery of similar DNA complexes. Interestingly, the location of transfected cells was consistent with the location to which DNA was tethered. Cationic polymers are a leading class of transfection reagents, which results from their ability to complex DNA, and their molecular diversity enables the physicochemical properties of the DNA complexes to be manipulated. PLL has been shown capable of condensing DNA into toroidal nanostructures with a radius of gyration less than 150 nm (23, 24). However, PLL does not generally produce extensive transfection, and several approaches have been undertaken to enhance its gene transfer efficiency. The internalization of PLL/DNA complexes can be enhanced by the covalent attachment of ligands, such as transferrin (29), folate (36), lectins (37), epidermal growth factor (EGF) (30), and RGD peptides (38). The ligand functions to direct the DNA complex to the cell surface by specifically binding to a receptor, thus increasing the quantity of DNA at the cell microenvironment (30). The approach described in this report utilizes a similar philosophy to enhance the concentration of DNA at the cell surface, with the primary difference being that the “ligand” in our system binds to a substrate, onto which cells can adhere. This strategy of nucleic acid complexes binding to the extra-

Figure 7. Expression levels by surface-mediated transfection after 96 h of incubation. The expression levels for (a) HEK293T and (b) NIH/3T3 were determined using an assay for β-galactosidase activity. The bulk controls were performed as previously described. The symbol * indicates statistical significance in transfection levels (p < 0.05) relative to both bulk delivery conditions.

cellular matrix to enhance gene transfer has been employed by retroviruses. The different domains of fibronectin can separately support the binding of cells (through integrin receptors) and retroviruses, which results in colocalization of the target cell and the virion and may lead to an enhancement of gene transfer (39). Synthetic systems have also been developed that increase cellsurface concentrations of DNA by adsorbing DNA complexes to the surface (35, 40, 41) or by releasing or entrapping DNA within a polymeric scaffold (32, 34). The molecular weight of the cationic polymer impacts the ability of the polymer to complex with DNA and, ultimately, the ability of complexes to bind to the surface. Previous reports with polylysine have identified that the optimal molecular weight of the unmodified polymer represents a balance between effective condensation and cytotoxicity. Relative to low-molecular weight, the highmolecular weight PLL forms tighter and smaller condensates that are more resistant to the effects of salt concentration and sonication (24, 42). However, cytotoxicity has been found to be inversely related to particle size, potentially due to the high molecular weight PLL (22). Our results with the peptides K150 and K19 on DNA condensation and the electrophoretic mobility are con-

Surface-Tethered DNA Complexes

sistent with these reports and suggest that the peptide K150 has a higher affinity for DNA than does K19. The attachment of biotin groups affected the ability of polylysine to complex with DNA, as evidenced by the change in the amount of polylysine required to completely eliminate the electrophoretic mobility of the DNA (Figure 2a,c). For complexes formed with low numbers of biotin groups (28 mol biotin/mol DNA), the mixture of biotinylated, high-molecular weight PLL (K150-B) and nonbiotinylated low molecular weight PLL (K19) proved to be most effective combination for the binding of DNA to the substrate. The increased binding of DNA complexes to the substrate with K150-B as the source of biotin groups was hypothesized to result from the degree of polymerization increasing the probability that the biotin groups are incorporated into the complex and that the incorporated biotins are available for binding. The studies on the elimination of electrophoretic mobility suggest that K150-B and K19 have a similar binding affinity for DNA (Figure 2a,d); thus, mixtures of the two peptides with DNA are not expected to favor either peptide for DNA complexation. The increased binding of the K150-B/K19/ DNA complexes to the substrate relative to either K150/ DNA or K19/DNA suggests that both the biotinylated and nonbiotinylated peptides were incorporated into the complex. The complexes are completely condensed, indicating that K19 is involved with complexation (Figure 2g). Additionally complexes are bound to the substrate at levels greater than control, suggesting that biotinylated K150 is present in the complex. Our results are consistent with previous studies with EGF modified polylysine (30). EGF was attached to a high molecular weight polylysine with the inclusion of a spacer arm, which produced a more nativelike EGF-receptor binding. Conversely, the nonbiotinylated peptide employed had a low degree of polymerization (K19), which was thought to have a reduced affinity for DNA and was thus less likely to displace the biotinylated polylysine. The number of functional groups incorporated into the DNA complex was found to determine the surface densities of DNA and influence the ability of cells to internalize the DNA. The number of tethers within the complex was regulated through combinations of biotinylated and nonbiotinylated peptide. Increases in the number of tethers per complex led to more surface-associated DNA; however, greater transfection was observed with fewer tethers per complex. This observed relationship between the number of functional groups, the surface quantities of DNA, and the transfection may be explained by the strength of adhesion between the complexes and the surface. Insight regarding the binding of DNA complexes to the surfaces can be obtained from the models that have been developed to describe the binding of multivalent ligands to multivalent receptors (43, 44). An increased number of functional groups in the complex would provide more sites for surface-association and thus an increase in the adhesion strength relative to complexes with fewer functional groups. The increased adhesion strength would likely increase the amount of DNA that would remain following the washing procedure. Nevertheless, the increased adhesion strength between the complex and the surface may limit cellular internalization. The DNA must be internalized by the cell and transported to the nucleus for transfection, which may be inhibited if the DNA complex is tightly associated with the surface. The ratio of lysine/nucleotide that has been used for DNA complexation can range from approximately 0.5 to

Bioconjugate Chem., Vol. 13, No. 3, 2002 627

6.0. An increase in this ratio produces a more complete condensation of the DNA (e.g., Figure 2) for improved protection against degradation. Additionally, nonspecific adsorption of the complexes to the cell surface has been shown to increase as the lysine/nucleotide ratio increased (30). Ratios between 0.9 and 1.3 are expected to have minimal effects due to charge interaction and are more effective for studies with specific cell targeting through attached ligands. The charge ratio used for these studies (5.5) takes advantage of the nonspecific interaction between the DNA complex and the cell. In addition to complexing the DNA, the polylysine may act to modulate the cellular interactions with the surface. Polylysine coating is a commonly used surface treatment for in vitro cell culture experiments. An additional advantage of the increased lysine/nucleotide ratio is the increased number of biotin groups that could associate with the complex and tether the complexes to the surface. The increased lysine/nucleotide ratio may also reduce the strength of interaction between the surface bound polylysine and the DNA complex, which could facilitate cellular internalization. An interesting result from these studies is that the expression levels increased significantly between 48 and 96 h for the surface mediated DNA delivery but not for the bulk delivery. This result suggests that the DNA is not internalized only at the initial exposure to the DNA, but increasing amounts may be internalized as the cells are adhered to the surface. These effects may result from either a decrease in the number of tethers with time, which would enable more complexes to be internalized over time, or from an increased stability of the complexes. The number of tethers per complex may decrease due to release of the complex from the surface, such as by dissociation of the DNA complex from the biotinylated PLL that is surface bound, or by degradation of polylysine by cell-secreted proteases. Alternatively, the increased gene expression may arise from an increased stability of the complexes over time. DNA complexed with polylysine is known to aggregate over time in the presence of serum (45, 46), which may decrease the transfection competence of the complex over time. However, complexes tethered to the surface would be prevented from aggregating and would thus retain their transfection competence. The concept of increased stability due to surface tethering has been demonstrated for proteins, where the tethering of transferrin or insulin to poly(methyl methacrylate) resulting in higher levels of activity that either soluble or physically adsorbed proteins (47). Additionally, tethering of EGF to poly(ethyleneoxide)-based hydrogels has also been found to increase its stability and activity when compared to physically absorbed EGF (48). In conclusion, we have described a process to enhance gene transfer through a surface-mediated delivery of DNA complexes. DNA complexes are tethered to surfaces that support cell adhesion through functional groups attached to the cationic polymers. Cells cultured on these surfaces are transfected by the surface-associated DNA and express the transgene. Importantly, transfected cells are localized to the sites at which DNA is tethered, potentially providing a mechanism to spatially regulate gene delivery. This approach may be applied to the development of microarrays of living cells expressing predefined genes for research into functional genomics or to the field of tissue engineering as a means to precisely direct cellular functions within a developing tissue.

628 Bioconjugate Chem., Vol. 13, No. 3, 2002 ACKNOWLEDGMENT

The authors gratefully acknowledge the efforts of Felicia M. Bogdan for technical assistance and Terry L. Sheppard for helpful discussions. Funding for this project was provided by a grant from the NSF (BES-0092701). T.S. was supported by Northwestern University’s NIH Predoctoral Training Grant (5-T32 6M 08449-06). LITERATURE CITED (1) Rohrer, D. K., and Kobilka, B. K. (1998) G protein-coupled receptors: functional and mechanistic insights through altered gene expression. Physiol Rev. 78, 35-52. (2) Stark, G. R., and Gudkov, A. V. (1999) Forward genetics in mammalian cells: functional approaches to gene discovery. Hum. Mol. Genet. 8, 1925-1938. (3) Silverman, L., Campbell, R., and Broach, J. R. (1998) New assay technologies for high-throughput screening. Curr. Opin. Chem. Biol. 2, 397-403. (4) Schena, M., Heller, R. A., Theriault, T. P., Konrad, K., Lachenmeier, E., and Davis, R. W. (1998) Microarrays: biotechnology’s discovery platform for functional genomics. Trends Biotechnol. 16, 301-306. (5) Ziauddin, J., and Sabatini, D. M. (2001) Microarrays of cells expressing defined cDNAs. Nature 411, 107-110. (6) Anderson, W. F. (1998) Human gene therapy. Nature 392, 25-30. (7) Bonadio, J., Goldstein, S. A., and Levy, R. J. (1998) Gene therapy for tissue repair and regeneration. Adv. Drug Del. Rev. 33, 53-69. (8) Segura, T., and Shea, L. D. (2001) Materials for Non-Viral Gene Delivery, in Annual Review of Materials Research (S. I. Stupp, Ed.), pp 25-46, Palo Alto. (9) Brown, M. D., Schatzlein, A. G., and Uchegbu, I. F. (2001) Gene delivery with synthetic (non viral) carriers. Int. J. Pharm. 229, 1-21. (10) Boussif, O., Delair, T., Brua, C., Veron, L., Pavirani, A., and Kolbe, H. V. (1999) Synthesis of polyallylamine derivatives and their use as gene transfer vectors in vitro. Bioconjugate Chem. 10, 877-883. (11) Gonzales, H., Hwang, S. J., and Davis, M. E. (1999) New class of polymers for the delivery of macromolecular therapeutic. Bioconjugate Chem. 10, 1068-1074. (12) Murphy, J. E., Uno, T., Hamer, J. D., Cohen, F. E., Dwarki, V., and Zuckermann, R. N. (1998) A combinatorial approach to the discovery of efficient cationic peptoid reagents for gene delivery. Proc. Natl. Acad. Sci. U.S.A. 95, 1517-1522. (13) van de Wetering, P., Moret, E. E., Schuurmans-Nieuwenbroek, N. M., van Steenbergen, M. J., and Hennink, W. E. (1999) Structure-activity relationships of water-soluble cationic methacrylate/methacrylamide polymers for nonviral gene delivery. Bioconjugate Chem. 10, 589-597. (14) McKenzie, D. L., Kwok, K. Y., and Rice, K. G. (2000) A potent new class of reductively activated peptide gene delivery agents. J. Biol. Chem. 275, 9970-9977. (15) Katayose, S., and Kataoka, K. (1997) Water-soluble polyion complex associates of DNA and poly(ethylene glycol)-poly(Llysine) block copolymer. Bioconjugate Chem. 8, 702-707. (16) Choi, J. S., Lee, E. J., Choi, Y. H., Jeong, Y. J., and Park, J. S. (1999) Poly(ethylene glycol)-block-poly(L-lysine) dendrimer: novel linear polymer/dendrimer block copolymer forming a spherical water-soluble polyionic complex with DNA. Bioconjugate Chem. 10, 62-65. (17) Fischer, D., Bieber, T., Li, Y., Elsasser, H. P., and Kissel, T. (1999) A novel nonviral vector for DNA delivery based on low molecular weight, branched polyethylenimine: effect of molecular weight on transfection efficiency and cytotoxicity. Pharm. Res. 16, 1273-1279. (18) Boussif, O., Lezoualc’h, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., and Behr, J. P. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc. Natl. Acad. Sci. U.S.A. 92, 7297-7301. (19) Blessing, T., Kursa, M., Holzhauser, R., Kircheis, R., and Wagner, E. (2001) Different strategies for formation of

Segura and Shea pegylated EGF-conjugated PEI/DNA complexes for targeted gene delivery. Bioconjugate Chem. 12, 529-537. (20) Zhang, Z. Y., and Smith, B. D. (2000) High-generation polycationic dendrimers are unusually effective at disrupting anionic vesicles: membrane bending model. Bioconjugate Chem. 11, 805-814. (21) Qin, L., Pahud, D. R., Ding, Y., Bielinska, A. U., KukowskaLatallo, J. F., Baker, J. R., Jr., and Bromberg, J. S. (1998) Efficient transfer of genes into murine cardiac grafts by Starburst polyamidoamine dendrimers. Hum. Gene Ther. 9, 553-560. (22) Duguid, J. G., Li, C., Shi, M., Logan, M. J., Alila, H., Rolland, A., Tomlinson, E., Sparrow, J. T., and Smith, L. C. (1998) A physicochemical approach for predicting the effectiveness of peptide-based gene delivery systems for use in plasmid-based gene therapy. Biophys. J. 74, 2802-2814. (23) Golan, R., Pietrasanta, L. I., Hsieh, W., and Hansma, H. G. (1999) DNA toroids: stages in condensation. Biochemistry 38, 14069-14076. (24) Wadhwa, M. S., Collard, W. T., Adami, R. C., McKenzie, D. L., and Rice, K. G. (1997) Peptide-mediated gene delivery: influence of peptide structure on gene expression. Bioconjugate Chem. 8, 81-88. (25) Goula, D., Remy, J. S., Erbacher, P., Wasowicz, M., Levi, G., Abdallah, B., and Demeneix, B. A. (1998) Size, diffusibility and transfection performance of linear PEI/DNA complexes in the mouse central nervous system. Gene Ther. 5, 712-717. (26) Midoux, P., Mendes, C., Legrand, A., Raimond, J., Mayer, R., Monsigny, M., and Roche, A. C. (1993) Specific gene transfer mediated by lactosylated poly-L-lysine into hepatoma cells. Nucleic Acids Res. 21, 871-878. (27) Putnam, D., Gentry, C. A., Pack, D. W., and Langer, R. (2001) Polymer-based gene delivery with low cytotoxicity by a unique balance of side-chain termini. Proc. Natl. Acad. Sci. U.S.A. 98, 1200-1205. (28) Baatz, J. E., Bruno, M. D., Ciraolo, P. J., Glasser, S. W., Stripp, B. R., Smyth, K. L., and Korfhagen, T. R. (1994) Utilization of modified surfactant-associated protein B for delivery of DNA to airway cells in culture. Proc. Natl. Acad. Sci. U.S.A. 91, 2547-2551. (29) Wagner, E., Zenke, M., Cotten, M., Beug, H., and Birnstiel, M. L. (1990) Transferrin-polycation conjugates as carriers for DNA uptake into cells. Proc. Natl. Acad. Sci. U.S.A. 87, 34103414. (30) Schaffer, D. V., and Lauffenburger, D. A. (1998) Optimization of cell surface binding enhances efficiency and specificity of molecular conjugate gene delivery. J. Biol. Chem. 273, 28004-28009. (31) Harbottle, R. P., Cooper, R. G., Hart, S. L., Ladhoff, A., McKay, T., Knight, A. M., Wagner, E., Miller, A. D., and Coutelle, C. (1998) An RGD-oligolysine peptide: a prototype construct for integrin-mediated gene delivery [see comments]. Hum. Gene Ther. 9, 1037-1047. (32) Shea, L. D., Smiley, E., Bonadio, J., and Mooney, D. J. (1999) DNA delivery from polymer matrixes for tissue engineering. Nat. Biotechnol. 17, 551-554. (33) Fang, J., Zhu, Y. Y., Smiley, E., Bonadio, J., Rouleau, J. P., Goldstein, S. A., McCauley, L. K., Davidson, B. L., and Roessler, B. J. (1996) Stimulation of new bone formation by direct transfer of osteogenic plasmid genes. Proc. Natl. Acad. Sci. U.S.A. 93, 5753-5758. (34) Bonadio, J., Smiley, E., Patil, P., and Goldstein, S. (1999) Localized, direct plasmid gene delivery in vivo: prolonged therapy results in reproducible tissue regeneration. Nat. Med. 5, 753-759. (35) Luo, D., and Saltzman, W. M. (2000) Enhancement of transfection by physical concentration of DNA at the cell surface. Nat. Biotechnol. 18, 893-895. (36) Gottschalk, S., Cristiano, R. J., Smith, L. C., and Woo, S. L. (1994) Folate receptor mediated DNA delivery into tumor cells: potosomal disruption results in enhanced gene expression. Gene Ther. 1, 185-191. (37) Stewart, A. J., Pichon, C., Meunier, L., Midoux, P., Monsigny, M., and Roche, A. C. (1996) Enhanced biological activity of antisense oligonucleotides complexed with glycosylated poly-L-lysine. Mol. Pharmacol. 50, 1487-1494.

Surface-Tethered DNA Complexes (38) Harbottle, R. P., Cooper, R. G., Hart, S. L., Ladhoff, A., McKay, T., Knight, A. M., Wagner, E., Miller, A. D., and Coutelle, C. (1998) An RGD-oligolysine peptide: a prototype construct for integrin-mediated gene delivery. Hum. Gene Ther. 9, 1037-1047. (39) Williams, D. A. (1999) Retroviral-fibronectin interactions in transduction of mammalian cells. Ann. N.Y. Acad. Sci. 872, 109-113; discussion 113-104. (40) Bielinska, A. U., Yen, A., Wu, H. L., Zahos, K. M., Sun, R., Weiner, N. D., Baker, J. R., Jr., and Roessler, B. J. (2000) Application of membrane-based dendrimer/DNA complexes for solid-phase transfection in vitro and in vivo. Biomaterials 21, 877-887. (41) Zheng, J., Manuel, W. S., and Hornsby, P. J. (2000) Transfection of cells mediated by biodegradable polymer materials with surface-bound polyethyleneimine. Biotechnol. Prog. 16, 254-257. (42) Adami, R. C., Collard, W. T., Gupta, S. A., Kwok, K. Y., Bonadio, J., and Rice, K. G. (1998) Stability of peptidecondensed plasmid DNA formulations. J. Pharm. Sci. 87, 678-683. (43) Perelson, A. S., Macken, C. A., Grimm, E. A., Roos, L. S., and Bonavida, B. (1984) Mechanism of cell-mediated cyto-

Bioconjugate Chem., Vol. 13, No. 3, 2002 629 toxicity at the single cell level. VIII. Kinetics of lysis of target cells bound by more than one cytotoxic T lymphocyte. J. Immunol. 132, 2190-2198. (44) Lauffenburger, D. A., and Lindermam, J. J. (1993) Receptors Models for binding, trafficking, and signaling, Oxford Unversity Press, New York. (45) Tang, M. X., and Szoka, F. C. (1997) The influence of polymer structure on the interactions of cationic polymers with DNA and morphology of the resulting complexes. Gene Ther. 4, 823-832. (46) Lucas, P., Milroy, D. A., Thomas, B. J., Moss, S. H., and Pouton, C. W. (1999) Pharmaceutical and biological properties of poly(amino acid)/DNA polyplexes. J. Drug Target. 7, 143156. (47) Ito, Y., Liu, S. Q., and Imanishi, Y. (1991) Enhancement of cell growth on growth factor-immobilized polymer film. Biomaterials 12, 449-453. (48) Kuhl, P. R., and Griffith-Cima, L. G. (1996) Tethered epidermal growth factor as a paradigm for growth factorinduced stimulation from the solid phase. Nat. Med. 2, 1022-1027.

BC015575F