Department of Biopharmaceutical Sciences and Pharmaceutical Chemistry, School of Pharmacy, University of California, San Francisco, CA, USA

Correspondence to: F Szoka, Department of Biopharmaceutical Sciences and Pharmaceutical Chemistry, School of Pharmacy, University of California, San Francisco, CA 94143-0446, USA

We introduce an injectable system for the formation of a biodegradable DNA-containing implant that releases DNA over a 2-month period to provide a robust and prolonged gene expression at the site. Sustained delivery of the appropriate plasmid DNA resulted in sustained expression of luciferase, the persistent appearance of secreted alkaline phosphatase in the serum and small blood vessel formation in the vicinity of the implant from the delivery of the development endothelial locus-1 gene. Local expression of development endothelial locus-1 protein promotes the development of blood vessels to meet the metabolic demands of new tissue and is a paradigm for the delivery of other growth factors that act locally to aid tissue regeneration. This delivery system involves simple preparation procedures and can be injected directly into the site, hence should be a useful approach to plasmid-based gene transfer for vaccination and tissue engineering.

Gene Therapy 2002 9, 1230-1237. doi:10.1038/sj.gt.3301786

biomaterials; gene therapy/delivery; glycofurol; injectable implants; medical device; plasmid DNA; tissue engineering

Introduction

In the foreseeable future, DNA will be available as a pharmaceutical for gene replacement, vaccination and tissue engineering.¹ To facilitate this development, biodegradable polymers have been introduced as implantable matrices for the sustained release of genes to augment local gene transfer. These DNA-containing implants have been fabricated from biodegradable polymers in a range of shapes and sizes: from microcapsules and microspheres to rods, films, and three-dimensional matrices.^2,3,4,5 All of these implants are solids that are formed outside the body to precise dimensions, and then inserted into the body for gene delivery. An alternative to ex vivo matrix formation has been proposed for protein delivery,^6,7,8,9 where the biodegradable implant forms in situ upon injection. In this case, the injectable implant system is comprised of a water-insoluble biodegradable polymer dissolved in a pharmaceutically acceptable water-miscible solvent (glycofurol) and contains a suitable biologically active agent. Upon injection the water-miscible solvent diffuses away from the polymer solution and the polymer coagulates or precipitates to form a solid polymeric implant. The biologically active agent is encapsulated within the polymer matrix as it solidifies. After solidification, the therapeutic agent is then released by similar mechanisms as those for the solid pre-formed implants.^7,8

In this study, we developed a monolithic gene delivery system where the implant is formed following injection to the site of need. Plasmids containing either the firefly Photinus pyralis luciferase gene (pCMV-Luc), secreted human placental alkaline phosphatase (pAP), or developmental endothelial locus (pDL-1), were delivered by this polymeric implant. The genes, efficiently transfected cells in mice and the encoded proteins were robustly expressed or mediated a biological response for a prolonged period.

Results

Plasmid DNA incorporation in the polymer device

Recently, an injectable biodegradable drug delivery system was developed,^7,8 that is administered via a standard syringe and needle. Experiments to incorporate and release plasmid DNA from such in situ forming polymer implants were performed to test whether DNA can be encapsulated and if plasmid release could be sustained over a prolonged time period.

Calf thymus DNA or pCMV-Luc plasmid DNA encoding the firefly Photinus pyralis luciferase gene were dissolved in water and mixed with the poly (lactide-co-glycolide) (PLGA) dissolved in glycofurol. A 10% w/w solution of PLGA 75:25 (inherent viscosity (_i) = 0.59 dl/g) in glycofurol containing 500 g plasmid DNA per ml formed a clear viscous solution that could be readily extruded through a 21-G needle. The PLGA dissolved in glycofurol, solidified in situ upon injection of the homogenous polymer-glycofurol solution containing DNA into phosphate buffered saline (PBS). Matrices containing plasmid DNA were created from two different PLGA copolymers that have varying degradation rates.⁷ Plasmid DNA could be readily encapsulated in implants of PLGA 75:25 (_i = 0.24 dl/g, lower molecular weight polymer), and PLGA 75:25 (_i = 0.59 dl/g, higher molecular weight polymer) up to 500 g/ml of the glycofurol solution.

To ascertain the location of the DNA in the solid polymer, we visualized the distribution of DNA stained with ethidium bromide (EtBr) during formation of the device (Figure 1a) using confocal microscopy. DNA fluorescence was higher at the edges of the implant and lower, but more uniform across the core of the implant. During the solidification process, there is a process of convection and diffusion of glycofurol out of the polymer matrix. This mechanism is probably responsible for the accumulation of DNA at the edges of the matrix as the polymer coagulates, as well as for part of the initial fast release before solidification of the matrix. The DNA distribution in the center of the implant appeared to be uniform as indicated by the absence of large aggregates. However, we could not determine if the DNA underwent chain entanglement with the PLGA or if it separated into a DNA-rich phase within the polymer matrix. Interconnecting channels or compartments containing DNA were formed inside the implants, as indicated by z-axis cross-section confocal microscope images Figure 1b. Nevertheless, other channels do not appear to be connected, suggesting that the DNA entrapped within them would be released only after degradation of the polymeric matrix.

In devices stained with EtBr 3 h after formation, there was a ring of fluorescence consisting of aggregates of DNA connected to the surface but the central regions were devoid of fluorescence (data not shown). This indicates DNA in the interior of the device is initially isolated from the exterior solution since the DNA is not accessible to the EtBr. Thus, release at longer periods will be controlled by degradation of the polymer, as well as by diffusion of the DNA through water-filled channels in the device.

Release of DNA from the polymer device

A sustained release of plasmid DNA into an excess of PBS was observed for implants formed from either polymer mixture, with times for total release of plasmid ranging from 10 days (_i = 0.24 dl/g) to more than 60 days (_i =0.59 dl/g) Figure 1c. DNA release was more rapid from the low molecular weight polymer but in both cases, the DNA was released before complete erosion of the polymeric matrix. All subsequent studies utilized the _i = 0.59 dl/g PLGA for plasmid encapsulation and delivery.

The structural integrity of the released plasmid was examined using gel electrophoresis of samples recovered from the PBS after various days of exposure to PBS Figure 1d. The migration pattern upon electrophoresis of unencapsulated pCMV-Luc and released plasmids were identical, indicating that the encapsulation techniques or glycofurol solvent did not degrade or result in linearization of DNA. This is in contrast to other methods which encapsulate DNA,^10,11 where a transformation of the DNA from a supercoiled to an open circular form was reported.

Transfection activity of DNA released from the polymer device

The transfection competence of plasmid released from the matrix was assessed by transfecting CV-1 cells in vitro with the luciferase plasmid collected from the release study, and comparing it to the level of gene expression from plasmid that was not exposed to the encapsulation process (Figure 2). Cells transfection was mediated by PEI as described in the experimental protocol. Gene expressions at two different concentrations of DNA were within ±20% of the value obtained with the standard (unincorporated) plasmid. A linear regression of the transfection activity versus time of release to detect if there was a trend towards DNA inactivation was not significant (P > 0.3).

In vivo gene transfection from the injectable polymeric devices

The ability of the plasmid released from injectable polymeric implants to transfect cells in vivo in mice was assessed using three independent reporter plasmids. The first experiment examined whether delivery of a plasmid encoding for luciferase could transfect the cells in the vicinity of the implant to produce the luciferase enzyme. Eleven and 28 days after injection the tissue samples adjacent to the polymeric implant in a radius of 1 cm were retrieved and analyzed for luciferase activity (Table 1). Both skin overlaying the implant and muscle and connective tissue underlying the implant showed substantial and significant (P < 0.001) luciferase activity. DNA released from the device produced over 1400 and ~4000 pg luciferase/g tissue after 11 and 28 days, respectively (Table 1). The luciferase activity from injection of naked DNA in glycofurol or naked DNA in saline induced less than 100 pg luciferase/g tissue. Furthermore, there was a statistically significant increase in transfection between 11 and 28 days in the muscle and connective tissue (P < 0.001), but not in the skin (P = 0.28), suggesting an increased transfection of cells in the muscle and connective tissue.

We next examined the ability of a plasmid encoding a secreted human placental alkaline phosphatase (SEAP) to mediate detectable levels of enzyme in the serum of athymic mouse. A pronounced increase in serum SEAP levels was evident 14 days after administration and high levels were sustained for the duration of the experimental period (Figure 3). SEAP expression levels in serum of animals injected with the implant containing the plasmid was at least 10-fold higher (P < 0.001) at each time-point (2.0-40.0 ng/ml) than in animals that received plasmid under other conditions (lower than 0.2 ng/ml) Figure 3.

We then examined whether delivery of a plasmid encoding for development endothelial locus-1 protein (Del-1) could elicit a physiological response after 2, 4 or 9 weeks of exposure to the implant. Del-1 is a novel extracellular matrix protein that under appropriate conditions enhances vascularization of tissues.^12,13 Samples of naked DNA in saline, empty injectable implant and naked plasmid injected separately or injectable implant lacking the plasmid served as controls.

First, tissues containing devices releasing the pDL-1 demonstrated a visible increase in blood vessel formation around the implant site (Figures 4i and 5i). Controls did not lead to a visible increase in blood vessel formation Figure 4c and f, 5c. Moreover, no increase in blood vessel number was observed when other plasmids (pCMV-Luc or pAP) were delivered by the implants (data not shown).

Second, histologic examination of implants containing pDL-1 revealed small vessels containing erythrocytes in regions around the implant at 14 days Figure 4k, 28 days Figure 5kand 62 days (not shown). Control samples did not exhibit a significant increase in small blood vessel formation at any time-point examined Figure 4e and h, 5e. Sustained delivery of pDL-1 was most likely responsible for these effects, since other controls injected directly into the subcutaneous pockets showed no increase of blood vessels at any time examined.

Finally, the in situ formation of the polymer device was very well tolerated. Injection of the in situ forming implants resulted in a modest foreign body response that is typically observed around PLGA implants Figure 4d, g, and j, 5d, g. Implants containing DNA had a slightly more reactive fibrosis rim surrounding the implant Figures 4j and 5j. Interestingly, the polymers containing the DNA eroded from the site more rapidly than polymers lacking DNA. This was observed regardless of the gene encoded in the plasmid.

Discussion

We demonstrate in this study robust gene expression in mice from plasmid DNA released from a solid matrix device. The innovative aspect in this report is that the device is formed in situ from an injectable glycofurol solution of a biodegradable polymer and plasmid DNA. This makes the device easy to form in sites that do not readily lend themselves to the implantation of solid devices.

This approach is based upon implantable systems that have been developed previously for protein and drug delivery.^7,8 Mixing the DNA in aqueous solution with a polymer solution in a water-miscible solvent (glycofurol) avoids the usage of high shear methods that can fragment the DNA, resulting in loss of biological activity. The mixture can then be injected into a subject through a standard syringe and needle with a DNA encapsulation efficiency approaching 100%.

The method of formation permits the DNA to be released from the device in a continuous manner that depends upon the characteristics of the PLGA, the mass of material injected, and the amount of DNA Figure 1c. When low molecular weight PLGA is used to form the implant, DNA is completely released by day 10. When high molecular weight PLGA is used, DNA is completely released over a 60-day period Figure 1c. The release pattern exhibits a rapid phase, followed by a slower phase, a second rapid phase and then a second slower phase. The DNA is released before the polymer is completely degraded. DNA released from the device is unchanged Figure 1dand retains 100 ± 20% DNA transfection activity in vitro even when released at 36 days Figure 2.

This release pattern is a consequence of how the device forms when the glycofurol solution is injected. The initial burst release of DNA is due to convection and diffusion of solvent out of the polymer matrix. A polymer 'skin' forms, as well as pathways that connect pockets of DNA to the surface. The early phases of release involve DNA diffusing from these compartments to the external solution. The later phases of DNA release occur subsequent to bioerosion of the polymer skin. DNA is then able to diffuse from the previously isolated compartments to the outside fluid. Different profiles of release can be engineered by changing parameters, such as weight percent of the polymer in solution, ratio of lactide to glycolide, molecular weight, and gene loading level in the polymer.

Most importantly DNA released from the device mediates pronounced gene expression when implanted subcutaneously in mice. Three different gene products were expressed in two different mouse strains. First, tissues in contact with the device expressed substantial amounts of luciferase (Table 1). Both the overlying skin and the underlying musculature expressed significantly higher levels of luciferase at both 11 and 28 days after injection than was found from injection of luciferase plasmid DNA or injection of luciferase plasmid after the polymer has been injected and the device formed. The level of gene expression was significantly higher in the tissue assayed at 28 days than at 11 days (Table 1).

Second, secreted human placental alkaline phosphatase was expressed at a significantly higher level for 67 days after injection of the device than was found with any other treatment Figure 3.

Third, the pDL-1 gene encodes Del-1, a matrix protein that is believed to regulate vascular morphogenesis in embryonic development.^12,13 Del-1 is reported to promote endothelial cell adhesion and when expressed in an appropriate environment promotes blood vessel remodeling.^12,13 It has been suggested that these properties may be useful in promoting angiogenesis in animals. Indeed, the developmentally regulated endothelial locus-1 gene released from the device led to the formation of an increased number of larger blood vessels in the vicinity of the device Figures 4i and 5i, whereas other forms of DNA administration caused no increase of blood vessels Figure 4c, f, and 5c, f.

Examination of histological sections of representative samples of tissue, taken from the vicinity of the injections, revealed a modest foreign body reaction and a slight cellular infiltrate surrounding all of the implants at day 14 Figure 4d, gand day 28 Figure 5d, g. This type of reaction is normal, anticipated and known to be associated with PLGA implants. This reaction was slightly more noticeable in pDL-1 containing implants. Very importantly, in the animals that received pDL-1 in the implant, there was a striking increase in structures that appeared to be well-formed small blood vessels that contained erythrocytes Figures 4k and 5k. These structures were rarely observed in animals receiving other treatments: DNA in saline, DNA plus implant, or PLGA alone.

If the exciting potential for tissue engineering is to be realized, a variety of scaffolds will be needed to serve both as a framework for the new tissue, as well as a source of the requisite growth factors. Pre-formed polymeric systems for gene delivery have been administered subcutaneously,¹⁰ intravenously,¹⁴ intramuscularly,^11,15 or orally^16,17 with promising results. These systems used various techniques to fabricate delivery systems from biodegradable polymers. PLGA polymeric matrices containing plasmid DNA for tissue engineering applications,¹⁰ fabricated with a gas foaming/particulate leaching process.¹⁸ PLGA microspheres containing DNA were prepared by a number of groups.^2,19 All these techniques have limitations; they are either complicated to fabricate, employ toxic solvents or inactivate plasmid DNA during the encapsulation process.

The approach presented in this paper has several advantages over exiting systems: (1) it is easy to administer and avoids surgery; (2) high DNA encapsulation efficiency; (3) the fabrication process is simple and does not require toxic solvent and high shear force; (4) other bioactive molecules can be co-encapsulated with the DNA; (5) the process uses pharmaceutically acceptable excipients; (6) manufacturing procedures required to produce a sterile, stable and reproducible DNA solid device are minimized. The ability to inject the polymer-DNA solution, makes it an ideal tool to direct reproducible tissue regeneration in small or irregular shaped cavities, such as can be found in nerve regeneration in the spinal cord, periodontal ligament detachment, and alveolar bone resorption. Moreover, the capability to control the time-frame for release of the plasmid enables the system to be used for applications in DNA vaccines.

Experimental protocols

Formation of injectable implant solution

We used 75:25 copolymers of D, L-lactide-co-glycolide (PLGA; Resomers RG 752 and RG 755, Boehringer Ingelheim Chemicals, Wallingford, CT, USA). The PLGA (_i = 0.24 dl/g or 0.59 dl/g) was dissolved in glycofurol in a weight ratio of 20:80 (200 mg PLGA in 800 mg glycofurol), at 37°C for 10 min. Calf thymus DNA (Sigma, St Louis, MO, USA), pCMV-Luc, pAP1166.157 (SEAP plasmid) or pDL 1575 (Del-1 plasmid) kindly supplied by Valentis (Woodlands, TX, USA), was dissolved in sterile deionized water at a concentration of 13 g/l. This solution was dissolved to 7.69% v/v in glycofurol (76.9 l plasmid-water in 923.1 l glycofurol), and then poured at 1:1 ratio into the PLGA-glycofurol solution. The mixture was rotated gently at 37°C for 30 min to form a homogenous solution.

Determination of the release characteristics in vitro

Devices used to obtain in vitro release profiles were prepared by injecting 100 g of the PLGA-glycofurol-pCMV-Luc solution through a 21-G needle directly into 10 ml of PBS (pH 7.4) contained in a vial (20 ml capacity). The vials were placed in an incubator shaker (New Brunswick Scientific, Edison, NJ, USA) at 37°C and shaken at 150 r.p.m. A 1 ml sample of the release medium was withdrawn, the remaining PBS replaced with 10 ml of fresh PBS and the samples stored at 4°C until assayed. DNA was quantified using picogreen assay as described.²⁰

Confocal microscopy to determine DNA location in the device in vitro

The DNA in the device was stained in two ways: to stain DNA throughout the device, the glycofurol mixture was injected into 10 ml PBS that contained 1 l of a 10 mg/ml EtBr solution (Gibco-BRL, Gaithersburg, MD, USA). In the second method, the glycofurol mixture was injected into PBS and 3 h later it was transferred to the EtBr-PBS solution and stained for 3 h. At the end of incubation, the implant was removed and inserted into 10 ml PBS vial for 1 min to wash non-intercalated EtBr from the implant. Three or 6 h after formation, the implant was removed and sectioned at 200 m with a Vibratome 1000 (Technical Products International, St Louis, MO, USA). Images were obtained on a BioRad laser scanning confocal microscope (MRC-1024, with a Nikon Diaphot 200 inverted microscope, using LaserSharp software) as described.²¹

In vitro transfection

In vitro transfection was assessed on the basis of luciferase expression mediated by complexation of the plasmid with 25 kDa PEI (Aldrich, Milwaukee, WI, USA) at a charge ratio of 2.3:1 (±) as described.^22,23 Results are the mean ± s.d. of at least three replicates.

In vivo studies

All animals were studied in accordance with guidelines established by the National Institute of Health Guidelines for the Care and Use of Laboratory animals, and with the approval of the Committee on Animal Research at the University of California, San Francisco. The ability of the released plasmids to transfect cells in vivo was assessed by injection of 100 l of the PLGA-glycofurol solution containing plasmid DNA (50 g DNA/mouse) into the subcutaneous flank tissue of CD-1 mice (for pCMV-Luc and pDL1) or nu/nu CD-1 athymic mice (for pAP) through a 21-G needle. Control mice (n = 6 in each group) were injected with 100 l of either 50 g DNA dissolved in glycofurol, DNA dissolved in saline, DNA dissolved in saline injected 10 min after the injection of PLGA-glycofurol solution lacking the DNA, or injected with PLGA-glycofurol solution lacking the plasmid.

Determination of luciferase activity in mouse tissue

To measure luciferase expression from animals receiving luciferase DNA, mice were killed and the skin, connective tissue and muscle in a 1 cm radius around the s.c. implantation were removed, weighted and frozen at -80°C. Luciferase activity in tissue was measured as described.²³

Determination of SEAP enzyme in serum

Blood was collected by retro-orbital puncture on a weekly basis starting at 14 days after DNA administration. Heat labile alkaline phosphatase was inactivated by incubating serum samples at 65°C for 30 min, and the samples were assayed for SEAP activity, using the Phospha-Light chemiluminescent assay (Tropix, Bedford, MA, USA) and Opticompt luminometer. SEAP concentration in serum was assayed relative to a standard curve of a standard SEAP enzyme (Tropix) diluted in control serum.

Determination of angiogenic effect from Del-1 expression

To measure the effect of murine Del-1 expression on blood vessel growth, animals were killed on days 14, 28 and 62 after treatment. The skin, connective tissue and muscle within a radius of 1 cm of the implant were removed. The tissue samples were fixed in 10% paraformaldehyde solution at 4°C for at least 1 day. Representative tissue samples were dissected and embedded in paraffin. Histologic sections were stained using hematoxylin and eosin at Biopathology Sciences (South San Francisco, CA, USA). Representative samples of injection sites/polymers at 14, 28 and 62 days were examined by light microscopy.

Statistical analysis

Statistical analysis was performed using StatView software. The ANOVA analysis employed a post hoc comparison to analyze the significance of protein expression among the various groups.

Acknowledgements

We thank Yael Eliaz for assistance with the statistical analysis, Michael Coleman and Liz Wilson for helpful comments on the manuscript and Edward Dy for technical assistance. We thank Valentis Inc. (Burlingame, CA, USA) for the gift of the plasmid DNA used in this research. We gratefully acknowledge financial support from NIH DK 46052 and the State of California Tobacco-Related Disease Research Program 8IT-0138. Dr Szoka has a financial interest in and serves as a consultant to Valentis, a biotechnology company developing gene medicines. We also thank Donald McDonald for access to the vibratome used to section the implants.

1 Langer R. Drug delivery and targeting. Nature (London) 1998; 392: 5-10. MEDLINE

2 Luo D, Woodrow-Mumford K, Belcheva N, Saltzman WM. Controlled DNA delivery systems. Pharm Res 1999; 16: 1300-1308. MEDLINE

3 Chasin M. Biodegradable polymers for controlled drug delivery. Hollinger JO (eds); Biomedical Applications of Synthetic Biodegradable Polymers CRC Press, 1995, pp 1-15.

4 Bonadio J, Smiley E, Patil P, Goldstein S. Localized, direct plasmid gene delivery in vivo: prolonged therapy results in reproducible tissue regeneration. Nat Med 1999; 5: 753-759. Article MEDLINE

5 Bonadio J. Tissue engineering via local gene delivery. J Molec Med 2000; 78: 303-311.

6 Dunn RLet al. . Sustained release of cisplatin in dogs from an injectable implant delivery system. J Bio Comp Poly 1996; 11: 286-300.

7 Eliaz RE, Kost J. Characterization of a polymeric PLGA-injectable implant delivery system for the controlled release of proteins. J Biomed Mat Res 2000; 50: 388-396.

8 Eliaz RE, Wallach D, Kost J. Delivery of soluble tumor necrosis factor receptor from in situ forming PLGA implants: in vivo. Pharma Res 2000; 17: 1546-1550.

9 Royals MAet al. . Biocompatibility of a biodegradable in situ forming implant system in rhesus monkeys. J Biomed Mat Res 1999; 45: 231-239.

10 Shea LD, Smiley E, Bonadio J, Mooney DJ. DNA delivery from polymer matrices for tissue engineering. Nat Biotechnol 1999; 17: 551-554. Article MEDLINE

11 Labhasetwar VAet al. . DNA controlled-release coating for gene transfer: transfection in skeletal and cardiac muscle. J Pharm Sci 1998; 87: 1347-1350. Article MEDLINE

12 Hidai Cet al. . Cloning and characterization of developmental endothelial locus-1: an embryonic endothelial cell protein that binds the alphavbeta3 integrin receptor. Genes Dev 1998; 12: 21-33. MEDLINE

13 Penta Ket al. . Del-1 induces integrin signaling and angiogenesis by ligation of alphaVbeta3. J Biol Chem 1999; 274: 11101-11109. MEDLINE

14 Mao HQet al. . Chitosan-DNA nanoparticles as gene carriers: synthesis, characterization and transfection efficiency. J Contr Rel 2001; 70: 399-421.

15 Truong-Le VL, August JT, Leong KW. Controlled gene delivery by DNA-gelatin nanospheres. Hum Gene Ther 1998; 9: 1709-1717. MEDLINE

16 Mathiowitz Eet al. . Biologically erodable microspheres as potential oral drug delivery systems. Nature 1997; 386: 410-414. MEDLINE

17 Roy K, Mao HQ, Huang SK, Leong KW. Oral gene delivery with chitosan-DNA nanoparticles generates immunologic protection in a murine model of peanut allergy. Nat Med 1999; 5: 387-391. Article MEDLINE

18 Harris LD, Kim B-S, Mooney DJ. Open pore biodegradable matrices formed with gas foaming. J Biomed Mater Res 1998; 42: 396-402. Article MEDLINE

19 Wang DQ, Robinson DR, Kwon GS, Samuel J. Encapsulation of plasmid DNA in biodegradable poly(D,L-lactic-co-glycolic acid) microspheres as a novel approach for immunogene delivery. J Contr Rel 1999; 57: 9-18.

20 Choi S-J, Szoka FC. Fluorometric determination of deoxyribonuclease I activity with PicoGreen. Anal Biochem 2000; 281: 95-97. Article MEDLINE

21 Uyechi LS, Gagne L, Thurston G, Szoka FC. Mechanism of lipoplex gene delivery in mouse lung: binding and internalization of fluorescent lipid and DNA components. Gene Therapy 2001; 8: 828-836. MEDLINE

22 Tang MX, Szoka FC. The influence of polymer structure on the interactions of cationic polymers with DNA and morphology of the resulting complexes. Gene Therapy 1997; 4: 823-832. MEDLINE

23 Barron LG, Uyechi LS, Szoka FC. Cationic lipids are essential for gene delivery mediated by intravenous administration of lipoplexes. Gene Therapy 1999; 6: 1179-1183. MEDLINE

Figure 1 Cross-sectional view of DNA embedded in matrix and in vitro release characteristics of DNA from Matrix. (a) Reconstructed cross-sectional view of PLGA injectable implant containing calf thymus DNA labeled with EtBr and (b) z-series cross-sectional view of the same implant showing the DNA distribution within the polymer matrix. (c) The cumulative release of plasmid DNA from 75:25 (_i = 0.24 dl/g) (), and 75:25 (_i = 0.59 dl/g) (). Values represent mean and standard deviation (n = 3). (d) Photograph of an agarose gel on which plasmid DNA (pCMV-Luc) was analyzed by electrophoresis. Lane 1, molecular weight marker (HindIII); lane 2, unincorporated plasmid DNA; lane 3, plasmid released between 1 h and day 1; lane 4, plasmid released between day 1 and day 2; lane 5, plasmid released between day 2 and day 5; lane 6, plasmid released between day 5 and day 7; lane 7, plasmid released between day 7 and day 14; lane 8, plasmid released between day 14 and day 28; lane 9, plasmid released between day 28 and day 36. The lanes have an equal loading of ~1 g plasmid. Original magnification of (a) and (b) is ´40.

Figure 2 Luciferase expression from in vitro released DNA. In vitro transfection activity of luciferase plasmid DNA released at different times from the implant: unincorporated plasmid (black bar); plasmid released up to day 1 (white bar); plasmid released between day 1 and day 2 (cross-striped bar); plasmid released between day 2 and day 5 (gray bar); plasmid released between day 5 and day 7 (dotted bar); plasmid released between day 7 and day 14 (diagonal striped bar); plasmid released between day 14 and day 28 (cross-striped bar); plasmid released between day 28 and day 36 (horizontal bar). A linear regression analysis to detect if there was a trend towards DNA inactivation as a function of transfection activity with time of release was not significant (P > 0.3; comparison to a slope = 0). Values represent mean and standard deviation (n = 3). Cell transfection was mediated by PEI as described in the Experimental protocol.

Figure 3 SEAP expression in serum following administration of SEAP plasmid in various formulations. Secreted human placental phosphatase alkaline (SEAP) activity in the serum of nu/nu mice at different times after injection of: DNA in saline (white bar); injectable implants containing the SEAP plasmid (black bar); injectable implant lacking the DNA followed by DNA in saline (diagonal striped bar); and injectable implant lacking the DNA (dotted bar). Asterisks indicate statistical difference at a level of P < 0.001 between implant containing DNA to every other treatment. Statistical analysis was performed using the software program StatView. ANOVA repeated measures. Values represent mean and standard deviation (n = 6).

Figure 4 Effect of polymer implant containing pDL-1 on blood vessel formation on day 14. In vivo response to release of plasmids encoding for Del-1 and controls after 2 weeks of administration. Photomicrograph of tissues (muscle/connective tissue and skin) within 1 cm in radius of site of injection of different treatments (a, c, f, i) followed by histological photomicrographs of tissue cross section: (a, b) DNA in saline; (c-e) DNA injected following polymer solution injection; (f-h) polymer solution lacking the DNA; (i-k) polymer solution containing DNA. For histological sections two magnifications of ´4 (b, d, g, j) and ´40 (e, h, k) have been used. Photomicrographs have labels for polymer (P), granulation tissue (G), muscle layer (M) and skin surface (S). Arrows indicate blood vessels.

Figure 5 Effect of polymer implant containing pDL-1 on blood vessel formation on day 28. In vivo response to release of plasmids encoding for Del-1 and controls after 4 weeks of administration. Labels are as in Figure 4. White arrow in panel (i) indicates formation of small blood vessels in the connective tissue adjacent to the implant. The tissue has the same circular shape as the implant (implant can be seen adhering to the connective tissue, labeled with (P).

Table1 Luciferase expression in tissues adjacent to implant