Femtosecond laser: a new intradermal DNA delivery method for efficient, long-term gene expression and genetic immunization

The FASEB Journal • Research Communication

Femtosecond laser: a new intradermal DNA delivery method for efficient, long-term gene expression and genetic immunization Evelyne Zeira,*,1 Alexandra Manevitch,*,1 Zakharia Manevitch,† Eli Kedar,§ Michal Gropp,* Nili Daudi,‡ Rimma Barsuk,‡ Menahem Harati,* Hagit Yotvat,* Philip J. Troilo,储 Thomas G. Griffiths II,储 Stephen J. Pacchione,储 Dana F. Roden,储 Zhutian Niu,储 Ofer Nussbaum,¶ Gideon Zamir,# Orit Papo,** Izhack Hemo,† Aaron Lewis,† and Eithan Galun*,2 *Goldyne Savad Institute of Gene Therapy, Hadassah-Hebrew University Hospital; †Department of Applied Physics, The Selim and Rachel Benin School of The Hebrew University of Jerusalem Engineering and Computer Sciences, Natural Science Faculty, and The Department of Ophthalmology, Hadassah Laser Center; ‡Liver Unit, Hadassah-Hebrew University Hospital; §The Lautenberg Center of Immunology, Hebrew University-Hadassah Medical School, Jerusalem, Israel; 储 Department of Biologics Safety Assessment, Merck Research Laboratories, Merck & Co., Inc., West Point, Pennsylvania, USA; ¶XTL Biopharmaceuticals Ltd., Rehovot, Israel; and #Departments of Surgery and of **Pathology, Hadassah-Hebrew University Hospital, Jerusalem, Israel A femtosecond laser beam gene transduction (SG-LBGT) system is described as a novel and efficient method of intradermal (i.d.) nonviral gene delivery in mice by permeabilizing cells utilizing femtosecond laser pulses. Using this approach, significant gene expression and efficient dermal transduction lasting for >7 months were obtained. The ability of this new DNA gene transfer method to enhance genetic vaccination was tested in BALB/C mice. A single i.d. injection of a plasmid (10 ␮g) containing the hepatitis B virus (HBV) surface antigen (HBsAg), followed by pulses of laser, induced high titers of HBsAg-specific antibodies lasting for >210 days and increased levels of IgG1, IgG2a, IFN␥, and IL-4, indicating the activation of both Th1 and Th2 cells. Moreover, mice vaccinated using the SG-LBGT followed by challenge with pHBV showed increased protection against viral challenge, as detected by decreased levels of HBV DNA, suggesting an efficient Th1 effect against HBV-infected replicating cells. Tumor growth retardation was induced in vaccinated mice challenged with an HBsAg-expressing syngeneic tumor. In most of the parameters tested, administration of plasmid followed by laser application was significantly more effective and prolonged than that of plasmid alone. Tissue damage was not detected and integration of the plasmid into the host genomic DNA probably did not occur. We suggest that the LBGT method is an efficient and safe technology for in vivo gene expression and vaccination and emphasizes its potential therapeutic applications for i.d. nonviral gene delivery.—Zeira, E., Manevitch, A., Manevitch, Z., Kedar, E., Gropp, M., Daudi, N., Barsuk, R., Harati, M., Yotvat, H., Troilo, P. J., Griffiths, T. G., II, Pacchione, S. J., Roden, D. F., Niu, Z., Nussbaum, O., Zamir, G., Papo, O., Hemo, I., Lewis, A., Galun, E. Femtosecond

ABSTRACT

3522

laser: a new intradermal DNA delivery method for efficient, long-term gene expression and genetic immunization. FASEB J. 21, 3522–3533 (2007) Key Words: gene delivery 䡠 vaccination 䡠 HBV 䡠 antitumor

Using gene therapy for the prevention and treatment of diseases has not yet reached its full potential. Delivering genetic materials to target cells with high efficiency is the key hurdle in using gene therapy for the development of new therapeutics. Although current viral vectors exhibit superior gene transfer efficacy, the safety and immunogenicity of viral-mediated gene delivery face major concerns. For these reasons, nonviral methods such as plasmid DNA gene delivery have become attractive alternatives for human gene therapy (1). Plasmid DNA raises less safety concerns than that associated with viral vectors; it is comparatively easy to produce, has negligible immunogenicity, and is also less expensive than viral vector preparation. However, major disadvantages of the current available methods for the transfer of naked DNA include low levels of transfection efficiency in vivo, lack of sustained expression, and even tissue damage (2). Therefore, technical improvements in transfection methodology are required before nonviral gene therapy can successfully be used as therapeutic tools in the clinic. 1

These authors contributed equally to this work. Correspondence: The Goldyne Savad Institute of Gene Therapy, Hadassah University Hospital, P.O. Box 12000, Jerusalem, 91120 Israel. E-mail: [email protected] doi: 10.1096/fj.06-7528com 2

0892-6638/07/0021-3522 © FASEB

In a preliminary report we described a novel permeabilization method using the laser beam gene transduction (LBGT) technique, which was shown to increase transfection efficacy of naked DNA in mouse muscle (3). Based on these findings and to enable the translation of this method for preclinical use, we further modified and optimized the femtosecond infrared laser system. The advantage of using femtosecond laser poration for effective permeation of cells under tissue surfaces is the minimal collateral damage and maximum efficacy encountered with this type of laser. Since the initial studies by Berns et al. (4, 5) of laser manipulation of cells, femtosecond laser poration has been documented to have this ability. This is the case for all lasers used, including the original nitrogen laser with its detrimental ultraviolet side effects, nanosecond lasers with their associated collateral damage, or even continuous wave lasers with their unacceptable heating. We demonstrate here that the second-generation femtosecond laser, with its mobile arm for precise guided targeting of the laser to the desired site of injection, is highly efficient for intradermal (i.d.) nonviral gene delivery. To further evaluate the therapeutic potential of this transfection method, we performed gene transfer experiments in a new hepatitis B virus (HBV) mouse model we developed (6, 7). We show that a single i.d. injection of naked DNA encoding HBsAg, followed by LBGT, elicited potent and long-lasting HBV-specific Th1 and Th2 immune responses. The enhanced responses were shown to provide effective protection against challenge with both HBV and an HBsAg-expressing syngeneic tumor.

MATERIALS AND METHODS Femtosecond laser system optimized for dermis gene delivery The first generation of femtosecond laser has been described (3). Coherent Radiation Mira titanium sapphire mode-locked laser emitting 200 fs pulses with a 76 MHz repetition rate was pumped by an argon ion laser (Coherent; Innova 200) operated at 12 W in a multiline mode. The operating wavelength is 780 nm. Laser powers of the objective plane were routinely measured, using a power meter (Nova display; Ophir Optronics, Tel Aviv, Israel). The beam delivery in this first version of our system was based on a beam scanner through an optical microscope that was not suitable for large animals and clinical studies. This is also the case in more recent reports applying the femtosecond laser for cell poration (8). In the present version of our system, the laser is transmitted through an articulated arm that is designed to freely illuminate specific regions of the animal. The end of the articulated arm, in closest proximity to the animal, was modified with a lens-based optical system shown in the insert of Fig. 1. The lens has a defined focal length that focuses the laser under the skin at a specific depth. The penetration depth was defined using two photon fluorescence and a micrometer. The end of the arm with the lens can be pressed against the skin so that accurate depth penetration of the focused laser beam is provided. We use a 50 ⫻ 0.45 numerical aperture objective. The relatively large numerical aperture allows the focal volume to be limited to a region of ⬃0.5 ␮m and permits nonlinear alteration, presumably poration of the cells that are in this focal volume. Thirty seconds after DNA injection, the anesthetized animal is placed under the objective and the site of injection is exposed to the laser beam through the objective. All this was designed and performed in an effort to develop the new apparatus into a clinical applicable system.

Figure 1. Femtosecond laser apparatus. Coherent Radiation Mira titanium sapphire mode-locked laser emitting 200 fs pulses with a 76 MHz repetition rate was pumped by an argon ion laser (Coherent; Innova 200) that was operated at 12 W in a multiline mode. The operating wavelength was 780 nm. The laser beams were transmitted via an articulated arm operated manually. Insert: The end of the articulated arm in closest proximity to the animal was modified with a lens-based system shown in the insert. The lens had a defined focal length that focused the laser under the skin at a specific depth. The end of the arm with the lens could be pressed against the skin in order to provide accurate depth penetration of the focused laser beam. The objective used was a 50 ⫻ 0.45 numerical aperture. The relatively large numerical aperture allowed the focal volume to be limited to a region of ⬃ 0.5 ␮m and permitted nonlinear alteration, presumably poration, of the cells that were in this focal volume. 30 s after DNA injection, the anesthetized animal was placed under the objective and the site of injection was exposed to the laser beam through the objective. LASER-FACILITATED DERMAL GENE DELIVERY AND GENETIC IMMUNIZATION

3523

In the optical path, a shutter was included to allow for a variable number of pulses of the laser beam at a depth of ⬃400 ␮m with an exposure time of 2 min and an average power of 30 mW energy for the mouse skin irradiation. The articulated arm allowed for rotation around a limb of the animal, for example. This mobility provided a ring of irradiation at a 400 ␮m radiation depth. Plasmid construction To measure relative gene expression levels after gene transfer by LBGT, a luciferase pCAG/Luc vector was constructed by cloning the CAG promoter (⫺SpeI-ECORI fragment of 1153 bp) (CAG promoter combines the human cytomegalovirus immediate-early enhancer and a modified chicken ␤-actin promoter including the first intron) from plasmid PBS-CAGM2nls-pA (kindly provided by Drs. P. Moullier and P. Chenuaud, Laboratoire Therapie Genetique, Nantes, France) into plasmid pGL3b (Promega, Madison, WI, USA) using SwaI. The resulting plasmid contained the luciferase transgene under transcriptional control of the CAG promoter. For construction of plasmid pCAG/hSeAp, the hSeAp expression cassette, including the SV40pA, was excised from pXL3010SeAp (generously provided by Dr. C. Trollet, Paris, France) using HindIII and SalI, and given a fragment of 1925 bp. The luciferase expression cassette, including the SV40pA, was excised with HindIII and SalI from the vector pCAG/Luc, and the hSeAp SV40pA expression cassette was inserted into the HindIII and SalI sites in pCAG(⫺Luc) to generate pCAG/ hSeAP vector. For ␤-galactosidase gene expression, we used the pCMV␤-gal encoding the lacZ gene (bearing the SV40-T antigen NLS) and driven by the CMVIE promoter (kindly provided by Dr. T. Lanigan, Vector Core Labs, Ann Arbor, MI, USA). For immunization studies, a plasmid encoding HBsAg under the control of the CMVIE promoter (pRc/ CMV-HBs(S) was obtained from Aldevron (Fargo, ND, USA). To study the antiviral effect of the LBGT-induced immunization, we used the full-length HBV carrying plasmid pHBVadwHTD, a head-to-tail dimmer of wild-type HBV, adw subtype, cloned into the ECORI site of pGEM-7 as described (6, 7). All plasmids were amplified in Escherichia coli JM109 and prepared with a Qiagen Endo-Free plasmid Giga kit (Qiagen GmbH, Cologne, Germany). In vitro gene transfection and expression The SG-LBGT was initially assessed for its transfection potential in vitro; these were experiments aimed to show the transfection potential prior to in vivo studies. HEK 293 T cells were seeded at 5 ⫻ 104 cells/well on glass coverslips in a 24-well plate, in 0.5 ml of complete medium (DMEM supplemented with 10% fetal calf serum) containing 0.5 ␮g plasmid DNA vector pCAG/EGFP encoding the enhanced green fluorescent protein (EGFP) gene. The coverslips with adherent cells were taken out of the well and placed onto a slide. The cells were then irradiated with laser pulses at a power of 15–30 mW for an exposure of 2 ms, and depth penetration was varied from 1 to 4 ␮. The coverslips were then replaced into the tissue culture 24-well plate. Twenty four hours after transfection, fluorescence was detected using a confocal Laser scanning microscope LSM410 (Zeiss, Jena, Germany) with Plan-neofluar ⫻40 (oil-Zeiss) lens. EGFP expression was detected using the fluorescein isothiocyanate channel. Animal injection and immunization protocols In all experiments, 6- to 7-wk-old, pathogen-free (SPF) female BALB/c mice (Harlan Laboratories, Jerusalem, Israel) were 3524

Vol. 21

November 2007

used. Mice were anesthetized with a single intraperitoneal dose of xylazine (11 ␮g/g) and ketamine hydrochloride (110 ␮g/g). Plasmids were injected into the dermis of the shaved trunk skin on the dorsal side of mice using a 301/2 gauge needle and a 1 cc insulin syringe. The site of injection was prelabeled, marking the area of irradiation to be with the dimensions of 5 ⫻ 5 mm2. Each mouse received 1 to 12 ␮g of plasmid DNA in 30 –50 ␮l of sterile PBS. For immunization, groups of 4 –10 mice each were injected i.d. once with the plasmid DNA encoding the HBsAg described above, with or without LBGT, for 30 s after the injection. Mice injected with the plasmid only (without laser beam pulses) and mice injected with a commercial HBsAg vaccine (Engerix®-1 ␮g/50 ␮l), (Smith Klein Beecham, Brussels, Belgium) as described previously (9) served as controls. Sera were collected before the first immunization and at several time points thereafter and analyzed for antibody and HBsAg levels. The animals were fed commercial diets and water ad libitum. Animals were kept in a 12 h light-dark cycle at constant temperature and humidity. The Institutional Animal Welfare Committee (NIH approval number OPRR-A01– 5011) approved all animal experiments. All animal experiments were performed according to national regulations and institutional guidelines. In vivo quantification of gene expression To detect and quantify continuously expressed genes in live animals, we used the CCCD imaging system as described (10). For histological and ␤-galactosidase activity analysis, formalinfixed, paraffin-embedded skin samples were cut into sections 4 ␮m thick, deparaffinized in xylene, and dehydrated through a series of decreasing concentrations of ethanol. Sections of skin tissue around the site of injection were excised and ␤-galactosidase activity was measured using the whole-mount method as described previously (11). Immunohistochemical assessment of apoptosis by TUNEL Mice were injected i.d. with 10 or 12 ␮g of plasmid DNA, followed by LBGT. Twenty-four hours later, skin tissues around the site of injection were excised, paraffin embedded, and the degree of apoptosis was examined in 4 ␮m sections using the Deadend Fluorometric Tunel system (Promega). The fluorescein-12-dUTP-labeled DNA was visualized directly by confocal microscopy (3). Integration assay The gel purification-based integration assay is described elsewhere in detail (12). Mice were injected i.d with 10 ␮g of plasmid pCAG/Luc, followed by LBGT. Eight months later the skin tissues at the site of injection were removed and processed for genomic DNA using the genomic DNA isolation kit, following the manufacturer’s protocol (Gentra Systems, Minneapolis, MN, USA). The DNA preparations (pool of two) from the skin were subjected to four successive rounds of preparative, conventional agarose gel electrophoresis; varying the percent and type of agarose, the type of running buffer, and the voltage of each round was carried out in an effort to remove all potential forms of the extrachromosomal plasmid. The high molecular weight genomic DNA was excised from the final gel, eluted from the agarose, quantitated, and subjected to TaqMan® QPCR. The QPCR was directed at a 58 bp amplicon within the luciferase gene. Both pre-gel and 4⫻ gel-purified skin DNA were subjected to QPCR. As determined by titration of the plasmid into TE buffer, the assay sensitivity (LOD) was established at ⬍4

The FASEB Journal

ZEIRA ET AL.

plasmid copies/␮g genomic DNA. Two negative controls were also run with the treated samples: control mouse DNA and control rat DNA that had been 4 ⫻ gel-purified. Both negative controls were negative in the QPCR assay. Table 1 represents the results of a pool of two skins of genomic DNA.

control wells (medium only). To measure cytokine secretion, splenocytes were cultured as described above for 3 days. Levels of IFN␥ and IL-4 in culture supernatants were assessed using a “sandwich” ELISA technique (PharMingen) as described (16).

Analysis of secreted proteins for gene expression measurement

Anti-viral effect of LBGT-induced immunization

Blood samples were collected from the retro-orbital plexus of anesthetized mice. Samples were centrifuged for 10 min at 14,000 g and serum was collected and frozen at ⫺20°C for storage until processed for analysis. Detection of the murinesecreted alkaline phosphatase was carried out with the Tropix® Phospha-Light™ System kit (Applied Biosystems, Foster City, CA, USA). To detect the HBsAg in the sera of mice, we used the AXSYM systems (Abbott GmbH Diagnostica, Wiesbaden-Delkenheim, Germany). Determination of serum antibody levels Detection of serum antibodies was performed according to standard protocols. Blood samples were obtained by retroorbital bleeding. Antibodies against HBsAg were tested in mouse sera using the commercially available test IMxAUSAB system (Abbott GmbH Diagnostica). Serum titers (in MIU/ ml) shown are the mean of at least five individual mice ⫾ sd. Sera were also analyzed for HBsAg-specific IgG1 and IgG2a antibodies, determined by an end point dilution ELISA assay as described (13) and slightly modified. Briefly, micro-ELISA plates (Nunc-Maxisorp, Nunc, Roskilde, Denmark) were coated by overnight incubation at 4°C with recombinant antigen HBsAg particles (BTG, Rehovot, Israel) at a concentration of 5 ␮g/ml in 50 ␮l of PBS. Plates were washed with PBS/0.5% Tween 20 and blocked with blocking buffer (PBS, 0.4% gelatin in PBS) for 2 h at room temperature. Sera (50 ␮l) were serially diluted in blocking buffer and added to the antigen-coated wells. Bound serum antibodies were detected using biotin-conjugated goat anti-mouse IgG1 (1:3000, SouthernBiotech, Birmingham, AL, USA) and goat anti-mouse IgG2a (1:3000, SouthernBiotech, USA). Avidin-alkaline phosphatase (1:1000, PharMingen, San Diego, CA, USA) was then added. Color was developed with the addition of p-nitro phenyl phosphate (Sigma-Aldrich, St. Louis, MO, USA) and absorbance at 405 nm was measured. Results are expressed as arbitrary units per milliliter. The reciprocal of the highest serum dilution yielding 50% of the maximal absorbance (maximum O.D.) obtained with a “standard” anti-HBsAg immune serum, after subtracting the blank (antigen only), was considered the ELISA antibody titer (U/ml; 1U⫽50% maximum OD) (14). Lymphocyte proliferation and cytokine secretion Spleens were removed 7 wk after immunization to make single-cell suspensions and processed for proliferation assay as described (15). Briefly, 100 ␮l of 2 ⫻ 106 cells/ml splenocytes in complete RPMI 1640 [supplemented with 10% fetal bovine serum (GIBCO, Grand Island, NY, USA, 100 U/ml penicillin, 100 ␮g streptomycin, and 2 mM l-glutamine (Biological Industries, Beit Ha’emek, Israel] was added to each well in 96-round bottom plates, stimulated wells received purified rHBsAg at a concentration of 0.5 ␮g/ml. Cells in all wells were cultured in a total volume of 200 ␮l of medium. After 4 days in culture, the cells were pulsed with thymidine (1 ␮Ci/well) for 18 h and harvested. The proliferative response is presented as Delta-cpm calculated as the mean cpm of the antigen-stimulated wells ⫺ the mean cpm of the

BALB/c mice were immunized i.d. with the HBsAg plasmid (10 ␮g), with or without the application of LBGT. Another group was injected i.d. with Engerix® (1 ␮g/50 ␮l) (9). On day 46, vaccinated mice received an intravenous hydrodynamic injection of the plasmid bearing the full length of the HBV mentioned above (12 ␮g pHBVadwHTD) in a total volume of 1.5 ml PBS. This plasmid was also injected into naive mice used for positive control while other naive mice were injected with the empty vector pGEM-7 as a negative control. On day 49, all mice were bled and the serum was tested for HBsAg, HBV-DNA, and anti-HBsAg antibody levels. Determination of HBV DNA in mouse sera and real-time PCR for quantitation of serum-associated HBV-DNA The HBV-DNA copy number in mouse sera was determined by real-time quantitative PCR. In brief, HBV-DNA was extracted from the mouse sera using the DNAzol-BD reagent (Molecular Research Center Inc., Cincinnati, OH, USA) as described (6) and resuspended in 10 ␮l of H2O. HBV-DNA was quantitated using Taq-Man real-time assay. Real-time PCR primers were designed as follows: sense primer (nucleotides 1778 to 1798, 21N) B1F-5⬘-GAG-GCT-GTA-GGC-ACA-AATTGG-3⬘; antisense primer (nucleotides 1821 to 1845, 25N) B1R-5⬘-AGA-TGA-TTA-GGC-AGA-GGT-GAA-AAA-G-3⬘. HBV probe: (nucleotides 1801 to 1820, 25N) B1P-5⬘-FAM-CTGCGC-ACC-AGC-ACC-ATG-CA-TAMRA-3⬘. The reaction was carried out in 50 ␮l of a PCR mixture containing 1 ⫻ Taq Pol. buffer, 7.5 mM MgCl2, 0.4 mM of each dNTP, 50 pmol of each primer, 12 pmol probe, 0.6 ␮M ROX (Bio Research Laboratories, Redmond, WA, USA), 0.1% BSA, and 2.5 U of Taq Pol (Promega). The PCR reaction was programmed for 5 min at 94°C, followed by 40 cycles of 30 s at 94°C and 1 min at 60°C. Reaction was performed using the ABI 7000 PCR machine and analyzed by tABI software. In vivo tumor protection Three groups of BALB/c mice (n⫽10/group) were immunized once i.d. as follows: 1) 10 ␮g HBsAg plasmid only; 2) 10 ␮g HBsAg plasmid followed 30 s later by pulse laser, 3) 1 ␮g Engerix®. Three weeks after immunization, mice were challenged subcutaneously in the left lateral flank with 1 ⫻ 105 CT26 (colon carcinoma) cells or the HBsAg gene stabletransfected CT26/S cells. Tumor volume was measured and calculated using the formula: V ⫽ length ⫻ width2 ⫻ 0.52. Statistical analysis Statistical differences were determined using the 2-tailed t test when two groups were compared and the nonparametric ANOVA test (Kruskal-Wallis) was used for comparisons of multiple groups (Dunn’s multiple comparisons test). Differences between groups were considered significant if P ⱕ 0.05.

LASER-FACILITATED DERMAL GENE DELIVERY AND GENETIC IMMUNIZATION

3525

RESULTS The femtosecond laser beam gene transduction device Our first-generation laser beam gene transduction (LBGT) method was developed for testing gene transduction in rodent models (3). To establish feasibility of this method for human clinical trials, this secondgeneration LBGT (SG-LBGT) apparatus was modified by the incorporation of a mobile articulated arm to target the laser to the desired site of gene transduction (see Materials and Methods). This permitted freely rotating the femtosecond laser beam around animal limbs and other extremities. We then optimized the parameters for gene transfer to the dermis of mice using the articulated arm and tested its potential for inducing DNA-mediated vaccination. The parameters critical for successful transduction, including the depth of laser action (200 ␮m to 400 ␮m), exposure time (2–3 min), and quantity of energy delivery, were optimized, from 15 to 40 mW and chosen to have a depth of ⬃400 ␮m, with an exposure for 2 min and an average power of 30 mW energy for the mouse skin irradiation (data not shown). The SG-LBGT system (Fig. 1) incorporates a femtosecond infrared mode-locked Ti:Sapphire laser as the illumination source. The fundamental basis of the action of this laser is most likely multiphoton transient alteration of cell membranes below the skin. The approach produces highly localized tissue poration with minimal heating, and thus minimal collateral damage of the tissue. The laser beam passes through an articulated arm for precise laser delivery. The laser light exiting the arm is focused by a lens that is integrally connected to the output of the arm. The flexibility of the arm should be suitable for gene delivery in humans. Femtosecond laser-assisted transfection of cells in culture To demonstrate the targeted specific transfection efficiency of DNA delivery mediated by LBGT in vitro prior

to in vivo experimentation, we used the HEK 293 T cells that were pulsed with the SG-LBGT in the presence of pCAG/EGFP in the medium. Ultrashort laser pulses with a power of 15–30 mW energy for an exposure time of 2 ms resulted in a successful transfection of individual cells at the specific targeting site of the beam (Fig. 2a, b), suggesting that the LBGT induces some kind of cell membrane permeabilization. However, additional studies are under way that should further elucidate the mechanism responsible for the enhanced transfection observed with LBGT. LBGT for dermal DNA delivery and gene expression in mice In addition to muscle, the skin is another attractive target for transient gene expression, especially genetic immunization, due to the presence of large numbers of antigen-presenting cells within the tissue enabling the development of specific Th1 and Th2 immune responses (17). To test the feasibility of this approach for genetic immunization, the i.d. gene delivery and resulting transgene protein expression were optimized using reporter genes in a mouse model. To optimize the laser parameters and analyze the stability of transgene expression after i.d. naked DNA injection, the pCAG/Luc construct was used as the reporter gene. Continuous luc expression was determined by the bioluminescence CCCD system (10). A single i.d. injection of 10 ␮g of pCAG/Luc, followed by laser beam illumination for 2 min, resulted in an increased transgene expression for up to an average of 625-fold on day 43 (2⫻106 integrated light units), relative to naked DNA injection, without the application of the laser beam (3⫻103 integrated light units). In mice injected with plasmid only, the level of expression was very low and dropped close to zero on day 43, whereas gene expression following LBGT was stable and continued for ⬎7 months, a time point selected to terminate experimentation for the assessment of vector integration (see Table 1) (Fig. 3A).

Figure 2. Cell transfection by the SG-LBGT in vitro. HEK 293 T cells were seeded in a 24-well plate at 5 ⫻ 104c/well and pulsed with the SG-LBGT in the presence of 0.5 ␮g pCAG/EGFP in the medium. Laser pulses of 2 ms with a power of 15–30 mW energy resulted in a successful transfection of individual cells (a, b). c) Cells not transfected. 3526

Vol. 21

November 2007

The FASEB Journal

ZEIRA ET AL.

After determining the optimal parameters for transfection efficiency in vivo, we analyzed the potential of this method for the expression of other proteins. The pCAG/hSeAp plasmid expressing human secreted alkaline phosphatase as well as the plasmid pCMV/ HBsAg carrying the HBsAg reporter gene was used to assess protein secretion. We also used the pCMVLacZ plasmid for ␤-galactosidase gene expression analysis in the dermal tissue. Successful i.d. transfection was confirmed by ␤-galactosidase staining on day 30 (Fig. 3B, a). Intradermal injection of pCAG/hSeAp, followed by LBGT, revealed a 3-fold increased expression relative to plasmid DNA injected without LBGT on day 4 (Fig. 3C, a). The i.d. injection of pCMV/HBsAg, followed by laser beam illumination, revealed a 14-fold in-

crease in the level of HBsAg in the plasma compared to mice injected with the plasmid only. However, the level of HBsAg expression of the plasmid injected, followed by laser, was sustained for just 1 day (Fig. 3C, b). The i.d. LBGT method induced only marginal levels of apoptosis as assessed by the TUNEL assay (Fig. 3D, a). The femtosecond laser-based DNA immunization: elicitation of humoral and cellular immune responses after i.d. delivery of naked DNA To determine the potential of the optimized SG-LBGT system for DNA vaccination, the system was tested in a mouse model of HBV replication. BALB/c mice were

Figure 3. A) LBGT dermal gene delivery and gene expression. After anesthesia, the mice were shaved on the back and disinfected, then 10 ␮g of pCAG/Luc was injected into the dermis of the dorsal trunk followed 30 s later by a laser beam application (focus 400 ␮m, exposure time 2 min, energy 30 mW). We followed the in vivo real-time luciferase expression using the CCCD bioluminescence imaging system. Presented are the log-integrated light units emitted from the dermis skin above background relative to injected naked DNA without laser beam application. On day 43 there was a significant expression enhancement by LBGT over the plasmid only (statistical difference was determined using the 2-tailed t test (P⬍0.02). The data show mean of integrated light units ⫻ 103 obtained from 6 mice per group ⫾ sd. The data shown are representative of 3 separate experiments, all of which showed comparable results B) Dermal laser gene (␤-gal) transduction day 30. 12 ␮g/30 ␮l pCMV/LacZ were injected into the dermis, followed by application of laser beam (as above). 30 days later, skin biopsies at the injection site were fixed and stained with X-gal for ␤-galactosidase expression in whole-mount preparation (b, d). The tissues were paraffin-embedded and the sections were counterstained with hematoxylin and eosin (a, c). The expressing cells are adipocytes, smooth muscle, and hair follicles (a). C) LBGT-induced gene expression: secreted proteins. a) Intradermal injection of plasmid pCMVhSeAP 12 ␮g/30 ␮l, followed by pulsed laser. Data show mean serum levels on day 4 of hSeAp obtained from 5 mice per group ⫾ sd. Statistical difference was determined using the 2-tailed t test (*P⫽0.005). b) Intradermal injection of plasmid pCMVHBsAg (12 ␮g/30 ␮l), followed by pulsed laser. Data represent serum HBsAg levels on day 1. D) Apoptosis in the dermis. Marginal apoptosis assessed by the TUNEL assay was observed in the dermis after plasmid injection, followed by LBGT (yellow nuclei, panel a compared with plasmid only, panel b). There were 21% of apoptotic cells in the plasmid ⫹ laser groups vs. a 15% of apoptotic cells in the groups of plasmid only. LASER-FACILITATED DERMAL GENE DELIVERY AND GENETIC IMMUNIZATION

3527

immunized once i.d. with 10 ␮g plasmid pCMV/HBsAg with and without laser beam illumination. In the experiment presented in Fig. 4, the mice were bled on days 21, 35, 48, 93, and on day 210, when the experiment ended. We conducted three different sets of experiments, and specific HBsAg antibody titers were detected in all of them on day 21 after injection (data not shown). In the laser-treated group, the antibody titers continued to increase ⬎7 months and were up to 215-fold higher on day 93 and 22-fold higher on day 210 than titers obtained by naked DNA administration without SG-LBGT (Fig. 4). To evaluate the immune response with different doses of DNA, mice were immunized with 1 ␮g, 3 ␮g or 10 ␮g. The mice were bled on days 21, 48, 76, and 156 and the levels of HBsAg antibodies were assessed. We found that the dose of DNA for long-term antibody response up to day 156 was 10 ␮g, the highest dose used (data not shown). We also compared the immune response induced by a single dose of LBGT-based DNA immunization with that elicited by two doses of a commercially available recombinant HBsAg vaccine, Engerix® as described previously (9) (Fig. 5). The results show that the levels of HBsAg antibodies induced by the plasmid ⫹ laser are significantly higher than those elicited by plasmid only, and similar to the response induced by two doses of Engerix®. To determine the effect of repeated plasmid administration on the immune response, mice were immunized i.d. with 10 ␮g of the plasmid pCMV/HBsAg. Three weeks later the mice were divided into two groups: one group received a booster injection of plasmid DNA i.d. with and without the laser, and the other served as a control. The anti-HBsAg antibody

Figure 4. Antibody response to HBsAg up to day 210. A single dose of 10 ␮g/30 ␮l plasmid pCMV-HBsAg was injected i.d. to BALB/c mice (n⫽5/group), followed by laser beam application 30 s later. The data represent specific serum antibodies to HBsAg up to 210 days relative to plasmid injection without the application of the laser beam. The HBsAg antibody titer was determined using the AUSAB EIA and represent the geometric means ⫾ sd. The difference between the two groups at all time points is statistically significant. Statistical difference was determined using the 2-tailed t test. *P ⫽ 0.014 on day 35, *P ⫽ 0.05 on day 48, *P ⫽ 0.03 on day 93, and *P ⫽ 0.04 on day 210. One of three similar experiments is shown. 3528

Vol. 21

November 2007

titers were determined up to day 56. Surprisingly, in all three experiments booster injections were not necessary and did not result in increased antibody titers (data not shown). To characterize the pattern of the immune response elicited by i.d. injection of plasmid DNA followed by laser pulse, HBsAg-specific IgG1 and IgG2a serum antibodies were determined. LBGT-mediated naked DNA vaccination elicited both IgG1 and IgG2a antibodies. Vaccination with the plasmid followed by laser, resulted in a 67-fold higher IgG1 antibody titer than the injected plasmid alone; the IgG2a levels were similar in the two groups (Fig. 6A), suggesting a combined Th2 and Th1 response, respectively. In addition, LBGT vaccination produced a 7-fold increase in specific cell proliferative response to HBsAg polypeptide compared with that generated by Engerix® as well as with the plasmid injected without LBGT (Fig. 6B). Cytokine measurements in splenocytes stimulated with the antigen in vitro showed increased production of IFN␥ and low production of IL-4 in cells obtained from mice vaccinated with the plasmid ⫹ laser. In contrast, a single dose of Engerix® did not elicit IFN␥ production. Protection from viral challenge To determine whether the immune response elicited after i.d. LBGT-induced vaccination could protect animals from viral challenge, we used the HBV small animal model (7). Mice were injected i.d. with either 10 ␮g plasmid bearing the HBsAg (with or without the application of LBGT) or 1 ␮g Engerix®. Mice were bled on days 21 and 39. On day 46, all mice received an i.v. hydrodynamic injection (12 ␮g/1.5 ml PBS) of the plasmid encoding the full length of HBV (pHBVadwHTD) (7). A group of naive mice was also injected i.v. with the same plasmid and used as a positive control. As a negative control, another group of naive mice was administered an empty vector (pGEM-7) using the same method. On day 49, all mice were bled and the serum was tested for HBsAg and HBV DNA. The levels of HBsAg in all three vaccinated groups were very low compared with those of the positive control (Fig. 7A). In the LBGT immunized group, the level of HBsAg was 3-fold lower than that of mice injected with plasmid only. The protective effect of Engerix® could be attributed to the Th2 response, which caused an efficient neutralization of HBsAg (Fig. 7A). In parallel with the decrease in viral HBsAg in the vaccinated mice, there was also a marked reduction in viral replication as determined by the decreased level of HBV DNA compared with the positive control (reduction of 91% for vaccination with plasmid ⫹ LBGT, 59% for vaccination with plasmid only, and 81% for vaccination with Engerix®, Fig. 7B). In vivo tumor immunity induced by LBGT HBV-DNA vaccination To further determine the effectiveness of the elicited immune response, the ability of the LBGT-based DNA

The FASEB Journal

ZEIRA ET AL.

Figure 5. A single LBGT-based DNA immunization is comparable to a dose and boost of a commercially available recombinant HBsAg vaccine (Engerix®). A single dose of 10 ␮g/30 ␮l plasmid pCMV-HBsAg was injected i.d. to BALB/c mice (n⫽5/group) with or without application of the laser beam 30 s later. A third group (n⫽6) was injected i.d. with 1 ␮g of the commercial vaccine, Engerix®. Presented are the geometric means ⫾ sd of 5 animals for the three groups. Statistical difference was determined using the Kruskal-Wallis Test (nonparametric ANOVA) (K-W) (*P⫽0.004 on day 69) and the Dunn’s multiple comparisons test (plasmid⫹laser-1 injection vs. plasmid only-1 injection *P⬍0.05). This experiment was repeated twice with similar results.

vaccine to immunoprotect against implantation of syngeneic HBsAg-transfected tumor cells was examined. We followed the previously described protocol (15). Groups of 10 mice each were vaccinated once with either 10 ␮g of pCMV/HBsAg (with and without LBGT) or the commercial Engerix® vaccine (1 ␮g). Three weeks after vaccination, five mice from each group were inoculated s.c. with either 1 ⫻ 105 HBsAgexpressing colon carcinoma CT26/S tumor cells or the parental nontransfected CT26 tumor cells (H2d, syngeneic to BALB/c mice). A significant reduction of CT26/S tumor growth was observed only in mice immunized with plasmid followed by LBGT (11-fold vs. plasmid only and 4-fold vs. Engerix® on day 21) (Fig. 8A). Tumor immunity was HBsAg-specific since all immunized groups showed little protection against the parental CT26 tumor (Fig. 8B). Integration of the plasmid into the host genomic DNA Genotoxicity is one of the main hurdles in current preclinical and clinical protocols. To assess this safety issue for the LBGT-DNA delivery system, the potential for integration into the host genomic DNA was assessed using a gel purification-based assay. Briefly, BALB/c mice were injected once i.d with 10 ␮g of plasmid pCAG/Luc. Genomic DNA from injection site skintreated tissues (8 months post-treatment) was subjected to multiple rounds of preparative agarose gel electro-

phoresis to remove free extrachromosomal plasmid DNA. The purified genomic DNA was then eluted from the gel and assayed for potentially integrated plasmid using a highly sensitive real-time fluorescence quantitative PCR (QPCR) assay. Levels of the vector DNA for skin tissues were also assessed in the genomic DNA prior to gel purification. The results of this integration assay (Table 1) show that in the treated tissues assayed, gel purification removed the vast majority of the extrachromosomal plasmid DNA that existed prior to the gels (⬎99.9%), indicating that the level of integration is negligible after laser treatment. The residual levels seen in the tissues could represent incomplete purification of extrachromosomal plasmid and/or actual integration. Even if all of the plasmid copies were integrated, the level of integration is 1 or 2 orders of magnitude less than the spontaneous rate of geneinactivating mutations using a worst-case-scenario analysis (12). TABLE 1. Summary of integration assay results Sample

Laser skin,a prior to gel purification Laser skin, after 4⫻ gel purification

Copies/␮gb

9603 8

a Skin was assayed 8 months post-treatment. bCopies of luciferase plasmid/1 ␮g mouse genomic DNA (representing ⬃150,000 cells), as determined by TaqMan® QPCR.

DISCUSSION Nonviral gene therapy has significant clinical potential. Several studies have demonstrated the efficiency of this approach in animal models for both gene correction therapy and for DNA vaccination against several viral and tumor-associated antigens (18 –20). However, when these techniques are used, the levels of gene expression have been disappointing and side effects are evident—in particular, tissue damage and pain. Given the apparent poor transduction efficiency, the search for delivery systems to increase the uptake and expression of plasmid DNA has been the subject of intense research in recent years. Cationic lipids, polymers, and other chemical agents are currently being investigated for delivering DNA into cells. However, their modest activity and associated toxicity have presented inherent problems in animal and clinical studies. Various physical methods are also being investigated, including electroporation, particle bombardment, ultrasound, and hydrodynamic pressure (2, 13, 21–23). We previously demonstrated that our laser beam gene transduction method (LBGT) provided an efficient and safe approach for muscle targeted gene expression in rodents (3). We have now modified our laser with a flexible arm and a specialized lens assembly for accurate, high-fluency depth irradiation so that the laser is better directed to sites of gene inoculation. We have investigated the LBGT as a potential approach for

LASER-FACILITATED DERMAL GENE DELIVERY AND GENETIC IMMUNIZATION

3529

Figure 6. A) Antibody isotype responses elicited by laser-based DNA immunization against hepatitis B. BALB/c mice (n⫽8/group) were immunized once as described in the legend to Fig. 4. Sera collected on day 48 after immunization were tested by ELISA for IgG1 and IgG2a antibodies to HBsAg. The results show elicitation of both IgG1 and IgG2a antibodies to HBsAg in both groups. Data represent the mean titers ⫾ sd. Statistical difference was determined using the 2-tailed t test (plasmid⫹laser vs. plasmid only, *P⫽0.03). Data represent one of the two experiments that showed comparable results. B) HBsAg-specific lymphoproliferative response to HBsAg. Two groups of mice (n⫽8 each) were injected i.d. with 10 ␮g pCMVHBsAg with or without laser beam application. A third group (n⫽8) was injected i.d. with 1 ␮g of the commercial vaccine, Engerix®. Splenocytes obtained on day 45 were pooled and incubated in triplicate with or without recombinant HBsAg at 0.5 ␮g/ml for 4 days. The proliferative response is presented as mean ⌬ cpm (mean cpm of antigen-stimulated wells ⫺ mean cpm of control wells) ⫾ sd. Statistical difference was determined using the K-W Test (*P⫽0.008) and the Dunn’s multiple comparisons test (for plasmid⫹laser vs. plasmid only *P⬍0.01 and vs. Engerix® *P⬍0.05). Data represent one of two experiments that showed comparable results. C) Cytokine secretion. Splenocytes harvested on day 45 after vaccination were pooled (n⫽4/group) and incubated for 3 days with 25 ␮g/ml of recombinant HBsAg. Supernatants were assayed for IFN␥ and IL-4 using a “sandwich” ELISA technique. The values show means ⫾ sd. Statistical difference was determined using the K-W test (*P⫽0.02 for the IFN␥), the Dunn’s multiple comparisons test (for plasmid⫹laser vs. Engerix® *P⬍0.01), the K-W test (*P⫽0.05 for the IL-4), and the Dunn’s multiple comparisons test (for plasmid⫹laser vs. plasmid only *P⬍0.05). This experiment was repeated twice with similar results.

delivery of genes into the mouse dermis. Luciferase expression in the mouse dermis after i.d. injection of the plasmid, followed by exposure to LBGT, was 625fold higher on day 43 than that of the skin injected with naked DNA alone. The transfection was remarkably stable, with expression lasting for ⬎7 months. The generation of the LBGT system will enable us to perform preclinical studies in large animals. We believe that this new generation could be applied to clinical needs for both long-term expression and vaccination. The current major limitation of the apparatus is its overall size. The size of the system will need to be significantly reduced. The mechanism underlying the increased transduction efficiency is still unknown. A probable mechanism involves multiphoton dielectric breakdown of materials in limited regions without the generation of much associated heat or collateral damage. Such a multiphoton effect could account for temporary poration of the tissue at the selected depth. Further experimentations are being conducted to better understand that exact mechanism of gene delivery using LBGT. 3530

Vol. 21

November 2007

Laser-based gene delivery has several advantages over other traditional gene delivery systems using physical methods such as ultrasound or electroporation (3, 24).The current study demonstrates that LBGT can enhance gene delivery by increasing and extending the duration of transgene expression while reducing the need for multiple administrations: A single i.d. injection of a plasmid (10 ␮g), followed by an application of laser beam, resulted in efficient dermal transduction and specific immune responses. The transfected cells in the dermis were mainly adipocytes, hair follicles, and, in some cases, reached smooth muscle cells. The reporter ␤-galactosidase activity remained detectable 30 days post-i.d. injection of a plasmid encoding lacZ, followed by laser pulses. The damage caused by needle injection and laser beam pulses was very marginal or negligible as tested by the TUNEL apoptosis assay. Lasers are controllable and can be precisely focused to the site of injection at a depth determined for the specific target on the selected layer of cells (25). Localized vector delivery to specific tissues, such as the skin, can reduce widespread distribution, decrease po-

The FASEB Journal

ZEIRA ET AL.

Figure 7. Active immunization with HBsAg plasmid: protection against viral challenge. A) Mice (n⫽5/group) were immunized i.d. with the plasmid with or without LBGT, as described in the legend of Fig. 4. Another group was immunized i.d. with Engerix® (1 ␮g). 46 days after immunization, vaccinated mice and a group of naive mice (used as positive control) were injected i.v. with the plasmid pHBVadwHTD. As a negative control, another group of naive mice was administered with the empty vector pGEM-7. On day 49, sera of all animals were checked for levels of HBsAg. The results show mean levels ⫾ sd. B) Antiviral effect of laser-induced immunization. The same sera were assayed for HBV-DNA by real-time PCR. The data represent mean values of copy number ⫻ 105 ⫾ sd. Statistical difference was determined using the K-W test (*P⫽0.05) and the Dunn’s multiple comparisons test (plasmid⫹laser vs. positive control *P⬍0.05).

tential toxicity to nontargeted cells, and reduce inflammation (23, 26). Considering these advantages, we further demonstrated the efficiency and efficacy of this method of nonviral delivery in a genetic immunization model. Immunization involved introduction of a plasmid DNA (10 –12 ␮g) encoding the HBsAg by a single i.d. dose, followed by pulses of LBGT. Although few cells could be transfected using this method, the amount of protein produced led to potent Th1 and Th2 immune responses without any additional adjuvant. Specific HBsAg antibodies were already detected after 3 wk and continued to increase over a period of 7 months. At all time points tested, the antibody levels obtained with plasmid followed by LBGT were significantly higher than those obtained with plasmid only. The SG-LGBT method also induced T cell activation with IFN␥ and IL-4 secretion by antigen restimulated splenocytes in vitro. Based on the IgG1 and IGg2a titers and the cytokine profile (Fig. 6), the immune

response induced by SG-LBGT was a mixed Th1 and Th2 response, which may prove to be beneficial in therapeutic tumor and infectious disease applications. The SG-LBGT method of vaccination also produced a specific cell proliferative response to HBsAg that was 7-fold higher than by the commercial vaccine, Engerix®. In addition, the lasermediated DNA immunization elicited HBsAg antibodies capable of neutralizing the virus and preventing infection after a single injection whereas the commercial vaccine required three vaccine doses, as shown in other studies (9). Nonviral gene therapy approaches targeting tumors represent another application that has been studied in animal models (27, 28). One of the goals of these applications is to enhance the host’s immune response against tumors in a prophylactic or a therapeutic setting (23). We showed that a single injection of a plasmid DNA encoding HBsAg, followed by LBGT, significantly inhibited tumor growth in mice implanted

Figure 8. Tumor protection induced by LBGT HBV-DNA vaccination. BALB/c mice (n⫽10/group) were immunized as described in the legend of Fig. 7A. Three wk after vaccination, 5 mice of each group were inoculated s.c. with 1 ⫻ 105 HBsAg-expressing CT26/S (A) or the parental nontransfected CT26 tumor cells (B). Data represent the mean values of tumor volume (cm3). Statistical difference was determined using the K-W test (*P⫽0.05) and the Dunn’s multiple comparisons test (plasmid⫹laser vs. plasmid only *P⬍0.05) on day 21. One of three similar experiments is shown. LASER-FACILITATED DERMAL GENE DELIVERY AND GENETIC IMMUNIZATION

3531

with a syngeneic HBsAg-transfected colon carcinoma tumor, suggesting the elicitation of tumor-specific, cellmediated immunity. Integration to the host’s genome is the primary safety concern for DNA vaccines (29, 30) since, by definition, it is insertional mutagenesis that has the potential to activate oncogenes or inactivate tumor suppressor genes. The integration assay performed on laser-treated skin showed that SG-LBGT-treatment carried a negligible risk of plasmid integration into the host cell genomic DNA. Collectively, these studies demonstrate that SG-LBGT is a safe method for nonviral naked DNA delivery into the dermis, and it has also proved to be efficient for long-term transgene expression. A single injection of plasmid DNA (10 ␮g) into the mouse dermis combined with laserenhanced gene delivery is sufficient for eliciting strong and long-lasting humoral and cellular immune responses against the protein encoded by the expression vector. Although the mechanism of the LBGT immunization method has not yet been elucidated, these results provide a proof-of-principle of its feasibility and suggest that this technique may improve the efficacy of long-lasting gene delivery and expression. The use of this SG-LBGT system with its mobile arm for precise guided targeting can be optimized so that gene delivery can also be accomplished in large animal models. Moreover, its successful application in genetic immunization in animal models makes this method attractive to implementation in clinical studies. We thank Ms. Mery Clausen for assistance in proofreading and editing of the manuscript and Ms. Meital Guy for secretarial assistance. E.G. is supported by the Israeli Ministry of Science through a grant to the National Gene Therapy Knowledge Center and through EC grants LSHBCT-2004 –512034 (MOLEDA) and LSHB-CT-2005– 018961 (INTHER). E.G. is also supported by grants from the Blum, the Harold Grinspoon, and the Horowitz and Wolfson Foundations. E.Z., E.K., A.L., and E.G. contributed to project conception and experimental design, acquisition, analysis, and interpretation of data and were involved in drafting and revising the final manuscript; E.Z., A.M. and Z.M. performed the experimental part and carried out all laser and animal experiments; M.G., participated in plasmid construction, generated stable-transfected clones, performed and analyzed in vitro cell culture assays, and drafted the manuscript; N.D. and R.B. carried out the serum studies and contributed to the gene expression studies and statistical analysis; M.H. and H.Y. participated in animal experimentation and contributed to the gene expression studies; P.J.T. participated in the design of the integration study, performed the statistical analysis, and helped draft the manuscript; T.G.G., S.J.P., D.F.R, Z.N., and O.N. performed and designed some of the experiments, designed PCR primers and probes, performed PCR and real-time PCR assays, and provided intellectual input to the work; G.Z. and O.P. were involved in histopathological and immunohistochemical studies and image analysis, and helped draft the manuscript; I.H. contributed to project conception and experimental design, and helped draft the manuscript. All authors read and approved the final manuscript. 3532

Vol. 21

November 2007

REFERENCES 1. 2.

3.

4.

5. 6.

7.

8.

9.

10.

11. 12.

13.

14.

15.

16.

17.

Wells, D. J. (2004) Gene therapy progress and prospects: electroporation and other physical methods. Gene Ther. 11, 1363–1369 Regnier, V., Tahiri, A., Andre´, N., Lemaıˆtre, M., Le Doan, T., and Pre´at, V. (2000) Electroporation-mediated delivery of 3⬘protected oligodeoxynucleotides to the skin. J. Control Release 67, 337–346 Zeira, E., Manevitch, A., Khatchatouriants, A., Pappo, O., Hyam, E., Darash-Yahana, M., Tavor, E., Honigman, A., Lewis, A., and Galun, E. (2003) Femtosecond infrared laser-an efficient and safe in vivo gene delivery system for prolonged expression. Mol. Ther. 8, 342–350 Berns, M. W., Aist, J., Edwards, J., Strahs, K., Girton, J., McNeill, P., Rattner, J. B., Kitzes, M., Hammer-Wilson, M., Liaw, L. H., et al. (1981) Laser microsurgery in cell and developmental biology. Science 213, 505–513 Berns, M. W., Wright, W. H., and Wiegand Steubing, R. (1991) Laser microbeam as a tool in cell biology. Int. Rev. Cytol. 129, 1– 44 Ilan, E., Burakova, T., Dagan, S., Nussbaum, O., Lubin, I., Eren, R., Ben-Moshe, O., Arazi, J., Berr, S., Neville, L., Yuen, L., Mansour, T. S., Gillard, J., Eid, A., Jurim, O., Shouval, D., Reisner, Y., and Galun, E. (1999) The hepatitis B virus-trimera mouse: a model for human HBV infection and evaluation of anti-HBV therapeutic agents. Hepatology 29, 553–562 Ketzinel-Gilad, M., Zauberman, A., Nussbaum, O., Shoshany, Y., Ben-Moshe, O., Pappo, O., Felig, Y., Ilan, E., Wald, H., Dagan, S., and Galun, E. (2006) The use of the hydrodynamic HBV animal model to study HBV biology and anti-viral therapy. Hepatol. Res. 34, 228 –237 Stevenson, D., Agate, B., Tsampoula, X., Fischer, P., Brown, C. T. N., Sibbett, W., Riches, A., Gunn-Moore, F., and Dholakia, K. (2006) femtosecond optical transfection of cells: viability and efficienty. Optics Express 14, 7125–7133 Ilan, Y., Nagler, A., Zeira, E., Adler, R., Slavin, S., and Shouval, D. (2000) Maintenance of immune memory to the hepatitis B envelope protein following adoptive transfer of immunity in bone marrow transplant recipients. Bone Marrow Transplant. 26, 633– 638 Honigman, A., Zeira, E., Ohana, P., Abramovitz, R., Tavor, E., Bar, I., Zilberman, Y., Rabinovsky, R., Gazit, D., Joseph, A., Panet, A., Shai, E., Palmon, A., Laster, M., and Galun, E. (2001) Imaging transgene expression in live animals. Mol. Ther. 4, 239 –249 Duguez, S., Feasson, L., Denis, C., and Freyssenet, D. (2002) Mitochondrial biogenesis during skeletal muscle regeneration. Am. J. Physiol. 282, E802–E809 Ledwith, B. J., Manam, S., Troilo, P. J., Barnum, A. B., Pauley, C. J., Griffiths, T. G., 2nd, Harper, L. B., Beare, C. M., Bagdon, W. J., and Nichols, W. W. (2000) Plasmid DNA vaccines: investigation of integration into host cellular DNA following intramuscular injection in mice. Intervirology 43, 258 –272 Kwissa, M., von Kampen, K., Zurbriggen, R., Gluck, R., Reimann, J., and Schirmbeck, R. (2000) Efficient vaccination by intradermal or intramuscular inoculation of plasmid DNA expressing hepatitis B surface antigen under desmin promoter/ enhancer control. Vaccine 18, 2337–2344 Van Slooten, M. L., Hayon, I., Babai, I., Zakay-Rones, Z., Wagner, E., Storm, G., and Kedar, E. (2001) Immunoadjuvant activity of interferon-gamma-liposomes co-administered with influenza vaccines. Biochim. Biophys. Acta 1531, 99 –110 Chow, Y. H., Chiang, B. L., Lee, Y. L., Chi, W. K., Lin, W. C., Chen, Y. T., and Tao, M. H. (1998) Development of Th1 and Th2 populations and the nature of immune responses to hepatitis B virus DNA vaccines can be modulated by codelivery of various cytokine genes. J. Immunol. 160, 1320 –1329 Weiss, L., Barak, V., Zeira, M., Abdul-Hai, A., Raibstein, I., Reich, S., Hirschfeld, E., Gross, D., and Slavin, S. (2002) Cytokine production in Linomide-treated nod mice and the potential role of a Th (1)/Th(2) shift on autoimmune and anti-inflammatory processes. Cytokine 19, 85–93 Spirito, F., Meneguzzi, G., Danos, O., and Mezzina, M. (2001) Cutaneous gene transfer and therapy: the present and the future. J. Gene Med. 3, 21–31

The FASEB Journal

ZEIRA ET AL.

18. 19. 20. 21. 22. 23. 24.

Haupt, K., Roggendorf, M., and Mann, K. (2002) The potential of DNA vaccination against tumor-associated antigens for antitumor therapy. Exp. Biol. Med. (Maywood) 227, 227–237 Donnelly, J. J., Ulmer, J. B., and Liu, M. A. (1995) Protective efficacy of intramuscular immunization with naked DNA. Ann. N. Y. Acad. Sci. 772, 40 – 46 Wolff, J. A., Malone, R. W., Williams, P., Chong, W., Acsadi, G., Jani, A., and Felgner, P. L. (1990) Direct gene transfer into mouse muscle in vivo. Science 247, 1465–1468 Pavselj, N., and Preat, V. (2005) DNA electrotransfer into the skin using a combination of one high- and one low-voltage pulse. J. Control Release 106, 407– 415 Lucas, M. L., Jaroszeski, M. J., Gilbert, R., and Heller, R. (2001) In vivo electroporation using an exponentially enhanced pulse: a new waveform. DNA Cell Biol. 20, 183–188 Vogel, J. C. (2000) Nonviral skin gene therapy. Hum. Gene Ther. 11, 2253–2259 Lee, W. R., Shen, S. C., Liu, C. R., Fang, C. L., Hu, C. H., and Fang, J. Y. (2006) Erbium:YAG laser-mediated oligonucleotide and DNA delivery via the skin: an animal study. J. Control Release 115, 344 –353

25. 26. 27.

28.

29. 30.

Tirlapur, U. K., and Konig, K. (2002) Targeted transfection by femtosecond laser. Nature 418, 290 –291 Glasspool-Malone, J., Somiari, S., Drabick, J. J., and Malone, R. W. (2000) Efficient nonviral cutaneous transfection. Mol. Ther. 2, 140 –146 Weichselbaum, R. R., Hallahan, D. E., Beckett, M. A., Mauceri, H. J., Lee, H., Sukhatme, V. P., and Kufe, D. W. (1994) Gene therapy targeted by radiation preferentially radiosensitizes tumor cells. Cancer Res. 54, 4266 – 4269 Kim, J. J., Yang, J. S., Dang, K., Manson, K. H., and Weiner, D. B. (2001) Engineering enhancement of immune responses to DNA-based vaccines in a prostate cancer model in rhesus macaques through the use of cytokine gene adjuvants. Clin. Cancer Res. 7, 882s– 889s Smith, H. A. (1994) Regulatory considerations for nucleic acid vaccines. Vaccine 12, 1515–1519 Robertson, J. S. (1994) Safety considerations for nucleic acid vaccines. Vaccine 12, 1526 –1528

LASER-FACILITATED DERMAL GENE DELIVERY AND GENETIC IMMUNIZATION

Received for publication October 17, 2006. Accepted for publication May 10, 2007.

3533