| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* 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;
|| 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
2Correspondence: The Goldyne Savad Institute of Gene Therapy, Hadassah University Hospital, P.O. Box 12000, Jerusalem, 91120 Israel. E-mail: eithang{at}hadassah.org.il
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 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 laser: a new intradermal DNA delivery method for efficient, long-term gene expression and genetic immunization.
Key Words: gene delivery • vaccination • HBV • antitumor
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. 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.
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 |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. 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 x 0.45 numerical aperture objective. The relatively large numerical aperture allows the focal volume to be limited to a region of 
0.5 µm 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.
|
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 µm 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 µm 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 ß-actin promoter including the first intron) from plasmid PBS-CAG-M2nls-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 pXL3010-SeAp (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 ß-galactosidase gene expression, we used the pCMVß-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 x 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 µg 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 µ. 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 x40 (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 used. Mice were anesthetized with a single intraperitoneal dose of xylazine (11 µg/g) and ketamine hydrochloride (110 µg/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 x 5 mm2. Each mouse received 1 to 12 µg of plasmid DNA in 30–50 µl 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 µg/50 µl), (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 ß-galactosidase activity analysis, formalin-fixed, paraffin-embedded skin samples were cut into sections 4 µm 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 ß-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 µg 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 µm 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 µg 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 4x 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 plasmid copies/µg genomic DNA. Two negative controls were also run with the treated samples: control mouse DNA and control rat DNA that had been 4 x 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.
|
Analysis of secreted proteins for gene expression measurement
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 murine-secreted alkaline phosphatase was carried out with the Tropix® Phospha-LightTM 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 retro-orbital 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 µg/ml in 50 µl 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 µl) 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 µl of 2 x 106 cells/ml splenocytes in complete RPMI 1640 [supplemented with 10% fetal bovine serum (GIBCO, Grand Island, NY, USA, 100 U/ml penicillin, 100 µg 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 µg/ml. Cells in all wells were cultured in a total volume of 200 µl of medium. After 4 days in culture, the cells were pulsed with thymidine (1 µCi/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 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)
.
Anti-viral effect of LBGT-induced immunization
BALB/c mice were immunized i.d. with the HBsAg plasmid (10 µg), with or without the application of LBGT. Another group was injected i.d. with Engerix® (1 µg/50 µl) (9)
. On day 46, vaccinated mice received an intravenous hydrodynamic injection of the plasmid bearing the full length of the HBV mentioned above (12 µg 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 µl 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-AAT-TGG-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-CTG-CGC-ACC-AGC-ACC-ATG-CA-TAMRA-3'. The reaction was carried out in 50 µl of a PCR mixture containing 1 x Taq Pol. buffer, 7.5 mM MgCl2, 0.4 mM of each dNTP, 50 pmol of each primer, 12 pmol probe, 0.6 µM 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 µg HBsAg plasmid only; 2) 10 µg HBsAg plasmid followed 30 s later by pulse laser, 3) 1 µg Engerix®. Three weeks after immunization, mice were challenged subcutaneously in the left lateral flank with 1 x 105 CT26 (colon carcinoma) cells or the HBsAg gene stable-transfected CT26/S cells. Tumor volume was measured and calculated using the formula: V = length x width2 x 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 
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. To establish feasibility of this method for human clinical trials, this second-generation 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 µm to 400 µm), 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 µm, 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. 2
a, 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 µg 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 (2x106 integrated light units), relative to naked DNA injection, without the application of the laser beam (3x103 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. 3
A).
|
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 ß-galactosidase gene expression analysis in the dermal tissue.
Successful i.d. transfection was confirmed by ß-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 increase 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 immunized once i.d. with 10 µg 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 µg, 3 µg or 10 µg. 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 µg, 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 µg 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 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. 6
A), 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 µg plasmid bearing the HBsAg (with or without the application of LBGT) or 1 µg Engerix®. Mice were bled on days 21 and 39. On day 46, all mice received an i.v. hydrodynamic injection (12 µg/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. 7
A). 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 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 µg of pCMV/HBsAg (with and without LBGT) or the commercial Engerix® vaccine (1 µg). Three weeks after vaccination, five mice from each group were inoculated s.c. with either 1 x 105 HBsAg-expressing 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. 8
A). 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 µg of plasmid pCAG/Luc. Genomic DNA from injection site skin-treated tissues (8 months post-treatment) was subjected to multiple rounds of preparative agarose gel electrophoresis 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 gene-inactivating mutations using a worst-case-scenario analysis (12)
.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
19
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
22
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 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 625-fold 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.
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 µg), 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 ß-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 potential 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 µg) 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 laser-mediated 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 with a syngeneic HBsAg-transfected colon carcinoma tumor, suggesting the elicitation of tumor-specific, cell-mediated 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 µg) into the mouse dermis combined with laser-enhanced 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.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication October 17, 2006. Accepted for publication May 10, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
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 µg/ml of recombinant HBsAg. Supernatants were assayed for IFN
