Particle-mediated gene transfer into murine livers using a newly developed gene gun

Transcription

1 (2000) 7, Macmillan Publishers Ltd All rights reserved /00 $ NONVIRAL TRANSFER TECHNOLOGY BRIEF COMMUNICATION using a newly developed gene gun S Kuriyama, A Mitoro, H Tsujinoue, T Nakatani, H Yoshiji, T Tsujimoto, M Yamazaki and H Fukui Third Department of Internal Medicine, Nara Medical University, 840 Shijo-cho, Kashihara, Nara , Japan Although particle-mediated gene transfer using gene gun technology has been applied for gene transfer into epidermis, applications of this technology to visceral tissues have not been well investigated. Although all helium gas-driven gene gun instruments have used macrocarriers to discharge DNA-coated microprojectiles so far, we used a newly developed gene gun instrument, in which a hammering bullet is used to discharge microprojectiles. With the gene gun, gold particles coated with lacz expression plasmid were discharged to murine livers. LacZ expression was induced much more profoundly in the liver by particle-mediated gene transfer than by simple plasmid injection and electropor- ation-mediated gene transfer. LacZ expression was broadly and randomly distributed throughout the bombarded livers, indicating that particle-mediated gene transfer can induce transgene expression even at relatively distant areas from the surface of the bombarded tissue. Furthermore, although transgene expression was at its peak on day 2 after the bombardment, it was still detectable even on day 28. These results indicate that particle-mediated gene transfer with a newly developed gene gun may provide a new approach to gene therapy for human diseases. (2000) 7, Keywords: gene gun; particle bombardment; liver; in vivo gene transfer; expression; lacz gene Although viral vectors or the vector-producing cells are being tested for in vivo gene transfer, 1 6 the use of nonviral in vivo gene transfer technologies has been much more limited. However, steady progress has been made in several areas of nonviral gene transfer methods. Particle-mediated gene transfer using a gene gun was originally developed for introducing DNA into plants This technology involves propelling high-density microparticles at high velocity to transfer DNA through the cell membrane. The gene gun technique provides a simple means of directly transferring genes not only into mammalian cells in vitro, but also into tissues of living animals. 11 Particle-mediated gene transfer techniques have been intensively investigated as a novel method for DNA vaccination against a wide variety of infectious diseases More recently, this technology has been employed as a method for cancer immunotherapy Although various designs of gene gun instruments using helium gas have been described previously, all devices use macrocarriers for accelerating DNA-coated microcarriers. 22,23 Furthermore, since particle-mediated in vivo gene transfer has been performed mainly to epidermis, applications of this technology to visceral tissues have not been well investigated. In the present study, we used a newly developed gene gun instrument that uses a hammering bullet but not a macrocarrier, and examined the applicability of the gene gun to murine livers for in vivo gene therapy. Correspondence: S Kuriyama Received 22 December 1999; accepted 28 February 2000 In the previously devised gene gun instruments (Figure 1a), helium gas is used to pressurize the gas acceleration tube, which is closed at one end by a rupture disk. The rupture disk is capable of maintaining the helium in the tube until a preset pressure is reached, at which time the disk will break. The resulting shock wave accelerate Figure 1 Schematic representation of gene guns with a macrocarrier (a) and with a hammering bullet (b). Bold arrows represent direction of the helium flow. Closed circles represent discharged DNA-coated gold particles. (a) Gas acceleration tube; (b) rupture disks before and after the bombardment; (c) macrocarriers with DNA-coated gold particles before and after the bombardment; (d) stopping screen; (e) target tissues; (f) discharge tube; (g) hammering bullets before and after the bombardment; (h) vibration plate with DNA-coated gold particles; (i) stopping plate.

2 a plastic membrane (macrocarrier) carrying high-density microparticles into a mesh stainless steel stopping screen. The screen allows the microparticles to pass through and reach the target tissue, but retains the macrocarrier. In this system, the shock wave also reaches the target tissue, resulting in the damage of the target tissue. The gene gun instrument (PIGG-X; Nippon Medical & Chemical Instruments, Osaka, Japan) used in the present study also utilizes helium gas but does not contain a macrocarrier (Figure 1b). A hammering bullet is used instead of a macrocarrier. The motive force to accelerate the hammering bullet is generated by release of a high-pressure burst of helium gas and carries the hammering bullet to the vibration plate made of titanium through the discharge tube. By impact of the hammering bullet to the vibration plate, DNA-coated gold particles affixed to the vibration plate are discharged into the target tissue. Since the pressurized gas does not come into direct contact with DNA-coated gold particles, it does not damage, scatter or contaminate the target tissue. The operation procedure related to DNA transfer is highly standardized and simplified, and can be readily used. The pch110 plasmid, 24 which contains the bacterial galactosidase (lacz) gene under the transcriptional control of the simian virus 40 early promoter, and 12-weekold female ICR mice, weighing g, were used in the experiments. Animal experiments were performed with the approved protocols and in accordance with the recommendations for the proper care and use of laboratory animals. Mice were randomly separated into the naive control, direct plasmid injection, electroporationmediated gene transfer, particle-mediated gene transfer and mock treatment groups. Each group consisted of five animals. Animals in the direct plasmid injection group were given an injection of 20 g of the pch110 plasmid contained in 100 l of phosphate-buffered saline into the left-lateral hepatic lobe. Animals in the electroporationmediated gene transfer group were also given an injection of 20 g of the pch110 plasmid into the left-lateral hepatic lobe, and then received the local delivery of direct current, square wave electric pulses (1000 V/cm, 99 s, 1 Hz, 8 pulses) through forceps-shaped electrodes firmly suppressing the left-lateral hepatic lobe as described previously. 25 The amount of plasmid DNA injected into the liver was decided, referring to the previous report. 26 For particle-mediated gene transfer into the liver, a vibration plate coated with 0.2 mg gold particles containing 0.8 g of the pch110 plasmid was set in the gene gun. The procedure for associating plasmid DNA and gold particles is described in the legend of Figure 2. DNA transfer was performed directly over the liver using a 150 lb/in 2 (psi) helium pulse to accelerate the hammering bullet. Animals in the mock treatment group were bombarded by the gene gun carrying a vibration plate coated with only gold particles. Two days after particle-mediated transfer of the lacz gene into left-lateral hepatic lobes of mice, animals were killed and their livers were removed. The left-lateral hepatic lobe was then sliced into 1-mm thick sections from the surface. Three sections were prepared for each mouse, because the depth of the left-lateral hepatic lobe of an ICR mouse was approximately 3 mm. As shown in Figure 2, background levels of -galactosidase activity of the whole left-lateral hepatic lobes of naive control animals were 144 ± 31 pg/mg protein. Levels of -galactosidase activity of the whole left-lateral hepatic lobes of Figure 2 Quantitative estimation of lacz gene expression in murine livers. For transferring plasmid DNA into the liver, laparotomy was carried out. Mice were given an intraperitoneal injection of 2.5% avertin (0.012 ml/g body weight). After the animals were completely anesthetized, the liver was exposed by making a central incision starting from the xiphoid process, extending approximately 1 cm toward the caudal surface of the mouse. The left-lateral hepatic lobe was exposed by drawing it out of the incision taking precautions not to cause any tissue damage. Left-lateral hepatic lobes of animals in the naive control, direct plasmid injection, electroporation-mediated gene transfer and mock treatment groups were homogenized and the -galactosidase activity in the homogenates was quantitatively estimated by a reporter gene assay (Galacto-Star; Tropix, Bedford, MA, USA) using chemiluminescent 1,2-dioxetane as a substrate. The procedures were performed according to the protocol provided by the manufacturer. A standard curve was generated by serially diluting the reagent grade -galactosidase (G-5635; Sigma, St Louis, MO, USA) in the lysis buffer provided by the manufacturer. -Galactosidase activity was standardized based on the protein content of the homogenates, using BioRad (Hercules, CA, USA) protein assay with bovine serum albumin as standard. For particle-mediated gene transfer using a newly developed gene gun, plasmid DNA and gold particles were associated as follows. Five milligrams of gold particles with 1.0- m diameter were suspended in 400 l of 100% ethanol and sonicated for 10 s in a 1.5-ml microtube. After centrifugation at r.p.m. for 1 s, the supernatant was discarded and the gold pellet was air-dried in a clean bench. Twenty micrograms of the pch110 plasmid DNA contained in 40 l of10mm Tris-HCl (ph 8.0) and 1 mm EDTA (ph 8.0) was added into the microtube followed by vortex. Four microliters of 3 m sodium acetate (ph 7.8) was added into the microtube followed by vortex, and 100 l of 100% ethanol was added followed by vortex. The microtube was then kept at 80 C for 30 min followed by centrifugation at r.p.m. at 4 C for 5 min. The supernatant was discarded and 500 l of 100% ethanol was added into the microtube followed by sonication for 3 s. Twenty microliters of the suspension containing DNA-coated gold particles was spread on a single side of each vibration plate in a clean bench, resulting in affixing 0.2 mg gold particles containing 0.8 g DNA per plate. After air-drying, the vibration plate was set in the gene gun. Left-lateral hepatic lobes of animals that received particle-mediated gene transfer were sliced from the surface into 1-mm thick sections, homogenized and subjected to the galactosidase assay. Right-lateral hepatic lobes of animals in the particlemediated gene transfer group were also subjected to the -galactosidase assay. Each group consisted of five animals. -Galactosidase activity of left-lateral hepatic lobes that received particle-mediated gene transfer was significantly higher than that of other liver samples at P using Welch s or Student s t test. There were no significant differences in galactosidase activity among the 1-mm thick sections of the left-lateral hepatic lobes that received particle-mediated gene transfer. There were no significant differences in -galactosidase activity among other liver samples. animals in the direct plasmid DNA injection and electroporation-mediated gene transfer groups were 177 ± 43 and 193 ± 35 pg/mg protein, respectively. Levels of galactosidase activity of the whole left-lateral hepatic lobes of animals in the mock group that received DNA- 1133

3 1134 uncoated gold particles into the left-lateral hepatic lobes were 131 ± 19 pg/mg protein. Conversely, left-lateral hepatic lobes of animals that received DNA-coated gold particles discharged by the gene gun exhibited considerable levels of -galactosidase activity. Levels of -galactosidase activity of the left-lateral hepatic lobes located from the surface to 1 mm in depth, from 1 to 2 mm in depth and from 2 to 3 mm in depth were 3831 ± 801, 4293 ± 1013 and 3968 ± 846 pg/mg protein, respectively. There were no significant differences in -galactosidase activity among the 1-mm thick liver sections, indicating that particle-mediated gene transfer resulted in broadly and randomly distributed transgene expression in bombarded hepatic lobes of mice. Conversely, levels of -galactosidase activity in non-bombarded right-lateral hepatic lobes of animals that received DNA-coated gold particles into left-lateral hepatic lobes were 124 ± 37 pg/mg protein, with no significant differences compared with those of naive control animals. It was reported that the simply direct injection of plasmid DNA containing a coding sequence of the firefly luciferase gene into rat livers resulted in detectable levels of transgene expression. 26 However, when we injected 20 g of a lacz expression plasmid, corresponding to 25 times of DNA used in particle-mediated in vivo gene transfer experiments, directly into the liver, levels of galactosidase activity of the injected hepatic lobes were not significantly higher than background levels. Furthermore, when 20 g of a lacz expression plasmid was injected directly into the liver followed by the delivery of direct current, square wave electric pulses to the injected hepatic lobes using forceps-shaped electrodes, levels of -galactosidase activity of electroporated hepatic lobes were not significantly higher than background levels. These results may be due to the rapid clearance of plasmid DNA directly injected into the liver by the ample blood supply in the liver, resulting in no significantly increased levels of -galactosidase activity. In contrast, it is supposed that accelerated DNA-coated gold particles can enter the cells in the liver without being washed away by the blood flow, resulting in considerable levels of transgene expression. It has been reported that compared with gene transfer by means of electroporation, lipofection and diethylaminoethyl dextran, particlemediated gene transfer could induce 50- to 240-fold higher levels of transgene expression in vitro. 23 Our results indicate that particle-mediated gene transfer into the liver appears to be much more promising than direct plasmid injection and electroporation-mediated gene transfer into the liver. Furthermore, particle-mediated gene transfer has the advantage of requiring a much lower amount of DNA compared with direct plasmid injection and electroporation-mediated gene transfer. To examine the duration of transgene expression in the liver, animals were killed at various time-points after particle-mediated gene transfer into left-lateral hepatic lobes, and levels of -galactosidase activity were estimated. As shown in Figure 3, levels of -galactosidase activity of bombarded hepatic lobes were at the peak on day 2 after particle-mediated gene transfer, with values being 4492 ± 807 pg/mg protein, followed by the decrease of transgene expression. Levels of -galactosidase activity of bombarded hepatic lobes on days 4, 7, 14 and 28 were 1984 ± 352, 749 ± 130, 526 ± 43 and 429 ± 62 pg/mg protein, respectively. Although levels of -galactosidase Figure 3 Time-course of lacz gene expression in the liver induced by particle-mediated gene transfer. Mice that received particle-mediated transfer of the lacz gene into left-lateral hepatic lobes were killed at various timepoints. Each group consisted of five animals. The lobes were homogenized and subjected to the -galactosidase assay. There were significant differences in -galactosidase activity between days 2 and 4, between days 4 and 7, between days 7 and 14, and between days 14 and 28, at P 0.001, P 0.001, P 0.01 and 0.01 P 0.05, respectively using Welch s or Student s t test. activity in the liver were significantly different among the estimated time-points, the difference between days 14 and 28 was not substantial. Microscopic histological examination of the liver bombarded by the newly developed gene gun revealed that the procedure did not cause any apparent tissue damage at all the time-points examined (data not shown). Expression of the lacz gene of bombarded hepatic lobes was then examined histochemically by the X-gal staining. After particle-mediated gene transfer, animals were killed at the previously described time-points, and bombarded left-lateral hepatic lobes were sliced into 35- m thick sections. A considerable number of clearly outlined X-gal staining-positive areas were observed along with diffusely stained areas in bombarded left-lateral hepatic lobes on day 2 after particle-mediated gene transfer at a low magnification (Figure 4a). The diffuse signal observed may be due to the diffusion of -galactosidase from transduced cells or cell clusters to neighboring cells. Histochemical examination at a high magnification revealed that X-gal staining positive cells involved parenchymal hepatocytes (Figure 4b). However, we could not exclude the possibility that non-parenchymal septal cells were also transduced with the lacz gene, because the gene gun system is considered to transfer an exogenous gene to all types of cells. There were no X-gal staining positive areas in non-bombarded hepatic lobes (data not shown). X-gal staining positive areas were decreased both in number and in size with the passage of the time. However, X-gal staining positive areas stained less outlined and weaker compared with those on day 2 were observed even on day 28 after particle-mediated gene transfer (Figure 4c). We could not exclude the possibility that the strong -galactosidase activity that was observed

4 a b c simple plasmid DNA injection, in which gene transfer and expression occurred only along with the track of the injection needle. 27 Efficient gene transfer and expression in relatively extended macroscopic organ areas makes in vivo particle-mediated gene transfer an attractive approach for local cancer gene therapy. Our results suggest that particle-mediated in vivo gene transfer may be effective in transfecting a therapeutic gene to a sufficient proportion of cells in tumors accessible to the gene gun. Gene transfer by a gene gun, which uses plasmid DNA as the vector, has several clinically relevant advantages over gene transfer using viral vectors. First, large quantities of stable and highly purified plasmid DNA can be readily and inexpensively obtained. Second, the gene of interest is transcribed without inducing immunological interference from viral vector proteins, which may permit repeated in vivo gene transfer. Third, there is less likelihood of recombination events with the cellular genome, eliminating the risk of the insertional mutagenesis that is associated with the use of viral vectors. Fourth, because there are fewer size constraints than with current viral vectors, plasmid vectors can carry larger genes. Fifth, it is feasible to allow co-delivery of two or more different plasmid constructs into tissues by a gene gun instrument, which may enhance the therapeutic effect. Lastly, effects of new plasmid constructs can be examined rapidly in vivo. Further investigations have to be performed to estimate the parameters, such as the pressure for accelerating the hammering bullet, number of bombardment, DNA amount mixed with gold particles. By optimizing these parameters, the efficacy of particle-mediated gene transfer may be improved further. Further improvements to gene gun technology may provide a new approach to gene therapy for human diseases. Acknowledgements The authors are grateful to D Yong Mei, I Kijihana and A Kijihana of Nippon Medical & Chemical Instruments for technical assistance with the gene gun. This work was supported in part by a Grand-in-Aid for Scientific Research (B ) from the Ministry of Education, Sciences, Sports and Culture of Japan Figure 4 X-gal staining of the liver that received particle-mediated transfer of the lacz gene. Animals were killed on day 2 (a and b) and on day 28 (c) after particle-mediated gene transfer into the liver. The livers were removed, sliced into 35- m thick sections and incubated overnight at 37 C in an X-gal (5-bromo-4-chloro-3-indolyl- -galactosidase; GIBCO, Grand Island, NY, USA) reaction mixture to examine lacz gene expression as described previously. 28 A representative picture is shown in the figure (original magnification, a and c 40, b 400). in restricted areas on day 28 was due to integration of the plasmid vector. Transgene-expressing cells were broadly and randomly distributed throughout the bombarded hepatic lobe. This broad distribution of transgene expression is different from expression reported after References 1 Cao G et al. Complete regression of established murine hepatocellular carcinoma by in vivo tumor necrosis factor gene transfer. Gastroenterology 1997; 112: Cao G et al. Effective and safe gene therapy for colorectal carcinoma using the cytosine deaminase gene directed by the carcinoembryonic antigen promoter. 1999; 6: Kuriyama S et al. Complete cure of established murine hepatocellular carcinoma is achievable by repeated injections of retrovirus carrying the herpes simplex virus thymidine kinase gene. 1999; 6: Kuriyama S et al. Transient cyclophosphamide treatment before intraportal readministration of an adenoviral vector can induce re-expression of the original gene construct in rat liver. Gene Therapy 1999; 6: Cao G et al. Analysis of the human carcinoembryonic antigen promoter core region in colorectal carcinoma-selective cytosine deaminase gene therapy. Cancer Gene Ther 1999; 6: Kuriyama S et al. Immunomodulation with FK506 around the time of intravenous re-administration of an adenoviral vector

5 1136 facilitates gene transfer into primed rat liver. Int J Cancer 2000; 85: Klein TM, Arentzen R, Lewis PA, Fitzpatrick-McElligott S. Transformation of microbes, plants and animals by particle bombardment. Biotechnology 1992; 10: Christou P. Rice transformation: bombardment. Plant Mol Biol 1997; 35: Komari T et al. Advances in cereal gene transfer. Curr Opin Plant Biol 1998; 1: Vasil IK, Vasil V. Transformation of wheat via particle bombardment. Meth Mol Biol 1999; 111: Yang N-S, Sun WH. Gene gun and other non-viral approaches for cancer gene therapy. Nature Med 1995; 1: Whalen RG, Davis HL. DNA-mediated immunization and the energetic immune response to hepatitis B surface antigen. Clin Immunol Immunopathol 1995; 75: Haynes JR et al. Particle-mediated nucleic acid immunization. J Biotechnol 1996; 44: Robinson HL, Torres CA. DNA vaccines. Semin Immunol 1997; 9: Lai WC, Bennett M. DNA vaccines. Crit Rev Immunol 1998; 18: Roth JA, Cristiano RJ. Gene therapy for cancer: what have we done and where are we going? J Natl Cancer Inst 1997; 89: Rosenberg SA. Development of cancer immunotherapies based on identification of the genes encoding cancer regression antigens. J Natl Cancer Inst 1996; 88: Gilboa E. Immunotherapy of cancer with genetically modified tumor vaccines. Semin Oncol 1996; 23: Yang N-S, Sun WH, McCabe D. Developing particle-mediated gene transfer technology for research into gene therapy of cancer. Mol Med Today 1996; 2: Mahvi DM, Sheehy MJ, Yang N-S. DNA cancer vaccines: a gene gun approach. Immunol Cell Biol 1997; 75: Dachs GU, Dougherty GJ, Stratford IJ, Chaplin DJ. Targeting gene therapy to cancer: a review. Oncol Res 1997; 9: Sanford JC, Smith FD, Russell JA. Optimizing the biolistic process for different biological applications. Meth Enzymol 1993; 217: Heiser WC. Gene transfer into mammalian cells by particle bombardment. Anal Biochem 1994; 217: Hall CV, Jacob PE, Ringold GM, Lee F. Expression and regulation of Escherichia coli lacz gene fusions in mammalian cells. J Mol Appl Genet 1983; 2: Kuriyama S et al. Antitumor effect of electrochemotherapy on colorectal carcinoma in an orthotopic mouse model. Int J Oncol 1999; 14: Heller R et al. In vivo gene electroinjection and expression in rat liver. FEBS Lett 1996; 389: Malone RW et al. Dexamethasone enhancement of gene expression after direct hepatic DNA injection. J Biol Chem 1994; 269: Kuriyama S et al. A potential approach for gene therapy targeting hepatoma using a liver-specific promoter on a retroviral vector. Cell Struct Funct 1991; 16: