Ultrasound in Med. & Biol., Vol. 31, No. 5, pp. 693–702, 2005 Copyright © 2005 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/05/$–see front matter
doi:10.1016/j.ultrasmedbio.2005.01.015
● Original Contribution INSONATION FACILITATES PLASMID DNA TRANSFECTION INTO THE CENTRAL NERVOUS SYSTEM AND MICROBUBBLES ENHANCE THE EFFECT YOSHINOBU MANOME,* NAOTO NAKAYAMA,† KIYOSHI NAKAYAMA,‡ and HIROSHI FURUHATA† *Department of Molecular Cell Biology, Institute of DNA Medicine, Research Center for Medical Science and † Medical Engineering Laboratory, Research Center for Medical Science, Jikei University School of Medicine, Tokyo, Japan; and ‡Department of Electrical and Electronic Engineering, Faculty of Science and Engineering, Sophia University, Tokyo, Japan (Received 23 August 2004; revised 19 January 2005; in final form 27 January 2005)
Abstract—Many of the diseases which affect the central nervous system are intractable to conventional therapies and therefore require alternative treatments such as gene therapy. Therapy requires safety, since the central nervous system is a critical organ. Choice of nonviral vectors such as naked plasmid DNA may have merit. However, transfection efficiencies of these vectors are low. We have investigated the use of 210.4 kHz ultrasound and found that 5.0 W/cm2 of insonation for 5 s most effectively transfected a plasmid DNA into culture slices of mouse brain (147.68-fold increase compared with 0 W/cm2 of insonation for 5 s). The effect was reinforced by combination with echo contrast agent, Levovist. One hundred fifty mg/mL of Levovist significantly increased gene transfection by ultrasound (5.23-fold when insonated at 5.0 W/cm2 for 5 s). When DNA was intracranially injected, Levovist also enhanced gene transfection in newborn mice (4.49-fold increase when insonated at 5.0 W/cm2 for 5 s). Since ultrasound successfully transfected naked plasmid DNA into the neural tissue and Levovist enhanced the effect, this approach may have a significant role in gene transfer to the central nervous system. (E-mail:
[email protected]) © 2005 World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasound, Plasmid, DNA, Gene therapy, Microbubbles, Central nervous system.
vivo, the viral vector arouses concern in terms of invoking strong adverse effects, such as immune responses, inflammation and/or carcinogenesis (Kang and Tisdale 2004). On the contrary, the nonviral vector is less hazardous. It lacks proteins that cause immune responses in the host. In addition, the vector itself is less toxic to the host, due to the lack of cytopathic effects attributable to viral infection. The nonviral vector also does not contain unnecessary genes, such as the viral structure genome. Furthermore, it is easier to prepare, has no gene–size limitations that affect delivery and can deliver genes into quiescent cells (Coonrod et al. 1997). Thus, the nonviral vector has several advantages. Among nonviral approaches, naked DNA transfer has recently been highlighted, since progress has been made in delivery to muscle in combination with electroporation or to the liver with intravascular delivery (Hartikka et al. 2001; Herweijer et al. 2001; Maruyama et al. 2002; Somiari et al. 2000). It also has a strong potential for use in gene therapy, since this approach becomes more significant
INTRODUCTION Merit exists for gene therapy in the treatment of many congenital and acquired diseases that are resistant to conventional therapies. Recent advancement of molecular biology in the medical field allows for the practical application of gene therapy with various candidate genes. While the development of therapeutic genes has progressed remarkably, present delivery systems have potential limitations to their clinical application (Nathwani et al. 2004). In principle, two major kinds of vectors are available; one is viral and the other is nonviral. The viral vector has the advantage of efficiency of gene delivery and is used for most clinical trials. However, despite its superior efficiency in gene transfer both in vitro and in
Address correspondence to: Yoshinobu Manome, M.D. Ph.D. Department of Molecular Cell Biology, Institute of DNA medicine, Research Center for Medical Science, Jikei University School of Medicine, 3-25-8 Nishishinbashi, Minato-ku, Tokyo, Japan 105-8461. E-mail:
[email protected] 693
694
Ultrasound in Medicine and Biology
when combined with a better strategy for treatment, such as combination with ribozyme, antisense or RNA interference (Pardridge 2004; Yu et al. 2002). Nevertheless, naked plasmid DNA gene transfer has been used mainly for genetic immunization studies, due to its low transfection efficiency (Herweijer and Wolff 2003). We have utilized ultrasound as a mechanical force and reported that ultrasound irradiation, called insonation, increased the efficiency of transfection into colon carcinoma cells both in vitro and in vivo (Manome et al. 2000). Since insonation achieved efficient in vivo transfection of naked plasmid DNA, the method will prove beneficial in gene therapy, especially for critical organs such as the central nervous system or the cardiovascular system. In this study, we attempted to evaluate the efficiency of gene transfer by ultrasound on the central nervous system cells. Recent studies demonstrated that microbubbles of echo contrast material enhanced gene transfection by ultrasound (Bekeredjian et al. 2003; Christiansen et al. 2003; Li et al. 2003; Lu et al. 2003; Pislaru et al. 2003). Therefore, we also attempted to evaluate the effect of microbubbles on gene transfection to the nervous system. MATERIALS AND METHODS Insonation Ultrasound was generated by a wave synthesizer, wave factory WF 1943 (NF electronic instruments, Yokohama, Japan). The amplifier and ultrasound probe were designed by one of the authors (H.F.) and constructed by Honda Electric Inc., Toyohashi, Japan. Transducer is made of piezoceramics and the diameter of the probe was 5 mm. This apparatus could emit continuous wave of 210.4 kHz ultrasound up to 5.08 W/cm2 measured by force balance method in water using ultrasound power meter (UPM-DT-1, Ohmic Instruments Co., Easton, MD, U.S.A.). In this setting, ultrasound intensity of 5 W/cm2 corresponds to 0.274 MPa. The insonation method to the culture slice is shown in Fig. 1. The distance between probe tip and culture slice was 2 mm and the slice was insonated passing through phosphate-buffered saline, with or without plasmid. Since transducer-generated plane wave and ultrasound was unfocused, the culture slice could be uniformly insonated. Insonation was performed at 23°C. This insonation at 5.0 W/cm2 for 5 s increased temperature of 4.19 ⫾ 0.79°C (n ⫽ 12) in the inner dish. Plasmid used for transfection The pCAG-Luc plasmid was provided by Kiyotsugu Yoshida (Tokyo Medical and Dental University, Tokyo, Japan). In this vector, CMV-IE enhancer was coupled with chicken -actin promoter (CAG promoter)
Volume 31, Number 5, 2005
Fig. 1. Schematic diagram of insonation to culture slice. Culture slice was directly incubated in plasmid DNA solution with or without microbubbles and insonated from the top of the inner dish of the culture insert. After insonation, the culture slice on the inner dish was returned to the six-well plate and further cultivated.
which drives the firefly luciferase (Luc) gene. The plasmid was prepared with alkali-SDS method with double cesium ultracentrifugation. Since the amounts of luciferase protein could be easily determined, this plasmid was used for the quantitative analysis of transgene expression. Expression of the luciferase gene was measured by the luciferase assay (Stratagene, La Jolla, CA. U.S.A.) with the luminometer, Lumat LB9501 (Berthold Japan, Tokyo). In contrast, the CMV-GFP plasmid (provided by Michiko Watanabe, Jikei University School of Medicine, Tokyo, Japan), in which the CMV promoter drives the modified green fluorescent protein gene, was used for histologic determination of the transfected cell type, since transfected cells can be identified by the confocal laser microscope (LSM510, Zeiss, Gottingen, Germany). Animal studies The ICR/Jcl mice were purchased from Clea Japan (Tokyo, Japan). All the animal experiments were performed under the guidelines of the Animal Care Facility in Jikei University School of Medicine. To obtain culture slices of the mouse brain, 1 to 4 – d– old ICR/Jcl mice were decapitated and the cerebra were quickly isolated under antiseptic conditions and sliced by a microslicer (DTK-1000, D. EM, Kyoto, Japan) at a thickness of 300 m. After preparation, cerebral slices were cultured on semiporous membranes (Millicell-CM, 60 mm, Millipore, Bedford, MA, USA) (Stoppini et al. 1991) and placed in six-well culture dishes with 1 mL of culture medium composed of 25% heat-inactivated horse serum, 25% Hanks’s balanced salt solution (HBSS) and 50% minimum essential medium (MEM) without glutamate (Invitrogen Japan, Tokyo, Japan) (Kudo 2000). We cul-
US facilitates DNA transfection ● Y. MANOME et al.
tured the slices for three to five weeks before starting the experiments. The cultivation reduced the slice thickness to about 100 m by eliminating dead cells induced by tissue slicing and enriched neurons with few layers of pyramidal cells (Okada et al. 1995). Levivost was purchased from Schering Japan (Osaka, Japan) and mixed with naked plasmid DNA just before experiments and, in all the experiments, insonation was carried out within 10 min after the preparation of Levovist. Each culture slice was incubated with 60 g of plasmid DNA in 0.3 mL of solution and exposed to various intensities of ultrasound (Fig. 1). Slices were returned to the incubator and incubated for an additional 48 h. Tissue lysates obtained under the supplier’s protocol were subjected to the luciferase assay. In the in vivo experiment, 50 g of naked plasmid DNA in 5 l of solution with 0 or 150 mg/mL of Levovist was directly injected into the right hemisphere of the cerebrum of newborn mice. Animals were anesthetized with 1.5% isoflurane (1– chloro–2, 2, 2–trifluoroethyl difluoromethyl ether, Dainabot, Osaka, Japan) using the MA110 anesthetic system with the Forawick vaporizer (Muraco Medical Co., Tokyo, Japan). Plasmid was injected with a 29-gauge–needle at a site 1.3 mm posterior and 1 mm to the right of the bregma, at the depth of 1 mm. After injection, since the skull is soft and floppy, mice were transcranially insonated with the same ultrasound apparatus. Forty-eight hours later, the right hemisphere of brain was excised, lysed (Manome et al. 1998) and subjected to the luciferase assay. All the assays were conducted in triplicate. Scanning of the culture slice surface for morphologic examination After insonation, the culture slice was fixed with 1.2% glutaraldehyde. Specimens were dehydrated by the CO2 critical-point drying method, coated by sputtering with gold and palladium and scanned by an electron microscope (JSM-5800LV, Japan Electro Optical Laboratory, Tokyo, Japan) for observation of the cell surface. Observations were accomplished by a panel experts in scanning electron microscope in Jikei University School of Medicine (Hideki Saito and Yoshiaki Hataba, Ph.D. at Jikei University School of Medicine) and to exclude the artifacts of sampling, the experiment was performed at least three times with six samples in each group and scanned more than six times. Cellular damage in culture slices Cellular damage in culture slices was monitored by the disruption of membrane integrity using densitometric measurement of the uptake of propidium iodide (PI). This measurement is one of the most reliable methods for evaluation of cellular death, especially in organotypic culture slices, and has been widely used to quantify a
695
neuronal cell death (Adembri et al. 2004; Bonde et al. 2003; Cho et al. 2004; Dehghani et al. 2003; Kristensen et al. 2003). Uptake of PI directly correlates with lactate dehydrogenase (LDH) efflux to the culture medium, ordinary Nissl cell staining, staining by the neurodegenerative marker fluoro-jade, neuronal microtubule degeneration by immunohistochemical staining for microtubule–associated protein 2 (MAP2) and Timm sulphide silver staining for heavy metal alterations and has been used as a good indicator of cell death, especially in organotypic brain slice cultures (Noraberg et al. 1999). Cultures were exposed to 2 M of PI, at which concentration PI is harmless to neurons (Macklis and Madison 1990) and the uptake recorded with a digital CCD camera (Coolpix955, Nikon, Japan) under a fluorescent microscope (MZFLIII, Leica Microsystems, Wetzler, Germany). PI uptake was quantified by the histogram, channel red analysis of Photoshop software (version 5.0-J, Adobe Systems Inc, San Jose, CA, USA) (Yoshinaga et al. 2003). Statistic analysis Statistical analysis was performed with two–sample t–tests. Significance was determined at p values less than the 0.05 level. Statistics were computed using the Statview statistic package (ver. 5.0, SAS Institute Inc, Berkeley, CA, USA). RESULTS Ultrasound improves transfection efficiency of naked plasmid DNA We first examined the chronological change of transgene expression. Mouse cerebral culture slices were insonated with pCAG–Luc plasmid and subjected to the luciferase assay. Expression of marker protein increased 24 h after insonation (Fig. 2). It rapidly augmented and the maximal expression was observed at 48 h. Subsequently, expression gradually decreased with time, suggesting that expression of the transgene was mostly transient. Since maximal expression was obtained at 48 h after insonation, all the other experiments were assayed at 48 h. We next examined the relationships of ultrasound intensities, durations and gene transfer efficiencies. The activities of luciferase were nearly undetectable at 0 W/cm2 of insonation (Fig. 3). When we gradually increased the intensity of ultrasound from 0 to 5.0 W/cm2, the transfection efficiencies at 5 or 10 s correspondingly increased. Interestingly, the peak transfection was found at 5 s of insonation at 5.0 W/cm2. Unlike 5.0 W/cm2, both intensities of 3.8 and 4.4 W/cm2 showed the highest transfection at 10 s and the peak transfection was not observed in these conditions. Transfection efficiencies decreased at 20 s at all the intensities. This is
696
Ultrasound in Medicine and Biology
Volume 31, Number 5, 2005
that gene expression was observed mainly not in neurons. However, the transgene was coexpressed in some neuronal cells (Fig. 5).
Fig. 2. Chronological change of transgene expression after insonation. Culture slices were insonated with 5.0 W/cm2 for 5 s with pCAG-Luc plasmid. After insonation, the slices were returned to the incubator and incubated at indicated hours before being subjected to the luciferase assay. Luciferase activity was increased at 24 h after insonation. The maximal expression was observed at 48 h after insonation. Results are expressed as the mean of four experiments and the bars indicate the standard error of the mean.
Levovist enhances transfection by ultrasound of culture slices Since we confirmed an affirmative response of ultrasound in the transfection of plasmid DNA into brain culture slices at the level of approximately 5.0 W/cm2 for 5 s, we attempted to increase the efficacy of ultrasound by using microbubbles. We tested the echo contrast agent Levovist and chose insonation at 4.4 and 5.0 W/cm2 for 5 and 10 s for the experiment. Levovist was used at concentrations of 0, 150 and 300 mg/mL. Without insonation, microbubbles did not transfect plasmid DNA into culture slices at any of the incubation durations (0, 5, 10 and 20 s) (Fig. 6a). While Levovist itself did not deliver plasmid DNA into the slices, microbubbles enhanced the transfection of DNA when insonated. The effect of microbubbles at 4.4 W/cm2 and at 5.0 W/cm2 of insonation is demonstrated in Fig. 6b. Levovist at the concentration of 150 mg/mL significantly enhanced gene transfection both at 5 s (p ⬍ 0.05, 13.69 –fold increase) and 10 s (p ⬍
attributable to the noxious effect of insonation on the slices, as shown later. Evidence of sonoporation on the culture slice surface The mechanism of transfection of plasmid DNA into target cells may derive from an observation that insonation causes the formation of transient holes in the cell surface (sonoporation), thereby resulting in the transition of plasmid DNA from outside into the cytoplasm (Miller et al. 2002). To investigate whether such a morphologic change can occur in the central nervous cells, culture slices were insonated and their surface was monitored by a scanning electron microscope (Fig. 4). Unlike the uninsonated control slices, insonation perforated the cellular membrane and small holes appeared on the cell surface. These results were consistent with data obtained using cultured human muscle cells (Taniyama et al. 2002). The size of the hole varied in each specimen, depending on the insonation time, intensity or slice condition; however, longer insonation caused severe damage as well as disruption of the cellular membrane. Transgene expression in glial and neuronal cells We used the pCMV–GFP plasmid and confirmed expression of the transgene in culture slices by confocal laser microscopy. Immunohistological staining with anti NeuN (1:500, Chemicon, Temecula, CA, USA) revealed
Fig. 3. Effect of intensities and durations of insonation on transgene expression in brain slice cultures. Culture slices were incubated with pCAG–Luc plasmid and insonated with various intensities and durations of ultrasound. Luciferase activity was nearly undetectable at insonation at 0 W/cm2. Compared with 0 W/cm2, intensities of 3.8, 4.4 and 5.0 W/cm2 facilitated gene transfection. Significant differences were demonstrated between 0 and 3.8 W/cm2 (20 s; p ⬍ 0.05: n ⫽ 6 each), 4.4 W/cm2 (10 s; p ⬍ 0.005: n ⫽ 15 and 12) and 5.0 W/cm2 (5 s; p ⬍ 0.005: n ⫽ 6 and 12, 10 s; p ⬍ 0.05: n ⫽ 15 and 12). Results are expressed as the mean of at least six experiments (maximum ⫽ 15) and the bars indicate the standard error of the mean.
US facilitates DNA transfection ● Y. MANOME et al.
697
300 mg/mL. Insonation of 5 s was more effective than 10 s. Longer insonation increases cell damage, but a higher concentration of Levovist does not Although we demonstrated that insonation transfected plasmid DNA into culture slices and microbubbles enhanced this effect, we also realized that the results obtained were to some extent different from our initial prediction. These experiments clearly indicated that gene transfection occurred in the first 5 or 10 s and that longer insonation abolished the efficiency. In addition, the higher concentration of Levovist did not guarantee a higher efficiency of gene transfection in enhancement of ultrasound. We attempted to identify the factors that determined these differences. One reason might derive from the toxic effect of insonation on the slices. To address whether or not longer insonation causes cellular damage in slice cultures, we measured cell death of the slices after insonation. Culture slices were insonated with various durations of ultrasound and their cell death was measured by the uptake of propidium iodide (PI) (Fig. 7). Except for 0 W/cm2, PI uptake increased according to the insonation time. Longer insonation caused more PI uptake in the culture cells. There were differences between the results at 10 and 20 s of insonation (p ⬍ 0.005, 3.8 and 4.4 W/cm2 and p ⬍ 0.001, 5.0 W/cm2). Insonation for a 20-s duration caused severe cell death. Differences between 0 and 20 s were significant at all the intensities (p ⬍ 0.0001). Since longer insonation caused cell death in the culture slices, this phenomenon might explain attenuated transgene expressions at the longer insonation
Fig. 4. Morphologic change in the culture slice surface after insonation. Culture slices were insonated, fixed and their surfaces were examined by a scanning electron microscope. a) Noninsonated cells demonstrated a clear cell surface, while b) Insonation caused morphologic changes in the cellular membrane. c) Longer insonation disrupted the cellular membrane and generated larger holes as well as further ablation.
0.05, 6.23–fold increase) at 4.4 W/cm2. This concentration also facilitated the transfection at 5 s of 5.0 W/cm2 (p ⬍ 0.0005, 5.23-fold increase). The highest gene transfection was observed at 150 mg/mL of Levovist with insonation at 5.0 W/cm2 for 5 s. It was not possible to achieve this level of gene expression without microbubbles. The concentration of 150 mg/mL of Levovist enhanced gene transfer better than
Fig. 5. Colocalization of transgene and cell marker. GFP expression was observed by confocal imaging (green color). a) Most of the transfected cells were not positive for NeuN (red color). b) Colocalization of GFP and NeuN was observed, however, in a small number of neuronal cells.
698
Ultrasound in Medicine and Biology
Volume 31, Number 5, 2005
measured at 3.8, 4.4 and 5.0 W/cm2. Unlike the previous experiment, we did not observe markedly increased cell death at the higher concentration of Levovist. There were no differences between the results at Levovist concentrations of 0, 150 and 300 mg/mL. Levovist at 300 mg/mL is hyperosmolar. However, even at the concentration of 300 mg/mL, Levovist was not toxic in the culture slices. To ascertain that the effective dose of Levovist did not cause delayed cell damage, we also evaluated the PI taken up at 120 h after administration. Both insonations at 5.0 W/cm2 for 5 s with or without 150 mg/mL of Levovist did not confer detectable cellular damage (Fig. 9).
Levovist enhances transfection by ultrasound in vivo As observed in Fig. 6b, 150 mg/mL of Levovist significantly increased gene transfection in culture slices in vitro. Encouraged by the result, we examined the effect of Levovist on in vivo gene transfection. The plasmid was injected with 0 or 150 mg/mL of Levovist into the right hemisphere of newborn mice and irradiated with 5.0 W/cm2 of insonation for 5 s. As controls, mice were also treated with or without plasmid that had not been insonated. In the noninsonated control groups, we observed no difference in transgene expression between nonplasmid (vehicle only) and plasmid (plasmid only) treated groups (Fig. 10), left and middle, Levovist 0 mg/mL). This result indicated that plasmid was not trans-
Fig. 6. Effect of Levovist on gene transfection a) The effect of Levovist on gene transfection was evaluated. Basically, Levovist itself did not increase uptake of the plasmid into culture slices without insonation. While longer exposure of Levovist (10 and 20 s) slightly enhanced gene transfection, the effect was modest and not significant. Results are demonstrated as the mean of three experiments and bars indicate the standard error of the mean. b) Effect of Levovist on gene transfection by ultrasound. Slices were incubated with plasmid DNA and insonated with Levovist. Compared with 0 mg/mL, 150 mg/mL of Levovist significantly enhanced gene transfection of ultrasound at 5 and 10 s of 4.4 W/ cm2 (both, p ⬍ 0.05) and 5 s of 5.0 W/cm2 (p ⬍ 0.0005). Three hundred mg/mL of Levovist also facilitated the effect of ultrasound at 5 and 10 s at 4.4 W/cm2 (both p ⬍ 0.05) and 10 s at 5.0 W/cm2 (p ⬍ 0.005). Results are demonstrated as the mean of at least seven experiments (maximum ⫽ 12) and bars indicate standard error of mean.
time. Correspondingly, we measured cell death at different concentrations of Levovist. Culture slices were exposed to 0, 150 and 300 mg/mL of Levovist and insonated for 5 (Fig. 8a) or 10 s (Fig. 8b). Cell death was
Fig. 7. Cellular damage caused by insonation in brain culture slices. Culture slices were insonated at various intensities of ultrasound for 0, 5, 10 or 20 s, then further cultivated for 48 h. After exposure to PI, cellular damage in culture slices was evaluated with the uptaken PI. Insonation at the 20-s duration caused significant cellular membrane disruption (0 vs. 3.8, 4.4 or 5.0 W/cm2 of insonation; p ⬍ 0.001). Results are demonstrated as the mean of at least four experiments (maximum ⫽ 7) and the bars indicate the standard error of the mean.
US facilitates DNA transfection ● Y. MANOME et al.
699
Safety evaluation of the animals One mouse with shallow anesthesia underwent generalized seizure for 5 min after insonation at 5 W/cm2 for 5 s with Levovist. However, none of the other 22 mice demonstrated abnormal symptoms including similar seizure attack. All the mice were healthy and did not demonstrate abnormal behavior, late onset epilepsy, neurologic deficits, malnutrition or growth retardation. The mice developed in good health and produced healthy offspring without detectable abnormalities. DISCUSSION Ultrasound induces cell membrane porosity (Tachibana et al. 1999) and has been shown to improve the efficiency of nonviral vector delivery in a wide variety of tissues such as muscle, blood vessels, cartilage, dental pulp, as well as in malignancies including prostate and colon cancers (Anwer et al. 2000; Hosseinkhani et al. 2002; Huber and Pfisterer 2000; Lawrie et al. 2000; Lawrie et al. 1999; Lawrie et al. 2003; Lu et al. 2003; Manome et al. 2000; Miura et al. 2002; Schratzberger et al. 2002; Taniyama et al. 2002; Taniyama et al. 2002; Zarnitsyn and Prausnitz 2004). This wide choice of target tissues or organs makes ultrasound an attractive technology for practical gene delivery. Following (Shimamura et al.’s 2004) report, we added the central nervous system to the list of target organs. Most diseases of the central nervous system are difficult to manage with current therapies; hence, the Fig. 8. Cellular damage by Levovist in brain culture slices. Culture slices were treated with 0, 150 and 300 mg/mL of Levovist and insonated for a) 5 or b) 10 seconds. Whereas gene transfection efficiencies varied among the concentrations of 0, 150 and 300 mg/mL of Levovist, no differences in cell death at 48 h after insonation were displayed in this experiment. Results are demonstrated as the mean of three experiments and the bars indicate standard error of mean.
fected by the injection alone. Furthermore, when plasmid was injected into the central nervous system, Levovist itself did not significantly deliver plasmid DNA to the surrounding brain tissue without insonation (middle, Levovist 0 and 150 mg/mL). The result was consistent with the data obtained from culture slices. On the contrary, when mice were insonated at 5.0 W/cm2 for 5 s, insonation induced transfection of plasmid DNA in the brain tissue even without the use of Levovist (middle and right, Levovist 0, p ⬍ 0.001). When plasmid DNA was injected with 150 mg/mL of Levovist, Levovist significantly enhanced ultrasound gene transfer in vivo (right, Levovist 0 and 150 mg/mL, p ⬍ 0.005, 4.59 –fold increase).
Fig. 9. Late cellular damage by insonation and Levovist in brain culture slices. Culture slices were treated with 0 and 5.0 W/cm2 for 5 s and cellular damage was sequentially monitored by the uptake PI. There were no differences among insonations at 0 W/cm2, 5.0 W/cm2 and 5.0 W/cm2 with 150 mg/mL of Levovist. As a control group, nine slices were treated with methanol for 5 min and subjected to the assay. Significant PI uptake was observed in this group (death control). Results are demonstrated as the mean of at least three experiments (maximum ⫽ 7) and the bars indicate the standard error of the mean.
700
Ultrasound in Medicine and Biology
Fig. 10. In vivo transfection of pCAG-Luc plasmid DNA into the central nervous system. Newborn mice were injected with pCAG–Luc plasmid and insonated at the 0 or 5.0 W/cm2 for 5 s. The effects of Levovist were compared by expressed luciferase activities. As a control, a group treated only with plasmid-carrying vehicle was also examined (left). Similar to the result obtained from culture slices, Levovist did not facilitate the uptake of plasmid DNA in itself (center). On the contrary, Levovist significantly enhanced the transfection when insonated (p ⬍ 0.005). Results are demonstrated as the mean of at least six experiments (maximum ⫽ 8) and the bars indicate the standard error of the mean.
development of better methods of treatment or therapy is urgent. The advantage of the central nervous system for gene therapy is that methods of drug or vector delivery have been established. Intrathecal or transarterial injections are used for drug delivery to the central nervous system. More directly, drugs and genes can be delivered by stereotactic surgery or utilization of the Ommaya reservoir. These methods have been commonly used for patients and lesions are directly accessible by the procedures. One advantage of insonation is its capability to transmit mechanical power or force to a distant area of the human body. Ultrasound can be focused on a site where gene delivery is required. This may be a benefit for gene delivery in deep-seated lesions. The authors previously implanted a malignant brain tumor in the caudate nucleus of the cerebrum of rats infected with adenoviral vectors encoding -galactosidase with a stereotactic device and found that the transgene was expressed nonselectively into the larger area of surrounding
Volume 31, Number 5, 2005
brain tissue (Manome et al. 1998; Parr et al. 1997). While stereotactic surgery offers efficient gene delivery to the central nervous system, nonselective expression will remain a major problem for clinical application. By focusing the ultrasound, mechanical power can be intensified specifically on the target area and selective expression can be expected. This method could also reduce cellular damage to adjacent unfocused tissue. In this study, we constructed an apparatus for gene transfection and demonstrated that insonation at 3.8 to 5.0 W/cm2 for 5 or 10 s enhanced the transfection of a plasmid DNA into culture slices of mouse brain. Among these conditions, we found that insonation at 5.0 W/cm2 for 5 s extensively facilitated transfection into the slices. In this case, we observed small cavitations on the surface membrane of the cells. By calculation, the mechanical index, or MI, of the intensity was 0.612 (Abbott 1999). This value is slightly higher than 0.547, at which insonation was also effective in transducing plasmid DNA into colon cancer cells with a different ultrasound probe (1 MHz, 20 W/cm2) (Manome et al. 2000). In addition, the current system is superior from the aspect of less heat generation. Thermal index category soft tissue, or TIS, (⫽0.935) was far smaller than that of the previous study (⫽18.70) (Abbott 1999; Barnett 2001; Manome et al. 2000). Hence, the condition we determined has advantages in ultrasound gene transfection into cells. Recently, echo contrast microbubbles have been used to increase the efficiency of ultrasonic gene transfection. The microbubbles of air or gas suspended in liquid can be collapsed intentionally by insonation and this collapse generates strong forces that can destroy the integrity of the adjoining cellular membrane. A recent study has demonstrated that gentle linear bubble oscillation is sufficient to achieve rupture of lipid membranes (Marmottant and Hilgenfeldt 2003). Currently, air–filled albumin microspheres (Albunex), nonshell type granules composed of 99.9% galactose and 0.1% palmitic acid (Levovist), and an albumin-shelled agent composed of the fluorocarbon gas octafluoropropane (Optison) are used as echo contrast agents in clinical ultrasound diagnosis. Among them, Optison has been shown to increase the gene transfer in cardiovascular (Bekeredjian et al. 2003; Lawrie et al. 2000; Miura et al. 2002; Taniyama et al. 2002) and skeletal muscle (Lu et al. 2003; Pislaru et al. 2003; Taniyama et al. 2002) cells. In this study, we used Levovist. Kudo et al. (2002) reported that microbubbles exposed to ultrasound caused mechanical stress that acted on the cells and bubble collapse was responsible for cell membrane damage. In this setting, Feril et al. (2003) reported that when Levovist, Optison and YM454 were compared, Levovist was shown to have the least effect on ultrasound-induced apoptosis and cell lysis. In fact, Miller et al. (2003) reported that Optison
US facilitates DNA transfection ● Y. MANOME et al.
induced significant cell killing in vitro. Although Lu et al. (2003) reported that Optison reduced insonation tissue damage in muscles, we chose Levovist in our initial attempt to transduce the central nervous cells, as those cells necessitate the greatest precautions. Interestingly, Lu et al. (2003) reported that Optison alone significantly caused transgene expression in muscle in vivo. We did not observe the effect of spontaneous gene delivery by microbubbles alone in brain culture slices nor in the in vivo mouse brain with Levovist. It is unclear whether this discrepancy arises from differences in the microbubbles (Optison vs. Levovist), target tissues (muscles vs. central nervous cells) or animal ages (four–week vs. newborn mouse). So far, we have tested only Levovist for the central nervous system. The efficiency of microbubbles was evaluated under limited conditions. Numerous patients suffering from central nervous diseases are waiting for more effective treatment. Despite the limited conditions, we demonstrated that insonation facilitated the uptake of plasmid DNA into the vulnerable central nervous system cells and that microbubbles enhanced the uptake. Further optimization of insonation conditions, as well as the selection of microbubbles, will lead to the development of safer and more efficient gene therapy for the central nervous system. Acknowledgments—The authors thank Yoshihisa Kudo of the Tokyo University of Pharmacy and Life Science for his helpful suggestions on the slice culture experiments. We also thank Utako Fukushima and Yuko Abe for their expert technical assistance. This work was supported by a grant-in-aid for “Bio-Venture Research Fund Project Aid,” from the Ministry of Education, Science, and Culture of Japan.
REFERENCES Abbott JG. Rationale and derivation of MI and TI-a review. Ultrasound Med Biol 1999;25:431– 441. Adembri C, Bechi A, Meli E, et al. Erythropoietin attenuates posttraumatic injury in organotypic hippocampal slices. J Neurotrauma 2004;21:1103–1112. Anwer K, Kao G, Proctor B, et al. Ultrasound enhancement of cationic lipid-mediated gene transfer to primary tumors following systemic administration. Gene Ther 2000;7:1833–1839. Barnett SB. Intracranial temperature elevation from diagnostic ultrasound. Ultrasound Med Biol 2001;27:883– 888. Bekeredjian R, Chen S, Frenkel PA, Grayburn PA, Shohet RV. Ultrasound-targeted microbubble destruction can repeatedly direct highly specific plasmid expression to the heart. Circulation 2003; 108:1022–1026. Bonde C, Sarup A, Schousboe A, et al. GDNF pre-treatment aggravates neuronal cell loss in oxygen-glucose deprived hippocampal slice cultures: a possible effect of glutamate transporter up-regulation. Neurochem Int 2003;43:381–388. Cho S, Liu D, Fairman D, et al. Spatiotemporal evidence of apoptosismediated ischemic injury in organotypic hippocampal slice cultures. Neurochem Int 2004;45:117–127. Christiansen JP, French BA, Klibanov AL, Kaul S, Lindner JR. Targeted tissue transfection with ultrasound destruction of plasmidbearing cationic microbubbles. Ultrasound Med Biol 2003;29:1759 –1767.
701
Coonrod A, Li FQ, Horwitz M. On the mechanism of DNA transfection: efficient gene transfer without viruses. Gene Ther 1997;4: 1313–1321. Dehghani F, Hischebeth GT, Wirjatijasa F, et al. The immunosuppressant mycophenolate mofetil attenuates neuronal damage after excitotoxic injury in hippocampal slice cultures. Eur J Neurosci 2003;18:1061–1072. Feril LB, Jr., Kondo T, Zhao QL, et al. Enhancement of ultrasoundinduced apoptosis and cell lysis by echo-contrast agents. Ultrasound Med Biol 2003;29:331–337. Hartikka J, Sukhu L, Buchner C, et al. Electroporation-facilitated delivery of plasmid DNA in skeletal muscle: plasmid dependence of muscle damage and effect of poloxamer 188. Mol Ther 2001;4: 407– 415. Herweijer H, Wolff JA. Progress and prospects: naked DNA gene transfer and therapy. Gene Ther 2003;10:453– 458. Herweijer H, Zhang G, Subbotin VM, et al. Time course of gene expression after plasmid DNA gene transfer to the liver. J Gene Med 2001;3:280 –291. Hosseinkhani H, Aoyama T, Ogawa O, Tabata Y. Ultrasound enhancement of in vitro transfection of plasmid DNA by a cationized gelatin. J Drug Target 2002;10:193–204. Huber PE, Pfisterer P. In vitro and in vivo transfection of plasmid DNA in the Dunning prostate tumor R3327-AT1 is enhanced by focused ultrasound. Gene Ther 2000;7:1516 –1525. Kang EM, Tisdale JF. The leukemogenic risk of integrating retroviral vectors in hematopoietic stem cell gene therapy applications. Curr Hematol Rep 2004;3:274 –281. Kristensen BW, Noraberg J, Zimmer J. The GABAA receptor agonist THIP is neuroprotective in organotypic hippocampal slice cultures. Brain Res 2003;973:303–306. Kudo N, Miyaoka T, Okada K, Yamamoto K, Niwa K. Study on mechanisms of cell damage caused by microbubbles exposed to ultrasound. IEEE Ultrasonics Symposium Proceedings 2002;1351– 1354. Kudo Y. Application of organotypic cultures to researches on functional synapses. Tankaku Kakusan Koso 2000;45:387–392. Lawrie A, Brisken AF, Francis SE, et al. Microbubble-enhanced ultrasound for vascular gene delivery. Gene Ther 2000;7:2023–2027. Lawrie A, Brisken AF, Francis SE, et al. Ultrasound enhances reporter gene expression after transfection of vascular cells in vitro. Circulation 1999;99:2617–2620. Lawrie A, Brisken AF, Francis SE, et al. Ultrasound-enhanced transgene expression in vascular cells is not dependent upon cavitationinduced free radicals. Ultrasound Med Biol 2003;29:1453–1461. Li T, Tachibana K, Kuroki M. Gene transfer with echo-enhanced contrast agents: comparison between Albunex, Optison, and Levovist in mice—initial results. Radiology 2003;229:423– 428. Lu QL, Liang HD, Partridge T, Blomley MJ. Microbubble ultrasound improves the efficiency of gene transduction in skeletal muscle in vivo with reduced tissue damage. Gene Ther 2003;10:396 – 405. Macklis JD, Madison RD. Progressive incorporation of propidium iodide in cultured mouse neurons correlates with declining electrophysiological status: a fluorescence scale of membrane integrity. J Neurosci Methods 1990;31:43– 46. Manome Y, Kunieda T, Wen PY, et al. Transgene expression in malignant glioma using a replication-defective adenoviral vector containing the Egr-1 promoter: activation by ionizing radiation or uptake of radioactive iododeoxyuridine. Hum Gene Ther 1998;9: 1409 –1417. Manome Y, Nakamura M, Ohno T, Furuhata H. Ultrasound facilitates transduction of naked plasmid DNA into colon carcinoma cells in vitro and in vivo. Hum Gene Ther 2000;11:1521–1528. Marmottant P, Hilgenfeldt S. Controlled vesicle deformation and lysis by single oscillating bubbles. Nature 2003;423:153–156. Maruyama H, Higuchi N, Nishikawa Y, et al. High-level expression of naked DNA delivered to rat liver via tail vein injection. J Gene Med 2002;4:333–341. Miller DL, Dou C, Song J. DNA transfer and cell killing in epidermoid cells by diagnostic ultrasound activation of contrast agent gas bodies in vitro. Ultrasound Med Biol 2003;29:601– 607.
702
Ultrasound in Medicine and Biology
Miller DL, Pislaru SV, Greenleaf JE. Sonoporation: mechanical DNA delivery by ultrasonic cavitation. Somat Cell Mol Genet 2002;27: 115–134. Miura S, Tachibana K, Okamoto T, Saku K. In vitro transfer of antisense oligodeoxynucleotides into coronary endothelial cells by ultrasound. Biochem Biophys Res Commun 2002;298:587–590. Nathwani AC, Benjamin R, Nienhuis AW, Davidoff AM. Current status and prospects for gene therapy. Vox Sang 2004;87:73– 81. Noraberg J, Kristensen BW, Zimmer J. Markers for neuronal degeneration in organotypic slice cultures. Brain Res Brain Res Protoc 1999;3:278 –290. Okada M, Sakaguchi T, Kawasaki K. Correlation between anti-ubiquitin immunoreactivity and region-specific neuronal death in Nmethyl-D-aspartate-treated rat hippocampal organotypic cultures. Neurosci Res 1995;22:359 –366. Pardridge WM. Intravenous, non-viral RNAi gene therapy of brain cancer. Expert Opin Biol Ther 2004;4:1103–1113. Parr MJ, Manome Y, Tanaka T, et al. Tumor-selective transgene expression in vivo mediated by an E2F-responsive adenoviral vector. Nat Med 1997;3:1145–1149. Pislaru SV, Pislaru C, Kinnick RR, et al. Optimization of ultrasoundmediated gene transfer: comparison of contrast agents and ultrasound modalities. Eur Heart J 2003;24:1690 –1698. Schratzberger P, Krainin JG, Schratzberger G, et al. Transcutaneous ultrasound augments naked DNA transfection of skeletal muscle. Mol Ther 2002;6:576 –583.
Volume 31, Number 5, 2005 Shimamura M, Sato N, Taniyama Y, et al. Development of efficient plasmid DNA transfer into adult rat central nervous system using microbubble-enhanced ultrasound. Gene Ther 2004;11:1532– 1539. Somiari S, Glasspool-Malone J, Drabick JJ, et al. Theory and in vivo application of electroporative gene delivery. Mol Ther 2000;2:178 – 187. Stoppini L, Buchs P-A, Muller D. A simple method for organotypic cultures of nervous tissue. J Neurosci Nethods 1991;37:173–182. Tachibana K, Uchida T, Ogawa K, Yamashita N, Tamura K. Induction of cell-membrane porosity by ultrasound. Lancet 1999;353:1409. Taniyama Y, Tachibana K, Hiraoka K, et al. Development of safe and efficient novel nonviral gene transfer using ultrasound: enhancement of transfection efficiency of naked plasmid DNA in skeletal muscle. Gene Ther 2002;9:372–380. Taniyama Y, Tachibana K, Hiraoka K, et al. Local delivery of plasmid DNA into rat carotid artery using ultrasound. Circulation 2002;105: 1233–1239. Yoshinaga H, Watanabe M, Manome Y. Possible role of nicaraven in neuroprotective effect on hippocampal slice culture. Canadian J Physiology and Pharmacology 2003;81:683– 689. Yu JY, DeRuiter SL, Turner DL. RNA interference by expression of short-interfering RNAs and hairpin RNAs in mammalian cells. Proc Natl Acad Sci U S A 2002;99:6047– 6052. Zarnitsyn VG, Prausnitz MR. Physical parameters influencing optimization of ultrasound-mediated DNA transfection. Ultrasound Med Biol 2004;30:527–538.