- Email: [email protected]
Targeted gene delivery to the synovial pannus in antigen-induced arthritis by ultrasound-targeted microbubble destruction in vivo
Ultrasonics 65 (2016) 304–314
Contents lists available at ScienceDirect
Ultrasonics journal homepage: www.elsevier.com/locate/ultras
Targeted gene delivery to the synovial pannus in antigen-induced arthritis by ultrasound-targeted microbubble destruction in vivo Xi Xiang, Yuanjiao Tang, Qianying Leng, Lingyan Zhang, Li Qiu ⇑ Department of Ultrasound, West China Hospital of Sichuan University, Chengdu, Sichuan, China
a r t i c l e
i n f o
Article history: Received 1 July 2015 Received in revised form 9 September 2015 Accepted 16 September 2015 Available online 26 September 2015 Keywords: Ultrasound-mediated microbubble destruction Gene transfection Enhanced green fluorescent protein Synovial pannus Antigen-induced arthritis
a b s t r a c t The purpose of this study was to optimize an ultrasound-targeted microbubble destruction (UTMD) technique to improve the in vivo transfection efficiency of the gene encoding enhanced green fluorescent protein (EGFP) in the synovial pannus in an antigen-induced arthritis rabbit model. A mixture of microbubbles and plasmids was locally injected into the knee joints of an antigen-induced arthritis (AIA) rabbits. The plasmid concentrations and ultrasound conditions were varied in the experiments. We also tested local articular and intravenous injections. The rabbits were divided into five groups: (1) ultrasound + microbubbles + plasmid; (2) ultrasound + plasmid; (3) microbubble + plasmid; (4) plasmid only; (5) untreated controls. EGFP expression was observed by fluorescent microscope and immunohistochemical staining in the synovial pannus of each group. The optimal plasmid dosage and ultrasound parameter were determined based on the results of EGFP expression and the present and absent of tissue damage under light microscopy. The irradiation procedure was performed to observe the duration of the EGFP expression in the synovial pannus and other tissues and organs, as well as the damage to the normal cells. The optimal condition was determined to be a 1-MHz ultrasound pulse applied for 5 min with a power output of 2 W/cm2 and a 20% duty cycle along with 300 lg of plasmid. Under these conditions, the synovial pannus showed significant EGFP expression without significant damage to the surrounding normal tissue. The EGFP expression induced by the local intra-articular injection was significantly more increased than that induced by the intravenous injection. The EGFP expression in the synovial pannus of the ultrasound + microbubbles + plasmid group was significantly higher than that of the other four groups (P < 0.05). The expression peaked on day 5, remained detectable on day 40 and disappeared on day 60. No EGFP expression was detected in the other tissues and organs. The UTMD technique can significantly enhance the in vivo gene transfection efficiency without significant tissue damage in the synovial pannus of an AIA model. Thus, this could become a safe and effective non-viral gene transfection procedure for arthritis therapy. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Rheumatoid arthritis (RA) is a systemic autoimmune disease that is characterized by chronic joint inflammation. Synovial pannus development is the main pathological basis for the joint lesions and cartilage destruction in RA. Patients experience the onset of this disease in youth and middle age. RA is characterized by a chronic course and persistent progression despite treatment and it is a refractory disease that often leads to limb disability. Therefore, researchers are actively searching for effective treatments for RA. The administration of cytokine antagonists, such as ⇑ Corresponding author at: Department of Ultrasound, West China Hospital of Sichuan University, No. 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, China. Tel.: +86 18980602044; fax: +86 02885423192. E-mail address: [email protected] (L. Qiu). http://dx.doi.org/10.1016/j.ultras.2015.09.011 0041-624X/Ó 2015 Elsevier B.V. All rights reserved.
antagonists of tumor necrosis factor a (TNF-a) and interleukin 1 (IL-1), can significantly improve RA symptoms [1–3]. However, these treatments only have a short effective time, and repeated administration is needed. The treatments are also given longterm and are thus associated with high costs. High-dose systemic injections or repeated intra-articular injections of the drugs are required to achieve effective local therapeutic concentrations in the joints. This feature of the therapy results in a large economic burden and in patient suffering. Recently, gene therapy has been proposed for treatment of RA [4,5]. There are numerous advantages of gene therapy. For instance, this therapy can be locally administered and provide stable and continuous in vivo expression of the therapeutic molecules. Therefore, gene therapy through the transfection and in vivo expression of therapeutic molecules may be a promising treatment method for RA [4,5]. Effective gene transfer
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
could relieve invasive joint synovitis, thereby improving the symptoms of rheumatoid arthritis. In addition, local gene expression therapy is not associated with the side effects that can be caused by systemic drug administration. Furthermore, the gene expression remains relatively constant. Thus, the chance of adverse effect is reduced because this method does not cause the temporary excessive drug concentrations associated with articular injections. The pathological changes observed in joint diseases are closely associated with the synovial tissues. The synovial cells are targeted in pathological conditions, and can also become participants in pathogenesis through multiple pathways. For example, these cells can pathologically secrete such substances as TNF-a, IL-1, the cell adhesion molecule ICAM-1 and reactive oxygen species. Furthermore, the persistent development and deterioration of joint damage in RA is associated with continuously amplified inflammation through the autocrine and paracrine release of cytokines. Therefore, the cells of the synovial tissues are ideal targets for gene therapy [6,7]. The current challenge is to select the optimal gene transfection method that minimizes the side effects and maximizes the effectiveness of successful gene delivery. At present, the vectors that have been applied for gene therapy can be broadly divided into the viral and the non-viral vectors. Each of these methods has their own limitations. Ultrasound-targeted microbubble destruction (UTMD) is a novel method for non-viral gene transfection. The UTMD technique compensates for the deficiencies of the viral and non-viral vectors, which exhibits greater transfection efficiency, specificity and safety. The studies of using UTMD for gene transfection include in vitro cell culture and in vivo heart, liver, blood vessel, kidney, skeletal muscle and retina models [8–10]. However, no studies of gene transfection using UTMD in synovial pannus of an RA model have been reported. The aim of this study is to optimize the UTMD technique to improve the in vivo transfection efficiency of the gene encoding enhanced green fluorescent protein (EGFP) in the synovial pannus of antigen-induced arthritis rabbit model.
305
Fig. 1. The exact injection site of knee joint in a rabbit.
Switzerland). The plasmids were transformed into DH5acompetent cells (Invitrogen, Carlsbad, CA, USA), and then extracted and purified according to the instructions of the plasmid extraction kit (Qiagen, Hilden, Germany). The concentration of the resulting plasmid DNA with absorbance at 260 nm (A260) was measured using spectrophotometry (1 lg/ll). The OD260 to OD280 ratio was between 1.8 and 2.1, which indicated that the plasmid DNA was not contaminated. The product was stored at 20 °C for later use. 2.3. Preparation of the SonoVue/DNA complexes
2. Materials and methods
The SonoVue microbubble suspensions (Bracco Imaging, Milano, Italy) were prepared at a density microbubbles of 2 108–5 108/ml and a concentration of 5 mg/ml using a 0.9% saline solution. Next, 100 ll of SonoVue was mixed with 100 ll of the plasmid DNA solution and stored at 4 °C just prior to each experiment.
2.1. Antigen-induced arthritis rabbit model
2.4. Animal preparation and experimental procedures
New Zealand white rabbits that weighted 2.5–3.0 kg with 1:1 ratio of male: female were used in this study. All of the rabbits were kept in a room with the appropriate temperature and humidity and fed with standard laboratory chow and water. All of the experiments were approved by the Sichuan University Research Committee for Animal Research. Four milligrams of chicken ovalbumin (OVA, Sigma, St Louis, MO, USA) were dissolved in 0.5 ml of Freund’s Complete Adjuvant (F-5881) (Sigma Chemical Co., St. Louis, MO, USA) and 0.5 ml of phosphate-buffered saline (PBS). One milliliter of this emulsion was percutaneously injected into five interscapular areas in the rabbit. The same injections were repeated every seven days for 3 weeks in total. Five days after the third injection, 10 mg of OVA was dissolved in 0.5 ml of normal saline and then injected into both knees of the rabbits [11,12], the exact injection site is shown in Fig. 1. In the fourth week after the OVA articular injection, ultrasound imaging was performed to evaluate the AIA lesions based on the presence of synovial thickening. The rabbits with AIA lesions were randomly divided into various experimental groups.
The rabbits were anesthetized with an intraperitoneal injection of 2% sodium pentobarbital (40 mg/kg) and fixed on the experimental table. The skin fur on the knee was removed, and coupling gel was spread on the skin. An ultrasound therapy device (Sonic Master ES-2, OG Giken, Japan) with a probe frequency of 1 MHz, transducer area of 10 cm2, and the pulse repetition frequency of 100 Hz was used in this study. The output power (1 W/cm2, 2 W/ cm2), duty cycle (20%, 30%, 50%) and time (2 min, 5 min,10 min) were adjusted throughout the experiments. The local articular injections were performed by injecting the various solution depending on the different groups with a 1 ml syringe into the knee joint cavity. UTMD was performed on the injected area immediately after the injection. The intravenous injections of contrast agent were performed by slowly and steadily injecting the solution into the ear vein. UTMD was performed on the knee after the initiation of the injection.
2.2. Plasmid DNA amplification, extraction and purification The pEGFP-C1 plasmid (4.2 kb) is an enhanced green fluorescent protein (EGFP) and purchased from Roche Applied Science (Basel,
2.5. Experimental series 2.5.1. Experiments to identify the optimal ultrasound pulse conditions Three experimental tests were performed. A mixture containing 200 ll (300 lg) of EGFP plasmid and 200 ll of SonoVue was injected into the knees bilaterally in each rabbit. The injections were immediately followed with an ultrasound pulse. The rabbits
306
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
were sacrificed 5 days after the ultrasound pulse. The optimal ultrasound pulse conditions were identified. Test 1 was to show the transfection efficiency with different power outputs by using 6 rabbits divided into two groups (3 in each). With the same 20% duty cycles and 5 min treatment time for both groups, the 1 W/cm2 output power was applied for the first group and 2 W/cm2 for the second group. Test 2 was to show the transfection efficiency with different duty cycles. Nine rabbits were selected and randomly divided into three groups. The 20% duty cycle was applied in the first group 30% in the second group, and 50% in the third group. The pulse duration for all three groups was 5 min. The optimal power output determined by the results of test 1 was used in each group. Test 3 was to show the transfection efficiency with different pulse durations. Nine rabbits were selected and randomly divided into three groups. The 2 min, 5 min, and 10 min pulse durations was applied in each group, respectively. The optimal power output determined by test 1 and the optimal duty cycle determined by test 2 were used in each group. 2.5.2. Experiments to compare injection approaches Six rabbits were selected and randomly divided into two groups to see which injection approach was more effectiveness for gene transfection. The optimal ultrasound parameters determined by the first experiment was used in these tests. A mixed solution containing 200 ll (200 lg) of EGFP plasmid and 200 ll of SonoVue was injected into both knees of the first group rabbits while 1 ml of mixed solution containing 0.5 ml (500 lg) of EGFP plasmid and 0.5 ml of Sonovue was divided into equal two doses (0.5 ml each) and two doses were slowly injected into the ear veins of the second group of rabbits at 30 min interval. Finally, another three rabbits received no treatment as the untreated controls. The rabbits were sacrificed 5 days after ultrasonic irradiation. 2.5.3. Experiments to determine the optimal plasmid dosage Nine rabbits were selected and randomly divided into three groups (3 in each group) to determine the optimal dosage of the plasmid for gene transfection. The optimal ultrasound parameters determined by the first experiment was used in these tests. A mixed solution containing EGFP plasmids and Sonovue (1:1 ratio) was injected into both knees of each rabbit immediately followed with an ultrasound irradiation. A 100 ll (300 lg), 200 ll (200 lg), and 300 ll (300 lg) of plasmid was given in each group accordingly. The rabbits were sacrificed 5 days after ultrasound irradiation. 2.5.4. Comparison of the transfection efficiency between the experimental and control groups Twelve rabbits with AIA were randomly divided into five groups: (1) ultrasound + microbubbles + plasmid (US + MB + PL), (2) ultrasound + plasmid (US + PL), (3) microbubble + plasmid (MB + PL), (4) plasmid only (PL) and (5) untreated controls. The optimal ultrasound parameters and plasmid dosage as well as injection approach determined by the above experiments were applied to these tests for evaluation of the synovial EGFP expression in each group. 2.5.5. Measurement of the synovial EGFP expression duration, EGFP expression in other organs and normal tissue damage Fifteen rabbits divided into 5 groups were injected with a mixture of SonoVue and plasmid mixture into the joint of the knees bilaterally. The optimal ultrasound parameters and plasmid dosage
determined by the above experiments were applied in these tests for evaluation of the synovial EGFP expression duration, EGFP expression in other organs and normal tissue damage in each group. Three rabbits randomly selected from 15 rabbits were sacrificed at 5, 10, 20, 40 and 60 days after the ultrasound irradiation. The EGFP expression in the synovial joint, liver and spleen was evaluated using fluorescent technique and the subcutaneous tissue and muscle damage in the irradiated area were assessed using an H&E stained specimen inspection. 2.6. Pathological observation The rabbits were sacrificed using an intravenous injection of an overdose of sodium pentobarbital. Longitudinal skin incisions were made in the middle of the knee until the knee joint was exposed. The synovial tissue was completely taken out using a scalpel. The surrounding subcutaneous and muscle tissues in the irradiated area were also selectively collected. Cryosectioning: The freshly sampled synovial tissue was immediately placed in a cryostat for 15 min. The tissue was subsequently trimmed, embedded and fixed. The samples were cryosectioned into 7-lm thick cross-sections at 90-lm intervals and the sections were later stained by 4, 6-diamidino-2-phenylindole (DAPI). Ten areas in each section were randomly selected and observed using a fluorescent-converted microscope (Nikon TE 2000U, Nikon, Japan) under 200 magnification. The sections with the EGFP expression were reviewed and scored using computerized image analysis (Image-Pro Plus; Media Cybernetics). The data were expressed as the fluorescence intensity (mean density) and mean value from three measurements. The liver and spleen were also sampled and stained to evaluate any fluorescence EGFP expression. Hematoxylin and eosin (H & E) staining: In order to determine whether the tissues were damaged or not, the synovial membrane, as well as the subcutaneous and muscle tissues, were fixed in 10% formalin for 24 h. Next, the tissues were embedded in paraffin, cut into 5-lm sections and stained with H & E. The sections also underwent further immunohistochemical staining. The areas showing the strongest GFP expression were scored using computerized image analysis (Image-Pro Plus; Media Cybernetics). The data were expressed as the GFP-staining intensity (mean density). The mean value of three measurements was used. The pathological examinations were performed by two experienced pathologists (JY and XH), who were blind to the experimental groups and conditions. 2.7. Statistical analysis The SPSS 18.0 software (IBM company, USA) was used for statistical analysis. Wilcoxon rank sum tests were used to compare between the mean values of two groups. ANOVA analysis was applied to compare between the mean values of multiple groups. A two-sided test was used, and a P < 0.05 indicated that the difference was statistically significant. 3. Results 3.1. Experiments of the ultrasound pulse conditions The test 1 study demonstrated that, under the same injection dosage, duty cycle and pulse time, the EGFP expression produced by 2 W/cm2 of ultrasound output power was higher than that produced by 1 W/cm2 (P < 0.05) (Fig. 2). No damage in the normal tissue was observed under the light microscope. Thus, the 2 W/ cm2 of the power output was selected for following experiments. The test 2 study showed that, under the same power output and
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
307
Fig. 2. Comparison of the transfection efficiency of different power outputs. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. Comparison of these two groups revealed that the EGFP expression produced by 2 W/cm2 of ultrasonic output power was greater than that produced by 1 W/cm2. ⁄ P < 0.05.
pulse time, the EGFP expression produced by the 20% of duty cycle was greater than that produced by the 30% or the 50% duty cycles (P < 0.05). At a 50% duty cycle, partial muscle necrosis and the infiltration of inflammatory cells in the irradiated area were observed under the light microscope (Fig. 3). Thus, the 20% duty cycle was selected as an optimal parameter. The test 3 study demonstrated that, under the same power output and duty cycle, a 10 min ultrasonic pulse duration resulted in greater EGFP expression compared to 2 and 5 min (P < 0.05). However, skin erythema was observed in the 10-min group. A partial muscle necrosis and the infiltration of inflammatory cells in the irradiated area were noticed in 10 min group under light microscopy (Fig. 4). Thus, 5 min was defined as the optimal pulse duration.
3.2. Injection method experiments Under the same power output, duty cycle and pulse duration, the EGFP expression efficiency produced by untreated controls was rather weak with a significant difference compared to the local injections group and intravenous injections group, and those produced by the local articular injections was greater than that produced by intravenous injections (P < 0.05) (Fig. 5). In addition, the local injections required a less concentrated microbubble and plasmid solution than the intravenous injections. 3.2.1. Experiments of plasmid dosage There was a positive correlation between the tested plasmid dosages and the EGFP expressions (P < 0.05). Since no evidence of
Fig. 3. Comparison of the transfection efficiency of different duty cycles. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. The three groups were compared. The EGFP expression produced by the 20% duty cycle was greater than those produced by the 30% or the 50% duty cycles. ⁄ P < 0.05. Panel c is a cross-section of normal muscle (H & E; 400), and panel d is an image of the results from the 50% duty cycle. We found partial muscle necrosis accompanied by the infiltration of inflammatory cells (H & E; 400).
308
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
Fig. 4. Comparison of the transfection results of different ultrasound pulse durations. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. The three groups were compared. A longer ultrasonic irradiation duration resulted in greater GFP expression. ⁄P < 0.05. Panel c is a longitudinal section of normal muscle (H & E staining; 200), and panel d shows the results for the 10-min irradiation condition. We identified partial muscle necrosis and inflammatory cell infiltration (H & E staining; 400).
Fig. 5. Comparison of different injection methods. a depicts the fluorescence intensity, and b depicts the immunohistochemistry experiment. The EGFP expression in the local articular injection group was greater than that of the intravenous injection group. And the EGFP expression in the untreated controls group was significant weaker than that of the above two groups. ⁄P < 0.05.
injury was found in the normal tissue (Fig. 6) 300 lg of plasmid was considered as the optimal dosage for local joint injection. 3.2.2. Comparison of the transfection efficiency between the experimental and control groups According to initial optimization tests, the experimental groups were injected with a mixture of microbubbles and 300 lg of plasmids along with the ultrasound parameters of 2 W/cm2 output power, a 20% duty cycle and a 5 min ultrasound irradiation for gene transfection efficiency experiments. The results showed that the EGFP transfection efficiencies of the US + MB + PL group, the US + PL group, MB + PL group and PL group were much greater than
that of the untreated controls group, and the former two groups were greater than that of the later three groups. In particular, the US + MB + PL group showed the greatest EGFP expression among the other groups (P < 0.05). The EGFP expression levels of the MB + PL group and the plasmid-only group were minimal with no statistically significant differences between these two groups (P > 0.05) (Figs. 7–9). 3.3. Duration of EGFP expression in the synovial pannus EGFP expression in the US + MB + PL group was observed in the synovial pannus on days 5, 10, 20 and 40. The expression reached a
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
309
Fig. 6. Comparison of transfection efficiency of different quantities of plasmids. a depicts the fluorescence intensity, and b depicts the immunohistochemistry experiment. Three groups were compared. Greater plasmid quantity results in greater GFP expression. ⁄P < 0.05.
peak level on day 5 and gradually declined after the peak and disappeared on day 60 (Fig. 10). 3.4. Detection of EGFP expression in the synovial pannus and other organs Green fluorescence was visible in the synovial lining and lower synovium in the US + MB + PL group. Brown particles were also observed in the lining layer and lower synovium in the immunohistochemistry experiment. No EGFP expression was observed in the liver and spleen (Fig. 11). 4. Discussion The research on the treatments for RA has made considerable progress in the past 10 years. However, current treatment approaches have been restricted as a result of high cost, severe adverse effects and the need for repeated injections. Gene therapy has potential advantages over the currently available treatments because of its high selectivity and persistent effects on the targeted gene. The goal of RA gene therapy is to transfect synovial cells with genes encoding therapeutic proteins and then the synovial cells subsequently synthesize and secrete these anti-arthritic molecules. In this way, the joint tissue synthesizes special materials that treat its own inflammation. The advantages of this approach include: (1) The therapeutic agents are directly and immediately synthesized in
the diseased areas. Thus, the issues related to drug transfer and side effects on the non-targeted tissues observed in the other treatments are ameliorated. (2) Gene therapy is highly applicable for treating chronic diseases. The genes transfected into the synovial lining maintain their activity for extended durations. This feature greatly benefits the treatment of arthritis [4,13]. The synovial pannus in patients with RA reflects diseased synovial tissue. This pannus lesion exhibits features of active proliferation, high metabolism, rich vasculature and strong invasiveness, which contribute to the pathogenesis of the joint disease and to cartilage destruction. In addition, the synovial cells secrete TNFa, IL-1 and other cytokines under inflammatory conditions, which contributes to the continued development and deterioration of joint damage in RA. Therefore, these synovial cells should be the targeted cells for gene transduction [6,7]. There are two different strategies for transferring genes into synovial cells. The direct method consists of an intra-articular injection of vectors that can be transfected into the synovial cells in situ. In contrast, the indirect method consists of removing the synovial membrane from the joint and transfecting the cell in vitro, and then these transfected cells are later re-introduced to the joint [14]. The direct method is simpler and less technical demanding when compared to indirect method. Thus, the direct method may be more suitable for a wide variety of clinical applications. At present, the vectors used for gene therapy can be broadly divided into viral and non-viral vectors. The viral vectors have been successfully utilized for in vivo gene transfection in synovial
Fig. 7. Comparison of the transfection efficiency between different groups. a depicts the fluorescence intensity, and b depicts the immunohistochemistry experiment. The EGFP transfection efficiencies of the US + MB + PL group, the US + PL group, the MB + PL group and the PL group were significantly higher than that of the untreated controls group, and the former two groups were higher than that of the later three groups The EGFP transfection efficiency of the US + MB + PL group was significantly higher than that of the US + PL group. Furthermore, the EGFP expression of the MB + PL group and the PL group were both notably low. ⁄P < 0.05.
310
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
Fig. 8. Fluorescence microscopy examination of EGFP expression. a, c, e, g and i depict fluorescence images, and b, d, f, h and j depict DAPI images. a and b depict the results of the US + MB + PL group, c and d depict the results of the UL + PL group, e and f depict the results of the MB + PL group, g and h depict the results of the PL group, and i and j depict the results of the untreated controls group. The US + MB + PL group showed the brightest fluorescence (magnification, 200).
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
311
Fig. 9. Immunohistochemical analysis of EGFP expression. a depicts the results of the US + MB + PL group, b depicts the results of the UL + PL group, c depicts the results of the MB + PL group, d depicts the results of the PL group and e depicts the results of the untreated controls group. The US + MB + PL group revealed the greatest positive immunoreactivity (magnification, 400).
Fig. 10. Comparison of transfection efficiency for different detection periods. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. EGFP expression was observed in the synovial pannus on days 5, 10, 20 and 40. The expression levels peaked on day 5, and no GFP expression was detected on day 60.
312
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
Fig. 11. Detection of EGFP expression under fluorescence microscopy. Panels a and c depict fluorescence images, and panels b and d depict DAPI images. Panels a and b depict the results of the liver, and panels c and d depict the results of the spleen. No fluorescence was detected in the liver or the spleen (magnification, 200).
membranes with relatively high transfection efficiency [15,16]. However, there are still many limitations associated with treatments using viral vectors. The treatments can trigger an immune response in the body and increase the risk of insertional mutagenesis that leads to tumorigenicity and toxicity. Vectors have also been associated with limited capacity and low preparation titers. The methods involving non-viral vectors, such as plasmids, liposomes and others can avoid these limitations. However, the efficiency of the gene expression is not sufficiently high, and the duration of the gene expression is relatively short using non-viral approach. Thus, the exploration and development of transduction components and technology are priorities of current research. The UTMD technique is a new concept and approach for gene transfection. Using ultrasound irradiation to break the microbubbles in the targeted tissue at a specific time and location could improve targeted gene therapy. UTMD research studies have been conducted in a variety of in vitro cells and in vivo tissues and organs [17–21]. A few studies have applied ultrasound in combination with plasmids and microbubbles to transfect the gene in normal synovial membranes [22,23]. However, no studies have been conducted using UTMD technique for gene transfection in the synovial pannus of the AIA model. Our research is focused on the UTMDmediated enhancement of gene transfection in the synovial pannus and also aimed to identify the optimal transfection conditions for a new non-viral gene transfection method for gene therapy of the arthritis.
4.1. Optimal UTMD transfection conditions for the synovial pannus Ultrasonic cavitation can improve gene transfection efficiency and also induce cytotoxic effects. Specifically, this technique can damage vascular endothelium and other tissue cells. Several
independent studies reported that ultrasonic cavitation can damage to vascular endothelial cells and causes capillary rupture and bleeding [24,25], lead to cell lysis [26,27], and activates platelets [28]. Therefore, in the UTMD technique, the ultrasonic intensity and consequent tissue damage must be tested and balanced in order to establish the optimal conditions for gene therapy of the arthritis. The study revealed that EGFP expression without damage the irradiated normal tissue was significantly increased at 2 W/cm ultrasound output compared to 1 W/cm. The highest transfection efficiency was achieved with a 20% duty cycle while a 50% duty cycle caused muscle injury in the irradiated area. Furthermore, the transfection efficiency was significantly enhanced with increasing ultrasound irradiation duration while a 10 min ultrasound irradiation caused skin erythema and muscle injury. Through a series of experiments, the optimal parameters and conditions for synovial gene transfection was an ultrasonic output power of 2 W/cm2, a duty cycle of 20%, and an ultrasonic irradiation duration of 5 min with local articular injection and a plasmid dosage of 300 lg. Under these conditions, relatively high gene transfection efficiency was achieved without damaging the normal tissue in the irradiated area.
4.2. Comparison of the transfection efficiency By injecting expression plasmids carrying reporter genes into the knees of rabbits and rats Yovandich et al. [29] demonstrated that synovial tissue is capable of rapidly uptaking injected plasmid DNA. However, the transfection efficiency was low. A study by Taniyama et al. has showed that the UTMD technique can effectively promotes gene transfection [30]. By transfecting human vascular smooth muscle and endothelial cells in vitro with a
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
luciferase-encoding plasmid DNA, they found that the transfection activity was 70 times greater in the US treatment group and 7000– 8000 times higher in the microbubble + US group when compared to plasmid only group. Furthermore, in the in vivo experiments, the luciferase activity of the group treated with plasmid + microbubble + US was approximately 1000 times higher than the group treated with the plasmid only. Similar results were obtained in the present study. We found that the EGFP transfection efficiency of the group treated with ultrasound combined with the microbubbles and plasmids was significantly higher than that of the other three groups. Ultrasound irradiation alone could also facilitate gene transfection, but the transfection efficiency was lower than that of the group treated with ultrasound and the microbubbles. Therefore, the UTMD technique significantly enhances the gene transfection efficiency in the synovial pannus. 4.3. Duration of the EGFP expression in the synovial membrane Joint inflammation often lasts for several weeks. Previous studies have found that the synovial gene expression achieved through an articular injection of the plasmid alone lasts for only 2–3 days [29,31]. This expression duration does not meet the requirements for articular gene therapy. The existing RA treatments are mainly associated with the transfection of TNF-a, IL-1 and other cytokine antagonists. Thus, if the expression duration is excessive, the treatment can cause excessive physiological regulation and immune system dysfunction. Yang et al. have found that the gene expression induced by the UTMD technique can last for weeks or even months in different tissue and cell types [32,33]. In our study, the gene expression in the synovial membrane reached to the peak on the fifth day and gradually declined until day 40. No significant fluorescence expression was detected in the synovium on day 60. Therefore, the UTMD-transfected genes can persist locally for a significant period of time but not for excessive durations. Thus, this feature is fitted well with the requirements for RA gene therapy. 4.4. Examination of EGFP expression in the synovial pannus and other organs Synovial lesions involve the cells in the synovial lining as well as the inflammatory cells in the deep synovium. These cells secrete large amounts of inflammatory cytokines and thereby promote inflammatory processes. Therefore, the optimal gene therapy method should be associated with a certain penetration capacity and should be able to carry genes into the deep tissues to achieve the complete treatment effect. Fluorescence and immunohistochemistry experiments both showed that the GFP expression could be found in the synovial lining and the deep synovium of the UTMD-transfected synovial pannus. These results indicate that this method has a certain penetration capacity of gene transfection. Previous study has shown that applying gene therapy in a single joint has mitigating effects on the arthritis of the other joints [34]. Our study confirmed that genes could be successfully transfected into the deep synovium. Because the deep synovium is rich with microvessels, the proteins released from the genetically modified synovial cells can be absorbed into the bloodstream through these vessels. Thus, the treatment could be effective not only locally but also potentially systemically. This mechanism may explain the relief of the lesions in the other joints. Since RA often manifests as a polyarticular disease, this gene transfection method is useful for systemic treatment. With the exception of the synovial pannus, no fluorescence was observed in the liver, spleen and other organs in the present study, which indicates that the synovial gene transfection via the UTMD method is expressed locally without systematic effect.
313
5. Conclusions In conclusion, our study demonstrated that the optimal UTMD conditions for enhancing synovial pannus transfections include 2 W/cm2 power output, 20% duty cycle, 5-min ultrasound irradiation duration, 300 lg of plasmid with a locally articular injection of the microbubble and plasmid mixture. The UTMD technique significantly enhances the plasmid transfection efficiency in the synovial pannus of the AIA model. The targeted expression exists for approximately 40 days and does not damage the surrounding normal tissue. This UTMD technique could became a safe and effective non-viral gene transfection method for the gene therapy of the arthritis. Acknowledgements This study was supported by General Projects of National Natural Science Foundation of China (approval number: 81271585) and Sichuan Science and Technology Plan (approval number: 2015SZ0125). References [1] C. Roubille, V. Richer, T. Starnino, C. McCourt, A. McFarlane, P. Fleming, et al., The effects of tumour necrosis factor inhibitors, methotrexate, non-steroidal anti-inflammatory drugs and corticosteroids on cardiovascular events in rheumatoid arthritis, psoriasis and psoriatic arthritis: a systematic review and meta-analysis, Ann. Rheum. Dis. 74 (3) (2015) 480–489. [2] J.P. Leombruno, T.R. Einarson, E.C. Keystone, The safety of anti-tumour necrosis factor treatments in rheumatoid arthritis: meta and exposure-adjusted pooled analyses of serious adverse events, Ann. Rheum. Dis. 68 (2009) 1136–1145. [3] R. Alten, J. Gomez-Reino, P. Durez, A. Beaulieu, A. Sebba, G. Krammer, et al., Efficacy and safety of the human anti-IL-1b monoclonal antibody canakinumab in rheumatoid arthritis: results of a 12-week. Phase II, dose-finding study, BMC Musculoskelet Disord. 12 (2011) 153. [4] S. Fabre, F. Apparailly, Gene therapy for rheumatoid arthritis: current status and future prospects, BioDrugs 25 (2011) 381–391. [5] C. Castro-Villegas, C. Pérez-Sánchez, A. Escudero, I. Filipescu, M. Verdu, P. RuizLimón, et al., Circulating miRNAs as potential biomarkers of therapy effectiveness in rheumatoid arthritis patients treated with anti-TNFa, Arthritis. Res. Ther. 17 (2015) 49. [6] D. Hammaker, D.L. Boyle, G.S. Firestein, Synoviocyte innate immune responses: TANK-binding kinase-1 as a potential therapeutic target in rheumatoid arthritis, Rheumatology (Oxford) 51 (2012) 610–618. [7] T. Laragione, M. Brenner, A. Mello, M. Symons, P.S. Gulko, The arthritis severity locus Cia5d is a novel genetic regulator of the invasive properties of synovial fibroblasts, Arthritis. Rheum. 58 (2008) 2296–2306. [8] Z.P. Shen, A.A. Brayman, L. Chen, C.H. Miao, Ultrasound with microbubbles enhances gene expression of plasmid DNA in the liver via intraportal delivery, Gene Ther. 15 (2008) 1147–1155. [9] A. Aoi, Y. Watanabe, S. Mori, M. Takahashi, G. Vassaux, T. Kodama, Herpes simplex virus thymidine kinase-mediated suicide gene therapy using nano microbubbles and ultrasound, Ultrasound Med. Biol. 34 (2008) 425–434. [10] K. Un, S. Kawakami, R. Suzuki, K. Maruyama, F. Yamashita, M. Hashida, Enhanced transfection efficiency into macrophages and dendritic cells by a combination method using mannosylated lipoplexes and bubble liposomes with ultrasound exposure, Hum. Gene Ther. 21 (2010) 65–74. [11] T. Okura, K. Marutsuka, H. Hamada, T. Sekimoto, T. Fukushima, Y. Asada, et al., Therapeutic efficacy of intra-articular adrenomedullin injection in antigeninduced arthritis in rabbits, Arthritis. Res. Ther. 10 (2008) R133. [12] L. Qiu, Y. Jiang, L. Zhang, L. Wang, Y. Luo, Ablation of synovial pannus using microbubble-mediated ultrasonic cavitation in antigen-induced arthritis in rabbits, Rheumatol. Int. 32 (12) (2012) 3813–3821. [13] J.L. Roybal, M. Endo, A. Radu, P.W. Zoltick, A.W. Flake, Early gestational gene transfer of IL-10 by systemic administration of lentiviral vector can prevent arthritis in a murine model, Gene Ther. 18 (2011) 719–726. [14] M.J. Park, H.S. Park, M.L. Cho, H.J. Oh, Y.G. Cho, S.Y. Min, et al., Transforming growth factor b-transduced mesenchymal stem cells ameliorate experimental autoimmune arthritis through reciprocal regulation of Treg/Th17 cells and osteoclastogenesis, Arthritis. Rheum. 63 (2011) 1668–1680. [15] S.W. Tas, J. Adriaansen, N. Hajji, A.C. Bakker, G.S. Firestein, M.J. Vervoordeldonk, et al., Amelioration of arthritis by intraarticular dominant negative Ikk beta gene therapy using adeno-associated virus type 5, Hum. Gene Ther. 17 (2006) 821–832. [16] L.R. Goodrich, B.D. Brower-Toland, L. Warnick, P.D. Robbins, C.H. Evans, A.J. Nixon, Direct adenovirus-mediated IGF-I gene transduction of synovium induces persisting synovial fluid IGF-I ligand elevations, Gene Ther. 13 (17) (2006) 1253–1262.
314
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
[17] H. Guo, J.C. Leung, L.Y. Chan, A.W. Tsang, M.F. Lam, H.Y. Lan, et al., Ultrasoundcontrast agent mediated naked gene delivery in the peritoneal cavity of adult rat, Gene Ther. 14 (2007) 1712–1720. [18] J.M. Escoffre, A. Zeghimi, A. Novell, A. Bouakaz, In-vivo gene delivery by sonoporation: recent progress and prospects, Curr. Gene. Ther 13 (1) (2013) 2– 14. [19] M. Tomizawa, F. Shinozaki, Y. Motoyoshi, T. Sugiyama, S. Yamamoto, M. Sueishi, Sonoporation: Gene transfer using ultrasound, World J. Methodol. 3 (4) (2013) 39–44. [20] L. Pinto de Carvalho, D. Takeshita, B.A. Carillo, B.C. Garcia Lisbôa, G. Molina, A. Beutel, et al., Hydrodynamics- and ultrasound-based transfection of heart with naked plasmid DNA, Hum. Gene Ther. 18 (2007) 1233–1243. [21] A. Delalande, C. Leduc, P. Midoux, M. Postema, C. Pichon, Efficient gene delivery by sonoporation is associated with microbubble entry into cells and the clathrin-dependent endocytosis pathway, Ultrasound Med. Biol. 41 (7) (2015) 1913–1926. [22] Y. Negishi, Y. Tsunoda, Y. Endo-Takahashi, Y. Oda, R. Suzuki, K. Maruyama, et al., Local gene delivery system by bubble liposomes and ultrasound exposure into joint synovium, J Drug Deliv. 2011 (2011) 203986. [23] M. Saito, O. Mazda, K.A. Takahashi, Y. Arai, T. Kishida, M. Shin-Ya, et al., Sonoporation mediated transduction of pDNA/siRNA into joint synovium in vivo, J. Orthop. Res. 25 (2007) 1308–1316. [24] E.L. Carstensen, S. Gracewski, D. Dalecki, The search for cavitation in vivo, Ultrasound Med. Biol. 26 (9) (2000) 1377–1385. [25] D. Dalecki, C.H. Raeman, S.Z. Child, D.P. Penney, R. Mayer, E.L. Carstensen, The influence of contrast agents on hemorrhage produced by lithotripter fields, Ultrasound Med. Biol. 23 (1997) 1435–1439.
[26] D.L. Miller, J. Quddus, Lysis and sonoporation of epidermoid and phagocytic monolayer cells by diagnostic ultrasound activation of contrast agent gas bodies, Ultrasound Med. Biol. 27 (2001) 1107–1113. [27] D.L. Miller, R.M. Thomas, Thresholds for hemorrhages in mouse skin and intestine induced by lithotripter shock waves, Ultrasound Med. Biol. 21 (1995) 249–257. [28] S.L. Poliachik, W.L. Chandler, P.D. Mourad, R.J. Ollos, L.A. Crum, Activation, aggregation and adhesion of platelets exposed to high intensity focused ultrasound, Ultrasound Med. Biol. 27 (2001) 1567–1576. [29] J. Yovandich, B. O’Malley Jr., M. Sikes, F.D. Ledley, Gene transfer to synovial cells by intra-articular administration of plasmid DNA, Hum. Gene Ther. 6 (1995) 603–610. [30] Y. Taniyama, K. Tachibana, K. Hiraoka, T. Namba, K. Yamasaki, N. Hashiya, et al., Local delivery of plasmid DNA into rat carotid artery using ultrasound, Circulation 105 (2002) 1233–1239. [31] I. Nita, S.C. Ghivizzani, J. Galea-Lauri, G. Bandara, H.I. Georgescu, P.D. Robbins, et al., Direct gene delivery to synovium. An evaluation of potential vectors in vitro and in vivo, Arthritis Rheum. 39 (1996) 820–828. [32] K. Nishida, M. Doita, T. Takada, K. Kakutani, H. Miyamoto, T. Shimomura, et al., Sustained transgene expression in intervertebral disc cells in vivo mediated by microbubble-enhanced ultrasound gene therapy, Spine (Phila Pa 1976) 31 (2006) 1415–1419. [33] L. Yang, Y. Shirakata, K. Tamai, X. Dai, Y. Hanakawa, S. Tokumaru, et al., Microbubble-enhanced ultrasound for gene transfer into living skin equivalents, Dermatol. Sci. 40 (2005) 105–114. [34] A. Inoue, K.A. Takahashi, O. Mazda, Y. Arai, M. Saito, T. Kishida, et al., Comparison of anti-rheumatic effects of local RNAi-based therapy in antigen induced arthritis rats using various cytokine genes as molecular targets, Mod. Rheumatol. 19 (2009) 125–133.
Contents lists available at ScienceDirect
Ultrasonics journal homepage: www.elsevier.com/locate/ultras
Targeted gene delivery to the synovial pannus in antigen-induced arthritis by ultrasound-targeted microbubble destruction in vivo Xi Xiang, Yuanjiao Tang, Qianying Leng, Lingyan Zhang, Li Qiu ⇑ Department of Ultrasound, West China Hospital of Sichuan University, Chengdu, Sichuan, China
a r t i c l e
i n f o
Article history: Received 1 July 2015 Received in revised form 9 September 2015 Accepted 16 September 2015 Available online 26 September 2015 Keywords: Ultrasound-mediated microbubble destruction Gene transfection Enhanced green fluorescent protein Synovial pannus Antigen-induced arthritis
a b s t r a c t The purpose of this study was to optimize an ultrasound-targeted microbubble destruction (UTMD) technique to improve the in vivo transfection efficiency of the gene encoding enhanced green fluorescent protein (EGFP) in the synovial pannus in an antigen-induced arthritis rabbit model. A mixture of microbubbles and plasmids was locally injected into the knee joints of an antigen-induced arthritis (AIA) rabbits. The plasmid concentrations and ultrasound conditions were varied in the experiments. We also tested local articular and intravenous injections. The rabbits were divided into five groups: (1) ultrasound + microbubbles + plasmid; (2) ultrasound + plasmid; (3) microbubble + plasmid; (4) plasmid only; (5) untreated controls. EGFP expression was observed by fluorescent microscope and immunohistochemical staining in the synovial pannus of each group. The optimal plasmid dosage and ultrasound parameter were determined based on the results of EGFP expression and the present and absent of tissue damage under light microscopy. The irradiation procedure was performed to observe the duration of the EGFP expression in the synovial pannus and other tissues and organs, as well as the damage to the normal cells. The optimal condition was determined to be a 1-MHz ultrasound pulse applied for 5 min with a power output of 2 W/cm2 and a 20% duty cycle along with 300 lg of plasmid. Under these conditions, the synovial pannus showed significant EGFP expression without significant damage to the surrounding normal tissue. The EGFP expression induced by the local intra-articular injection was significantly more increased than that induced by the intravenous injection. The EGFP expression in the synovial pannus of the ultrasound + microbubbles + plasmid group was significantly higher than that of the other four groups (P < 0.05). The expression peaked on day 5, remained detectable on day 40 and disappeared on day 60. No EGFP expression was detected in the other tissues and organs. The UTMD technique can significantly enhance the in vivo gene transfection efficiency without significant tissue damage in the synovial pannus of an AIA model. Thus, this could become a safe and effective non-viral gene transfection procedure for arthritis therapy. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Rheumatoid arthritis (RA) is a systemic autoimmune disease that is characterized by chronic joint inflammation. Synovial pannus development is the main pathological basis for the joint lesions and cartilage destruction in RA. Patients experience the onset of this disease in youth and middle age. RA is characterized by a chronic course and persistent progression despite treatment and it is a refractory disease that often leads to limb disability. Therefore, researchers are actively searching for effective treatments for RA. The administration of cytokine antagonists, such as ⇑ Corresponding author at: Department of Ultrasound, West China Hospital of Sichuan University, No. 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, China. Tel.: +86 18980602044; fax: +86 02885423192. E-mail address: [email protected] (L. Qiu). http://dx.doi.org/10.1016/j.ultras.2015.09.011 0041-624X/Ó 2015 Elsevier B.V. All rights reserved.
antagonists of tumor necrosis factor a (TNF-a) and interleukin 1 (IL-1), can significantly improve RA symptoms [1–3]. However, these treatments only have a short effective time, and repeated administration is needed. The treatments are also given longterm and are thus associated with high costs. High-dose systemic injections or repeated intra-articular injections of the drugs are required to achieve effective local therapeutic concentrations in the joints. This feature of the therapy results in a large economic burden and in patient suffering. Recently, gene therapy has been proposed for treatment of RA [4,5]. There are numerous advantages of gene therapy. For instance, this therapy can be locally administered and provide stable and continuous in vivo expression of the therapeutic molecules. Therefore, gene therapy through the transfection and in vivo expression of therapeutic molecules may be a promising treatment method for RA [4,5]. Effective gene transfer
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
could relieve invasive joint synovitis, thereby improving the symptoms of rheumatoid arthritis. In addition, local gene expression therapy is not associated with the side effects that can be caused by systemic drug administration. Furthermore, the gene expression remains relatively constant. Thus, the chance of adverse effect is reduced because this method does not cause the temporary excessive drug concentrations associated with articular injections. The pathological changes observed in joint diseases are closely associated with the synovial tissues. The synovial cells are targeted in pathological conditions, and can also become participants in pathogenesis through multiple pathways. For example, these cells can pathologically secrete such substances as TNF-a, IL-1, the cell adhesion molecule ICAM-1 and reactive oxygen species. Furthermore, the persistent development and deterioration of joint damage in RA is associated with continuously amplified inflammation through the autocrine and paracrine release of cytokines. Therefore, the cells of the synovial tissues are ideal targets for gene therapy [6,7]. The current challenge is to select the optimal gene transfection method that minimizes the side effects and maximizes the effectiveness of successful gene delivery. At present, the vectors that have been applied for gene therapy can be broadly divided into the viral and the non-viral vectors. Each of these methods has their own limitations. Ultrasound-targeted microbubble destruction (UTMD) is a novel method for non-viral gene transfection. The UTMD technique compensates for the deficiencies of the viral and non-viral vectors, which exhibits greater transfection efficiency, specificity and safety. The studies of using UTMD for gene transfection include in vitro cell culture and in vivo heart, liver, blood vessel, kidney, skeletal muscle and retina models [8–10]. However, no studies of gene transfection using UTMD in synovial pannus of an RA model have been reported. The aim of this study is to optimize the UTMD technique to improve the in vivo transfection efficiency of the gene encoding enhanced green fluorescent protein (EGFP) in the synovial pannus of antigen-induced arthritis rabbit model.
305
Fig. 1. The exact injection site of knee joint in a rabbit.
Switzerland). The plasmids were transformed into DH5acompetent cells (Invitrogen, Carlsbad, CA, USA), and then extracted and purified according to the instructions of the plasmid extraction kit (Qiagen, Hilden, Germany). The concentration of the resulting plasmid DNA with absorbance at 260 nm (A260) was measured using spectrophotometry (1 lg/ll). The OD260 to OD280 ratio was between 1.8 and 2.1, which indicated that the plasmid DNA was not contaminated. The product was stored at 20 °C for later use. 2.3. Preparation of the SonoVue/DNA complexes
2. Materials and methods
The SonoVue microbubble suspensions (Bracco Imaging, Milano, Italy) were prepared at a density microbubbles of 2 108–5 108/ml and a concentration of 5 mg/ml using a 0.9% saline solution. Next, 100 ll of SonoVue was mixed with 100 ll of the plasmid DNA solution and stored at 4 °C just prior to each experiment.
2.1. Antigen-induced arthritis rabbit model
2.4. Animal preparation and experimental procedures
New Zealand white rabbits that weighted 2.5–3.0 kg with 1:1 ratio of male: female were used in this study. All of the rabbits were kept in a room with the appropriate temperature and humidity and fed with standard laboratory chow and water. All of the experiments were approved by the Sichuan University Research Committee for Animal Research. Four milligrams of chicken ovalbumin (OVA, Sigma, St Louis, MO, USA) were dissolved in 0.5 ml of Freund’s Complete Adjuvant (F-5881) (Sigma Chemical Co., St. Louis, MO, USA) and 0.5 ml of phosphate-buffered saline (PBS). One milliliter of this emulsion was percutaneously injected into five interscapular areas in the rabbit. The same injections were repeated every seven days for 3 weeks in total. Five days after the third injection, 10 mg of OVA was dissolved in 0.5 ml of normal saline and then injected into both knees of the rabbits [11,12], the exact injection site is shown in Fig. 1. In the fourth week after the OVA articular injection, ultrasound imaging was performed to evaluate the AIA lesions based on the presence of synovial thickening. The rabbits with AIA lesions were randomly divided into various experimental groups.
The rabbits were anesthetized with an intraperitoneal injection of 2% sodium pentobarbital (40 mg/kg) and fixed on the experimental table. The skin fur on the knee was removed, and coupling gel was spread on the skin. An ultrasound therapy device (Sonic Master ES-2, OG Giken, Japan) with a probe frequency of 1 MHz, transducer area of 10 cm2, and the pulse repetition frequency of 100 Hz was used in this study. The output power (1 W/cm2, 2 W/ cm2), duty cycle (20%, 30%, 50%) and time (2 min, 5 min,10 min) were adjusted throughout the experiments. The local articular injections were performed by injecting the various solution depending on the different groups with a 1 ml syringe into the knee joint cavity. UTMD was performed on the injected area immediately after the injection. The intravenous injections of contrast agent were performed by slowly and steadily injecting the solution into the ear vein. UTMD was performed on the knee after the initiation of the injection.
2.2. Plasmid DNA amplification, extraction and purification The pEGFP-C1 plasmid (4.2 kb) is an enhanced green fluorescent protein (EGFP) and purchased from Roche Applied Science (Basel,
2.5. Experimental series 2.5.1. Experiments to identify the optimal ultrasound pulse conditions Three experimental tests were performed. A mixture containing 200 ll (300 lg) of EGFP plasmid and 200 ll of SonoVue was injected into the knees bilaterally in each rabbit. The injections were immediately followed with an ultrasound pulse. The rabbits
306
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
were sacrificed 5 days after the ultrasound pulse. The optimal ultrasound pulse conditions were identified. Test 1 was to show the transfection efficiency with different power outputs by using 6 rabbits divided into two groups (3 in each). With the same 20% duty cycles and 5 min treatment time for both groups, the 1 W/cm2 output power was applied for the first group and 2 W/cm2 for the second group. Test 2 was to show the transfection efficiency with different duty cycles. Nine rabbits were selected and randomly divided into three groups. The 20% duty cycle was applied in the first group 30% in the second group, and 50% in the third group. The pulse duration for all three groups was 5 min. The optimal power output determined by the results of test 1 was used in each group. Test 3 was to show the transfection efficiency with different pulse durations. Nine rabbits were selected and randomly divided into three groups. The 2 min, 5 min, and 10 min pulse durations was applied in each group, respectively. The optimal power output determined by test 1 and the optimal duty cycle determined by test 2 were used in each group. 2.5.2. Experiments to compare injection approaches Six rabbits were selected and randomly divided into two groups to see which injection approach was more effectiveness for gene transfection. The optimal ultrasound parameters determined by the first experiment was used in these tests. A mixed solution containing 200 ll (200 lg) of EGFP plasmid and 200 ll of SonoVue was injected into both knees of the first group rabbits while 1 ml of mixed solution containing 0.5 ml (500 lg) of EGFP plasmid and 0.5 ml of Sonovue was divided into equal two doses (0.5 ml each) and two doses were slowly injected into the ear veins of the second group of rabbits at 30 min interval. Finally, another three rabbits received no treatment as the untreated controls. The rabbits were sacrificed 5 days after ultrasonic irradiation. 2.5.3. Experiments to determine the optimal plasmid dosage Nine rabbits were selected and randomly divided into three groups (3 in each group) to determine the optimal dosage of the plasmid for gene transfection. The optimal ultrasound parameters determined by the first experiment was used in these tests. A mixed solution containing EGFP plasmids and Sonovue (1:1 ratio) was injected into both knees of each rabbit immediately followed with an ultrasound irradiation. A 100 ll (300 lg), 200 ll (200 lg), and 300 ll (300 lg) of plasmid was given in each group accordingly. The rabbits were sacrificed 5 days after ultrasound irradiation. 2.5.4. Comparison of the transfection efficiency between the experimental and control groups Twelve rabbits with AIA were randomly divided into five groups: (1) ultrasound + microbubbles + plasmid (US + MB + PL), (2) ultrasound + plasmid (US + PL), (3) microbubble + plasmid (MB + PL), (4) plasmid only (PL) and (5) untreated controls. The optimal ultrasound parameters and plasmid dosage as well as injection approach determined by the above experiments were applied to these tests for evaluation of the synovial EGFP expression in each group. 2.5.5. Measurement of the synovial EGFP expression duration, EGFP expression in other organs and normal tissue damage Fifteen rabbits divided into 5 groups were injected with a mixture of SonoVue and plasmid mixture into the joint of the knees bilaterally. The optimal ultrasound parameters and plasmid dosage
determined by the above experiments were applied in these tests for evaluation of the synovial EGFP expression duration, EGFP expression in other organs and normal tissue damage in each group. Three rabbits randomly selected from 15 rabbits were sacrificed at 5, 10, 20, 40 and 60 days after the ultrasound irradiation. The EGFP expression in the synovial joint, liver and spleen was evaluated using fluorescent technique and the subcutaneous tissue and muscle damage in the irradiated area were assessed using an H&E stained specimen inspection. 2.6. Pathological observation The rabbits were sacrificed using an intravenous injection of an overdose of sodium pentobarbital. Longitudinal skin incisions were made in the middle of the knee until the knee joint was exposed. The synovial tissue was completely taken out using a scalpel. The surrounding subcutaneous and muscle tissues in the irradiated area were also selectively collected. Cryosectioning: The freshly sampled synovial tissue was immediately placed in a cryostat for 15 min. The tissue was subsequently trimmed, embedded and fixed. The samples were cryosectioned into 7-lm thick cross-sections at 90-lm intervals and the sections were later stained by 4, 6-diamidino-2-phenylindole (DAPI). Ten areas in each section were randomly selected and observed using a fluorescent-converted microscope (Nikon TE 2000U, Nikon, Japan) under 200 magnification. The sections with the EGFP expression were reviewed and scored using computerized image analysis (Image-Pro Plus; Media Cybernetics). The data were expressed as the fluorescence intensity (mean density) and mean value from three measurements. The liver and spleen were also sampled and stained to evaluate any fluorescence EGFP expression. Hematoxylin and eosin (H & E) staining: In order to determine whether the tissues were damaged or not, the synovial membrane, as well as the subcutaneous and muscle tissues, were fixed in 10% formalin for 24 h. Next, the tissues were embedded in paraffin, cut into 5-lm sections and stained with H & E. The sections also underwent further immunohistochemical staining. The areas showing the strongest GFP expression were scored using computerized image analysis (Image-Pro Plus; Media Cybernetics). The data were expressed as the GFP-staining intensity (mean density). The mean value of three measurements was used. The pathological examinations were performed by two experienced pathologists (JY and XH), who were blind to the experimental groups and conditions. 2.7. Statistical analysis The SPSS 18.0 software (IBM company, USA) was used for statistical analysis. Wilcoxon rank sum tests were used to compare between the mean values of two groups. ANOVA analysis was applied to compare between the mean values of multiple groups. A two-sided test was used, and a P < 0.05 indicated that the difference was statistically significant. 3. Results 3.1. Experiments of the ultrasound pulse conditions The test 1 study demonstrated that, under the same injection dosage, duty cycle and pulse time, the EGFP expression produced by 2 W/cm2 of ultrasound output power was higher than that produced by 1 W/cm2 (P < 0.05) (Fig. 2). No damage in the normal tissue was observed under the light microscope. Thus, the 2 W/ cm2 of the power output was selected for following experiments. The test 2 study showed that, under the same power output and
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
307
Fig. 2. Comparison of the transfection efficiency of different power outputs. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. Comparison of these two groups revealed that the EGFP expression produced by 2 W/cm2 of ultrasonic output power was greater than that produced by 1 W/cm2. ⁄ P < 0.05.
pulse time, the EGFP expression produced by the 20% of duty cycle was greater than that produced by the 30% or the 50% duty cycles (P < 0.05). At a 50% duty cycle, partial muscle necrosis and the infiltration of inflammatory cells in the irradiated area were observed under the light microscope (Fig. 3). Thus, the 20% duty cycle was selected as an optimal parameter. The test 3 study demonstrated that, under the same power output and duty cycle, a 10 min ultrasonic pulse duration resulted in greater EGFP expression compared to 2 and 5 min (P < 0.05). However, skin erythema was observed in the 10-min group. A partial muscle necrosis and the infiltration of inflammatory cells in the irradiated area were noticed in 10 min group under light microscopy (Fig. 4). Thus, 5 min was defined as the optimal pulse duration.
3.2. Injection method experiments Under the same power output, duty cycle and pulse duration, the EGFP expression efficiency produced by untreated controls was rather weak with a significant difference compared to the local injections group and intravenous injections group, and those produced by the local articular injections was greater than that produced by intravenous injections (P < 0.05) (Fig. 5). In addition, the local injections required a less concentrated microbubble and plasmid solution than the intravenous injections. 3.2.1. Experiments of plasmid dosage There was a positive correlation between the tested plasmid dosages and the EGFP expressions (P < 0.05). Since no evidence of
Fig. 3. Comparison of the transfection efficiency of different duty cycles. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. The three groups were compared. The EGFP expression produced by the 20% duty cycle was greater than those produced by the 30% or the 50% duty cycles. ⁄ P < 0.05. Panel c is a cross-section of normal muscle (H & E; 400), and panel d is an image of the results from the 50% duty cycle. We found partial muscle necrosis accompanied by the infiltration of inflammatory cells (H & E; 400).
308
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
Fig. 4. Comparison of the transfection results of different ultrasound pulse durations. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. The three groups were compared. A longer ultrasonic irradiation duration resulted in greater GFP expression. ⁄P < 0.05. Panel c is a longitudinal section of normal muscle (H & E staining; 200), and panel d shows the results for the 10-min irradiation condition. We identified partial muscle necrosis and inflammatory cell infiltration (H & E staining; 400).
Fig. 5. Comparison of different injection methods. a depicts the fluorescence intensity, and b depicts the immunohistochemistry experiment. The EGFP expression in the local articular injection group was greater than that of the intravenous injection group. And the EGFP expression in the untreated controls group was significant weaker than that of the above two groups. ⁄P < 0.05.
injury was found in the normal tissue (Fig. 6) 300 lg of plasmid was considered as the optimal dosage for local joint injection. 3.2.2. Comparison of the transfection efficiency between the experimental and control groups According to initial optimization tests, the experimental groups were injected with a mixture of microbubbles and 300 lg of plasmids along with the ultrasound parameters of 2 W/cm2 output power, a 20% duty cycle and a 5 min ultrasound irradiation for gene transfection efficiency experiments. The results showed that the EGFP transfection efficiencies of the US + MB + PL group, the US + PL group, MB + PL group and PL group were much greater than
that of the untreated controls group, and the former two groups were greater than that of the later three groups. In particular, the US + MB + PL group showed the greatest EGFP expression among the other groups (P < 0.05). The EGFP expression levels of the MB + PL group and the plasmid-only group were minimal with no statistically significant differences between these two groups (P > 0.05) (Figs. 7–9). 3.3. Duration of EGFP expression in the synovial pannus EGFP expression in the US + MB + PL group was observed in the synovial pannus on days 5, 10, 20 and 40. The expression reached a
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
309
Fig. 6. Comparison of transfection efficiency of different quantities of plasmids. a depicts the fluorescence intensity, and b depicts the immunohistochemistry experiment. Three groups were compared. Greater plasmid quantity results in greater GFP expression. ⁄P < 0.05.
peak level on day 5 and gradually declined after the peak and disappeared on day 60 (Fig. 10). 3.4. Detection of EGFP expression in the synovial pannus and other organs Green fluorescence was visible in the synovial lining and lower synovium in the US + MB + PL group. Brown particles were also observed in the lining layer and lower synovium in the immunohistochemistry experiment. No EGFP expression was observed in the liver and spleen (Fig. 11). 4. Discussion The research on the treatments for RA has made considerable progress in the past 10 years. However, current treatment approaches have been restricted as a result of high cost, severe adverse effects and the need for repeated injections. Gene therapy has potential advantages over the currently available treatments because of its high selectivity and persistent effects on the targeted gene. The goal of RA gene therapy is to transfect synovial cells with genes encoding therapeutic proteins and then the synovial cells subsequently synthesize and secrete these anti-arthritic molecules. In this way, the joint tissue synthesizes special materials that treat its own inflammation. The advantages of this approach include: (1) The therapeutic agents are directly and immediately synthesized in
the diseased areas. Thus, the issues related to drug transfer and side effects on the non-targeted tissues observed in the other treatments are ameliorated. (2) Gene therapy is highly applicable for treating chronic diseases. The genes transfected into the synovial lining maintain their activity for extended durations. This feature greatly benefits the treatment of arthritis [4,13]. The synovial pannus in patients with RA reflects diseased synovial tissue. This pannus lesion exhibits features of active proliferation, high metabolism, rich vasculature and strong invasiveness, which contribute to the pathogenesis of the joint disease and to cartilage destruction. In addition, the synovial cells secrete TNFa, IL-1 and other cytokines under inflammatory conditions, which contributes to the continued development and deterioration of joint damage in RA. Therefore, these synovial cells should be the targeted cells for gene transduction [6,7]. There are two different strategies for transferring genes into synovial cells. The direct method consists of an intra-articular injection of vectors that can be transfected into the synovial cells in situ. In contrast, the indirect method consists of removing the synovial membrane from the joint and transfecting the cell in vitro, and then these transfected cells are later re-introduced to the joint [14]. The direct method is simpler and less technical demanding when compared to indirect method. Thus, the direct method may be more suitable for a wide variety of clinical applications. At present, the vectors used for gene therapy can be broadly divided into viral and non-viral vectors. The viral vectors have been successfully utilized for in vivo gene transfection in synovial
Fig. 7. Comparison of the transfection efficiency between different groups. a depicts the fluorescence intensity, and b depicts the immunohistochemistry experiment. The EGFP transfection efficiencies of the US + MB + PL group, the US + PL group, the MB + PL group and the PL group were significantly higher than that of the untreated controls group, and the former two groups were higher than that of the later three groups The EGFP transfection efficiency of the US + MB + PL group was significantly higher than that of the US + PL group. Furthermore, the EGFP expression of the MB + PL group and the PL group were both notably low. ⁄P < 0.05.
310
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
Fig. 8. Fluorescence microscopy examination of EGFP expression. a, c, e, g and i depict fluorescence images, and b, d, f, h and j depict DAPI images. a and b depict the results of the US + MB + PL group, c and d depict the results of the UL + PL group, e and f depict the results of the MB + PL group, g and h depict the results of the PL group, and i and j depict the results of the untreated controls group. The US + MB + PL group showed the brightest fluorescence (magnification, 200).
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
311
Fig. 9. Immunohistochemical analysis of EGFP expression. a depicts the results of the US + MB + PL group, b depicts the results of the UL + PL group, c depicts the results of the MB + PL group, d depicts the results of the PL group and e depicts the results of the untreated controls group. The US + MB + PL group revealed the greatest positive immunoreactivity (magnification, 400).
Fig. 10. Comparison of transfection efficiency for different detection periods. Panel a depicts the fluorescence intensity, and panel b depicts the immunohistochemistry experiment. EGFP expression was observed in the synovial pannus on days 5, 10, 20 and 40. The expression levels peaked on day 5, and no GFP expression was detected on day 60.
312
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
Fig. 11. Detection of EGFP expression under fluorescence microscopy. Panels a and c depict fluorescence images, and panels b and d depict DAPI images. Panels a and b depict the results of the liver, and panels c and d depict the results of the spleen. No fluorescence was detected in the liver or the spleen (magnification, 200).
membranes with relatively high transfection efficiency [15,16]. However, there are still many limitations associated with treatments using viral vectors. The treatments can trigger an immune response in the body and increase the risk of insertional mutagenesis that leads to tumorigenicity and toxicity. Vectors have also been associated with limited capacity and low preparation titers. The methods involving non-viral vectors, such as plasmids, liposomes and others can avoid these limitations. However, the efficiency of the gene expression is not sufficiently high, and the duration of the gene expression is relatively short using non-viral approach. Thus, the exploration and development of transduction components and technology are priorities of current research. The UTMD technique is a new concept and approach for gene transfection. Using ultrasound irradiation to break the microbubbles in the targeted tissue at a specific time and location could improve targeted gene therapy. UTMD research studies have been conducted in a variety of in vitro cells and in vivo tissues and organs [17–21]. A few studies have applied ultrasound in combination with plasmids and microbubbles to transfect the gene in normal synovial membranes [22,23]. However, no studies have been conducted using UTMD technique for gene transfection in the synovial pannus of the AIA model. Our research is focused on the UTMDmediated enhancement of gene transfection in the synovial pannus and also aimed to identify the optimal transfection conditions for a new non-viral gene transfection method for gene therapy of the arthritis.
4.1. Optimal UTMD transfection conditions for the synovial pannus Ultrasonic cavitation can improve gene transfection efficiency and also induce cytotoxic effects. Specifically, this technique can damage vascular endothelium and other tissue cells. Several
independent studies reported that ultrasonic cavitation can damage to vascular endothelial cells and causes capillary rupture and bleeding [24,25], lead to cell lysis [26,27], and activates platelets [28]. Therefore, in the UTMD technique, the ultrasonic intensity and consequent tissue damage must be tested and balanced in order to establish the optimal conditions for gene therapy of the arthritis. The study revealed that EGFP expression without damage the irradiated normal tissue was significantly increased at 2 W/cm ultrasound output compared to 1 W/cm. The highest transfection efficiency was achieved with a 20% duty cycle while a 50% duty cycle caused muscle injury in the irradiated area. Furthermore, the transfection efficiency was significantly enhanced with increasing ultrasound irradiation duration while a 10 min ultrasound irradiation caused skin erythema and muscle injury. Through a series of experiments, the optimal parameters and conditions for synovial gene transfection was an ultrasonic output power of 2 W/cm2, a duty cycle of 20%, and an ultrasonic irradiation duration of 5 min with local articular injection and a plasmid dosage of 300 lg. Under these conditions, relatively high gene transfection efficiency was achieved without damaging the normal tissue in the irradiated area.
4.2. Comparison of the transfection efficiency By injecting expression plasmids carrying reporter genes into the knees of rabbits and rats Yovandich et al. [29] demonstrated that synovial tissue is capable of rapidly uptaking injected plasmid DNA. However, the transfection efficiency was low. A study by Taniyama et al. has showed that the UTMD technique can effectively promotes gene transfection [30]. By transfecting human vascular smooth muscle and endothelial cells in vitro with a
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
luciferase-encoding plasmid DNA, they found that the transfection activity was 70 times greater in the US treatment group and 7000– 8000 times higher in the microbubble + US group when compared to plasmid only group. Furthermore, in the in vivo experiments, the luciferase activity of the group treated with plasmid + microbubble + US was approximately 1000 times higher than the group treated with the plasmid only. Similar results were obtained in the present study. We found that the EGFP transfection efficiency of the group treated with ultrasound combined with the microbubbles and plasmids was significantly higher than that of the other three groups. Ultrasound irradiation alone could also facilitate gene transfection, but the transfection efficiency was lower than that of the group treated with ultrasound and the microbubbles. Therefore, the UTMD technique significantly enhances the gene transfection efficiency in the synovial pannus. 4.3. Duration of the EGFP expression in the synovial membrane Joint inflammation often lasts for several weeks. Previous studies have found that the synovial gene expression achieved through an articular injection of the plasmid alone lasts for only 2–3 days [29,31]. This expression duration does not meet the requirements for articular gene therapy. The existing RA treatments are mainly associated with the transfection of TNF-a, IL-1 and other cytokine antagonists. Thus, if the expression duration is excessive, the treatment can cause excessive physiological regulation and immune system dysfunction. Yang et al. have found that the gene expression induced by the UTMD technique can last for weeks or even months in different tissue and cell types [32,33]. In our study, the gene expression in the synovial membrane reached to the peak on the fifth day and gradually declined until day 40. No significant fluorescence expression was detected in the synovium on day 60. Therefore, the UTMD-transfected genes can persist locally for a significant period of time but not for excessive durations. Thus, this feature is fitted well with the requirements for RA gene therapy. 4.4. Examination of EGFP expression in the synovial pannus and other organs Synovial lesions involve the cells in the synovial lining as well as the inflammatory cells in the deep synovium. These cells secrete large amounts of inflammatory cytokines and thereby promote inflammatory processes. Therefore, the optimal gene therapy method should be associated with a certain penetration capacity and should be able to carry genes into the deep tissues to achieve the complete treatment effect. Fluorescence and immunohistochemistry experiments both showed that the GFP expression could be found in the synovial lining and the deep synovium of the UTMD-transfected synovial pannus. These results indicate that this method has a certain penetration capacity of gene transfection. Previous study has shown that applying gene therapy in a single joint has mitigating effects on the arthritis of the other joints [34]. Our study confirmed that genes could be successfully transfected into the deep synovium. Because the deep synovium is rich with microvessels, the proteins released from the genetically modified synovial cells can be absorbed into the bloodstream through these vessels. Thus, the treatment could be effective not only locally but also potentially systemically. This mechanism may explain the relief of the lesions in the other joints. Since RA often manifests as a polyarticular disease, this gene transfection method is useful for systemic treatment. With the exception of the synovial pannus, no fluorescence was observed in the liver, spleen and other organs in the present study, which indicates that the synovial gene transfection via the UTMD method is expressed locally without systematic effect.
313
5. Conclusions In conclusion, our study demonstrated that the optimal UTMD conditions for enhancing synovial pannus transfections include 2 W/cm2 power output, 20% duty cycle, 5-min ultrasound irradiation duration, 300 lg of plasmid with a locally articular injection of the microbubble and plasmid mixture. The UTMD technique significantly enhances the plasmid transfection efficiency in the synovial pannus of the AIA model. The targeted expression exists for approximately 40 days and does not damage the surrounding normal tissue. This UTMD technique could became a safe and effective non-viral gene transfection method for the gene therapy of the arthritis. Acknowledgements This study was supported by General Projects of National Natural Science Foundation of China (approval number: 81271585) and Sichuan Science and Technology Plan (approval number: 2015SZ0125). References [1] C. Roubille, V. Richer, T. Starnino, C. McCourt, A. McFarlane, P. Fleming, et al., The effects of tumour necrosis factor inhibitors, methotrexate, non-steroidal anti-inflammatory drugs and corticosteroids on cardiovascular events in rheumatoid arthritis, psoriasis and psoriatic arthritis: a systematic review and meta-analysis, Ann. Rheum. Dis. 74 (3) (2015) 480–489. [2] J.P. Leombruno, T.R. Einarson, E.C. Keystone, The safety of anti-tumour necrosis factor treatments in rheumatoid arthritis: meta and exposure-adjusted pooled analyses of serious adverse events, Ann. Rheum. Dis. 68 (2009) 1136–1145. [3] R. Alten, J. Gomez-Reino, P. Durez, A. Beaulieu, A. Sebba, G. Krammer, et al., Efficacy and safety of the human anti-IL-1b monoclonal antibody canakinumab in rheumatoid arthritis: results of a 12-week. Phase II, dose-finding study, BMC Musculoskelet Disord. 12 (2011) 153. [4] S. Fabre, F. Apparailly, Gene therapy for rheumatoid arthritis: current status and future prospects, BioDrugs 25 (2011) 381–391. [5] C. Castro-Villegas, C. Pérez-Sánchez, A. Escudero, I. Filipescu, M. Verdu, P. RuizLimón, et al., Circulating miRNAs as potential biomarkers of therapy effectiveness in rheumatoid arthritis patients treated with anti-TNFa, Arthritis. Res. Ther. 17 (2015) 49. [6] D. Hammaker, D.L. Boyle, G.S. Firestein, Synoviocyte innate immune responses: TANK-binding kinase-1 as a potential therapeutic target in rheumatoid arthritis, Rheumatology (Oxford) 51 (2012) 610–618. [7] T. Laragione, M. Brenner, A. Mello, M. Symons, P.S. Gulko, The arthritis severity locus Cia5d is a novel genetic regulator of the invasive properties of synovial fibroblasts, Arthritis. Rheum. 58 (2008) 2296–2306. [8] Z.P. Shen, A.A. Brayman, L. Chen, C.H. Miao, Ultrasound with microbubbles enhances gene expression of plasmid DNA in the liver via intraportal delivery, Gene Ther. 15 (2008) 1147–1155. [9] A. Aoi, Y. Watanabe, S. Mori, M. Takahashi, G. Vassaux, T. Kodama, Herpes simplex virus thymidine kinase-mediated suicide gene therapy using nano microbubbles and ultrasound, Ultrasound Med. Biol. 34 (2008) 425–434. [10] K. Un, S. Kawakami, R. Suzuki, K. Maruyama, F. Yamashita, M. Hashida, Enhanced transfection efficiency into macrophages and dendritic cells by a combination method using mannosylated lipoplexes and bubble liposomes with ultrasound exposure, Hum. Gene Ther. 21 (2010) 65–74. [11] T. Okura, K. Marutsuka, H. Hamada, T. Sekimoto, T. Fukushima, Y. Asada, et al., Therapeutic efficacy of intra-articular adrenomedullin injection in antigeninduced arthritis in rabbits, Arthritis. Res. Ther. 10 (2008) R133. [12] L. Qiu, Y. Jiang, L. Zhang, L. Wang, Y. Luo, Ablation of synovial pannus using microbubble-mediated ultrasonic cavitation in antigen-induced arthritis in rabbits, Rheumatol. Int. 32 (12) (2012) 3813–3821. [13] J.L. Roybal, M. Endo, A. Radu, P.W. Zoltick, A.W. Flake, Early gestational gene transfer of IL-10 by systemic administration of lentiviral vector can prevent arthritis in a murine model, Gene Ther. 18 (2011) 719–726. [14] M.J. Park, H.S. Park, M.L. Cho, H.J. Oh, Y.G. Cho, S.Y. Min, et al., Transforming growth factor b-transduced mesenchymal stem cells ameliorate experimental autoimmune arthritis through reciprocal regulation of Treg/Th17 cells and osteoclastogenesis, Arthritis. Rheum. 63 (2011) 1668–1680. [15] S.W. Tas, J. Adriaansen, N. Hajji, A.C. Bakker, G.S. Firestein, M.J. Vervoordeldonk, et al., Amelioration of arthritis by intraarticular dominant negative Ikk beta gene therapy using adeno-associated virus type 5, Hum. Gene Ther. 17 (2006) 821–832. [16] L.R. Goodrich, B.D. Brower-Toland, L. Warnick, P.D. Robbins, C.H. Evans, A.J. Nixon, Direct adenovirus-mediated IGF-I gene transduction of synovium induces persisting synovial fluid IGF-I ligand elevations, Gene Ther. 13 (17) (2006) 1253–1262.
314
X. Xiang et al. / Ultrasonics 65 (2016) 304–314
[17] H. Guo, J.C. Leung, L.Y. Chan, A.W. Tsang, M.F. Lam, H.Y. Lan, et al., Ultrasoundcontrast agent mediated naked gene delivery in the peritoneal cavity of adult rat, Gene Ther. 14 (2007) 1712–1720. [18] J.M. Escoffre, A. Zeghimi, A. Novell, A. Bouakaz, In-vivo gene delivery by sonoporation: recent progress and prospects, Curr. Gene. Ther 13 (1) (2013) 2– 14. [19] M. Tomizawa, F. Shinozaki, Y. Motoyoshi, T. Sugiyama, S. Yamamoto, M. Sueishi, Sonoporation: Gene transfer using ultrasound, World J. Methodol. 3 (4) (2013) 39–44. [20] L. Pinto de Carvalho, D. Takeshita, B.A. Carillo, B.C. Garcia Lisbôa, G. Molina, A. Beutel, et al., Hydrodynamics- and ultrasound-based transfection of heart with naked plasmid DNA, Hum. Gene Ther. 18 (2007) 1233–1243. [21] A. Delalande, C. Leduc, P. Midoux, M. Postema, C. Pichon, Efficient gene delivery by sonoporation is associated with microbubble entry into cells and the clathrin-dependent endocytosis pathway, Ultrasound Med. Biol. 41 (7) (2015) 1913–1926. [22] Y. Negishi, Y. Tsunoda, Y. Endo-Takahashi, Y. Oda, R. Suzuki, K. Maruyama, et al., Local gene delivery system by bubble liposomes and ultrasound exposure into joint synovium, J Drug Deliv. 2011 (2011) 203986. [23] M. Saito, O. Mazda, K.A. Takahashi, Y. Arai, T. Kishida, M. Shin-Ya, et al., Sonoporation mediated transduction of pDNA/siRNA into joint synovium in vivo, J. Orthop. Res. 25 (2007) 1308–1316. [24] E.L. Carstensen, S. Gracewski, D. Dalecki, The search for cavitation in vivo, Ultrasound Med. Biol. 26 (9) (2000) 1377–1385. [25] D. Dalecki, C.H. Raeman, S.Z. Child, D.P. Penney, R. Mayer, E.L. Carstensen, The influence of contrast agents on hemorrhage produced by lithotripter fields, Ultrasound Med. Biol. 23 (1997) 1435–1439.
[26] D.L. Miller, J. Quddus, Lysis and sonoporation of epidermoid and phagocytic monolayer cells by diagnostic ultrasound activation of contrast agent gas bodies, Ultrasound Med. Biol. 27 (2001) 1107–1113. [27] D.L. Miller, R.M. Thomas, Thresholds for hemorrhages in mouse skin and intestine induced by lithotripter shock waves, Ultrasound Med. Biol. 21 (1995) 249–257. [28] S.L. Poliachik, W.L. Chandler, P.D. Mourad, R.J. Ollos, L.A. Crum, Activation, aggregation and adhesion of platelets exposed to high intensity focused ultrasound, Ultrasound Med. Biol. 27 (2001) 1567–1576. [29] J. Yovandich, B. O’Malley Jr., M. Sikes, F.D. Ledley, Gene transfer to synovial cells by intra-articular administration of plasmid DNA, Hum. Gene Ther. 6 (1995) 603–610. [30] Y. Taniyama, K. Tachibana, K. Hiraoka, T. Namba, K. Yamasaki, N. Hashiya, et al., Local delivery of plasmid DNA into rat carotid artery using ultrasound, Circulation 105 (2002) 1233–1239. [31] I. Nita, S.C. Ghivizzani, J. Galea-Lauri, G. Bandara, H.I. Georgescu, P.D. Robbins, et al., Direct gene delivery to synovium. An evaluation of potential vectors in vitro and in vivo, Arthritis Rheum. 39 (1996) 820–828. [32] K. Nishida, M. Doita, T. Takada, K. Kakutani, H. Miyamoto, T. Shimomura, et al., Sustained transgene expression in intervertebral disc cells in vivo mediated by microbubble-enhanced ultrasound gene therapy, Spine (Phila Pa 1976) 31 (2006) 1415–1419. [33] L. Yang, Y. Shirakata, K. Tamai, X. Dai, Y. Hanakawa, S. Tokumaru, et al., Microbubble-enhanced ultrasound for gene transfer into living skin equivalents, Dermatol. Sci. 40 (2005) 105–114. [34] A. Inoue, K.A. Takahashi, O. Mazda, Y. Arai, M. Saito, T. Kishida, et al., Comparison of anti-rheumatic effects of local RNAi-based therapy in antigen induced arthritis rats using various cytokine genes as molecular targets, Mod. Rheumatol. 19 (2009) 125–133.