Author + information
- Received December 11, 2008
- Revision received September 3, 2009
- Accepted September 15, 2009
- Published online March 2, 2010.
- Jun-ichi Suzuki, MD, PhD⁎,‡,⁎ (, )
- Masahito Ogawa, MS⁎,‡,
- Kiyoshi Takayama, PhD‡,
- Yoshiaki Taniyama, MD, PhD§,
- Ryuichi Morishita, MD, PhD§,
- Yasunobu Hirata, MD, PhD⁎,†,
- Ryozo Nagai, MD, PhD† and
- Mitsuaki Isobe, MD, PhD†
- ↵⁎Reprint requests and correspondence:
Dr. Jun-ichi Suzuki, Department of Advanced Clinical Science and Therapeutics, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo, Tokyo, 113-8655 Japan
Objectives The purpose of this study was to investigate the efficiency of small interfering ribonucleic acid (siRNA) in murine arteries. We transfected it using a nonviral ultrasound-microbubble–mediated in vivo gene delivery system.
Background siRNA is an effective methodology to suppress gene function. The siRNA can be synthesized easily; however, a major obstacle in the use of siRNA as therapeutics is the difficulty involved in effective in vivo delivery.
Methods To investigate the efficiency of nonviral ultrasound-microbubble–mediated in vivo siRNA delivery, we used a fluorescein-labeled siRNA, green fluorescent protein (GFP) siRNA, and intercellular adhesion molecule (ICAM)-1 siRNA in murine arteries. Murine femoral arteries were injured using flexible wires to establish arterial injury.
Results The fluorescein-labeled siRNA and GFP siRNA showed that this nonviral approach could deliver siRNA into target arteries effectively without any tissue damage and systemic adverse effects. ICAM-1 siRNA transfection into murine injured arteries significantly suppressed the development of neointimal formation in comparison to those in the control group. Immunohistochemistry revealed that accumulation of T cells and adhesion molecule positive cells was observed in nontreated injured arteries, whereas siRNA suppressed accumulation.
Conclusions The nonviral ultrasound-microbubble delivery of siRNA ensures effective transfection into target arteries. ICAM-1 siRNA has the potential to suppress arterial neointimal formation. Transfection of siRNA can be beneficial for the clinical treatment of cardiovascular and other inflammatory diseases.
The most effective strategy for evaluating gene function is using ribozymes and ribonucleic acid (RNA) interference (1). Small interfering ribonucleic acids (siRNA) are powerful tools that suppress gene expression in cells (2). However, the therapeutic application of siRNA is dependent upon the development of delivery vehicles (3). In addition, such a delivery vehicle should be administered efficiently, safely, and repeatedly. However, there are only a few reports on effective and safe in vivo transfection of siRNA into target organs. Although viral vector systems are efficient for transfection of genes, such as short hairpin RNAs and molecular derivatives of siRNAs, there is concern that these systems may adversely affect clinical utility (4). As therapeutic ultrasound increases cell membrane permeabilization (5), the nonviral ultrasound method increases the transfection efficiency of decoy in vivo into arteries (6), and other organs. Although there are a few papers that demonstrate sonoporation using microbubble-promoted plasmid deoxyribonucleic acid/siRNA transduction to murine hearts (7) and to joint synovium (8), there have been few reports of the transfection of siRNA into target arteries using ultrasound-microbubble methods.
Recently, we reported decoy oligonucleotide transfection into the arteries using this methodology (9). Thus, we have used the fluorescein isothiocyanate (FITC)-labeled siRNA and a murine arterial injury model to study the transfection efficiency of siRNA using an ultrasound microbubble method into target arteries. Because intercellular adhesion molecule (ICAM)-1 plays a critical role for progression of inflammation and arterial neointimal formation (10), we examined the effects of ICAM-1 siRNA for prevention of neointimal formation after arterial injury using the ultrasound-microbubble–mediated transfection.
This is the first paper to demonstrate that the ultrasound microbubble method significantly increased siRNA transfection efficiency into arteries. We revealed that ICAM-1 siRNA transfection using this method has potential to prevent inflammatory-related cardiovascular diseases.
Preparation of siRNA-microbubble complexes
The 1,2 dioleoyl-3-trimethylammonium-propane (DOTAP) was purchased from Roche Diagnostics (Alameda, California). For each artery, 20-μg siRNAs in 90-μl transfection buffer was transferred into a sterile Eppendorf tube. In a separate sterile polystyrene tube, 50-μg DOTAP was mixed with 90-μl transfection buffer with 10-μl microbubble (Optison solution [human albumin microspheres], Mallinckrodt, Hazelwood, Missouri), and then the siRNA mixture was transferred to the polystyrene tube containing the DOTAP and incubated at room temperature for 30 min. The mixture should have turned cloudy, but no precipitates or aggregation were visible. Charge ratios of DOTAP and siRNAs varied from 1:2 to 2:1.
Arterial injury models
Male wild-type mice and transgenic mice constitutively overexpressing enhanced green fluorescent protein (GFP [C57BL/6, age 4 to 6 weeks, weight 20 to 25 g] 006567, Japan Charles River Laboratories, Atsugi, Kanagawa, Japan) were used in this study. We made an arterial injury model that was modified from the previous reports (11). Briefly, the femoral artery was looped and tied off with 6-0 silk sutures for temporary vascular control during the procedure. A transverse arteriotomy was made, and a flexible angioplasty guidewire (a curved 350-μm polished copper wire) was introduced and advanced 1 cm. Endothelial denudation injury of the artery was performed by use of wire withdrawal injury, and 3 passes were made along the artery. A sham operation (no wire injury) was also performed. The animals were fed a standard diet and water and were maintained in compliance with animal welfare guidelines of the Institute of Experimental Animals, Tokyo Medical and Dental University. The investigation conforms to the “Guide for the Care and Use of Laboratory Animals” published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996).
Transfection of FITC-labeled siRNA into injured arteries of wild-type mice
To study the in vivo transfection efficiency of siRNA, we used the FITC-labeled siRNA and the wild-type murine arterial injury model. FITC-labeled siRNA (siRNA Alexa Fluor 488) was purchased from Qiagen K.K. (Tokyo, Japan). Immediately after injury of right femoral arteries, 90 μl of the FITC-labeled siRNA (20 μg) plus microbubble (10 μl) mixture was incubated within the arterial lumen. The ultrasound transducer was held by hand, and it was attached directly to the artery. The ultrasound was irradiated for 20 s (1 MHz, 0.5 W/cm2, duty 50%) to the artery. The siRNA transfection was performed while the vessel was being ligated (n = 6). For control studies of injured arteries: 1) no siRNA, microbubble and ultrasound (n = 2); 2) non–FITC-labeled siRNA without microbubble and ultrasound (n = 4); 3) FITC-labeled siRNA without microbubble and ultrasound (n = 4); and 4) FITC-labeled siRNA with microbubble but no ultrasound irradiation (n = 6) were performed. Additionally, we compared the FITC siRNA transfection efficiency between de-endothelialized vessels (n = 6) and non–de-endothelialized vessels (n = 6) using the same ultrasound-microbubble method. The arteries and other organs were harvested 8 h after transfection; the transfection efficiency was judged using fluorescent microscopy. Systemic adverse effects were defined as death or significant body weight loss during the observation period.
Transfection of GFP siRNA into injured arteries of GFP transgenic mice
To study the chronic efficiency of siRNA, we used the GFP siRNA (AM4626, Ambion, Foster City, California) and the GFP transgenic murine arterial injury model. Immediately after injury of right femoral arteries of GFP transgenic mice, 90-μl of GFP siRNA (20 μg) plus microbubble (10 μl) mixture was incubated within the arterial lumen. The siRNA was simply mixed with the microbubbles. The flow in the femoral artery was temporarily disrupted during the ultrasound-microbubble destruction. Thus, the flow was similarly disrupted during the control experiments. The ultrasound was irradiated for 20 s (1 MHz, 0.5 W/cm2, duty 50%) to the artery (n = 4). For control studies: 1) no siRNA, microbubble and ultrasound (n = 2); 2) GFP siRNA without microbubble and ultrasound (n = 4); and 3) GFP siRNA with microbubble but no ultrasound irradiation (n = 4) were performed. Additionally, we compared GFP expression between GFP siRNA transfected vessels (n = 4) and remote vessels (n = 4) of GFP-transgenic mice. The arteries were harvested 7 days after transfection; the transfection efficiency was judged using fluorescent microscopy.
Design of ICAM-1 siRNA
We designed 7 ICAM-1 siRNA according to the murine ICAM-1 complementary deoxyribonucleic acid (cDNA) sequence (GeneBank #X52264.1) as follows, referring to the previous report (12): S1: 5′-AAG GGC TGG CAT TGT TCT CTA-3′; S2: 5′-AAC TGT GAA GTG TGA AGC CCA-3′; S3: 5′-AAG TTA CAG AAG GCT CAC GAG-3′; S4: 5′-AAG GAG ATC ACA TTC ACG GTG-3′; S5: 5′-AAA CAG CAG TGA CTC TGT GTC-3′; S6: 5′-AAC TGG AAG CTG TTT GAG CTG-3′; and S7: 5′-AAG GGA GCC AAG TAA CTG TGA-3′.
Mixed lymphocyte reaction for evaluation of ICAM-1 siRNA
We examined mixed lymphocyte reaction (MLR) to assess which ICAM-1 siRNA in selected sequences is most effective in the suppression of the immune reaction. The MLR was performed with responder lymph node cells from native C57BL/6J mice, and mitomycin-C–inactivated stimulator lymph node cells from naive BALB/c mice, as described previously (13,14). Each ICAM-1 siRNA was transfected into naive responder cells using the HVJ-liposome method (15). A total of 1 × 105responder cells were cultured with an equal number of mitomycin-C treated stimulator cells in 96-well plates in C/10 media. Proliferation was assessed at 37°C under 5% CO2on day 4. Cell proliferation was assessed with the Cell Counting Kit-8 (Dojindo, Tokyo, Japan), and the proliferation was evaluated using the optical density (16).
Transfection of ICAM-1, vascular cell adhesion molecule (VCAM)-1, and scrambled siRNA into injured arteries
To confirm the in vivo efficacy of ICAM-1 siRNA to prevent neointimal formation, we used the murine arterial injury model and the ultrasound microbubble method. We chose 1 of 7 ICAM-1 siRNAs after MLR evaluation for in vivo transfection. Immediately after arterial injury, 90 μl ICAM-1 siRNA (20 μg) plus microbubble (10 μl) mixture was incubated within the arterial lumen. The ultrasound was irradiated for 20 s (1 MHz, 0.5 W/cm2, duty 50%) to the artery. For control studies, 1) no siRNA without microbubble and ultrasound; 2) ICAM-1 siRNA without microbubble and ultrasound; and 3) ICAM-1 siRNA with microbubble but no ultrasound irradiation were performed. To compare the in vivo effects between ICAM-1 and VCAM-1 for neointimal formation, we performed VCAM-1 siRNA (SiRNA ID: s67996, Ambion, Foster City, California) transfection using the same animal models and methodologies. In regard to recent concerns that gene specific therapeutic benefits with siRNA may be the result of nonspecific off-target effects, scrambled siRNA (SiRNA ID: AM4611, Ambion) was also transfected using the same animal models and methodologies. The arteries were harvested on day 28.
Quantitative assessment of arterial neointima
Tissue sections were stained with hematoxylin and eosin and Elastica van Gieson. Light microscopic morphometric computer analysis was carried out. The severity of neointimal formation was quantitatively assessed in each artery, as described earlier (4). The areas encompassed by the outer medial layer (OML), the inner elastic lamina (IEL), and the lumen were measured, and the length of intimal thickening (IT) in relation to the length of medial wall was calculated by the formula: IT = (IEL − lumen area) / (OML − IEL).
Complete transverse sections of the arteries approximately 3 mm in length were obtained and stored in an optimum cutting temperature compound (Ted Pella, Inc., Redding, California). Serial sections (6 μm) were cut and dipped in cold acetone for 10 min. The sections were rehydrated in phosphate buffered saline (PBS) and incubated with 5% normal goat serum to avoid nonspecific reaction. The samples were incubated with primary antibodies against ICAM-1 (YN1/1.7), VCAM-1 (MK/2), CD4 (GK1.5), CD8 (53-6.7), and CD11b (M1/70 [Pharmingen, San Diego, California]), CD31 (H-300 [Santa Cruz, San Diego, California]), and proliferating cell nuclear antigen (PC10 [Sigma Aldrich, Tokyo, Japan]) for 12 h at 4°C. Biotin-conjugated antibodies were detected with an avidin-biotin-horseradish peroxidase complex (Nichirei, Tokyo, Japan) according to the manufacturer's instructions. Enzyme activity was detected with diaminobenzidine (0.5 mg/ml) with 0.05% NiCl in 50 mM Tris Buffer, pH 7.5 (4,12). We counted positive cell numbers per artery for CD4, CD8, CD11b, and proliferating cell nuclear antigen. Immunohistochemical analyses for ICAM-1, VCAM-1, and CD31 were performed by independent observers using qualitative scoring as previously reported (0, absent; 1, weak, focal; 2, weak, diffuse; 3, strong, focal; 4, strong, diffuse; and 5, very strong and diffuse). Scores uniformly fell within 1 grade of each other and were averaged (9).
Real-time polymerase chain reaction (RT-PCR)
RT-PCR was used to determine the messenger ribonucleic acid (mRNA) expression of ICAM-1 (assay ID: 00516023_m1, Applied Biosystems). To account for differences in cDNA preparation and cDNA amplification efficiency, the mRNA expression of the target gene was normalized by 18s ribosomal RNA (assay ID: 4308329). Quantitative data were calculated using the comparative CT (ΔΔCT) method (17).
All data are expressed as mean ± SD. Scores were compared among the groups using a Scheffe analysis of variance with Scheffe tests for post-hoc comparisons. Differences with values of p < 0.05 were considered significant.
Ultrasound-microbubble method enhanced FITC-labeled siRNA transfection into target arteries
At 8 h after transfection, enhanced expression of fluorescence in cell nuclei of the endothelial cells, medial vascular smooth muscle cells, and perivascular cells was observed in the ultrasound microbubble group. However, only faint fluorescence was observed in the target arteries of the control groups. The transfection efficiency of the FITC-labeled siRNA using the ultrasound-microbubble method was significantly increased compared to the control groups. To compare the transfection efficiency between de-endothelialized and non–de-endothelialized vessels, we used the same FITC siRNA and the ultrasound-microbubble method. We revealed that there was no statistical difference in the efficiency between the 2 groups (Fig. 1,Supplemental Figure 1). No dissemination of FITC-labeled siRNA in the nonirradiated organs was observed. No systemic adverse effect was observed in the mice during the observation period.
Ultrasound-microbubble–mediated GFP siRNA transfection suppressed GFP expression in target arteries
At 7 days after transfection, suppressed expression of GFP in cell nuclei of the endothelial cells, medial vascular smooth muscle cells, and perivascular cells was observed in the ultrasound-microbubble–mediated GFP siRNA transfection group. The partial expression of GFP was remaining because the bottom of the arteries could not receive enough ultrasound irradiation using this perivascular approach. In contrast, overexpressed GFP was observed in the target arteries of the control groups. The transfection efficiency of the GFP siRNA using the ultrasound-microbubble method was significantly increased compared with that of the control groups. We also observed GFP expression in the remote arteries of GFP transgenic mice. We revealed that limited transfection of siRNA did not alter expression in the remote arteries of the mice (Fig. 2).
ICAM-1 siRNA transfection with ultrasound-microbubble method could suppress neointimal formation
Before the in vivo study of ICAM-1 siRNA, we examined in vitro MLR to evaluate the silencing effect of several murine ICAM-1 siRNAs in the suppression of the immune reaction. All sequences had suppressive effects compared to the proliferation without siRNA administration in the assay (Fig. 3).Thus, we selected the most effective sequence (S1) among the 7 sequences for in vivo analysis.
On the basis of these results, we made murine arterial injury models. The ICAM-1 siRNA plus microbubble mixture was incubated within the arterial lumen and then ultrasound irradiated. In the 3 groups of controls, injured arteries on day 28 showed significantly thickened intima, whereas the ICAM-1 siRNA with microbubble and ultrasound group showed suppressed neointimal formation. There was significant difference between the treated group (ICAM-1 siRNA, microbubble and ultrasound) and the 3 control groups. The VCAM-1 siRNA had a statistically comparable effect to that of ICAM-1 siRNA in the prevention of neointimal formation. The scrambled siRNA did not suppress neointimal formation (Fig. 4).No neointimal formation was detected in native arteries and arteries from sham-operated mice.
To clarify the mechanism, we performed immunohistochemistry. ICAM-1 was much more intensely expressed in the thickened intima of the injured arteries in the control groups, whereas the expression was reduced in the arteries from mice that received the ICAM-1 siRNA with microbubble and ultrasound. VCAM-1 was also enhanced in the thickened intima of the injured arteries from the control groups, whereas the expression was reduced in the arteries from mice that received the treatment. The scores of ICAM-1 and VCAM-1 were significantly reduced after ICAM-1 siRNA with ultrasound treatment. The CD4, CD8, and CD11b positive cells were severely infiltrated into the thickened intima of the control arteries, while these positive cells were suppressed in the ICAM-1 siRNA in the microbubble-ultrasound–treated group. We compared re-endothelialization and cell proliferation between the ICAM-1 siRNA group and the nontreated group. Although cell proliferation was suppressed in the siRNA ICAM-1 transfected arteries compared with the vehicle-treated arteries, there was no difference in re-endothelialization between the 2 groups. No CD4, CD8, CD11b, ICAM-1, and VCAM-1 expression was enhanced in native arteries and arteries from sham-operated mice (Figs. 5 and 6).⇓
The RT-PCR revealed that the mRNA levels of ICAM-1 were elevated in vehicle-treated arteries after the wire injury, whereas siRNA ICAM-1 treatment significantly suppressed the levels (Fig. 7).
In this study, we demonstrated that the ultrasound-microbubble method enhanced the selective transfection efficiency of siRNA. Previous studies have indicated that the therapeutic application of siRNAs is largely dependent upon the development of a delivery vehicle that can efficiently deliver the siRNAs to target cells. Sioud et al. (18) indicated that they could deliver a fluorescein-labeled siRNA using cationic liposome-mediated intravenous and intraperitoneal administration in adult mice. Although they showed that the cationic liposome-mediated transfection could deliver siRNAs into various cell types, this method could not limit the transfection into targeted tissues; that may disseminate the siRNA in whole bodies for systemic adverse effects. However, as we have shown in this study, the ultrasound-microbubble method can deliver siRNA into target tissues and prevent systemic adverse effects because siRNA was not transfected into nonultrasound areas. The use of ultrasound-mediated microbubble destruction for gene therapy was pioneered by Shohet et al. (19). They revealed that ultrasound-mediated destruction of albumin-coated microbubbles is a promising method for the delivery of bioactive agents to the heart. Lindner et al. (20) also pioneered the use of microbubbles targeted to ICAM-1 and other markers of vascular endothelium. They formulated novel microbubbles that were targeted to endothelial cell adhesion molecules, including P-selectin and ICAM-1, for imaging inflamed tissues (20). Because these 2 laboratories pioneered the whole field of ultrasound-targeted microbubble destruction, both for gene delivery/therapy and diagnostic imaging, we could develop new techniques using siRNA to ICAM-1. Thus, the ultrasound-microbubble method is a novel, nonviral, effective, and safe method of in vivo gene transfer.
Using this methodology, we have revealed that in vivo transfection of siRNA against ICAM-1 significantly suppressed neointimal formation and inflammation without systemic adverse effects. As previously reported, activated leukocytes, neutrophils, monocytes, and platelets play a critical role in the initial stage of the inflammatory process after vascular injury. Firstly, platelets and fibrin are deposited on the injured arterial surface and activated platelets express adhesion molecules. Adhesion molecules mediate the initial tethering and rolling of leukocytes on platelets. Leukocytes then adhere firmly to the vessel surface through Mac-1 (CD11b/CD18) by a ligand-receptor interaction with fibrinogen to the glycoprotein IIb/IIIa receptor (21). Worthy of note, platelets will adhere to the de-endothelialized artery, and could also use ICAM-1 as a receptor for glycoprotein IIb/IIIa–bound fibrinogen (22). These inflammatory processes are followed by proliferation of vascular components, such as smooth muscle cells and extracellular matrix, leading to neointimal thickening.
Hill et al. (23) reported that silencing interleukin-12 production in dendritic cells using siRNA was effective in decreasing allostimulation in vitro and modulated the Th1 versus Th2 balance in vivo during a primary immune response. Although this approach is useful for systemic treatment, this cytokine treatment may not be beneficial for many cardiovascular diseases that need specific and limited treatments. Cell adhesion molecules are good therapeutic targets for suppression of various inflammatory and cardiovascular diseases.
We have revealed that E-selectin siRNA in vascular endothelium leads to inhibition of leukocyte adhesion in vitro (24). Because ICAM-1 plays a central role in the progression of cardiovascular diseases (25), immunomodulation of ICAM-1 using siRNA could be clinically useful without systemic immune modulation and adverse effects. However, the multiple roles of ICAM-1 suppressing injury-induced neointimal formation are still controversial. Yasukawa et al. (10) reported that intravenous administration of ICAM-1 monoclonal antibody significantly attenuated carotid artery neointimal formation in response to injury compared with that of control animals. However, the ICAM-1 antibody did not reduce the accumulation of monocytes/macrophages within the injured vessels, indicating the role of other ICAM-1–dependent events in restenosis (10). Manka et al. (26) reported no effect of ICAM-1 deficiency on macrophage accumulation in apolipoprotein-E knockout mice. We speculate that siRNA ICAM-1 reduced neointimal formation through suppression of T-cell accumulation and expression of other adhesion molecules, although macrophage accumulation plays an important role in the development of neointimal formation after arterial injury. Of note, siRNA against ICAM-1 inhibited VCAM-1 expression after injury. This finding may imply an indirect mechanism by which inflammatory cells adhering to ICAM-1 induce VCAM-1. Although the event was supported by previous investigations (27), the detailed in vivo mechanism of siRNA against ICAM-1 remains unclear. Thus, further investigation is needed to elucidate the mechanisms of siRNA ICAM-1 in cardiovascular diseases.
The ultrasound-microbubble delivery of siRNAs is effective for transfection into target arteries in vivo. Using this method, ICAM-1 siRNA is potent for the suppression of neointimal formation after coronary angioplasty. Thus, this nonviral transfection of siRNA may be beneficial for the clinical treatment of cardiovascular diseases and other inflammatory diseases.
The authors thank Ms. Noriko Tamura and Ms. Yasuko Matsuda for excellent technical assistance.
For a supplemental figure showing the representative finding of untreated injured femoral artery in wild-type mice on day 28 after injury, please see the online version of this article
Ultrasound-Microbubble–Mediated Intercellular Adhesion Molecule-1 Small Interfering Ribonucleic Acid Transfection Attenuates Neointimal Formation After Arterial Injury in Mice
This work was supported by grants from the Grant-in-Aid from the Japanese Ministry of Education, Science and Culture, and the Grant-in-Aid from the Japanese Ministry of Welfare.
- Abbreviations and Acronyms
- complementary deoxyribonucleic acid
- 1,2 dioleoyl-3-trimethylammonium-propane
- fluorescein isothiocyanate
- green fluorescent protein
- intercellular adhesion molecule
- mixed lymphocyte reaction
- messenger ribonucleic acid
- real-time polymerase chain reaction
- small interfering ribonucleic acid
- vascular cell adhesion molecule
- Received December 11, 2008.
- Revision received September 3, 2009.
- Accepted September 15, 2009.
- American College of Cardiology Foundation
- Taniyama Y.,
- Tachibana K.,
- Hiraoka K.,
- et al.
- Yasukawa H.,
- Imaizumi T.,
- Matsuoka H.,
- Nakashima A.,
- Morimatsu M.
- Lindner V.,
- Fingerle J.,
- Reidy M.A.
- Suzuki J.,
- Ogawa M.,
- Sagesaka Y.M.,
- Isobe M.
- Ogawa M.,
- Suzuki J.,
- Hishikari K.,
- Takayama K.,
- Tanaka H.,
- Isobe M.
- Futamatsu H.,
- Suzuki J.,
- Kosuge H.,
- et al.
- Shohet R.V.,
- Chen S.,
- Zhou Y.T.,
- et al.
- Massberg S.,
- Enders G.,
- Matos F.C.,
- et al.
- Hill J.A.,
- Ichim T.E.,
- Kusznieruk K.P.,
- et al.
- Manka D.,
- Collins R.G.,
- Ley K.,
- Beaudet A.L.,
- Sarembock I.J.