Author + information
- Received April 30, 2012
- Revision received July 6, 2012
- Accepted August 12, 2012
- Published online December 18, 2012.
- Hiroyuki Kawata, MD⁎,
- Yoshiko Uesugi, PhD†,
- Tsunenari Soeda, MD⁎,
- Yasuhiro Takemoto, MD⁎,
- Ji-Hee Sung, MD⁎,
- Kiyotaka Umaki, PhD‡,
- Keiji Kato, PhD‡,
- Kenichi Ogiwara, MD§,
- Keiji Nogami, MD§,
- Kenichi Ishigami, MD⁎,
- Manabu Horii, MD⁎,
- Shiro Uemura, MD⁎,
- Midori Shima, MD§,
- Yasuhiko Tabata, PhD† and
- Yoshihiko Saito, MD⁎∥,⁎ ()
- ↵⁎Reprints requests and correspondence:
Dr. Yoshihiko Saito, First Department of Internal Medicine, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan
Objectives The purpose of this study was to develop a new intelligent drug delivery system for intracoronary thrombolysis with a strong thrombolytic effect without increasing bleeding risk.
Background Rapid recanalization of an occluded coronary artery is essential for better outcomes in acute myocardial infarction. Catheter-based recanalization is widely accepted, but it takes time to transport patients. Although the current fibrinolytic therapy can be started quickly, it cannot achieve a high reperfusion rate. Recently, we generated nanoparticles comprising tissue-type plasminogen activator (tPA), basic gelatin, and zinc ions, which suppress tPA activity by 50% with 100% recovery by ultrasound (US) in vitro.
Methods The thrombus-targeting property of nanoparticles was examined by an in vitro binding assay with von Wilbrand factor and with a mouse arterial thrombosis model in vivo. The thrombolytic efficacy of nanoparticles was evaluated with a swine acute myocardial infarction model.
Results Nanoparticles bound to von Wilbrand factor in vitro and preferentially accumulated at the site of thrombus in a mouse model. In a swine acute myocardial infarction model, plasma tPA activity after intravenous injection of nanoparticles was approximately 25% of tPA alone and was recovered completely by transthoracic US (1.0 MHz, 1.0 W/cm2). During US application, plasma tPA activity near the affected coronary artery was recovered and was higher than that near the femoral artery. Although treatment with tPA alone (55,000 IU/kg) recanalized the occluded coronary artery in only 1 of 10 swine, nanoparticles containing the same dose of tPA with US achieved recanalization in 9 of 10 swine within 30 min.
Conclusions We developed an intelligent drug delivery system with promising potential for better intravenous coronary thrombolysis.
In the treatment of acute myocardial infarction (AMI), rapid coronary recanalization is essential for reducing infarct size and preserving cardiac function, leading to better prognosis. Although percutaneous coronary intervention is accepted as the appropriate emergent therapy because of its ability to achieve reliable recanalization, it takes time for transportation of patients to specialized hospitals, where cardiac catheterization can be performed. However, intravenous pharmacological thrombolysis is noninvasive, less expensive, and simpler than percutaneous coronary intervention and theoretically can be started immediately after an AMI is identified, even in an ambulance. With this approach, however, only 60% to 75% of patients achieve Thrombolysis In Myocardial Infarction (TIMI) flow grade 3 recanalization (1–4), which is very low in comparison with the rates accompanying percutaneous coronary intervention. There are also risks of hemorrhagic complications.
Accordingly, to enhance thrombolysis, many investigators, including our group, have combined locally applied ultrasound (US) and thrombolytic agents (sonothrombolysis), and have demonstrated its efficacy in experimental animals (5–7). However, a recent clinical study testing the efficacy of sonothrombolysis in AMI patients failed to show its superiority to conventional thrombolytic methods, both in terms of the recanalization rate and the occurrence of hemorrhagic complications (8).
Therefore, the new strategy that coronary recanalization therapy could be started immediately after the arrival of emergent medical services should be developed, such as an intelligent drug delivery system (DDS) for better thrombolysis, which encapsulates the thrombolytic agents to suppress their activity within the systemic circulation (stealth activity), delivers the encapsulated agents preferentially to the thrombus (thrombus targeting), releases the agents (controlled activity), and consequently facilitates local thrombolytic activity without systemic bleeding risk.
Given that fibrillar collagen binds to von Willebrand factor (vWF), which is expressed on the surface of platelets (9,10) and is a key component of platelet-rich thrombi, we took note of gelatin as the material of DDS encapsulating the thrombolytic agent for targeting the thrombus, because gelatin is produced by partial hydrolysis of collagen. Recently, we had developed a novel DDS, comprising tissue-type plasminogen activator (tPA), basic gelatin, and zinc ions that suppresses tPA activity, but is recovered by US in vitro (11). In the present study, we show that this DDS has stealth activity, is US controlled in vivo, and has thrombus-targeting properties in vitro and in vivo, thus demonstrating the intelligent design of this DDS. Furthermore, we evaluated the efficacy of this intelligent DDS on intravenous coronary thrombolysis in a swine model of AMI.
A detailed Methods section is available in the Online Appendix.
Preparation of the nanoparticles encapsulating tPA
Briefly, basic gelatin (100,000 MW, Nitta Gelatin, Inc., Osaka, Japan) and tPA (monteplase, Cleactor Injection, Eisai Co., Ltd., Tokyo, Japan) were mixed at the concentration of 20 mg/ml and 1.0 mg/ml, respectively, in 10 mM phosphate-buffered saline (pH 7.4). After agitation, zinc acetate (Nacalai Tesque, Inc., Kyoto, Japan) was added to the mixture of basic gelatin and tPA (at a final concentration of 5 mM) to stabilize and tighten their connection (Fig. 1A).
Assay of tPA activity in vitro and in vivo
To determine tPA activity in vitro, a fibrin clot lysis assay was performed on the basis of the plasminogen-rich fibrin plate method (12,13). The plasma tPA activity of rabbit and swine was measured with the synthetic substrate, chromozym tPA, according to the manufacturer's instructions (Roche Applied Science, Indianapolis, USA).
In vitro vWF binding assay
To evaluate whether vWF protein binds to tPA, basic gelatin, or nanoparticles, an enzyme immunoassay was performed with the use of rabbit antihuman vWF antibody (DAKO, Glostrup, Denmark).
Evaluation of tPA accumulation in the occluded vessel
For the evaluation of tPA accumulation at the thrombus site in the mouse arterial thrombosis model induced by ferric chloride methods, tPA was radioiodinated with iodide 125 (125I) according to the chloramine T method (14).
125I-labeled tPA alone (27,500 IU/kg; 0.2235 mg/kg) or nanoparticles containing the same dose of 125I-labeled tPA (n = 5 for each) in 100 μl phosphate-buffered saline were injected into the tail vein of the mouse 30 minutes after the induction of thrombotic occlusion in the left femoral artery (FA). Ten minutes later, the bilateral iliofemoral arteries at 2 cm length were excised, along with the blood inside the lumen, after ligation of both ends and all branches. The radioactivity of the iliofemoral artery was counted with a gamma counter (ARC-301B, Aloka, Tokyo, Japan).
Generation and evaluation of the transthoracic US device
On the basis of the in vitro and rabbit experiments, we tested US probes of 3 different frequencies (0.5, 1.0, and 2.0 MHz) and 1.0 W/cm2 in intensity. Four US probes of each frequency were combined in parallel to generate continuous-wave transthoracic US (TUS) devices for the swine model, as shown in Online Figure 1. Only the 1.0-MHz frequency (continuous wave, 1.0 W/cm2) device produced relatively uniform US fields both in a water bath test (data not shown) and over the swine heart, as quantified by an oscilloscope (Online Fig. 2).
Preparation of the swine AMI model
For the preparation of the swine AMI model, thrombotic occlusion of the left circumflex coronary artery (LCx) was induced in 60 female swine (35 to 45 kg) by balloon injury with distal balloon occlusion, as shown in Online Figure 3.
Thrombolysis in the swine AMI model
Thrombolysis was initiated by the injection of 27,500 IU/kg (0.2235 mg/kg) or 55,000 IU/kg (0.447 mg/kg) tPA, or nanoparticles containing the same doses of tPA via an ear vein over 5 min. Subsequently, continuous-wave US (1.0 MHz, 1.0 W/cm2) was applied transthoracically in the cases with TUS for up to 60 minutes (n = 10 for each treatment).
Evaluation of left ventricular ejection fraction after thrombolysis
At 60 min after the initiation of thrombolysis, left ventriculography at the left anterior oblique view was performed, and left ventricular ejection fraction was calculated.
Data were expressed as mean ± SD. Data distribution was assessed for normality by using the Kolmogorov-Smirnov test. Normally distributed data were analyzed by paired or unpaired Student t tests for 2-group comparisons and 1-factor analyses of variance with Fisher Protected Least Significant Difference correction for multigroup comparisons. Nonnormally distributed data were compared between groups by using the Kruskal-Wallis analysis and subsequent Mann-Whitney U test for multiple comparisons. The ratios of TIMI flow grade 2 and TIMI flow grade 3 in thrombolysis at each time point were compared between groups by using the Fisher exact test, and adjusted multiple comparisons for intergroup differences in the data of thrombolysis over the time course were not performed. Values of p <0.05 were considered to indicate statistical significance.
Schema of the new intelligent DDS
Figure 1A shows the schema of the intelligent DDS, which is characterized by both its thrombus-targeting property and its stealth activity of tPA reactivated by US. The nanoparticles are prepared by adding zinc acetate (final concentration, 5 mM) into the mixture of tPA and basic gelatin at concentrations of 0.5 mg/ml and 10 mg/ml, respectively. The nanoparticles are approximately 100 nm in diameter as measured by the dynamic light scattering method (13). tPA is released from the nanoparticles by application of US (continuous wave, 1.0 MHz, and 1.0 W/cm2).
tPA activity of the intelligent DDS in vivo
tPA activity was reduced by approximately 50% by the nanoparticle formulation, but was fully recovered by US application in vitro (Online Fig. 4) (11). In the present study, decreased tPA activity of the nanoparticles also was confirmed in rabbits. Plasma tPA activity immediately after nanoparticle injection was 71.4% lower than that after injection of the same dose of free tPA (Fig. 1B). Five minutes after the transcutaneous US application, tPA activity of the nanoparticles at the site of US application was recovered to the level similar to that after free tPA injection. Additionally, plasma tPA activity of the nanoparticles also was recovered by the US application 40 min after nanoparticle injection to the level higher than that after the injection of free tPA alone (Online Fig. 5).
Binding ability of the nanoparticles to vWF in vitro
To design nanoparticles to target sites with thrombus, we adopted basic gelatin, a heat-denatured collagen, because collagen binds to vWF, a key component of platelet-rich thrombi. As shown in Figure 1C, the binding ability of basic gelatin (10−6 M) and the nanoparticles (10−6 M) to vWF was as strong as that of antihuman vWF antibody (10−8 M).
Thrombus-targeting property of nanoparticles
Next, we evaluated whether nanoparticles target thrombi using radiolabeled tPA in vivo. Free 125I-labeled tPA alone or nanoparticles containing the same dose of 125I-labeled tPA were injected into the tail vein of a mouse with thrombotic occlusion in the left FA. The ratio of radioactivity of the thrombosed left FA to the control right FA was approximately 3-fold higher in the mice treated with the nanoparticles than in the mice treated with tPA alone (Fig. 1D), indicating that the nanoparticles have a higher affinity for thrombus than free tPA.
Thrombolytic efficacy of the intelligent DDS in a swine AMI model
Figure 2 shows the results of thrombolysis in the swine AMI model. In the cases using the intelligent DDS (intravenous injection of the nanoparticles containing 55,000 IU/kg [0.447 mg/kg] tPA followed by TUS application), thrombolysis with TIMI flow grade 2 or more was achieved in 9 of 10 swine within 30 min. In contrast, only 4 of 10 swine treated with 55,000 IU/kg (0.447 mg/kg) of tPA followed by TUS application (tPA plus TUS) and only 1 of 10 swine treated with 55,000 IU/kg (0.447 mg/kg) of tPA alone (Figs. 2A and 2B) achieved the same degree of flow. The intelligent DDS had a higher rate of TIMI flow grade 3 recanalization within 30 min (60%) in comparison with tPA plus TUS (30%) or tPA alone (0%) (Fig. 2B). Comparing the mean TIMI grade at 30 min among 3 groups, it was significantly higher in the swine treated with the intelligent DDS containing 55,000 IU/kg (0.447 mg/kg) of tPA than those treated with 55,000 IU/kg (0.447 mg/kg) tPA plus TUS or 55,000 IU/kg (0.447 mg/kg) tPA alone (2.50 ± 0.71 vs. 1.30 ± 1.34, p < 0.05, and 0.40 ± 0.70, p < 0.01, respectively) (Fig. 2C).
Unexpectedly, in the swine AMI model, low-dose tPA alone (27,500 IU/kg, 0.2235 mg/kg), equivalent to the standard clinical dose for intravenous thrombolytic therapy in humans with a reported coronary recanalization rate 60% (15), did not recanalize at all up to 60 min after the initiation of treatment. However, the intelligent DDS with 27,500 IU/kg (0.2235 mg/kg) tPA for 60 min achieved a TIMI flow grade 3 recanalization rate of 60% (Figs. 2D and 2E). Additionally, the intelligent DDS with low-dose tPA attained a significantly higher mean TIMI grade than tPA plus TUS at 30 min and 60 min (1.30 ± 1.16 vs. 0.10 ± 0.32, p < 0.01, and 2.00 ± 1.33 vs. 0.50 ± 0.97, p < 0.05, respectively) (Fig. 2F).
Figure 3 and the Online Videos 1, 2, and 3 show representative images of coronary angiography at 60 min in swine treated with tPA alone, swine treated with tPA plus TUS, and swine treated with intelligent DDS at the tPA dose of 55,000 IU/kg (0.447 mg/kg). Treatment with tPA alone failed to recanalize the occluded LCx. In the group treated with tPA plus TUS, although 6 of 10 swine achieved TIMI flow grade 3 after 60 min of treatment, 4 of those 6 had residual thrombi at the affected LCx. In contrast, the group treated with the intelligent DDS had no residual thrombi.
Plasma tPA activity in the swine AMI model
Figure 4 shows the time course of the plasma tPA activity expressed as the equivalent dose of tPA (in micrograms per milliliter). We drew blood samples proximal to the occluded LCx and FA in which TUS was not applied.
In the swine injected with 55,000 IU/kg (0.447 mg/kg) tPA, plasma tPA activity peaked at 30.09 ± 2.82 μg/ml equivalent immediately after injection and decreased gradually. The plasma tPA activity in swine treated with 55,000 IU/kg (0.447 mg/kg) of tPA plus TUS was similar. In both groups, the plasma tPA activity in the FA was similar to that in the LCx at each time point (data not shown).
The plasma tPA activity immediately after intravenous injection of the nanoparticles was approximately 75% lower before TUS application in the LCx and the FA (8.01 ± 0.89 μg/ml and 7.75 ± 1.00 μg/ml equivalent, respectively) than after free tPA injection (30.09 ± 2.82 μg/ml and 29.79 ± 1.13 μg/ml equivalent, p < 0.01) (Fig. 4A). After TUS application for 5 minutes (at 10 minutes), the plasma tPA activity in the LCx in swine treated with the nanoparticles increased to 32.58 ± 3.04 μg/ml equivalent, similar to that immediately after free tPA injection. However, the plasma tPA activity in the FA increased to a much lower level than that in the LCx (18.45 ± 4.41 μg/ml equivalent, p < 0.01). Treatment with the intelligent DDS maintained higher plasma tPA activity in the LCx for up to 30 minutes when compared with tPA alone or with tPA plus TUS (Fig. 4A).
The plasma tPA activity in the swine treated with low-dose tPA (27,500 IU/kg, 0.2235 mg/kg) is illustrated in Figure 4B. The persistently higher plasma tPA activity also was observed in swine treated with the low dose of tPA, as is the cases with the high dose of tPA (55,000 IU/kg, 0.447 mg/kg). However, in swine treated with the intelligent DDS, tPA activity between 10 and 60 min, despite a lower tPA dose, had a similarly shaped profile.
Left ventricular ejection fraction after thrombolysis
As shown in Figure 5, left ventricular ejection fraction at 60 min after thrombolysis in the swine treated with the intelligent DDS with 55,000 IU/kg tPA was significantly higher than in the swine treated with tPA alone (p < 0.05) and tended to be higher than that in the swine treated with tPA plus TUS.
This is the first report of the development and application of an intelligent DDS for intravenous coronary thrombolysis comprising tPA, basic gelatin, and zinc ions in a swine model of AMI. Our DDS is equipped with stealth activity, thrombus targeting, and US-controlled properties, which enables higher tPA activity at the affected coronary artery as opposed to the FA, a remote region, thus enhancing intravenous coronary thrombolysis in a swine model of AMI. With the present DDS, gelatin both covered the active site of tPA and targeted the nanoparticles to the thrombi, probably because of its ability to bind to vWF, and US helped both to release tPA and to enhance thrombolysis. Thus, the DDS is intelligent, unique, and promising.
In our swine model of AMI, treatment with tPA (monteplase) alone could not dissolve any thrombi in occluded coronary arteries within 60 min at the dose of 27,500 IU/kg, which achieves approximately 60% recanalization in humans (15) and could dissolve only 30% of thrombi at a dose of 55,000 IU/kg, suggesting our swine AMI model is relatively refractory to conventional thrombolytic therapy. Even in such a model, the intelligent DDS carrying 55,000 IU/kg and 27,500 IU/kg tPA could achieve TIMI flow grade 2 or 3 recanalization in 90% (within 30 min) and in 60% (within 60 min) of cases, respectively. TIMI flow grade 3 was achieved in 60% of swine within 30 min by the intelligent DDS carrying 55,000 IU/kg tPA. Additionally, by this thrombolytic method, left ventricular ejection fraction at 60 min was higher compared with other thrombolytic methods. Thus, the intelligent DDS has much stronger thrombolytic activity than conventional thrombolytic therapy, which probably leads to the preservation of cardiac function; it is promising in the clinical setting.
During last 2 decades, several investigators used US to improve thrombolytic methods without increased bleeding risk. In fact, transcutaneous or transthoracic low-frequency US facilitated thrombolysis in several experimental animal models (5–7), as in the present study (tPA plus TUS group in Fig. 2). However, in a double-blind, randomized, controlled, multicenter trial that consisted of 396 patients with ST-segment elevation myocardial infarction, thrombolytic agents plus low-frequency transcutaneous US failed to improve TIMI flow grade and ST-segment resolution at 60 min versus a thrombolytic agent alone (8).
Recently, we reported that nanoparticles comprising tPA, cationized gelatin, and anionic gelatin with polyethylene glycol suppressed tPA activity by approximately 55%, but this activity was fully recovered by US application (13). In the present study, we formed a new generation of nanoparticles comprising gelatin and tPA in the presence of zinc ions (5 mM), in which tPA activity was reduced to approximately 50% in vitro and to 25% to 30% in vivo of the equivalent dose of free tPA, and that was fully recovered by US application. It is well known that US exerts the effect of cavitation on tissue liquid, which induces an intense local pressure (16). This pressure would enable the dissociation of tPA from nanoparticles and would allow tPA to exert biological action. As shown in Online Figure 6 and Online Table 1, only injection of nanoparticles without US application had little effect on the thrombus in the rabbit arterial thrombosis model, whereas TIMI flow grade 2 and 3 recanalization was observed in 5 of 10 rabbits with injection of the same dose of free tPA. However, with transcutaneous US application, the nanoparticles recanalized the FA occluded by thrombus in all 10 rabbits within 30 min. These data suggests the tPA activity of the nanoparticles was suppressed and recovered by transcutaneous US application in vivo.
Another important issue is the development of a new device for TUS application. The new device, comprising 4 probes of 1.0 MHz in parallel (Online Fig. 1), created uniform US fields over the entire heart, as quantified by a catheter-type oscilloscope inserted into the left ventricle, right coronary artery, left anterior descending coronary artery, and LCx, as shown in Online Figure 2. Thus, it may lead to feasible release of tPA and favorable US-related effects, such as cavitation and drug permeation into the clot in any coronary artery.
In terms of safety, gelatin and zinc acetate have proven biosafety profiles because of their longstanding use in medical materials such as drug ingredients (17–19). We also investigated the direct effects of US on apoptosis and inflammation. As shown in Online Figure 7, gene expression of apoptotic and inflammatory markers was not induced by transcutaneous or transthoracic US applications for 60 min. These findings suggest that the present intelligent DDS is promising in the clinical setting.
In summary, we developed for the first time a US-controlled DDS for coronary thrombolysis with both stealth activity of thrombolytic agents and thrombus-targeting properties and demonstrated its efficacy in thrombolysis in a swine model of AMI. Our intelligent DDS may provide novel reperfusion therapy in AMI patients that theoretically can be started in an ambulance and is also applicable in the treatment of stroke, pulmonary thromboembolism, peripheral arterial thrombosis, and leg vein thrombosis.
For a supplemental Methods section, and supplemental figures, tables, references, and videos, please see the online version of this article.
This study was supported by the Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation (project no. 06-14). Dr. Saito reports that he has received research funding from Merck & Co., Inc., Takeda Pharmaceutical Company Limited, Novartis Pharma K.K., Daiichi Sankyo Company Limited, Mitsubishi Tanabe Pharma Corporation, Pfizer Japan Inc., Astellas Pharma, Inc., Baxter Limited, AstraZeneca K.K., Shionogi & Co., Ltd.; honoraria from Merck & Co., Inc., Takeda Pharmaceutical Company Limited, Novartis Pharma K.K., Daiichi Sankyo Company Limited, Mitsubishi Tanabe Pharma Corporation, Pfizer Japan Inc., and Otsuka Pharmaceutical Co., Ltd.; and donations from Merck & Co., Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute myocardial infarction
- drug delivery system
- femoral artery
- left circumflex coronary artery
- Thrombolysis In Myocardial Infarction
- tissue-type plasminogen activator
- transthoracic ultrasound
- von Willebrand factor
- iodide 125
- Received April 30, 2012.
- Revision received July 6, 2012.
- Accepted August 12, 2012.
- American College of Cardiology Foundation
- Schroder R.,
- Neuhaus K.L.,
- Leizorovicz A.,
- Linderer T.,
- Tebbe U.
- Lincoff A.M.,
- Topol E.J.
- Luo H.,
- Nishioka T.,
- Fishbein M.C.,
- et al.
- Siegel R.J.,
- Atar S.,
- Fishbein M.C.,
- et al.
- Hudson M.,
- Greenbaum A.,
- Brenton L.,
- et al.
- Herr A.B.,
- Farndale R.W.
- Wilbur D.S.,
- Hadley S.W.,
- Hylarides M.D.,
- et al.
- Suslick K.S.
- Anderson L.A.,
- Hakojarvi S.L.,
- Boudreaux S.K.