Intracerebral hemorrhage (ICH) is a common type of stroke (about 10%–15% of strokes) with high morbidity and mortality.1–3 Most of the survivors have sequelae of varying degrees of motor impairment, cognitive impairment, and speech and swallowing disorders. The treatment of ICH patients involves the urgent treatment of mass effect, aggressive blood pressure reduction, and correction of contributing coagulopathies to achieve hemostasis and keep the patient stabilized.4 However, few ICH patients benefit from existing therapies.5 Effective treatments for ICH patients are desperately needed and this depends on appropriate animal models.
An ideal animal model mimics the natural events of hemorrhagic stroke in human beings. Rodent, canine, feline, and primate models of ICH, using mechanical balloon compression, bacterial collagenase, and either donor or autologous blood injection, have been reported in the past.6–8 These models have been used to study different pathophysiological aspects of ICH and its clinical disease progression. However, some inevitable disadvantages are noted. Mechanical balloon compression cannot naturally model the spontaneous hemorrhage progress and the ICH associated pathophysiologic changes. Consequently, this method has little application.8 The external injection methods are widely acceptable but invasive, which can lead to untargeted brain parenchyma injury and extra inflammation.9 Hence, it is necessary to establish a non-invasive, reproducible, and stable animal model of spontaneous intracerebral hemorrhage.
SonoVue (sulfur hexafluoride microbubble (μBs) for injection, Bracco Diagnostics Inc., Milan, Italy), a pure blood pool agent, is currently a widely used ultrasound contrast agent. It is commercially available and clinically approved for disease diagnosis,10 with low solubility in blood; it is also non-toxic.11 SonoVue is a second-generation contrast agent composed of a sulfur hexafluoride core and a phospholipid coating. μBs are used as contrast enhancers in ultrasound imaging, drug/gene delivery vectors, thrombus reducers, oxygen gas carriers, and drug delivery across the BBB.12
Focused ultrasound (FUS) has been widely used as a non-invasive ablation technique with millimeter-sized focal regions for the destruction of tissue without disruption of surrounding tissue by either thermal or mechanical bio-effects.13 FUS has been applied to generate mechanical bioeffects by co-administering μBs to trigger the acoustic cavitation effect.14 With the application of sufficient acoustic pressure (>100 kPa), the μBs generate nonlinear oscillations, leading to non-inertial and/or inertial cavitation, which can produce significant microstreaming.15 This bioeffect can lead to cellular injury or tissue destruction and is reported to be helpful in the treatment of tumors,16,17 immunotherapy,18 BBB opening,19–21 and drug delivery.22
We were thus inspired to hypothesize that when μBs and FUS are co-administered at a desired region within the rats' brain, local spontaneous ICH may occur due to the vessel rupturing cavitation bioeffect. Therefore, this study aimed to introduce a non-invasive and reproducible ICH model in rats by using FUS with μBs and to delineate changes in neurologic outcome and histology in acute ICH conditions.
METHODS Experimental animalsThirty-three Sprague Dawley (SD) male rats (age 6–7 weeks, weight 200–250 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The animals were housed under specific pathogen-free (SPF) conditions, at 22 ± 2°C and 55 ± 10% relative humidity, with a 12 h light/dark cycle. We randomly divided thirty rats into three experimental groups using a random number table based on body weight. Six rats were treated with FUS without injection of μB (FUS group), fourteen were treated with FUS and injection of μBs (FUS + μBs group, F&B group), and ten were treated with FUS, injection of μBs and 6.0 mg/kg edaravone (FUS + μBs + Edaravone group, F&B&E group). Neither μBs nor FUS intervention was given to three rats (Sham group), and the remaining operations were consistent with those of the other groups. The protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of West China Hospital, Sichuan University (Approval No. 2019192A). All animals were acclimated to the laboratory for 1 week.
Acoustic setupsAn Agilent N9310A RF Signal Generator (Agilent Technology Co., Ltd, Santa Clara, USA) was used to generate the driving signals. The signals were tone burst sine waves at 1.1 MHz with a 20% duty cycle and a 200 Hz pulse repetition rate. The signals were subsequently amplified by an ENI Model 3100LA RF Power Amplifier (Rochester, USA). The amplified signal was transmitted to the transducer via an impedance matching network (Sonic Concepts Inc., Bothell, USA) which was set to its fundamental mode. An H102 transducer from Sonic Concepts fitted with a customized focusing cone directed the ultrasonic energy to the rat's head, forming an oval-shaped acoustic focus at 5 mm beneath the skull surface. The schematic of the acoustic setup is shown in Figure 1A.
Enlarge this image.FIGURE 1. (A) The connections of the device. (B) A plane diagram of the target site for ultrasound intervention. (C) A schematic image representing the target coordinates of FUS. (D) Schematic diagram of the intervention procedure. (E) Experimental timeline.
Rats were fasted for 10 h before surgery and anesthetized with isoflurane mixed with oxygen (5% for surgical induction, 2% to 2.5% for maintenance). After anesthesia, the rats were fixed in a stereotaxic apparatus. The cranial skin was sterilized and split, and then 3% hydrogen peroxide was used to uncover the periosteum, thus exposing the bregma of the skull. The anterior bregma was used as a point to mark the target site. According to the stereoscopic map of the rat brain, we identified the striatum as the target site of focused ultrasound. The coordinates were 3.0 mm right, 1.0 mm anterior, and 5.0 mm ventral from bregma at the skull surface using a stereotactic apparatus. The schematic image that represents the target coordinates of FUS is shown in Figure 1B,C.
Preparation of water sac and turning on the deviceThe bottom of the transducer was wrapped with plastic wrap, and the transducer was filled with pure water to form a water sac, which was then aligned to the marked position on the head of rat. After that, an ultrasonic coupling agent was applied between the adventitia of the water sac and the skull to make a tight connection.
With the help of a microinjection pump, all groups except for the FUS group were injected with 45 μg/ml of μBs (SonoVue®) via the tail vein of the rats, at an injection speed of 5 μl/s. The injection of μBs took a total of 80 s. The first 50 s were used to stabilize the concentration of μBs in the blood pool (0.8–1.0 ml/kg), and during the last 30 s the equipment was activated for ultrasound intervention. The parameters of the ultrasonic equipment are shown in Table 1. After ultrasound intervention, the rats in the F&B&E group were immediately injected with edaravone (6 mg/kg) via the tail vein. Thirty minutes after the ultrasound intervention, rats in each group were intraperitoneally injected with 10% sodium fluorescein (NaFL) at a dose of 4 ml/kg.
TABLE 1 Acoustic parameters of each group
| Group | Parameter | ||||
| Frequency (MHz) | Acoustic pressure (MPa) | Duty ratio (%) | Time (s) | Number | |
| Sham | – | – | – | – | 3 |
| FUS | 1.1 | 9 | 20 | 30 | 6 |
| FUS + μBs (F&B) | 1.1 | 9 | 20 | 30 | 14 |
| FUS + μBs + Edaravone (F&B&E) | 1.1 | 9 | 20 | 30 | 10 |
Twenty-four hours post ultrasound intervention, the rats were anesthetized with 3% pentobarbital sodium. The rat hearts were perfused with PBS solution, and the brain tissue was removed and observed for gross histology, especially at the ICH location. The number of red pixels in the gross histology was counted as a rough quantification of the hemorrhage. Subsequently, the brain tissue was cut into six coronal slices of 2 mm thickness. The slices underwent the following fluorescent imaging and the TTC staining for side-by-side comparison. BBB permeability was detected using the fluorescence imaging mode of an IVIS® Lumia series III imaging system (PerkinElmer, USA), and fluorescence intensity was quantized using the total radiant efficiency (TRE) value. The fluorescence imaging methods were detailed in our previous study.23 The same brain slices were subsequently stained with 2,3,5-triphenyl tetrazolium chloride (TTC), and the cerebral infarction volume of each group was analyzed with ImageJ software (National Institutes of Health, Bethesda, USA). Two brain tissue samples were randomly selected from the F&B group and the F&B&E group for HE staining to observe the morphological structure of the hemorrhagic area of the brain tissue. A schematic diagram of the intervention procedure and the experimental timeline are shown in Figure 1D,E.
Statistical analysisGraphPad Prism software (GraphPad Software Inc., San Diego, USA) version 9.0 was used for statistical analysis. Gaussian distributed data (Shapiro–Wilk test) were expressed as means ± standard deviation (mean ± SD). The TRE in each group was evaluated for statistical significance by the nonparametric Mann–Whitney test. Comparison of cerebral infarction volumes among groups was tested by the unpaired t test. p values <0.05 were considered statistically significant.
RESULTS Successfully established ratIn our research, none of the animals in any group died during the experiment. The mortality rate was zero. In addition, hematoma lesions were observed in all rats in the F&B group. In other words, the success rate of establishing the rat ICH model with our new method is 100%. In the F&B&E group, no hematoma lesions were observed in two rats, while hematoma was found in the remaining rats, but the size of the hematoma was smaller than that in the F&B group.
TRE values were 3.23 ± 0.66 × 1010 in the Sham group, 3.65 ± 0.92 × 1010 in the FUS group, and 1.55 ± 0.88 × 1011 in the F&B group. As shown in Figure 2A, BBB permeability was slightly increased in the FUS group compared with the Sham group. Compared to the FUS group, the BBB permeability was significantly enhanced in F&B group (Figure 2A,B, P = 0.0021). It can be seen from the coronal brain slices in Figure 2C that no cerebral hemorrhage symptoms occurred in the FUS group, while cerebral hemorrhage occurred in the F&B group, indicating that with the help of μBs, FUS can non-invasively create a rat ICH model. TTC staining showed no cerebral infarction in the sham group (Figure 2C). At the same time, the ratios of cerebral infarction volume to whole brain volume were 0.93% ± 0.67% in the FUS group and 3.13% ± 1.46% in the F&B group. The F&B group had a significantly larger cerebral infarction volume than that of the FUS group (Figure 2D, p = 0.0030). Since it is well known that hemorrhage causes excessive brain tissue damage,4 it is reasonable to assume that the extra infarction volume was caused by the hemorrhage induced by ultrasound, further indicating that hemorrhagic cerebral infarction occurred in the F&B group. In particular, we noted that the fluorescent intensity distribution correlates spatially with the tissue infarction location, in accordance with the ultrasound focus.
Enlarge this image.FIGURE 2. (A) Representative images of brain slices detected by the IVIS® Lumina III system in three groups. The relative fluorescence intensity is color graded as low (blue), medium (green), and high (red). (B) Quantification the fluorescence intensity of brain slices in three groups. (C) Representative images of brain slices before and after TTC staining in three groups. The red pixel in the red box is the cerebral hemorrhage lesion. (D) Cerebral infarction volume in two groups.
In Figure 3A, the red-framed brain slices of the F&B group and F&B&E group had different degrees of hemorrhage. The cerebral infarction volume in the F&B&E group was 1.90% ± 0.78%, which was significantly smaller than that in F&B group (Figure 3B, p = 0.0208). It indicated that the therapeutic effect of edaravone reduced cerebral infarction volume.
Enlarge this image.FIGURE 3. (A) Representative images of brain slices before and after TTC staining in two groups. The red pixel in the red box is the cerebral hemorrhage lesion. (B) Cerebral infarction volume in two groups. (C) Representative pictures of HE staining in two groups. Nerve fiber dissolution (green arrow), neuron necrosis (purple arrow), microglia (blue arrow), erythrocyte (red arrow). The corresponding HE stain results at different magnifications: 8× (bar = 1000 μm), 40× (bar = 200 μm), 200× (bar = 50 μm).
The HE staining results (Figure 3C) of brain tissue showed that nerve fibrinolysis, neuronal necrosis, microglia production and erythrocytes were found in both groups, but the area of the nerve fibrinolysis in the F&B group was larger than that in the F&B&E group, indicating that edaravone can reduce the nerve fibrinolysis and has a therapeutic effect in this model.
DISCUSSIONMicrobubbles expand continuously when subjected to ultrasonic waves and rupture when the sound pressure reaches a certain value. In this process, the ultrasonic cavitation effect causes damage or even rupture of the blood vessel walls. Under the same acoustic conditions, the BBB permeability of the FUS group and the F&B group were both increased, but the BBB permeability of the latter was significantly greater than that of the former, which was consistent with existing reports that μBs contribute to the increase in BBB permeability.24 BBB permeability is an important pathophysiological process during the stroke.25,26 At the same time, no intracerebral hemorrhage occurred in the FUS group, but different degrees of cerebral hemorrhage occurred when FUS and μBs are combined in the other two groups. Therefore, the presence of both the appropriate ultrasound field and the μBs are necessary for the hemorrhage to occur.
When ICH occurs, the hematoma compresses the surrounding tissue leading to ischemia and hypoxia in the compressed tissue, which causes oxidative stress and oxidative damage in the brain tissue.27,28 Previous studies have shown that edaravone protects neurons and vascular endothelial cells by scavenging free radicals (such as hydrogen peroxide and hydroxyl radicals) and inhibiting lipid peroxidation.29,30 Therefore, the reduced cerebral infarction volume in the F&B&E group is an indication that ICH indeed occurred in both μB-assisted FUS intervention groups. The results of HE staining showed that the areas of neuronal necrosis and nerve fibrinolysis in the edaravone treatment group were smaller than those in the group without the edaravone treatment, possibly because edaravone reduced the oxidative damage to the brain tissue and the local neurons were thus protected. This result indicates that μB-assisted FUS intervention not only successfully created a rat ICH model, but also that the cerebral infarction volume caused by the intervention can be improved by edaravone to some extent.
FUS has been used in the treatment of various solid tumors, such as liver cancer, breast cancer, prostate cancer, and malignant renal tumors.31–33 The main mechanisms of FUS treatment of tumors are thermal and mechanical effects.34 In our study, the ultrasound pulse sequence was designed to minimize the thermal effect. Moreover, the cerebral blood flow carried away some of the thermal deposits. Nevertheless, the inevitable residual thermal effect may lead to additional damage to the brain tissue between the focal point and the transducer.35 Because thermal damage is known to be irreversible, the portion of damage rescued with the edaravone therapy is proved to be nonthermal ICH. Hence, the cerebral infarct volume in this study consists of both the thermal effect-induced injury and the ICH-induced injury.
In follow-up studies, we will reduce the cerebral infarction volume caused by the thermal effect by adjusting the parameters of the device or preparing new μBs, and then optimize this model to make it a more reliable animal model for preclinical research of new drugs that treat spontaneous ICH-related diseases.
CONCLUSIONThis study reported a non-invasive, highly reproducible, and highly controllable rat ICH model prepared with μB-assisted FUS treatment. The ICH condition generated was found to be improved by the free radical scavenger edaravone, which may in future be used to explore the pathophysiological mechanisms of ICH-related diseases and in preclinical research into related new drugs. A residual thermal effect was documented and will be reduced in later research designs, such as by enhancing the maximum negative peak pressure of ultrasound or using improved nano-agents.
AUTHOR CONTRIBUTIONSZhihui Zhong, Qiang Lu, and Zhe Wu conceived the experiments. Yao He, Jie Yang, and Fengying Hu performed the experiments and drafted the manuscript. Min Liao, Yuru Nie, and Xiaoxia Zhu collected literature. Tao Zhang and Keer Song helped perform the analysis with constructive discussions. Qinxi Li, Xiaojie Li, and Chenghan Mei analyzed the data. All authors read and approved the final manuscript.
ACKNOWLEDGMENTSThis study was funded by the National Natural Scientific Foundation of China (82071349, 82027808, 82171952, 81771310) and West China Hospital of Sichuan University Discipline Excellence Development 1·3·5 Engineering Project (Interdisciplinary Innovation Project, ZYYC08005, ZYJC18041).
CONFLICT OF INTERESTThe authors have no competing interests to declare. Zhihui Zhong is an Editorial Board member of AMEM and a co-author of this article. To minimize bias, he was excluded from all editorial decision-making related to the acceptance of this article for publication.
ETHICS STATEMENTThe animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of West China Hospital, Sichuan University (Approval No. 2019192A).
© 2023. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Yang, Jie 2 ; Hu, Fengying 3 ; Liao, Min 2 ; Nie, Yuru 3 ; Zhu, Xiaoxia 2 ; Zhang, Tao 4 ; Song, Keer 5 ; Li, Qinxi 1 ; Li, Xiaojie 1 ; Chenghan Mei 1 ; Wu, Zhe 3 ; Lu, Qiang 2 ; Zhong, Zhihui 1
1 Laboratory of Nonhuman Primate Disease Modeling Research, State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu, China
2 Department of Medical Ultrasound, West China Hospital, Sichuan University, Chengdu, China
3 School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China
4 School of Bioscience and Technology, Chengdu Medical College, Chengdu, China
5 Franklin College of Arts and Science, University of Georgia, Athens, Georgia, USA