Low‐intensity pulsed ultrasound attenuates cardiac inflammation of CVB3‐induced viral myocarditis via regulation of caveolin‐1 and MAPK pathways

Abstract The aggressive immunological activity elicited by acute viral myocarditis contributes to a large amount of cardiomyocytes loss and poor prognosis of patients in clinic. Low‐intensity pulsed ultrasound (LIPUS), which is an effective treatment modality for osteoarthropathy, has been recently illustrated regulating the overactive inflammatory response in various diseases. Here, we aimed to investigate whether LIPUS could attenuate coxsackievirus B3 (CVB3) infection‐induced injury by coordinating the inflammatory response. Male BALB/c mice were inoculated intraperitoneally with CVB3 to establish the model of acute viral myocarditis. LIPUS treatment was given on Day 1, Day 1, 3 and Day 1, 3, 5 post‐inoculation, respectively. All mice were followed up for 14 days. Day 1, 3, 5 LIPUS treatment significantly improved the survival rate, attenuated the ventricular dysfunction and ameliorated the cardiac histopathological injury of CVB3‐infected mice. Western blotting analysis showed Day 1, 3, 5 LIPUS treatment decreased pro‐inflammatory cytokines, increased the activation of caveolin‐1 and suppressed p38 mitogen‐activated protein kinase (MAPK) and extracellular signal‐regulated kinase (ERK) signallings in heart tissue. RAW264.7 cells were treated with lipopolysaccharides (LPS) to simulate the augmented inflammatory response in vivo. LIPUS treatment on RAW264.7 inhibited the expression of pro‐inflammatory cytokines, activated caveolin‐1 and suppressed p38 MAPK and ERK signallings. Transfecting RAW264.7 with caveolin‐1 siRNA blunted the suppression of pro‐inflammatory cytokines and MAPK signallings by LIPUS treatment. Taken together, we demonstrated for the first time that LIPUS treatment attenuated the aggressive inflammatory response during acute viral myocarditis. The underlying mechanism may be activating caveolin‐1 and suppressing MAPK signallings.


| INTRODUCTION
Viral infection-induced myocarditis is one of the leading causes of acute heart failure and malignant ventricular arrhythmias among young patients. During the acute phase of the viral myocarditis, the virus invasion not only induced a direct viral cytopathic effect on cardiomyocytes, but also elicited the activation of host immune response. 1 It has been demonstrated that it was the immunemediated indirect myocardial injury but not virus-mediated direct destroy that resulted in a large amount of myocardium loss, contributing to the acute heart failure and even sudden death in clinic.
Therefore, regulation of the aggressive immunological response during this acute phase could be potential and effective therapy for viral myocarditis. [2][3][4][5] Low-intensity pulsed ultrasound (LIPUS) was a form of mechanical stimulation delivered via a special device and is widely applied as a treatment modality for osteoarthritis in clinical settings. 6,7 During past decades, emerging evidences revealed that LIPUS treatment could reduce the expression of pro-inflammtory cytokines, limit the infiltration of inflammatory cells and modulate the phenotype of inflammatory cells in a series of bone and joint diseases, and consequently accelerate the recovery from diseases. [8][9][10][11] Moreover, recent animal experiments demonstrated that LIPUS could also benefit several cardiovascular diseases, such as left ventricular remodelling induced by chronic myocardial ischaemia and transverse aortic constriction (TAC), where mechanical stimulation of caveolin-1 and regulation of its downstream intracellular signallings were critically involved. [12][13][14] However, so far the effect of LIPUS on acute viral myocarditis was never investigated. Based on the studies mentioned above, we doubted whether LIPUS treatment, which had an inflammatory modulating effect, could be utilized as a therapeutic method for the aggressive cardiac inflammation induced by CVB3 infection.
Caveolin-1, a major component of the plasma membrane microdomains, has been identified in cellular mechanotransduction system sensing the stimulation of LIPUS treatment. It was well known that caveolin-1 was expressed in various immune cells, including macrophages, dendritic cells and lymphocytes and acted as a potent immunomodulatory molecule via regulating the mitogen-activated protein kinases (MAPK) family members. [15][16][17][18] In this study, we aimed to investigate two unknown entities, whether the LIPUS treatment could regulate the aggressive immunological response of CVB3-induced myocarditis, and if so, to elucidate the underlying molecular mechanisms involved in the beneficial effect.

| Animal preparation
Male BALB/c mice (4 weeks) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd, China. All mice were kept under the specific pathogen-free conditions (24 ± 1°C, 45 ± 10% humidity) with a 12-hour light/dark cycle daily and food and water available ad libitum in the Wenzhou Medical University animal facilities. All animal experiments were approved by the Animal Ethics Committee of Wenzhou Medical University (number wydw2014-0058) and conformed to the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health.

| Myocarditis animal model
CVB3 was amplified by Hep2 cell and harvested. A 50% tissue culture infectious dose (TCID50) assay was used to determine the viral titer. At the age of 5 weeks old, mice were intraperitoneally injected with 0.2 mL phosphate buffered saline (PBS) containing 105 TCID50 CVB3 to induce viral myocarditis. Control mice were intraperitoneally injected with same dosage of PBS. We defined the day of virus inoculation as day 0.

| Low-intensity pulsed ultrasound therapy
The LIPUS therapy was delivered by an ultrasound device (Sonopuls 190; Enraf-Nonius BV, Rotterdam, the Netherlands) in accordance with the manufacturer's instructions. The parameters of LIPUS therapy adopted for both in vivo and in vitro experiments were based on the specific settings of the device and our preliminary study (Data S1), which were ultrasound frequency of 1 MHz, duty cycle of 20%, pulse repetition frequency of 100 Hz, output intensity of 0.5 W cm −2 , giving an intensity of spatial average and temporal average of 100 mW cm −2 . The beams were irradiated from the probe to the mice though an agar phantom gel. LIPUS was applied to the heart in the parasternal short-axis view at papillary muscle levels with the direction of transthoracic echocardiography for 20 minutes a day under inhalation anaesthesia with 1.2% isoflurane. No-LIPUS group underwent the same procedures including anaesthesia but without the LIPUS delivered by ultrasound device. In vitro experiments, LIPUS treatment was applied to the bottom of the cultured platelet via an agar phantom gel for 20 minutes.

| Group
All control mice were then assigned to two groups blindly, one group was given LIPUS device without ultrasound delivery on day 1, day 3, day 5 (Control group) and another group was given LIPUS device with low-intensity pulsed ultrasound delivery on day 1, day 3, day 5 (LIPUS group). All CVB3 infected mice was blindly assigned into four groups, the following treatment were given respectively: (a) VMC group: CVB3-infected mice were given LIPUS device without ultrasound delivery on day 1, day 3, day 5; (b) VMC + D1 LIPUS group: infected mice were given LIPUS device with ultrasound delivery on day 1 and without ultrasound delivery on day 3 and day 5; (c) VMC + D1, 3 LIPUS group: infected mice were given LIPUS device with ultrasound delivery on day 1, day 3 and without ultrasound delivery on day 5; (d) VMC + D1, 3, 5 LIPUS: infected mice were given LIPUS device with ultrasound delivery on day 1, day 3 and day 5. Six mice were selected randomly from each group and killed on days 7 and 14, the hearts of mice were obtained for histological and biochemical examinations, in addition, the blood of mice were collected and plasma level of Troponin I were measured. Preceding to sacrifice, all mice were anaesthetized with pentobarbital (100 mg kg −1 , administered intraperitoneally). Efficient anaesthesia was evaluated by pinching the hind paw, when sufficient sedation was achieved, the mice were killed by cervical dislocation.

| Survival rate
Forty mice extracted from each group were used to monitor survival rate. The survival of each group was observed up to 14 days.

| Echocardiographic evaluation
Transthoracic echocardiography was performed with vevo1100

HW/BW
The heart weight (HW), body weight (BW) and the ratio of HW to BW (HW/BW) were calculated on day 7 and day 14.

| Enzyme-linked immunosorbent assay (ELISA) measurement of Troponin I in plasma
The blood samples were centrifuged by 1000 g for 10 minutes at room temperature, plasma collected was then analysed by a mousespecific ELISA kit for Troponin I (Elabscience, E-EL-M0086c), following the manufacturer's instruction.

| Haematoxylin and eosin (HE) stain
Heart tissues were fixed in 10% formaldehyde, embedded in paraffin, sectioned into 5-μm-thick slices and stained with HE. The cardiac inflammatory infiltration was evaluated by observers who were blinded to the experimental groups. Pathological scores were given based on the following criteria: 0 = no lesion; 1 = lesion involving 25% of the myocardium; 2 = lesions involving 25% to 50% of the myocardium; 3 = lesions involving 50% to 75% of the myocardium and 4 = lesions involving 75% to 100% of the myocardium.

| Real-time polymerase chain reaction (PCR)
According to the manufacturer's protocol, total RNA was extracted from the heart tissue with the TRIzol Reagent (Invitrogen Corporation) and converted into cDNA with a RevertAid RT Reverse Transcription Kit (Thermo fisher). Real-time PCR was performed using SYBR Green I Master (Roche) by LightCycler ® 480 System (Roche).
The primer sequences applied in the procedure were shown as fol-

| Western blotting
The proteins were extracted from heart tissues and RAW264.7 cells.

| Immunofluorescence
Heart tissues were embedded in the optimal cutting temperature (OCT) and sectioned at a thickness of 5 μm. The sections were then stained with anti-CD68 primary antibody [dilution 1:500; Proteintech, 25747-1-AP] at 4°C overnight and FITC-conjugated goat antirabbit IgG [dilution 1:500; Yeasen, 33107ES60] at room temperature for 1 hour to identify monocytes/macrophages. The fluorescent dye 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) (beyotime, C1005) was finally added to the sections to mark the nuclei. Images were collected using an Olympus IX71 fluorescence microscope. RAW264.7 cells were seeded in 6-well tissue culture plate at a density of 5 × 10 5 per well. After adherence, the cell were treated with 100 ng mL −1 lipopolysaccharides (LPS) for 12 hours and then exposed to LIPUS therapy. After LIPUS delivery, the cells were stored for 6 hours in the same medium before protein extraction.

| Statistical methods
Data were expressed as means ± standard deviation (SD). Shapiro-Wilk test was used for normal distribution. If distribution of the data was normal, outcomes were compared among groups using 1-way ANOVA followed by the Dunnett multiple-comparison test.
If not, Wilcoxon's rank sum test should be applied instead.
The Kaplan-Meier method was applied to calculate survival rate.
The differences in the pathological scores were evaluated using Krusal-Wallis test. All the analyses were performed with SPSS 17.0 statistical software. A value of P < 0.05 was considered significant.

| Low-intensity pulsed ultrasound improved the left ventricular function in CVB3-infected mice
On day 14, transthoracic echocardiography was utilized to evaluate the left ventricular function of mice from each group. Compared with mice in Control group and LIPUS group, mice in VMC group exhibited an obvious deteriorated heart function, manifesting as enlarged left ventricular end-diastolic and end-systolic diameter and declined left ventricular ejection function (P < 0.05). Mice in VMC + D1 LIPUS group showed no apparent differences with mice in VMC group regarding these echocardiographic parameters. Compared with mice in VMC group, mice receiving D1, 3 and D1, 3, 5 LIPUS treatment presented an significant improvement in heart function with an increased left ventricular ejection fraction (P < 0.05). In addition, a more improvement of heart function was observed in VMC + D1, 3, 5 LIPUS group (P < 0.05) (Figure 2A-D).
The cardiac expression of brain natriuretic peptide (BNP) was also measured to monitor the heart function of mice from each group. On Day 14, Compared with Control group and LIPUS group, there was an increased expression of cardiac BNP in VMC group (P < 0.05). VMC + D1 LIPUS group exhibited no significant difference of BNP with VMC group. The BNP level was largely decreased in VMC + D1, 3 LIPUS and VMC + D1, 3, 5 LIPUS groups (P < 0.05) and a higher level of BNP decline was observed in VMC + D1, 3, 5 LIPUS group (P < 0.05) ( Figure 2E).

| LIPUS treatment increased the activation of caveolin-1 and decreased the phosphorylation of p38 MAPK and ERK in heart tissue of CVB3-infected mice
As previous studies revealed that LIPUS treatment could elicit a mechanical stimulation of caveolin-1, and MAPK signallings have been demonstrated critically involved in inflammatory response and regulated by caveolin-1, we further investigated the effect of LIPUS on cardiac expression of caveolin-1 and MAPK signallings in viral myocarditis. On day 7, compared with Control group and LIPUS group, there was an increased phosphorylation of caveolin-1, p38 MAPK and ERK in VMC group (P < 0.05). Though the expression of P-caveolin-1, P-p38 MAPK and P-ERK did not differ between VMC group and VMC + D1 LIPUS group, D1, 3 LIPUS and D1, 3, 5 LIPUS treatment significantly increased the phosphorylation of caveolin-1 and decreased the phosphorylation of p38MAPK and ERK (P < 0.05) ( Figure 5A). Additionally, D1, 3, 5 LIPUS treatment increased P-caveolin-1 and decreased P-p38 MAPK and P-ERK to a much larger extent. On day 14, there was still a higher expression of P-caveolin-1, P-p38 MAPK and P-ERK in VMC group compared with Control and LIPUS group (P < 0.05). Besides, VMC + D1, 3, 5 LIPUS group still exhibited a much higher expression of P-caveolin-1 and lower expression of P-p38 MAPK and P-ERK on day 14 (P < 0.05) ( Figure 5B).

| LIPUS treatment alleviated LPS-induced inflammatory response on RAW264.7 by increasing the activation of caveolin-1 and suppressing the phosphorylation of p38 MAPK and ERK
Given the catastrophic cardiac injury during acute viral myocarditis was caused by enlarged innate immunological activity, but not the CVB3 itself. Thus, in vitro study, the LPS was utilized on RAW264.7 to mimic the overactive immunological response during acute viral myocarditis. Compared with Control and LIPUS group, LPS treatment increased the expression of TNF-α and IL-6 (P < 0.05), however, LIPUS treatment decreased LPS-induced elevation of pro-inflammatory cytokines (P < 0.05) ( Figure 7A). LPS-treated RAW264.7 exhibited an increased expression of P-caveolin-1, P-p38 MAPK and P-ERK (P < 0.05), however, LIPUS treatment post-LPS inoculation led a more increase of P-caveolin-1 and higher suppression of P-p38 MAPK and P-ERK (P < 0.05) ( Figure 7B). The relationship between caveolin-1 and MAPK signallings under LIPUS treatment was further explored. After transfected with caveolin-1 siRNA, the expression of caveolin-1 in RAW264.7 was remarkably downregulated (P < 0.05) ( Figure 7C). With the knock-down of caveolin-1, the inhibitory effect of LIPUS on P-p38 MAPK, P-ERK and pro-inflammatory cytokines (TNF-α and IL-6) was significantly blunted (P < 0.05) ( Figure 7D).

| DISCUSSION
In this study, we first revealed that low-intensity pulsed ultrasound treatment could alleviate the drastic cardiac inflammatory response  Low-intensity pulsed ultrasound was a form of mechanical energy delivered via a special device and resulted in activation of mechanotransduction system of various cells. 19 Recently, LIPUS has been proven to ameliorate inflammatory activity in a series of animal experiments. Tatsuya Nakamura et al. once reported that LIPUS treatment alleviated synovitis in the knee joints of animal models for rheumatoid arthritis. 20 In acute muscle tissue injury, LIPUS treatment led to reductions in the number of neutrophils and M1 macrophages, and consequently attenuated tissue inflammation and fibrosis. 11 In TAC-induced heart failure mice model, LIPUS treatment were revealed ameliorating perivascular fibrosis and myocardial ischaemia and suppressing macrophage infiltration. 14 In this study, we found that LIPUS treatment on day 1, 3 and 5 after CVB3 inoculation remarkably increased animal survival, improved heart function,  LPS-induced production of inflammatory cytokines and phosphorylation of ERK and p38MAPK were decreased by LIPUS treatment. 8 In addition, recent studies indicated that p38 MAPK and ERK signallings played pivotal roles in the pathogenesis of viral myocarditis, and suppressing these signallings led to an inhibition of cardiac injury and benefited the recovery from viral myocarditis. [23][24][25] In our study, we also found that p38 MAPK and ERK signallings were activated during viral myocarditis, while LIPUS treatment suppressed these signallings and attenuated the excessive inflammatory response during the disease course. Intriguingly, some studies showed that MAPK signallings could be further activated after receiving mechanical stimulation from LIPUS, which were contrary to our findings. 13,26 It was noteworthy to mention that these stud- Several limitations should be mentioned for this study. First, as LIPUS treatment was performed within the first 5 days after virus inoculation, the effect of a relative longer course of LIPUS treatment on viral myocarditis was not investigated. However, based on our study, we confirmed the cardioprotective effect of LIPUS treatment F I G U R E 6 Day 1, 3, 5 LIPUS treatment inhibited macrophages infiltration in heart tissue of CVB3-infected mice. Representative immunofluorescence staining of CD68 + cells in heart tissue from each group (magnification ×200). n = 6 in each group. *P < 0.05 vs Control group, # P < 0.05 vs LIPUS group, & P < 0.05 vs VMC group,^P < 0.05 vs VMC + D1 LIPUS group, Δ P < 0.05 vs VMC + D1, 3 LIPUS F I G U R E 7 LIPUS treatment alleviated the LPS-induced inflammatory response on RAW264.7 by increasing the activation of caveolin-1 and suppressing the phosphorylation of p38 MAPK and ERK. A, LIPUS inhibited the pro-inflammatory cytokines (TNF-α and IL-6) expression of RAW264.7 treated with LPS. n = 6 in each group. *P < 0.05 vs Control group, # P < 0.05 vs LIPUS group, & P < 0.05 vs LPS group. B, LIPUS increased P-caveolin-1 and suppressed P-p38 MAPK and P-ERK of RAW264.7 treated with LPS. Immunofluorescence staining of P-caveolin-1 in control, LIPUS, LPS and LPS + LIPUS cells (magnification ×400). *P < 0.05 vs Control group, # P < 0.05 vs LIPUS group, & P < 0.05 vs LPS group. P-caveolin-1, P-p38 MAPK and P-ERK expression in control, LIPUS, LPS and LPS + LIPUS treatment cells. n = 6 in each group. *P < 0.05 vs Control group, # P < 0.05 vs LIPUS group, & P < 0.05 vs LPS group. C, Transfection of RAW264.7 with caveolin-1 siRNA downregulated the expression of caveolin-1. n = 6 in each group. *P < 0.05 vs Control group, # P < 0.05 vs RAW264.7 transfected with non-target siRNA. D, Transfection of RAW264.7 with caveolin-1 siRNA blunted the inhibitory effect of LIPUS on pro-inflammatory cytokines (TNF-α and IL-6) and P-p38 MAPK and P-ERK. n = 6 in each group. *P < 0.05 vs LPS group, # P < 0.05 vs LPS + LIPUS group given in the early phase of viral myocarditis. Second, we have not investigated LIPUS treatment on caveolin-1 knochout (KO) mice.
Though caveolin-1 KO mice were not utilized in vivo study to testify the LIPUS effect on viral myocarditis, caveolin-1 knockdown by small interfering RNA was performed in vitro to clarify the mechanism of LIPUS treatment. In addition, it has been reported that caveolin-1 KO mice exhibited defected innate immunity and overactive inflammatory activity in response to various stress. 28,29 Thus, the beneficial effects of LIPUS on viral myocarditis might be blunted in caveolin-1 KO mice. Third, the enlarged immunological response during the acute viral myocarditis was mimicked by LPS treatment on RAW264.7 in vitro but not by CVB3 infection. As an essential innate immune cell component, cardiac macrophages began to enrich at as early as day 3 but not immediately post-CVB3 infection.
Conditioned by the infected cardiac microenvironment but not CVB3, infiltrating macrophages initiated and promoted cardiac inflammation by secreting a series of pro-inflammatory cytokines and further dictating the subsequent adaptive immunity. It was worth mentioning that the virus itself only caused local immune response with limited cardiac injury, whereas, the character took responsibility for the large amount of myocardium loss and poor prognosis during viral myocarditis was the overactive immunological activity. From this standpoint, we thought that LPS was more reasonable than CVB3 to simulate the overactive immunological response. Fourth, in this study, we only investigated LIPUS treatment on macrophage in vitro, leaving its effect on other cells not explored. A wide range of cardiac resident cells (cardiomyocytes, endothelial cells, fibroblasts) were also induced inflammation by CVB3 infection and assisted in shaping the inflammatory response during viral myocarditis, 30 their inflammatory response may be also influenced by LIPUS treatment. However, as we aimed to investigate the LIPUS treatment on the aggressive immunological response during acute viral myocarditis, which was initiated and gone through by macrophages, thus we selected macrophages rather than other cells to do our explorations.

| CONCLUSION
In this study, we revealed that the LIPUS therapy attenuated the inflammatory response and improved the survival of viral myocarditis, where activation of caveolin-1 and regulation of MAPK signalling pathways were key mechanisms involved. Our study provided a new, promising and noninvasive strategy for the treatment of viral myocarditis.

CONF LICT OF I NTEREST
The authors declare that they have no conflict of interest.