Sonodynamic therapy reduces iron retention of hemorrhagic plaque

Abstract Intraplaque hemorrhage (IPH) plays a major role in the aggressive progression of vulnerable plaque, leading to acute cardiovascular events. We previously demonstrated that sonodynamic therapy (SDT) inhibits atherosclerotic plaque progression. In this study, we investigated whether SDT could also be applied to treat more advanced hemorrhagic plaque and addressed the underlying mechanism. SDT decreased atherosclerotic burden, positively altered atherosclerotic lesion composition, and alleviated iron retention in rabbit hemorrhagic plaques. Furthermore, SDT reduced iron retention by stimulating ferroportin 1 (Fpn1) expression in apolipoprotein E (ApoE)−/− mouse plaques with high susceptibility to IPH. Subsequently, SDT inhibited iron‐overload‐induced foam‐cell formation and pro‐inflammatory cytokines secretion in vitro. Moreover, SDT reduced levels of the labile iron pool and ferritin expression via the reactive oxygen species (ROS)‐nuclear factor erythroid 2‐related factor 2 (Nrf2)‐FPN1 pathway. SDT exerted therapeutic effects on hemorrhagic plaques and reduced iron retention via the ROS‐Nrf2‐FPN1 pathway in macrophages, thereby suggesting that it is a potential translational strategy for patients with advanced atherosclerosis in clinical practice.


INTRODUCTION
Currently, an existing challenge to antithrombotic treatment for atherosclerotic vascular diseases is the higher frequencies of intraplaque hemorrhage (IPH). 1,2 IPH plays a major role in the aggressive progression of atherosclerotic plaque, consequently leading to acute cardiovascular events. 3 The potent atherogenic stimulus caused by IPH is attributed to the deposition of erythrocyte lysis products, 4 which cause not only cholesterol deposits but also iron retention in plaques. 3 Ironmediated oxidative injury potentiates human atherogenesis 5 and increases the risk of plaque destabilization. 6 Additionally, free iron and iron-binding proteins derived from hemoglobin-degradation products increase the labile iron pool (LIP) in phagocytes of the plaque, especially in macrophages. The labile nature of LIP is revealed by its capacity to promote reactive oxygen species (ROS) generation via Fenton and Haber-Weiss reactions. 7 Ferroportin 1 (FPN1), an essential iron exporter identified in mammals, is regulated by hepcidin-mediated internalization and degradation. 8 FPN1 levels insufficient to maintain iron homeostasis in cells with iron overload results in excessive cyclic production of ROS, leading to oxidativestress injury. Therefore, FPN1 upregulation or FPN1-activity restoration might represent a novel approach to treating iron-loaded hemorrhagic plaque.
Established clinical anti-atherosclerotic strategies intended to control risk factors (e.g., antihypertensive medications and statins) have not manifested significant benefits in treating hemorrhagic plaque. 9,10 Iron-deprivation treatments (e.g., iron chelators and hepcidin inhibitors) reportedly exhibit anti-atherosclerotic effects 11,12 but can cause systemic imbalances in iron metabolism, leading to adverse effects, such as infection 13 and audiovisual-perception impairment. 14 Therefore, novel tissue-specific and efficient therapy is in great demand.
Sonodynamic therapy (SDT) has been proposed as a non-invasive approach to treating tumor and atherosclerosis. [15][16][17] SDT generates different concentrations of ROS based on the synergistic effects of sonosensitizers and ultrasound. The sonosensitizers used for SDT are mainly porphyrin derivatives. Among these, sinoporphyrin sodium (DVDMS) is a novel sonosensitizer isolated from Photofrin which has been approved by the Food and Drug Administration as a sensitizer.
DVDMS has higher chemical purity, better water solubility, stronger sonoactivity, and less skin sensitivity. 18,19 Therefore, DVDMS has priority in SDT application. Moreover, compared with tumor treatment, lower intensity of ultrasound is applied for atherosclerosis treatment. 20 Our previous studies showed that SDT promotes atherosclerotic plaque stability and regression by targeting macrophages. [15][16][17] Macrophages play a central role in iron metabolism 21 ; therefore, in the present study, we used animal models and in vitro experiments to clarify whether SDT rescued complicated plaques with IPH and modulated plaque-specific iron metabolism, as well as address the underlying mechanisms.

MATERIALS AND METHODS
A detailed description of the materials and methods is available in the Online Supplement.

Animals
Animal experiment protocols were approved by the Ethics Committee of Harbin Medical University. All applicable institutional and national guidelines for the care and use of animals were followed. Male New Zealand rabbits (age: 3-4 months; 2.5-3.0 kg) were purchased from Solarbio Bioscience & Technology Co., Ltd. (Shanghai, China).
Male apolipoprotein E (ApoE) −/− mice (age: 6 weeks) and 12-14-weekold C57BL/6 mice were purchased from Qingzilan Technology Co., Ltd. (Nanjing, China). All efforts were made to minimize animal suffering and reduce the number of animals used. Figure 1  Isolation of murine peritoneal macrophages and ironloaded macrophage formation C57BL/6 mice were intraperitoneally injected with 3% thioglycolate before isolating macrophages from the peritoneal cavity. After the mice were euthanized, 5 ml of ice-cold phosphate-buffered saline (PBS) containing 3% fetal bovine serum (FBS) was injected for peritoneal cavity perfusion. Peritoneal cells were harvested and plated on cell-culture dishes at appropriate concentrations. After 2 h, the medium was replaced with Roswell Park Memorial Institute (RPMI)-1,640 containing 12% FBS for adherent cell culture. Macrophages were incubated with 100 μM ferric ammonium citrate (FAC) for 12 h to allow iron-loaded macrophage (ILM) formation. Expanded methods are available in the Online Supplement.

SDT treatment
The ultrasound device was manufactured by Harbin Institute of Technology (Harbin, China). The ultrasonic transducer parameters were as follows: diameter, 35 mm; and resonance frequency, 1.0 MHz. DVDMS was used as the sonosensitizer for SDT.
The animals were kept away from light during and after SDT treatment for 24 h. At 4 h after intravenous DVDMS (4 mg/kg) administration, anesthetized animals were subjected to ultrasound for 15 min with an ultrasonic intensity of 1.5 W/cm 2 for rabbits and 0.4 W/cm 2 for mice, as previously described. 17,22 Mice in the hepcidin + SDT group were intraperitoneally administered 25 μg human hepcidin-25 dissolved in 100 μl PBS 1 h before SDT treatment, whereas mice in the hepcidin group received only hepcidin-25. The ultrasound was applied as F I G U R E 1 Sonodynamic therapy (SDT) reduces iron retention and exerts anti-atherosclerotic effects on rabbit hemorrhagic plaque. Red scale bar, 250 μm; black scale bar, 50 μm. *p < 0.05, **p < 0.01, *** p < 0.001 previously described. 15 In brief, for rabbits, the ultrasonic transducer was placed on the marked femoral artery through a degassed water column, whereas for mice, the ultrasonic transducer was placed under the neck through a degassed water column.
For the in vitro study, cells were plated on 35-mm Petri dishes that were placed on a degassed water bath 30 cm away from the ultrasonic transducer. Based on the optimized SDT parameters ( Figure S2(a)-(c)), ILMs were incubated with 0.2 μM DVDMS for 4 h, followed by irradiation with an ultrasound intensity of 0.2 W/cm 2 for 5 min.

Statistical analysis
All data are presented as the mean ± standard error of the mean. The normality test was performed to determine whether the data were normally distributed, and in cases where data were distributed normally, data were analyzed using a Student's t test, one-way analysis of variance (ANOVA), followed by Tukey's multiple comparison test or Dunnett's multiple comparison test, or two-way ANOVA, followed by Sidak's multiple comparison test, as appropriate. For data that were not normally distributed, a Kruskal-Wallis test, followed by Dunn's multiple comparison test, was used. Statistical analyses were performed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA, USA), and p < 0.05 was considered statistically significant.

SDT reduces iron retention and exerts antiatherosclerotic effects in rabbit hemorrhagic plaque
The rabbit IPH model was established as shown in Figure 1 SDT modifies iron metabolism in plaques of ApoE −/− mice Consistently, SDT treatment of plaques with high susceptibility to IPH in ApoE −/− mice exerted anti-atherosclerotic effects ( Figure S3). We then evaluated changes in levels of proteins associated with iron metabolism (ferritin, FPN1, and hepcidin) and nonheme iron in plaques of ApoE −/− mice. Ferritin, including H-and L-ferritin, reflects ironstorage levels. 23 We found that both H-and L-ferritin levels gradually

SDT reduces intracellular iron content of cultured ILMs by inducing Fpn1 expression
Murine peritoneal macrophages were identified by flow cytometry (Figure 4(a)). To investigate whether SDT directly affects iron metabolism in macrophages, ILMs were cultured and treated using optimized SDT parameters (see the schematic diagram in Figure 4  SDT attenuates iron overload in ILMs by activating ROS-Nrf2-FPN1 signaling SDT exerts its biological effects through ROS production. 24 As shown in Figure 6(a), the production of ROS in plaque increased instantly  Figure S7(a), and the Nfe2l2-binding motif is shown in Figure S7 Nrf2 is an ROS-activated transcription factor that interacts with the antioxidant response element (ARE) and plays a protective role during the anti-oxidation processes. 25 In the present study, SDT increased nuclear accumulation of Nrf2 in ILMs, which was effectively reversed by NAC pretreatment (Figure 6  To investigate the effect of SDT on hemorrhagic plaque, a rabbit IPH model was used. Although rabbit abdominal aortic IPH model was reported by delivering autologous erythrocytes into the plaques, 26 given the negative effect of intestinal gas on ultrasonic irradiation, this model is not suitable for SDT treatment. Therefore, in this study, we established a rabbit femoral artery IPH model. Glycophorin A in erythrocyte membranes scattered within the plaque confirmed successful model establishment. With the rabbit IPH model, we revealed that SDT exerts anti-atherosclerotic effects on hemorrhagic plaques, which is accompanied by the intraplaque iron retention reduction. Iron-mediated oxidative stress caused by IPH is a potent potentiator of plaque progression. 5 Our previous study indicated that SDT inhibits early stage atherosclerosis progression and upregulates heme oxygenase-1 (HO-1) levels. 17 HO-1, which is regulated by Nrf2, is a protective factor against atherogenesis but can still catabolize heme into free ferric ions. 25 However, in the murine model, we observed significant reduction of iron retention in advanced hemorrhagic plaque F I G U R E 6 Sonodynamic therapy (SDT) alleviates iron overload in iron-loaded macrophages (ILMs) by activating reactive oxygen species (ROS)-Nrf2-FPN1 signaling. (a) DCF fluorescence shows that SDT induced ROS generation in mouse plaques immediately (n = 4/group). Blue: nucleus; green: ROS. Scale bar, 25 μm. (b) Real-time monitoring shows that SDT promoted ROS generation in ILMs during ultrasonic irradiation.
(c) DCF fluorescence shows that SDT decreased ferric ammonium citrate (FAC)-induced increases in intracellular ROS levels at 24 h, with this activity inhibited by Fpn1 siRNA (siFPN1) (n = 5/group). (d) Western blot analysis and relative quantification show that ROS scavenger 4-hydroxy-TEMPO (Tempol) inhibited SDT-induced increases in FPN1 levels in mouse plaques (n = 3/group). (e) Calcein fluorescence sshows that ROS scavenger N-acetylcysteine (NAC) inhibited SDT-induced reductions in the intracellular labile iron pool (LIP) in ILMs (n = 5/group). (f) Real-time PCR analysis shows that SDT increased Fpn1 mRNA levels in ILMs, with this activity inhibited by NAC (n = 3/group). Fpn1 mRNA levels were quantified relative to Gapdh mRNA. (g) Western blot analysis and relative quantification show that NAC inhibited SDT-induced increases in FPN1 levels in ILMs (n = 3/group). (h) SDT-induced nuclear accumulation of Nrf2 observed by laser scanning confocal microscopy. Blue: nucleus; red: Nrf2. Scale bar, 20 μm. (i) Western blot analysis and relative quantification show increases in Nrf2 protein in the nucleus of ILMs after SDT, with this activity reversed by NAC (n = 3/group). (j) Real-time PCR analysis shows that Nrf2 siRNA (siNrf2) inhibited increases in SDT-induced Fpn1 mRNA levels in ILMs (n = 3/group). Fpn1 mRNA levels were quantified relative to Gapdh mRNA. (k) Western blot analysis and relative quantification show that siNrf2 inhibited SDT-induced increases in FPN1 levels in ILMs (n = 3/group). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 F I G U R E 7 A proposed theoretical model of sonodynamic therapy (SDT) stabilization of plaque with intraplaque hemorrhage (IPH). DVDMS accumulates in iron-loaded macrophages in hemorrhagic plaque. During ultrasonic irradiation, reactive oxygen species (ROS)-Nrf2-FPN1 signaling is active due to increases in intracellular ROS levels. This process promotes iron efflux in macrophages and plaque, resulting in inhibition of foamcell transformation, as well as alleviation of inflammatory responses but not in the media after SDT, which was reversed by hepcidin.
These results suggest that the predominant impact of SDT was due to upregulated Fpn1 expression in cells within the plaque, especially in the abundant macrophages recruited in response to IPH, which are the target cells of SDT. Moreover, this process is not accompanied by serum iron or serum hepcidin levels changing, suggesting that SDT modulated iron metabolism in a more tissue-specific manner rather than systemically, which will definitely decrease off-target effects and systemic side effects.
Mechanistically, SDT reduced iron overload via ROS-Nrf2-FPN1 signaling. ROS levels are temporarily increased during SDT, which can deplete cytosolic thiol and thus induce Kelch-like ECH-associated pro-  11 and increases pro-inflammatory cytokine secretion. 28 In addition, it is also suggested that the temporary explosive increase of ROS may activate the antioxidant stress pathway, thus inhibiting the oxidative stress caused by iron overload in macrophages.
Depending on whether SDT increases macrophage apoptosis, it can be divided into the following types: low-intensity SDT 15 and very low-intensity SDT, also known as nonlethal SDT (NL-SDT). 17 Although low-intensity ultrasound based on our previous study was used to treat rabbit hemorrhagic plaque, 22 no significant increase in the rate of macrophage apoptosis was observed in rabbit plaques following SDT treatment in this study ( Figure S8(a) and (b)). After reviewing the rabbit femoral artery ultrasound images, we found that the subcutaneous tissue above the artery in the rabbit model of this study was thicker than that observed in the early plaque model ( Figure S8(c) and (d)). The thickened subcutaneous tissue, which was due to the repeated surgical procedures in this study, could aggravate the attenuation of ultrasonic intensity. Therefore, NL-SDT parameters were applied to murine vulnerable-plaque model prone to IPH according to previous study. 17 Furthermore, in vitro study, the ultrasonic intensity of SDT (0.2-0.3 W/cm 2 ) resulting in elevated FPN1 levels was lower than that previously used to enhance cell apoptosis (0.5 W/cm 2 ). 16 Interestingly, SDT with 0.5 W/cm 2 ultrasonic intensity did not increase FPN1 levels. These results indicated that reduced iron retention in hemorrhagic plaque and the antiatherosclerotic effects were manifested by NL-SDT and different signaling pathways were activated in the presence of variable ultrasonic intensity. In addition, this low-energy SDT may avoid tissue damage caused by excessive ultrasound intensity, thus promoting its future clinical application.
A previous study showed that atherosclerotic carotid plaques obtained from men had a higher prevalence of IPH compared with those obtained from women, 29 and physiological iron loss before menopause in women has been proposed to be a cardiovascular protective factor. 30 However, the incidence of carotid IPH and plaque morphology in women with increasing age postmenopause becomes closer to that of men. 31,32 In this study, the animal model was established using only males, whereas the effect of SDT on hemorrhagic plaque in female animals was not verified. This is a limitation of this study. The effect of SDT on atherosclerotic plaque of individuals with different gender and different physiological periods needs to be further explored and verified in clinical research.
Our recently published clinical study (NCT03457662) showed that SDT rapidly reduced plaque inflammation and improved walking performance among patients with symptomatic peripheral artery disease. 33 This clinical study mainly focused on the treatment of patients with hypoechoic femoral artery plaque that usually tended to be lipid plaque. Although the clinical study is encouraging to promote the clinical practice of SDT on atherosclerosis treatment, the safety and efficacy of SDT have not yet been confirmed for the treatment of more advanced plaques such as hemorrhagic plaque. The promising results presented herein provided a more convincing rationale to conduct another clinical trial (NCT03871725) for assessing the efficacy and safety of SDT in patients with carotid hemorrhagic plaque, which is a more vulnerable condition. We expect that such studies will promote the rapid development of new approaches to treat vulnerable plaques without systemic side effects.

CONFLICT OF INTEREST
The authors declare no conflicts of interest. writing-review and editing.