Effects of Salvianolic Acid B/ginsenoside Rg1 Combination Against Chronic Pulmonary Embolism Induced by Polystyrene Microspheres

Pulmonary embolism (PE) is the most life-threatening complication of venous thromboembolism, but few effective treatments have been discovered to attenuate chronic PE currently. In this study, we investigated the protective effects of salvianolic acid B (SalB) and ginsenoside Rg1 (Rg1) combination (SalB/Rg1) on chronic PE and explored the potential mechanisms. The PE model was induced by 45 ìm polystyrene microspheres and 20 mg/kg of SalB/Rg1 was administered to PE rats intraperitoneally. A histopathological analysis of the lungs and heart was performed through hematoxylin and eosin staining and immunohistochemical analysis. The pulmonary index and right ventricular cardiomyocyte cross-sectional area were evaluated. SalB/Rg1 markedly downregulated pulmonary index, attenuated pulmonary interstitial changes, suppressed neutrophil infiltration, prevented collagen deposition, and inhibited MMP-9 activities in the lung. We also found that SalB/Rg1 improved right ventricular hypertrophy accompanied by reducing the cardiomyocyte cross-sectional area. These data suggest that SalB/Rg1 played a protective role against microsphere-induced PE and holds a high potential for the treatment of PE in the future.

Pulmonary embolism (PE) is a disease in which various emboli enter the pulmonary artery and prevent blood from flowing to the lungs. The annual morbidity of PE has revealed a rising tendency over time (WENDELBOE & RASKOB 2016;KONSTANTINIDES et al. 2019). PE is the third most common cardiovascular cause of death after myocardial infarction and stroke in China. PE patients still face the huge challenge of the deficiency of therapeutic medicines in clinic.
Chronic PE, in which residual clots in the pulmonary arteries continue to exist after acute PE, leads to pulmonary artery hypertension (BRESSER et al. 2004;NETO-NEVES et al. 2017). The current treatment for PE is focused on removing the mechanical obstruction through thrombolysis and anticoagulation. In certain cases, a surgical method, embolectomy, is also recommended (TOBA et al. 2008). The drug of thrombolysis includes recombinant tissue plasminogen activator (recombinant tPA), direct thrombin inhibitors, and factor Xa inhibitors (BARCO et al. 2020). Anticoagulation drugs include low-molecular-weight heparin and warfarin (YANG et al. 2020). These medical treatments are always used as a first aid and have been proved to improve survival. Symptoms of chronic PE included lung interstitial pathological changes, pulmonary inflammation, pulmonary fibrosis, and right ventricular (RV) hypertrophy.
Salvianolic acid B (SalB) and ginsenoside Rg1 (Rg1) are the representative active ingredients of Salviae miltiorrhizae and Panax notoginseng, respectively. Previous studies demonstrated the powerful effects of Rg1 against inflammation and fibrosis through various animal models (BAO et al. 2015). SalB was reported to hold antioxidant effects and prevent myocardial ischemia (QIAO & XU 2016). As the combination of SalB with Rg1, our previous studies have demonstrated that SalB/Rg1 held significant effects on cardio-protection with high safety (ZHAO et al. 2015;DENG et al. 2015a;DENG et al. 2015b). In the present study, we evaluate the effects of SalB/Rg1 on chronic PE.

Materials and Methods
Materials 56 healthy Wistar male rats (200-220 g) were purchased from the Shanghai Center of Experimental Animals, Chinese Academy of Sciences. All studies were approved by the Animal Care and Use Committee at the Shanghai Institute of Materia Medica (IACUC number: 2019-08-GDA-64). Polystyrene microspheres were purchased from Beijing Zhongke Thunder Co., Ltd.. Zoletil was purchased from France Vic Co., Ltd. Hematoxylin and eosin (HE) staining solution were purchased from Shanghai YiFan Biological Technology Co., Ltd.. CD68 antibody, CD44 antibody, MMP-9 antibody, goat anti-rabbit antibody (second antibody), and DAB kits were purchased from Boster Biological Technology Co., Ltd. A Masson detection kit was purchased from Nanjing Jiancheng Bioengineering Institute. Salvianolic acid B and ginsenoside Rg1 were purchased from Shanghai Yousi Biotechnology Co., Ltd.. The purity of the SalB and Rg1 was more than 98% detected by HPLC (Fig. 1S, Supplementary Material). Unless otherwise noted, reagents were purchased from Sinopharm Chemical Reagent Co., Ltd.

Establishment of PE model and SalB/Rg1 treatment
To confirm the establishment of the PE model by 45 µm microspheres successfully, rats were randomly divided into 2 groups: pulmonary embolism group (PE, n=10) and Control group (Control, n=10). Rats of from the PE group were injected with one million polystyrene microspheres/kg intravenously once a week for three consecutive weeks. Rats in the Control group were given the same volume of saline at the same time.
After we confirmed that 45 µm microspheres were suitable to induce PE, we evaluated the effects of SalB/Rg1 against PE. The rats were randomly divided into 3 groups: Control group (Control, n=10), pulmonary embolism group (PE, n=10), and SalB/Rg1 group (SalB/Rg1, n=7). Rats from the PE and SalB/Rg1 groups were injected with one million polystyrene microspheres/kg intravenously once a week for four consecutive weeks to induce PE, while the rats in control group were injected with saline at the same time. Then, rats from the Control and PE groups were administered an intraperitoneal injection of saline for four weeks, while rats from the SalB/Rg1 group were administrated an intraperitoneal injection of 20 mg/kg of SalB/Rg1 for four weeks after the second injection of microspheres.

Heart rate (HR) measurement
Rats were weighed and anesthetized with 30 mg/kg of zoletil then fixed on the experimental table. The electrodes were placed subcutaneously into the limbs. According to the connection mode of left upper limb red, left lower limb green, and right lower limb black. The HR of the rats was measured and analyzed using PowerLab software (AD Instruments, Australia).

Organ index and ventricular hypertrophy index assessment
After measurement of HR, the rats were euthanized with a lethal dose of 200 mg/kg of pentobarbital. The organs were weighed and tibial length was measured. The left or right pulmonary index was calculated as the ratio of left or right pulmonary weight to body weight. The myocardial hypertrophy index was calculated by the ratio of heart weight to tibial length (HW/TL). The ratio of right heart weight to left heart plus septum weight (RH/(LH+S)) were also assessed for ventricular hypertrophy.

Hematoxylin and eosin (H&E) staining
The fresh lung samples and right ventricle samples were fixed in 10% neutral-buffered paraformaldehyde for 48 hours and then were paraffin-embedded. 5 mm slices were cut for histopathological examination. After the slices were dewaxed and hydrated, they were stained with hematoxylin solution for 15 mins, differentiated with 1% alcohol hydrochloric acid for 2-3 seconds, then rinsed in distilled water. Next the slices were stained with eosin solution for 5 mins followed by dehydration with graded alcohol and clearing in xylene. Photomicrographs were taken using an Olympus BX51 microscope plus an Olympus DP71 CCD camera (Olympus, Tokyo, Japan).
The quantification of microspheres was based on H&E staining. There were a large number of smooth spherical vacuoles in the lung samples of the PE rats. The spherical vacuoles were identified as microspheres based on their diameter of about 45 µm. Eight visual fields were randomly obtained, and microspheres were counted according to the 8 fields. The average number of microspheres in the 8 fields was calculated for each rat.
The lung interstitium was stained purple in the H&E staining. The purple interstitial area of lung was quantified by Image-Pro Plus 6.0 software. The mean value of 8 fields per sample was calculated for every rat. The quantification of cardiomyocyte cross-sectional area (CCSA) was based on H&E staining. The 3 fields of per right ventricle sample were captured and 10 CCSA were quantified for per fields using Image-Pro Plus 6.0 software. The mean of CCSA in 3 fields was calculated for every rat.

Masson staining
Deparaffinized lung slices were stained with nuclear dyeing liquid for 1 min, washed with flushing fluid for 30 seconds, stained with plasma dyeing liquid for 1 min, washed with flushing fluid for 30 seconds, stained with color separation fluid for about 8 mins, drained and stained with counter dyeing liquid for about 5 mins, and washed with absolute alcohol. Photomicrographs were taken using an Olympus BX51 microscope plus an Olympus DP71 CCD camera (Olympus, Tokyo, Japan).

Immunohistochemical staining
Antigen was retrieved for paraffin-embedded slices in sodium citrate buffer (pH 6.0) for 20 mins using a microwave oven. After incubating with 3% H 2 O 2 for 30 mins, the slices were rinsed with water and incubated with the primary antibody (CD68, CD44, or MMP-9 antibody diluted 1:100 in 1% PBS) overnight at 4 o C. Then, the slices were rinsed and incubated with the goat anti-rabbit antibody for 1 h at 37 o C, then stained by DAB kit and hematoxylin. Photomicrographs were taken using an Olympus BX51 microscope plus an Olympus DP71 CCD camera (Olympus, Tokyo, Japan). The quantification of the positive staining area and integrated optical density (IOD) were based on Image-Pro Plus 6.0 software. Five images from each rat were analyzed and IOD/positive area was calculated to express the mean staining intensity.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 5.0 (GraphPad software, LA Jolla, CA, USA). Multiple-group comparisons were made using a oneway ANOVA test and data were performed with Bonferroni correction. A Student's t-test (two-tailed, paired) was used for two-group comparisons. All data were expressed as mean ± standard error, p<0.05 was statistically significant.

Results
First the PE model induced by 45 µm microspheres was set up. As shown in Fig. 1a, the right pulmonary index (RPI) of the PE group was higher in comparison with the Control group (3.64±0.47 vs. 2.34±0.40 mg/g, p<0.001), while the left pulmonary index (LPI) showed an increasing tendency in the PE group (Fig. 1b). Because the increase of RPI was more obvious than LPI, the histochemical observation of right lung tissues based on H&E staining was conducted, and the different appearance between alveoli and microspheres was in evidence. The shapes of alveoli were irregular, while the microspheres were round and smooth (Fig.1c). The number of microspheres (25.77±4.94 per field) was counted based on being round, smooth, and about 45 µm in size (Fig. 1d). RV hypertrophy is another important parameter during the development and progression of PE, and an increased tendency of RH/(LH+S) and HW/TL in the PE group was observed compared with the Control group (Fig. 1e,f). The representative pictures of H&E stain for RV were shown in Fig. 1g. The CCSA of RV was quantified per field (Fig. 1h), and the data of CCSA in the PE group was significantly higher than in the Control group (0.02±0.00 vs. 0.01±0.00 mm 2 , p<0.01).
After we set up the PE model successfully, the protective effects of SalB/Rg1 on the lungs were evaluated according the experimental scheme shown in Fig. 2. As shown in Fig. 3a and Fig. 3c, lung inflammation, thickened alveolar walls, increased lung interstitium, and decreased alveolar space were observed in the PE group compared to the Control group, and SalB/Rg1 ameliorated the interstitial structure of the lung significantly. Fig. 3d showed an increase of lung interstitial area in the PE group (0.91±0.05 vs. 0.59±0.05 mm 2 , p<0.001), and this increase was significantly reversed in the SalB/Rg1 group (0.66±0.17 vs. 0.91±0.05 mm 2 , p<0.001).
After H&E staining, inflammatory cell infiltration was also observed in the lung interstitium. The types of inflammatory cells involved and the efficiency of SalB/Rg1 were illustrated following. As two important inflammatory cells, macrophages and neutrophils were investigated through immunohistochemical staining. CD68 positive cells (macrophages) in the PE group were significantly increased compared to the Control group (0.20±0.04 vs. 0.01±0.01, p<0.001), while no regulation of SalB/Rg1 was found in comparison with the PE group ( Fig. 4a and 4b). The immunohistochemical stain for CD44 positive cells immunohistochemistry staining at a low magnification (scale bar: 100 µm) and the images below at a high magnification (scale bar: 50 µm). The arrows represent CD68 positive cells. b. The mean staining intensity was assessed for CD68 immunohistochemistry staining. c. The above images show representative CD44 immunohistochemistry staining at low magnification (scale bar: 100 µm) and the images below at a high magnification (scale bar: 50 µm). The arrows represent CD44 positive cells. d. The mean staining intensity was assessed for CD44 immunohistochemistry staining. All the values are expressed as mean ± SE. **p<0.01, ***p<0.001 versus Control; ### p<0.001 versus PE. n=10 in the Control and PE groups. n=7 in the SalB/Rg1 group.
Because lung inflammation affected the elasticity of alveoli and lead to pulmonary fibrosis, the impacts of SalB/Rg1 on pulmonary fibrosis were further evaluated. First, Masson staining was conducted to evaluate pulmonary fibrosis as shown in Fig. 5a. Blue collagen fibers were profoundly increased in the PE group compared with the Control group, and SalB/Rg1 markedly ameliorated the deposition of blue collagen fibers. Then to detect whether pulmonary fibrosis was related to matrix metalloproteinase-9 (MMP-9), an immunohistochemical stain for MMP-9 was conducted. As shown in Fig. 5b, MMP-9 positive cells were significantly obvious in the PE group compared to the Control group. SalB/Rg1 treatment down-regulated MMP-9 positive cells in comparison with the PE group significantly. After observation of the efficiency of SalB/Rg1 on the lung, the protective effect of SalB/Rg1 on the right heart was further investigated. PE resulted in an increase in RV cardiomyocyte size and RV hypertrophy. The representative photomicrographs of H&E staining for RV were shown in Fig. 6a. Fig. 6b showed that CCSA was significantly increased in the PE group in comparison to the Control group (0.02±0.00 vs. 0.01±0.00 mm 2 , p<0.001). And SalB/Rg1 significantly down-regulated CCSA compared to the PE group (0.01±0.00 vs. 0.02±0.00 mm 2 , p<0.05). Besides, RH/(LH+S) as an index of ventricular hypertrophy in the PE group was higher than that in the Control group. SalB/Rg1 also caused a decreasing tendency in RH/(LH+S) compared with the PE group in Fig. 6c.

Discussion
In the present study, PE was induced by 45 µm microspheres and characterized by increased LPI and RPI, enlarged lung interstitial area, aggravated in-flammatory cell infiltration, and fibrosis on the lung, along with cardiac hypertrophy on the heart. SalB/Rg1, a combination of SalB and Rg1, exerted protective effects not only on the lung, but also on the heart in PE rats.
Multiple mechanisms have been proposed for PE, among them inflammation is a crucial player. Neutrophils are classic inflammatory cells, and their levels are tightly associated with the extent of inflammation (TOBA et al. 2016). Neutrophils infiltrate the venous thrombus early and trigger a cascade of inflammatory reactions that can destroy surrounding tissues, facilitate micro thrombosis, and result in PE (PUJHARI et al. 2020). In our research, SalB/Rg1 attenuated the infiltration of neutrophils and suppressed pulmonary interstitial changes.
Matrix metalloproteinases (MMPs) have been viewed as a promising target in the search for cardiovascular disease applications (BAGGIO et al. 2020). Previous studies in our laboratory have shown that SalB functioned as a competitive inhibitor of MMP-9 and efficiently prevented cardiac remodeling, which was consistent with the present finding in PE (JIANG et al. 2010). The effect of MMP-9 deficiency was elucidated on deep venous thrombosis. MMP-9 modulated midterm vein wall collagen content, with a local inflammatory and pro-fibrotic environment (MUKHOPADHYAY et al. 2019). MMP-12 was also a valid target for pharmacological intervention, its inhibitor has significant therapeutic potential for the chronic obstructive pulmonary disease (MAGNUSSEN et al. 2011). In the current study, the protective effects of SalB/Rg1 on pulmonary inflammation and fibrosis may be correlated to its inhibition of MMP-9 activity. The inhibition of SalB/Rg1 on other MMPs remains to be further investigated. Chronic increase in pulmonary vascular resistance translates into elevated RV afterload during the development and progression of PE. To cope with the increased afterload, RV undergoes structural and functional changes, which ultimately causes RV dilatation and failure (STAM et al. 2019). RV hypertrophy is an abnormal enlargement or pathologic increase in muscle mass of the RV in response to pressure overload, most commonly due to severe lung disease. PE is a major source of morbidity and mortality, patient outcome depends on how well the RV can sustain the increased afterload caused by the embolic burden. In the present study, RV hypertrophy was attenuated by SalB/Rg1. This finding is consistent with our previous report that SalB/Rg1 prevented or delayed the progression of cardiac remodeling, ameliorated cardiac structure and function.
In summary, we have shown that SalB/Rg1 significantly improved PE induced by microspheres for the first time. SalB/Rg1 not only ameliorated the pulmonary changes of early PE, but also held the prevention on RV hypertrophy of post-PE, indicating the potential of SalB/Rg1 for further development.