The Carthamus tinctorius L. and Lepidium apetalum Willd. Drug Pair Inhibits EndMT through the TGFβ1/Snail Signaling Pathway in the Treatment of Myocardial Fibrosis

Background Myocardial fibrosis (MF) is an essential pathological factor for heart failure. Previous studies have shown that the combination of Carthamus tinctorius L. and Lepidium apetalum Willd. (C-L), two types of Chinese herbal medicine, can ameliorate MF after myocardial infarction (MI) in rats and inhibit the activation of myocardial fibroblasts. However, the mechanism of C-L in the treatment of MF remains unclear. Methods A rat model of MF with left anterior descending coronary ligation-induced MI was first established. Then, the effects of C-L on cardiac function, MF, and endothelial-to-mesenchymal transition (EndMT) were evaluated by the left ventricular ejection fraction (LVEF), serum N-terminal pro-brain natriuretic peptide (NT-proBNP) levels, Masson's trichrome staining, and immunohistochemical and immunofluorescence staining. Next, a hypoxia-induced cardiac microvascular endothelial cell (CMEC) model was established to observe the effects of C-L on EndMT. The supernatant of CMECs was collected and used to culture cardiac fibroblasts (CFs) and observe the effects of CMEC paracrine factors on CFs. Results Animal experiments indicated that C-L improves the cardiac function of rats after MI, inhibits the progression of EndMT and MF, and downregulates TGFβ1, Snail, and CTGF expression. Cell experiments showed that drug-loaded serum containing C-L inhibits the EndMT of CMECs under hypoxic conditions. The culture supernatant of CMECs grown under hypoxic conditions significantly activated CFs. After treatment with C-L, the activating factor for CFs in hypoxic CMEC culture supernatant was substantially downregulated, and the effect of the culture supernatant on CF activation was also reduced. However, TGFβ1 agonists inhibited the effects of C-L on CMECs and CFs. Conclusion Our data demonstrated that by regulating the TGFβ1/Snail pathway, C-L inhibits EndMT of CMECs and reduces the release of CF-activating factors in cells undergoing EndMT.


Introduction
Heart failure (HF), a life-threatening condition, currently afects approximately 64 million people worldwide [1]. After a pathological myocardial injury, collagen fbers will replace the defective myocardium and exacerbate HF. Tus, inhibition or reversal of myocardial fbrosis (MF) has become an important strategy for treating HF [2,3]. After pathological stimulation, fbroblasts transform into myofbroblasts, the major source of the extracellular matrix during MF, as these cells show signifcantly increased collagen synthesis [4]. Fibroblasts have been studied to ameliorate MF, but increasing evidence has shown that nonfbroblasts in the heart also play an important role in MF [5]. Under pathological conditions, nonfbroblasts, including cardiomyocytes and infammatory cells, afect fbroblast activity and function by secreting factors that induce fbroblast activation [6,7].
We previously reviewed the efects of cardiac microvascular endothelial cells (CMECs) on HF and concluded that CMECs also contribute to the progression of MF [8]. Under certain conditions (hypoxia, oxidative stress, infammation, abnormal fuid shear stress, etc.), the phenotypic transformation of endothelial cells involves the loss of endothelial-specifc markers, including platelet EC adhesion molecule-1 (CD31), Tie1, Tie2, VE-cadherin, and von Willebrand factor (VWF), as well as the acquisition of mesenchymal markers, including alpha smooth muscle actin (α-SMA), N-cadherin, and vimentin. Tis process is known as an endothelial-to-mesenchymal transition (EndMT) [9]. During this process, endothelial cells peel away and separate from the ordered endodermis, migrate into the parenchyma and lose their apical-basal polarity, becoming an essential part of the fbroblast pool during MF [10]. Studies have also shown that upregulation of the zinc fnger protein SNAI1 (Snail) induced by transforming growth factor beta (TGFβ) plays a crucial role in promoting EndMT in human umbilical vein endothelial cells (HUVECs) [11]. Upregulation of Snail not only promotes EndMT to generate a large number of fbroblasts but also induces fbroblasts to transition into myofbroblasts through connective tissue growth factor (CTGF) secreted by cells undergoing EndMT, which accelerates the development of MF [12]. However, the abovementioned studies were limited to HUVECs, and their applicability to CMECs is currently unknown. In addition, whether the TGFβ1/Snail signaling pathway inhibits the activation of cardiac fbroblasts (CFs) by regulating the paracrine process of CMECs after EndMT has not been clarifed.
Herbs and their main components are widely used in disease treatment worldwide [13]. Recently, many studies have confrmed the potential value of herbal medicine [14][15][16]. Traditional Chinese medicine (TCM) has been used to treat HF for thousands of years. In TCM theory, HF belongs to the "cardiac edema" category, which was frst noted in the Eastern Han dynasty in Zhang Zhongjing's Synopsis of Golden Chamber (Jin Gui Yao Lue). Dr. Zhang noted that the pathological basis of cardiac edema is "blood vessel stasis and obstruction caused by water pathogens attacking the heart." As a result, the TCM treatment of HF should "activate blood" and "promote urination." Carthamus tinctorius L. (Asteraceae, Chinese Pinyin: Hong Hua), which functions as a blood activator, was frst reported in the Synopsis of Golden Chamber and is often used by TCM doctors to promote blood circulation and expel blood stasis. Lepidium apetalum Willd. (Cruciferae, Chinese Pinyin: Ting Li Zi) is believed to promote diuresis and resolve edema, which was recorded in the Compendium of Materia Medica (Ben Cao Gang Mu) of the Ming dynasty. Both C. tinctorius L. and L. apetalum Willd., are used to treat HF, especially in combination [7,17]. In TCM theory, using two herbs in combination for the same therapeutic purpose is referred to as a "drug pair." However, the pharmacodynamic properties and critical targets of the C. tinctorius L. and L. apetalum Willd. drug pair (C-L) remain unclear.
Here, we studied the regulatory efect of C-L on EndMT of CMECs for the frst time and revealed the efect of profbrotic factors secreted by cells undergoing EndMT on CF activation.

Ultrahigh Performance Liquid Chromatography (UHPLC) Coupled with a Quadrupole-Orbitrap Mass Spectrometer.
A total of 100 μL of 0.5 g/mL C-L solution (see "2.8 preparation of the herbal solution" for the preparation method) was diluted with water ten times [high-performance liquid chromatography (HPLC) grade]. Next, 200 μL of diluent and 10 μL of internal standard (L-2-chlorophenylalanine, 0.06 mg/mL; prepared by methanol) were added and fltered with a 0.22-µm membrane. Ten, the sample was centrifuged at 8, 000 × g for 5 min, and the supernatant was subsequently extracted for analysis. Te component spectrum was analyzed using the Dionex Ultimate 3000 RS UHPLC) system together with a Q-Exactive plus quadrupole-Orbitrap mass spectrometer equipped with a heated electrospray ionization (ESI) source (Termo Fisher Scientifc, Waltham, MA, USA) in the ESI-positive and ESI-negative ion patterns. Te Acquity UPLC HSS T3 column also includes positive and negative modes. Finally, the separation was completed based on a binary gradient elution system including water and acetonitrile according to the following protocol: 0 min, 5% acetonitrile; 2 min, 5% acetonitrile; 4 min, 25% acetonitrile; 8 min, 50% acetonitrile; 10 min, 80% acetonitrile;  14 min, 100% acetonitrile; 15 min, 100% acetonitrile; 15.1 min, 5% acetonitrile; and 16 min, 5% acetonitrile. Te column's fow rate and temperature were set to 0.35 mL/min and 45°C, respectively. During the analysis, all operations were performed at 4°C. C-L components were identifed based on the mass spectrometric data of the standard substance by reviewing the references.

2.4.
Animal and Postinfarction HF Models. Sprague-Dawley (SD) male rats (specifc-pathogen free) weighing 190-210 g were purchased from Beijing Charles River Laboratories (Beijing, China, Certifcate No. 2021-0006). Te care and use of laboratory animals in this study followed the Guidelines for the Care and Use of Laboratory Animals (National Institutes of Health publication, 8 th edition, 2011). All operation procedures were reviewed and approved by the Animal Ethics Committee of the Afliated Hospital of Shandong University of Traditional Chinese Medicine (Permit number: 2021-33). All rats were raised in a room with constant temperature, humidity control, and a 12-h light/dark cycle. Standard food and water were also provided.
Te rats were randomly divided into 3 groups: a control group (n � 10), a sham operation group (n � 10), and a myocardial infarction (MI) model group. Te rats in the model group underwent left anterior descending branch (LAD) ligation, as documented in the references, to construct the MI model [18]. Specifcally, SD rats were anesthetized and ventilated by a ventilator (Harvard Apparatus). After the chest was opened and the heart was exposed, the LAD was ligated with a 7.0 surgical suture proximal to its main branching point. Changes in the electrocardiogram's ST segment were used to assess the presence of MI. Te rats in the sham operation group (sham) underwent a similar procedure without actual ligation in the LAD. One week after surgery, the rats' cardiac function was assessed with a Vetus 7 (Mindray, China) system equipped with a P10-4s and 14-18 MHz imaging transducer. Rats meeting the criteria of cardiac insufciency [left ventricular ejection fraction (LVEF) < 50%] were included in the study (n � 40) and randomly divided into 4 groups: the HF model, low-dose, medium-dose, and high-dose groups. Ten, the rats in the high-, medium-and low-dose C-L groups were treated with low, medium, and high C-L doses, respectively, by intragastric administration. Based on the human and animal surface area of the equivalent dose conversion ratio table, the C-L doses in the low, medium, and high doses of C-L groups were 4.8, 2.4, and 1.2 g/kg/day, respectively [19]. Te rats in the other groups received an equal volume of 0.9% sodium chloride solution. After four weeks of treatment, the rats were anesthetized with 2% pentobarbital sodium to obtain blood from the abdominal aorta in each group, and the left ventricular myocardium was also collected for the study.
2.5. Serum NT-proBNP. Blood samples were collected from the abdominal aorta and allowed to stand for 2 h. Te serum was collected after centrifugation. A biotin double antibody sandwich ELISA was used to determine the concentration of NT-proBNP in the serum. Te absorbance was measured at 450 nm using a Multiskan GO1510 microplate reader (Termo Fisher Scientifc, USA).

Echocardiographic Evaluation.
Echocardiographic measurements were performed at the end of the study period. Left ventricular function was assessed by LVEF and ventricular wall motion. All measurements were performed by an investigator blinded to the experimental groups.

Masson's Trichrome Staining and Immunohistochemical
Analysis. Te left ventricular tissues were dehydrated and embedded in parafn, sliced into 4 μm sections, and mounted on glass slides. Ten, the slices underwent dewaxing and hydration, hematoxylin staining, Ponceau S staining, phosphomolybdic acid hydrate staining, aniline blue staining, dehydration, and sectioning. Te sections were observed under a light microscope.

Preparation of Rat Drug-Loaded Serum Containing C-L.
Twenty male SD rats were used to prepare drug-loaded serum containing C-L. In short, the rats were intragastrically administered a moderate dose of C-L solution (2.4 g/ kg/day). On the 7 th day, blood was collected from the abdominal aorta in a sterile environment 2 h after administration. Ten, the serum was separated and inactivated in a 55°C water bath for 15 min and stored at −20°C for future use. For the establishment of an MI model in vitro, a hypoxic cell model was used in our study. Before diferent treatments were performed, the CMECs, which were grown to approximately 80% confuency, were placed in a hypoxia incubation chamber with 95% N 2 and 5% CO 2 . Te incubation time was 48 h, and all experiments were conducted in triplicate. After a hypoxic intervention, the hypoxic cells Evidence-Based Complementary and Alternative Medicine 3 were randomly divided into 4 groups: the hypoxia group and the low-dose, medium-dose, and high-dose C-L groups. Te drug-loaded serum containing C-L (C-L DS) was added to the culture medium of the low-, medium-, and high-dose groups and diluted to low (2.5%), medium (5%), and high (10%) concentrations, respectively. Subsequently, to study the molecular mechanism by which C-L improves MF, we cultured CMECs again and divided them into 4 groups: the control, hypoxia, C-L treatment, and SRI-011381 groups. Te cells in the control group were cultured under conventional conditions, and the cells of the other groups were used to construct a hypoxia model for 48 h. Ten, the CMECs in the C-L treatment group were treated with 10% C-L DS. Te cells in the SRI-011381 group was additionally treated with 10 μM SRI-011381 [dissolved in dimethyl sulfoxide (DMSO) and then further attenuated with cell culture medium to the consistency required for in vitro experiments]. Te treatment duration was 24 h. After treatment, the culture supernatant from the CMECs in each group was collected and used to culture CFs for 24 h.

CF Viability Assay.
CF viability was examined with CCK-8 assays according to the manufacturer's protocols. Briefy, 100 µL of the cell suspension was seeded in a 96-well plate (5,000 cells per well), and then, the experimental treatment was performed after the cells reached confuence. Te culture medium, which contained CCK-8 reagent and 10% base solution, was placed in an incubator for 2 h, and the absorbance was measured at 450 nm.
2.14. Western Blotting Analysis. Myocardial tissues (50 mg) or cells (5 × 10 5 ) from each group were homogenized and lysed in RIPA lysis bufer, and the protein concentration was determined by a BCA kit. Normalized proteins were loaded on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and transferred onto a polyvinylidene fuoride (PVDF) membrane. Next, 5% skim milk was used to block the membrane, and the membrane was incubated with primary antibodies at 4°C overnight. Ten, the membrane was incubated with a secondary antibody. Western blot analysis was conducted by a FluorChem FC3 Gel Imager System (ProteinSimple, USA). Te signal density was quantifed by ImageJ (version 1.8.0).

Statistical Analysis and Diagram
Generation. All data were analyzed by GraphPad Prism 7 and are expressed as the mean ± standard deviation (SD). Te signifcance of diferences among three or more experimental groups was calculated by one-way analysis of variance. A value of p < 0.05 was considered statistically signifcant. Diagrams were generated using GraphPad Prism 7 and https://www. bioinformatics.com.cn/, a free online data analysis and visualization platform.  Table 1 of the Supplementary Materials.

C-L Improves Rat Cardiac Function after MI.
After 28 days of C-L treatment, cardiac function and serum NT-proBNP levels of the rats with MI were assessed, and the degree of left ventricular fbrosis was evaluated in the pathological sections. Te results indicated that the LVEF of the rats in the MI group was signifcantly reduced compared to that of the rats in the control group, and echocardiography revealed that the anterior wall of the heart (left ventricular wall) of the rats in the model group was signifcantly weakened (Figures 2(a)and 2(b)). Te serum NT-proBNP level of the model group rats signifcantly increased (Figure 2(e)). However, no signifcant diference was observed between the sham and control groups. For Masson staining, the myocardium of the rats in the control and sham operation groups was bright red without a prominent fbrotic area. In contrast, the myocardium of the rats in the model group had a large area of fbrotic lesions (blue area). Tese fndings indicated that a large amount of collagen fber synthesis occurred in the rat myocardium after MI (Figures 2(c) and 2(d)). However, the rats treated with C-L showed a signifcant improvement in LVEF, ventricular wall motion, NT-proBNP expression, and degree of MF. Te efect was most signifcant in the high-dose group (Figure 2(e)). Although the level of NT-proBNP in the lowdose group rats was not signifcantly diferent from that in the model group, the LVEF and MF area substantially improved.

C-L Protects the Myocardium and CMECs in the Rats with MI.
To determine the efect of C-L treatment on myocardial tissue, we used H&E staining to assess myocardial tissue damage. Te H&E staining results indicated that the myocardial tissue of the control and sham operation groups was orderly and that the cell morphology was normal. In contrast, the myocardial tissue of the rats with MI showed a large amount of scar tissue, and myocardial cells exhibited disordered arrangements. After C-L treatment, the scar tissue in the rat myocardium was signifcantly reduced, and the regularity and cell integrity of the myocardium was substantially improved (Figure 3(a)). CD31 is a specifc marker of endothelial cells, and vimentin is a marker of fbroblasts. In immunohistochemical staining, signifcant expression of CD31 on microvessels in the control and sham groups was noted, but vimentin staining was negative (Figures 3(b)-3(e)). However, CD31 expression was signifcantly decreased after MI, whereas vimentin expression was strongly positive. Tese results suggested that the EndMT process exists in myocardial microvascular endothelial cells after MI, but this process was signifcantly inhibited after C-L treatment.

C-L Inhibits EndMT in CMECs.
Recent studies have shown that EndMT-derived CFs play an essential role in the progression of MF [20]. VWF and α-SMA are specifc markers of endothelial cells and mesenchymal cells (mainly fbroblasts in the myocardium). Immunofuorescence staining was used to evaluate the EndMT process of CMECs. Te results indicated that VWF fuorescence (red) was strongly positive and that α-SMA fuorescence (green) was negative in the myocardial microvessels of the control group Evidence-Based Complementary and Alternative Medicine and the sham operation group (Figure 4(a)). However, the fuorescence of myocardial microvessels in the rats with MI showed the opposite pattern to that in the control group, and the fuorescence ratio of VWF and α-SMA was signifcantly lower than that of the control group (Figure 4). After the C-L treatment, the fuorescence ratio of VWF and α-SMA dramatically increased in the low-dose, medium-dose, and high-dose groups. Tese results suggested that the myocardium of the rats with MI undergoes signifcant EndMT. Tis process results in the pathological conversion of CMECs to CFs. Moreover, C-L signifcantly inhibited the EndMT of CMECs. Tese results are consistent with the detection of CD31 and α-SMA protein expression in rat myocardial tissue (Figure 5(a)).

C-L Inhibits EndMT and the Expression of Profbrotic
Factors. TGFβ1-mediated upregulation of Snail has been identifed as a critical factor in EndMT and promotes the progression of tissue fbrosis. In addition, Snail overexpression promotes the secretion of CTGF in cells undergoing EndMT and the transformation of fbroblasts into myofbroblasts. We assessed the expression of TGFβ1, Snail, CTGF, CD31, and α-SMA in the rat myocardium by western blotting. Te results indicated that TGFβ1, Snail, CTGF, and α-SMA expression was signifcantly upregulated in the rats with MI, whereas CD31 expression was inhibited (Figures 5(a)-5(f )). In addition, quantitative reverse transcription-polymerase chain reaction (qRT-PCR) indicated that TGFβ1, Snail, and CTGF were highly expressed at the transcriptional level (Figures 5(g)-5(i)). Te obtained results revealed high expression of profbrotic factors and a large area of endothelial cell injury in the myocardium of the rats after MI. However, in the myocardium of the rats treated with C-L, all the above-mentioned efects were inhibited, and the inhibitory efect correlated with the dose of C-L. Terefore, we hypothesized that the inhibitory efect of C-L on MF and EndMT is related to the regulation of TGFβ1, Snail, and CTGF expression.

C-L Inhibits EndMT in CMECs.
Hypoxia after MI is an essential inducer of MF [21]. We constructed a hypoxic model of CMECs to simulate the environment of myocardial tissue after MI and observed the efect of C-L on EndMT by CD31 and α-SMA immunofuorescence staining of CMECs. However, CD31 fuorescence in CMECs was weakened under hypoxic conditions, whereas positive α-SMA fuorescence was observed. Tese results indicated the existence of EndMT under hypoxic conditions. Notably, this hypoxiainduced change in CD31 and α-SMA expression was inhibited in the CMECs treated with drug-loaded serum containing C-L. In addition, we assessed the mRNA expression of type I collagen and type III collagen in CMECs by qRT-PCR. Te mRNA expression levels of type I collagen and type III collagen in CMECs signifcantly increased under hypoxic conditions but were signifcantly reduced in the CMECs treated with C-L compared with those in the model group (Figures 7(a) and 7(b)). Tese results suggested that hypoxia activates the EndMT of CMECs, which can transform CMECs into mesenchymal cells (mainly fbroblasts) and enhance collagen synthesis. However, C-L inhibits the EndMT process and collagen synthesis of CMECs.

C-L Inhibits EndMT in CMECs via the TGFβ1/Snail
Pathway. To confrm that C-L ameliorates EndMT by inhibiting the TGFβ1/Snail pathway, we added SRI-011381 (an activator of TGFβ1) [22] to the culture medium of hypoxic CMECs treated with the optimal therapeutic concentration (10%) of drug-loaded serum containing C-L. Immunofuorescence staining of CD31 and α-SMA in CMECs showed that enhancement of the TGFβ1/Snail pathway signifcantly inhibited the anti-EndMT efect of C-L on CMECs under hypoxic conditions (Figures 7(c)-7(e)). Tese results indicated that the C-L-mediated inhibition of EndMT in CMECs under hypoxic conditions is dependent on the inhibition of the TGFβ1/Snail signaling pathway. Previous studies have shown that the Smad signaling pathway is essential for EndMT [23]. Terefore, we also examined Smad2/3 expression. Smad2/3, TGFβ1, Snail, and CTGF protein expression levels were assessed by western blotting. TGFβ1, Snail, and CTGF mRNA expression levels were assessed by qRT-PCR. Te results indicated that C-L strongly inhibits the upregulated expression of Smad2/3, TGFβ1, Snail, and CTGF induced by hypoxia in CMECs. However, after the addition of a TGFβ1 activator, this efect of C-L was attenuated (Figures 8(a)-8(h)). Moreover, the high collagen mRNA expression in CMECs under hypoxic conditions was inhibited by C-L, and this efect was weakened after the upregulation of TGFβ1 (Figures 8(i) and 8(j)).

C-L Can Inhibit the Efect of CMECs on CF Activation.
During CF promotion of MF, collagen fber synthesis in the activated cells (myofbroblasts) was multiplied several times compared with that under normal conditions; consequently, myofbroblasts are the main efector cells of MF [24]. We collected the culture supernatant of each group of CMECs from the previous experiment, added them to the culture medium of CFs, and observed the efect on the proliferation and collagen synthesis of CFs (Figure 9(a)). Te proliferative activity of CFs was assessed using CCK-8 assays, and we found that the culture supernatant of CMECs under hypoxic conditions could signifcantly increase CF proliferation (Figure 9(b)). In addition, qRT-PCR indicated that the culture supernatant of the hypoxia-treated CMECs could substantially increase the transcription level of the CF collagen gene (Figures 9(c) and 9(d)). However, the culture supernatant of the CMECs in the control group did not exhibit these efects, indicating that CMECs under hypoxia could secrete some profbrotic factors into the culture medium, driving the activation of CFs. Nevertheless, this promotion of CF activation was signifcantly inhibited in the culture supernatant of the CMECs treated with C-L. Terefore, we hypothesized that C-L inhibits the activation of CFs by inhibiting the paracrine pathway of CMECs.  3.9. C-L Inhibition of Paracrine-Mediated CF Activation by the TGFβ1/Snail Pathway. We measured TGFβ1 and CTGF levels in the cell supernatant using ELISAs to evaluate the ability of CMECs to secrete profbrotic factors. Te results indicated that the levels of TGFβ1 and CTGF signifcantly increased in the supernatant of CMECs under hypoxic conditions and that C-L signifcantly inhibited the secretion of profbrotic factors by CMECs. Tis inhibitory efect was strongly weakened after activating the TGFβ1/Snail pathway (Figures 9(e) and 9(f )). Terefore, we believe that C-L inhibits the secretion of profbrotic factors by CMECs under hypoxic conditions by inhibiting the TGFβ1/Snail pathway. We also observed that the ability of CMECs to secrete profbrotic factors was regulated by TGFβ1/Snail and that C-L inhibited the activation of CFs mediated by the culture supernatant of hypoxia-treated CMECs. Terefore, the culture supernatant of CMECs in the SRI-011381 group and other groups was collected to treat CFs, and we investigated whether TGFβ1/Snail activation afected this inhibitory efect. We found that the culture supernatant of the hypoxiatreated CMECs stimulated the activation of CFs and that C-L inhibited this activation. However, the efect of C-L was signifcantly weakened after the enhancement of the TGFβ1/ Snail pathway (Figures 9(g)-9(i)). Tese results indicated that C-L inhibits the paracrine efect of fbroblast activation factors in CMECs by inhibiting the TGFβ1/Snail pathway under hypoxic conditions and ultimately inhibits CF activation.

Discussion
MF is a compensatory process induced by high levels of myocardial cell death after MI. Tis process primarily flls the areas of the ventricular wall destroyed by partial myocardial death [25,26]. However, the pathological scar tissue formed by MF exhibits poor relaxation and contraction abilities as well as poor compliance. Many fbrotic lesions promote the remodeling of the ventricular wall, enlarge the heart cavity, and seriously afect the systolic function of the ventricular wall, ultimately aggravating the progression of HF. Terefore, antifbrotic therapy is critical for improving cardiac insufciency. Currently, some anti-MF drugs have been applied in clinical practice. However, regardless of whether the drugs have been used clinically for a long time (spironolactone, lisinopril, etc.) or are undergoing assessment in preliminary clinical trials (pirfenidone, doxycycline, etc.), their efcacy in reducing the formation of myocardial fbers is limited [27,28]. Terefore, we need to further explore the pathological mechanism of MF and identify new and efective anti-MF strategies to achieve the clinical goal of efective control, even the reversal of HF.  Evidence-Based Complementary and Alternative Medicine profbrotic factors (TGFβ1, CTGF) and collagen (type I collagen and type III collagen) [29,30]. By assessing profbrogenic factors in infarcted myocardial tissue, we found that TGFβ1, CTGF, and TIMP1 expression in the myocardium of the rats with MI treated with the Luhong formula was substantially downregulated, and all three of these factors could promote fbroblast activation [31][32][33][34].
To further explore the anti-MF mechanism of the Luhong formula, we selected the C-L drug pair for pathological study. Te results indicated that C-L signifcantly inhibited the degree of MF both in mice after MI and in rats after angiotensin II (Ang II) treatment, and the mechanism was related to the inhibition of TGFβ1 expression. Notably, Ang II increased the proliferation of endothelial cells and CFs and promoted TGFβ1 secretion in both types of cells, but this efect was inhibited using C-L [35,36]. Previous studies have shown that Ang II enhances the proliferative activity of endothelial cells by promoting EndMT. In addition, CTGF secretion activates CFs, but this efect mainly depends on the autocrine release of CTGF by fbroblasts rather than other nonfbroblasts [33,37]. In this study, we demonstrated that C-L could signifcantly inhibit EndMT in CMECs from the rats with MI, reduce the extracellular secretion of profbrotic factors (TGFβ1 and CTGF) from EndMT-derived cells, and reduce the cellular activity and collagen synthesis of EndMT-derived cells and normal CFs. Tese results suggested that the mechanism by which C-L inhibits MF is closely related to EndMT and the paracrine pathways that regulate CMECs.
In addition to their role in protecting blood vessels, CMECs are closely related to MF. After stimulation with hypoxia or inducible factors, activated CMECs express adhesion factors that recruit infammatory cells in circulating blood and secrete cytokines, as well as promote the infltration of infammatory cells and MF [38]. In addition, the transformation of microvascular endothelial cells into fbroblasts and the main mesenchymal cells of the myocardium are important pathological mechanisms of MF [39]. Te transcriptional repressor Snail is a member of the zinc fnger protein family and mainly inhibits the transcription of target genes by binding to specifc sites in promoters of target genes after entering the nucleus [40]. Under certain conditions, endothelial cells transition into mesenchymal cells (EndMT) regulated by Snail [11]. Snail activation induces endothelial cells to transition into mesenchymal cells (especially fbroblasts), and EndMT-derived CFs synthesize a large amount of collagen, ultimately resulting in pathological collagen fber deposition and fbrotic lesions. Tis process has also been confrmed in other studies on MF [12,41]. TGFβ1, a cytokine that can strongly promote fbrosis, was shown to play a critical role in various fbrotic diseases, including MF [42]. Recent studies have shown that Snail expression is regulated by TGFβ1. Myocardin-related transcription factor (MRTF-B) is upregulated by TGFβ1, and it binds to an SP1 element (SP1near) in the promoter of Snail to promote Snail expression, and Snail continues to induce EndMT [20,43]. Tus, by inhibiting the TGFβ1/Snail pathway, EndMT inhibition is an important therapeutic strategy for MF.
However, current studies have not investigated whether the TGFβ1/Snail pathway promotes the activation of CFs through paracrine signaling in addition to promoting EndMT. In this study, we found that C-L can signifcantly inhibit the degree of EndMT, decrease the collagen synthesis of cells undergoing EndMT and CFs, and reduce the activation of CFs. Tese efects were related to TGFβ1 and Snail expression levels in the CMECs and were signifcantly inhibited during C-L treatment. However, after the addition of a TGFβ1 activator in hypoxic CMECs, the antagonistic efect of C-L against EndMT was strongly inhibited via upregulation of TGFβ1 and Snail, and the inhibition of collagen synthesis was also weakened. In addition, we found that the ability of C-L to inhibit CF activation was weakened by the enhancement of the TGFβ1/Snail pathway. Terefore, we believe that C-L inhibits EndMT of CMECs and the paracrine pathway of cells undergoing EndMT that promotes the activation of CFs, playing a role in the treatment of MF, and this phenomenon is achieved by regulating the TGFβ1/Snail signaling pathway ( Figure 10).

Conclusion
In summary, we found that TGFβ1/Snail pathwaymediated EndMT is a core mechanism of C-L therapy for MF. Experimental studies showed that C-L regulates the EndMT process by inhibiting the TGFβ1/Snail signaling pathway and inhibits the activation of CFs induced by EndMT by regulating the paracrine pathway, ultimately playing an antifbrotic role. Tis study further elucidated the pharmacological mechanism of C-L in treating MF in HF and provided a new strategy for the treatment of MF. Nevertheless, our study has limitations. CMECs promote the activation of CFs through various factors in the paracrine pathway; however, we did not determine whether CTGF is a crucial factor. In addition, the results were not further verifed by in vivo gene knockout experiments, and the main pharmacological components of C-L have not been determined. In future work, to discover the main components of C-L that ameliorate MF, we will provide more rigorous results using gene knockout animal models in accordance with pharmacological studies on various components contained in C-L.

Data Availability
Te original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

Conflicts of Interest
Te authors declare that they have no conficts of interest.  Figure 10: Schematic diagram of C-L treatment of MF. C-L efectively blocked the TGFβ1/Snail pathway, inhibited EndMT and profbrogenic factor release in CMECs, and eventually attenuated hypoxia-induced MF.