Elucidating the Novel Mechanism of Ligustrazine in Preventing Postoperative Peritoneal Adhesion Formation

Postoperative peritoneal adhesion (PPA) is a major clinical complication after open surgery or laparoscopic procedure. Ligustrazine is the active ingredient extracted from the natural herb Ligusticum chuanxiong Hort, which has promising antiadhesion properties. This study is aimed at revealing the underlying mechanisms of ligustrazine in preventing PPA at molecular and cellular levels. Both rat primary peritoneal mesothelial cells (PMCs) and human PMCs were used for analysis in vitro. Several molecular biological techniques were applied to uncover the potential mechanisms of ligustrazine in preventing PPA. And molecular docking and site-directed mutagenesis assay were used to predict the binding sites of ligustrazine with PPARγ. The bioinformatics analysis was further applied to identify the key pathway in the pathogenesis of PPA. Besides, PPA rodent models were prepared and developed to evaluate the novel ligustrazine nanoparticles in vivo. Ligustrazine could significantly suppress hypoxia-induced PMC functions, such as restricting the production of profibrotic cytokines, inhibiting the expression of migration and adhesion-associated molecules, repressing the expression of cytoskeleton proteins, restricting hypoxia-induced PMCs to obtain myofibroblast-like phenotypes, and reversing ECM remodeling and EMT phenotype transitions by activating PPARγ. The antagonist GW9662 of PPARγ could restore the inhibitory effects of ligustrazine on hypoxia-induced PMC functions. The inhibitor KC7F2 of HIF-1α could repress hypoxia-induced PMC functions, and ligustrazine could downregulate the expression of HIF-1α, which could be reversed by GW9662. And the expression of HIF-1α inhibited by ligustrazine was dramatically reversed after transfection with si-SMRT. The results showed that the benefit of ligustrazine on PMC functions is contributed to the activation of PPARγ on the transrepression of HIF-1α in an SMRT-dependent manner. Molecular docking and site-directed mutagenesis tests uncovered that ligustrazine bound directly to PPARγ, and Val 339/Ile 341 residue was critical for the binding of PPARγ to ligustrazine. Besides, we discovered a novel nanoparticle agent with sustained release behavior, drug delivery efficiency, and good tissue penetration in PPA rodent models. Our study unravels a novel mechanism of ligustrazine in preventing PPA. The findings indicated that ligustrazine is a potential strategy for PPA formation and ligustrazine nanoparticles are promising agents for preclinical application.


Introduction
Postoperative peritoneal adhesion (PPA) is a major clinical complication after open surgery or laparoscopic procedure. It may cause a range of complications including acute bowel obstruction and chronic adhesion symptoms, such as abdominal pain or female infertility [1]. PPA may increase the subsequent reoperation rate, prolong the time of hospitalization, and increase medical costs. A recent retrospective cohort study performed in Scotland showed that 72,270 (17.6%) surgical patients among 12,687 patients who were readmitted within 5 years for adhesion-related problems were possibly associated with adhesions, 9436 (13.1%) were potentially complicated by adhesion, and 2527 (3.5%) were directly related to adhesion [2]. And PPA has brought heavy economic impacts on individuals, families, and the whole society [3].
Peritoneal mesothelial cells (PMCs), as the main cells of the functional peritoneum, are involved in the pathogenesis of adhesion formation [4]. Injured PMCs can induce profibrotic cytokine expression [5], i.e., vascular endothelial growth factor (VEGF) and connective tissue growth factor (CTGF), which can lead to inflammatory response and even result in peritoneal adhesion [6]. Existing studies [7,8] found that hypoxia caused by tissue injury is the most important factor in adhesion formation. Under the oxygen-deficient condition, adhesion molecules elevated, cytoskeleton reorganized, epithelial markers lost, and mesenchymal markers acquired, leading to excess accumulation of extracellular matrix (ECM) and epithelial-mesenchymal transition (EMT) [9,10]. However, the exact pathogenesis of hypoxia-induced PMC driving adhesion formation is precisely unclear. Besides, with the development of material science and life science, mounting barriers or biological materials and antiadhesion drugs have gained much attention in adhesion prevention. Each material or drug has its advantages and disadvantages [11][12][13]. Hence, it is also urgent to identify novel and more effective agents in preventing PPA.
Fortunately, nanoparticle application with unique advantages is considered as a promising approach in disease prevention and treatment, such as postoperative adhesions [14] and peritoneal fibrosis [15]. Intriguingly, our previous study confirmed that the bioactive alkaloid ligustrazine has significant effectiveness in preventing adhesion formation [16,17]. Ligustrazine is the active ingredient that is extracted from the root of natural herbal Ligusticum chuanxiong Hort (Umbelliferae). It has a hot spot in the field of cardiovascular, tumor, and other diseases [18]. Recently, ligustrazine had also attracted considerable attention in the inflammatory response [19]. Hence, we hypothesized that there might be a close relation between ligustrazine and the pathological EMT process in hypoxia-induced PMCs. In this study, we set out to explore the underlying molecular mechanisms of ligustrazine in preventing PPA both in vivo and in vitro. Our findings might generate new sights into the therapeutic strategies for PPA and good prospects for the preclinical application of ligustrazine nanoparticles (LN).

Material and Methods
2.1. Chemicals and Antibodies. Ligustrazine (C 8 H 12 N 2 , purity > 98%) was obtained from Tokyo Chemical Industry (Japan). 2.2. Cell Culture. Rat primary PMCs (RPMCs) were isolated based on previous studies [17,20]. Briefly, after the Sprague-Dawley (SD) rats were sacrificed with cervical dislocation, the abdominal skin was prepared and sterilized under aseptic conditions. 20 ml 0.25% trypsin-0.016% EDTA was injected into the peritoneal cavity for 20 min. During this period, the abdomen of the rats could be kneaded, and then, 10 ml complete medium with 10% FBS was injected into the abdominal cavity to end the digestion. The peritoneal fluid was collected, followed by centrifugation at 150 × g for 10 min. Cells were suspended and cultured in RPMI 1640 medium containing 10% FBS, 1% penicillin, and streptomycin in the incubator under 5% CO 2 at 37°C. The third-generation cells were used for the following research as previously described [17]. Besides, human PMCs (HPMCs and HMrSV5) were donated by Professor Sheng's team of the first clinical medical college in Nanjing University of Chinese Medicine. HPMCs were cultured as above. To induce hypoxia, PMCs were seeded under hypoxic conditions, which were maintained by pumping a mixture of gases (1% O 2 , 94% N 2 , and 5% CO 2 ) into a hypoxia chamber at 37°C.
2.13. Biolayer Interferometry (BLI) Assay. BLI is an alloptical method to characterize blinding affinity between small molecules and WT or MT forms that was carried out on an Octet RED 96 system (ForteBio, USA) [22]. In brief, 200 μl sample volumes per well were added in opaque 96well plates. His-tagged PPARγ (100 μg/ml) was loaded onto Ni-nitrilotriacetic acid biosensors, which were preequilibrated in kinetics buffer (1× PBS with 0.02% Tween-20) at 30°C. The biosensors were brought in the kinetics buffer to baselines for 60 s. Then, a concentration gradient of ligustrazine solution (500, 250, 125, 62.5, and 31.25 μM in PBS buffer with 0.02% Tween-20 and 0.5% DMSO) was immobilized onto Ni-nitrilotriacetic acid biosensors to develop kinetics analysis. An equal amount of DMSO was added into wells as the controls. After the association (60 s) and dissociation steps (90 s), the association and dissociation curves were fitted, and data were analyzed by Octet Data Analysis software. The affinity constant (K D ) was measured as the ratio between the dissociation rate constant (K off ) and the association rate constant (K on ); that is, K D = K off /K on .
2.14. Animal Model Preparation and Group Assignment. A total of thirty-six adult male SD rats (weighing 200 ± 20 g) were provided from the Qinglongshan Experimental Animal Breeding Farm (Nanjing, China). Rats were randomly divided into six groups (n = 6), that is, the sham, model, LN, sodium hyaluronate (SH), polylactic acid (PLA), and ligustrazine (LZ) groups. They were housed under standard conditions of controlled temperature (22 ± 2°C) with a reverse 12/12 h light/dark cycle (lights off at 06:00 AM) and had free access to tap water and food. The experiments were approved by the Laboratory Animal Management Committee of Nanjing University of Chinese Medicine (No. ACU171112).
The model preparation was established by previous studies [23,24]. Briefly, rats were stopped feeding for about 12 h. After anesthesia with 1~1.5% isoflurane, a 1.5~2 cm midline incision was made after shaving and disinfection. The cecum was placed on wet gauze and scraped by dry gauze until serosal petechiae appeared on the intestinal surfaces. After exposure to air for 5 min, the cecum was replaced into the abdominal cavity and the abdominal wall was sutured. The sham group was only given laparotomy. LN was prepared by our team as previously described [25]. A proper amount of ligustrazine was dissolved in 0.25% poloxamer solution to prepare 1 mg/ml ligustrazine solution, that is, LZ. PLA nanoparticles were prepared using the same preparation methods of LN. In the LN, SH, PLA, and LZ groups, 5 ml/kg LN, 0.5 ml/kg SH, 0.5 ml/kg PLA, and 1 mg/ml LZ were applied to the abraded peritoneum and its surrounding areas before closing the peritoneal cavity, respectively. The animal grouping and treatment are listed in Table 1. On the 7 th day after the operation, rats were anesthetized with 1~1.5% isoflurane. And an inverted Ushaped incision was used to open the abdominal cavity. The cecum including adhesive sites was collected for further study. After hemostasis was done completely, the abdominal wall was closed. All rats were sacrificed with cervical dislocation.
2.15. Macroscopic Evaluation. The adhesions were blindly evaluated and scored by two authors, individually. The adhesion scoring system was based on a five-stage adhesion score [26,27], as shown in Table 2.

Histopathological Examination.
A portion of the cecum tissues was dissected and fixed in 4% formaldehyde for 24 h. The tissues were dehydrated, embedded, and cut into sections using a routine method. The sections were then used for hematoxylin-eosin (HE) and Masson staining. Images were visualized under a microscope (Leica DM2500, Germany). The represented graphs were presented.
2.17. Immunohistochemical Assay. The cecum tissues were fixed in 4% formaldehyde for 24 h, following a series of normal procedures of deparaffinization with xylene and dehydration with alcohol step by step. And then they were embedded in paraffin wax and cut into 4 μm thickness sections. These sections were immunostained with primary antibodies against CTGF, VCAM-1, and MMP2 overnight at 4°C. The second antibody and color were all conducted according to the diaminobenzidine tetrahydrochloride (DAB) kit's instructions. Views were randomly visualized under a microscope (Leica DM2500, Germany).
2.18. Immunofluorescence Assay. The tissue slides were baked at 65°C for 1 h, blocked in 1% bovine serum albumin, and incubated with the primary antibodies against PPARγ and HIF-1α overnight at 4°C. After three washes with PBS, these slides were incubated with fluorescent secondary antibody for 30 min. For RPMCs or HPMCs, the cells were treated with vehicle or ligustrazine at different concentrations. After incubation with primary and second antibodies, the nuclei of cells were stained with DAPI for 5 min. The images of sections and cells were blindly observed with a microscope (Leica DMi8, Germany).

2.19
. qRT-PCR Assay. Total RNA from PMCs or cecum tissues was prepared with TRIzol reagent (Invitrogen, USA) and used to synthesize the first-strand cDNA with Hifair®Ш 1st Strand cDNA Synthesis Kit (11141ES60, Yeasen, China). The qRT-PCR was conducted using Hieff® qPCR SYBR® Green Master Mix (High Rox) (11203ES08, Yeasen, China), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) 4 Oxidative Medicine and Cellular Longevity was used as a control. The primer sequences were presented in Supplementary Table S2. 2.20. Western Blot Analysis. Total crude proteins were prepared from PMCs or cecum tissues by using routine protocols. Briefly, 15 μl loading buffer of each group was added into sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The separated proteins were transferred onto PVDF membranes, which were then incubated with primary antibodies overnight at 4°C. After three washes with TBST, the membranes were incubated with secondary antibodies at room temperature for 80 min. Finally, the band visualization was viewed by the Chemiluminescence Imaging System (Bio-Rad, USA).

Bioinformatics
Analysis. The mRNA microarray expression profile datasets were retrieved and downloaded from the GEO database (available online: http://www.ncbi .nlm.nih.gov/geo) by searching the following keywords: "RNA", "peritoneal adhesion" or "abdominal adhesion", and "Mus musculus" (organism). After screening, one mRNA expression dataset GSE4715 was selected for analysis. The data were downloaded and subjected to Morpheus online tool (https://software.broadinstitute.org/morpheus/). The significantly expressed genes were identified when the signal to noise > 1 or signal to noise < −1. GO annotation and KEGG enrichment were conducted by the Database for Annotation, Visualization, and Integrated Discovery (DAVID) [29] when P value < 0.05 was considered as a screening threshold.

Statistics.
All results were expressed as the mean ± standard deviation. The data among multigroups were analyzed by one-way analysis of variance (ANOVA) with LSD test, and the data between two groups were analyzed using t-test. The Kruskal-Wallis test was used for nonnormally distributed continuous data. P < 0:05 was considered significantly different.

Ligustrazine Suppresses the Production of Profibrotic
Cytokines in Hypoxia-Induced PMCs. PMCs were cultured under hypoxic conditions at four different time (6, 12, 24, and 48 h), respectively. It showed that cell viability was decreased depending on the different exposure time (Figure 1(a)). Because hypoxic conditions for 24 h could reduce the viability of 50% PMCs, this exposure time was used in the subsequent study. VEGF and CTGF are the two pivotal profibrotic cytokines in PMCs. We preliminarily found that ligustrazine reduced the protein expression and mRNA levels of the two profibrotic cytokines in a concentration-dependent . And the supernatant levels of VEGF and CTGF were concentration-dependently decreased (Figures 1(e) and 1(f)). Taken together, the results suggested that ligustrazine could suppress the production of profibrotic cytokines in hypoxia-induced PMCs.

Ligustrazine Inhibits the Expression of Migration and
Adhesion-Associated Molecules in Hypoxia-Induced PMCs.
Cell migration and adhesion are key steps towards peritoneal adhesion, during which many adhesion molecules are involved in. VCAM-1 and ICAM-1 are identified as two important biomarkers in cell migration and adhesion [6]. The protein expression of both adhesion contributors was decreased by ligustrazine in a concentration-dependent The cecum was exposed in the air for 5 min 2

Model
The cecum was scraped by dry gauze in the air until serosal petechiae appeared on the intestinal surfaces (lasting for 5 min) 3 Ligustrazine nanoparticles (LN) The cecum was scraped by dry gauze in the air until serosal petechiae appeared on the intestinal surfaces (lasting for 5 min), and then, 5 ml/kg LN was applied to the abraded peritoneum and its surrounding areas before closing the peritoneal cavity 4 Sodium hyaluronate (SH) The cecum was scraped by dry gauze in the air until serosal petechiae appeared on the intestinal surfaces (lasting for 5 min), and then, 0.5 ml/kg SH was applied to the abraded peritoneum and its surrounding areas before closing the peritoneal cavity 5 Polylactic acid (PLA) The cecum was scraped by dry gauze in the air until serosal petechiae appeared on the intestinal surfaces (lasting for 5 min), and then, 0.5 ml/kg PLA was applied to the abraded peritoneum and its surrounding areas before closing the peritoneal cavity The cecum was scraped by dry gauze in the air until serosal petechiae appeared on the intestinal surfaces (lasting for 5 min), and then, 1 mg/ml LZ was applied to the abraded peritoneum and its surrounding areas before closing the peritoneal cavity

Ligustrazine Represses the Expression of Cytoskeleton
Proteins in Hypoxia-Induced PMCs. Cytoskeleton acts a meaningful role in cellular function, and it is closely related to the activation of the fibrinolysis system in PMCs [30]. To determine the exact roles of ligustrazine on cytoskeletal   Oxidative Medicine and Cellular Longevity change of PMCs under hypoxia, the cytoskeleton protein expression of vinculin was measured at a concentrated point. We found that ligustrazine could significantly reduce the expression of vinculin, evidenced by immunofluorescence analysis (Figure 3(a)). It also demonstrated that ligustrazine concentration-dependently decreased the expression of vinculin, evidenced by Western blot and qRT-PCR analysis (Figures 3(b) and 3(c)). Taken together, the results revealed that ligustrazine could repress the expression of cytoskeleton proteins in hypoxia-induced PMCs.

Ligustrazine Restricts Hypoxia-Induced PMCs to Obtain
Myofibroblast-Like Phenotypes. Cytoskeletal change might give us important information about the critical function of cells [31]. And the reorganization of F-actin microfilament plays an important role in the cytoskeleton. Herein, the cytoskeleton was stained with F-actin using phalloidin, which is represented by a green color in Figure 4(a). It demonstrated that ligustrazine could significantly reduce the formation of actin stress fibers. Considering the central roles of cofilin in cytoskeletal stability [32], we also detected the levels of the biomarker cofilin and its phosphorylation status under hypoxic conditions, which indicated that p-cofilin was significantly decreased by ligustrazine, evidenced by immunofluorescence assay (Figure 4(b)). The results were also in line with those of Western blot analysis. It indicated that the protein expression of p-cofilin was decreased by ligustrazine in a concentration-dependent manner (Figure 4(c)). It is reported that PMCs after the peritoneal injury can transform into myofibroblasts, which is the key step of the developmental progress of peritoneal adhesion or fibrosis [33,34]. FSP1 acts as an important contributor to fibroblast and myofibroblast [35]. The immunofluorescence results demonstrated that ligustrazine could markedly decrease the expression of FSP1 (Figure 4(d)). Western blot and qRT-PCR analysis showed that the protein expression and mRNA levels of FSP1 were downregulated in a concentrationdependent manner (Figures 4(e) and 4(f)). Besides, intracellular Ca 2+ levels were concentration-dependently decreased, as presented by flow cytometry assay (Figures 4(g) and 4(h)). Collectively, the findings revealed that ligustrazine could restrict hypoxia-induced PMCs to obtain myofibroblastlike phenotypes.

Ligustrazine Inhibits Hypoxia-Induced PMC ECM
Deposition and EMT Transition. ECM remodeling and EMT transition are critical steps toward the accumulation of myofibroblasts in the pathological process of peritoneal fibrosis, during which many genetic transcripts are involved [33]. ECM-associated molecules (MMP2 and TIMP-1) were measured at a concentrated point. It suggested that protein expression and mRNA levels of MMP2 and TIMP-1 were concentration-dependently downexpressed by ligustrazine (Figures 5(a)-5(c)). Meanwhile, the mesothelial-related phenotypic biomarkers (E-cadherin and cytokeratin 18) and mesenchymal-related phenotypic biomarkers (Snail 3.6. Ligustrazine Suppresses Hypoxia-Induced PMC Functions by Activating PPARγ. Our previous study has shown that ligustrazine has positive effects on postoperative peritoneal adhesion through activating the inflammatory signaling path-way [16]. PPARγ is regarded as a target biomarker that has a direct effect on PMC biology [36]. To determine the exact role of the nuclear receptor in the biological function of ligustrazine on hypoxia-induced PMC transition, the nuclear protein expression and mRNA levels of PPARγ were preliminarily measured using Western blot and qRT-PCR. It indicated that ligustrazine concentration-dependently increased the protein expression and mRNA levels of PPARγ (Figures 6(a) and 6(b)), which was in line with the results of immunofluorescence analysis (Figure 6(c)). GW9662 as an antagonist of PPARγ [37] was used to further verify the critical function of PPARγ. The results of flow cytometry analysis showed that the ligustrazine-driven nuclear distribution of PPARγ was reversed by GW9662 ( Figure 6(d)). We found that GW9662 reversed the suppression effect of ligustrazine on the production of profibrotic cytokines (VEGF and CTGF) (Figure 6(e)). And GW9662 restored the inhibitory effect of  Collectively, the results demonstrated that the suppression effect of ligustrazine on hypoxia-induced PMC functions was achieved by regulating PPARγ.    14 Oxidative Medicine and Cellular Longevity transcription factor involved in the pathophysiologic process of peritoneal injury [38]. To this point, the HIF-1α inhibitor KC7F2 (50 μM) was applied to determine the transcriptional role of HIF-1α in the EMT process of PMCs. It indicated that KC7F2 downregulated the production of the profibrotic cytokines, restricted the expression of migration and adhesion-associated molecules, and decreased the expression of cytoskeleton protein (Figures 7(a)-7(e)). Moreover, KC7F2 suppressed FSP1 expression and inhibited ECM recruitment and EMT transition (Figures 7(f) and 7(g)).

The Suppression Effect of Ligustrazine Is Achieved by the
The above data indicated that the HIF-1α played key roles in hypoxia-induced PMC functions. The mRNA levels and protein expression of HIF-1α were downregulated by ligustrazine, which were reversed by GW9662 (Figures 7(h) and      17 Oxidative Medicine and Cellular Longevity 7(i)). It demonstrated that the HIF-1α expression suppressed by ligustrazine was achieved by regulating PPARγ. It reported that the transcriptional activity of PPARγ was regulated by the recruitment of NCoR or SMRT [39]. In our study, we found that the mRNA levels and protein expression of HIF-1α inhibited by ligustrazine were dramatically reversed after transfection with si-SMRT in HPMCs (Figures 7(j)-7(m)). However, there was no change in the si-NCoR group. Taken together, it showed that the suppression effect of ligustrazine was achieved by the activated PPARγ on the transrepression of SMRT-mediated HIF-1α.

18
Oxidative Medicine and Cellular Longevity emphasized the importance of hydrophobic bonds in ligustrazine. Altogether, these results suggested that there were binding sites between ligustrazine and PPARγ, thereby leading to a series of subsequent biological processes. All discoveries revealed that ligustrazine is directly bound to PPARγ, and mutant 3 displayed the weakest interaction with ligustrazine than the other four mutants, suggesting the Val 339/Ile 341 residue was critical for the binding of PPARγ to ligustrazine.

PPARγ Pathway Identification by Bioinformatics
Analysis. The GSE4715 expression profile dataset consists of 6144 gene probes. A total of 274 differently expressed genes were identified, among which 115 were upregulated and 159 were downregulated. The 274 genes were then conducted using DAVID for biological process annotation and KEGG pathway enrichment. The GO term assignment analysis indicated that the enriched GOs in the biological process included retinoic acid receptor signaling pathway (GO: 0048384), Golgi to plasma membrane protein transport (GO: 0043001), collagen fibril organization (GO: 0030199), and serine family amino acid metabolic process (GO: 0009069), as viewed in Figure 10(a). Based on the pathway enrichment analysis, the altered genes were involved in the PPAR signaling pathway (top 1) and ECM-receptor interaction ( Figure 10(b)). These findings further verified the critical roles of the PPAR pathway in the biological process of PPA.

LN Attenuates Peritoneal Adhesion in PPA Rats by
Inhibiting PMC Functions. Loading capacity and entrapment efficiency are two pivotal indicators to measure the properties of nanoparticles. The PLA nanoparticles loaded with ligustrazine had high loading capacity and entrapment efficiency, as we reported previously [25]. It also had uniform spherical morphology with a smooth surface and good dispersivity (Figure 11(a)). The distribution of particle size is about 200 nm. And LN had stable and sustained release behavior, evidenced by the in vitro drug release assay [25]. The outstanding features of LN are the prominently sustained release effect and good stability for tissue penetration via a positive targeting mechanism, which may be a novel agent for preventing PPA. We found that the incision of all rats was primary healing, without infection or other complications during the model preparation. No rat death was observed, and there was no significant difference in body weight among the six groups. Two rats in the model group had dark black cecum, which might have intestinal obstruction or necrosis. One rat in the PLA group had swelling cecum, which was smelly after incision. All rats were sacrificed on the 7 th day after successfully modeling. The adhesion scores and frequency of different grades among groups are displayed in Figures 11(b) and 11(c). The adhesion scores were listed as follows: model > PLA > LZ > SH> LN > Sham. Compared with the sham group, the model group showed a severe peritoneal adhesion with a significantly higher adhesion score. In comparison with the model group, the LN group had a markedly lower adhesion score. The representative images of peritoneal adhesion in six groups are presented in Figure 11(d). Masson staining analysis showed that massive inflammatory cells and collagen fibers were found in the model group compared with the sham group. The LN group displayed a decreased fibrin thickness in comparison with the model group, which revealed fewer collagen fiber depositions (Figure 11(d)).
The protein expression and mRNA levels of the profibrotic cytokines (VEGF and CTGF), migration and adhesionassociated molecules (VCAM-1 and ICAM-1), and ECMassociated molecules (MMP2 and TIMP-1) were elevated in the model group. After LN treatment, they were all significantly downregulated in the LN group (Figures 11(e)-11(k)). The results were in line with those of immunohistochemical analysis (Figure 11(l)). Additionally, we found that the expression of PPARγ was increased and HIF-1α was downregulated after LN treatment, as evidenced by the immunofluorescence results (Figure 11(m)). Together, these findings demonstrated that LN significantly attenuated peritoneal adhesion by suppressing the production of profibrotic cytokines, inhibiting the expression of migration and adhesion-associated molecules, ECM deposition, and EMT transition, which were in line with the results of cell experiments.

Discussion
In this study, we initially demonstrated the specific roles of ligustrazine in preventing PPA at molecular levels. Our findings indicated that ligustrazine could significantly reverse hypoxia-induced PMC EMT-like changes on the cellular levels and in the rodent models. The benefit of ligustrazine on ECM remodeling and EMT transition is contributed to the activation of PPARγ on transrepression of HIF-1α in an SMRT-dependent manner. Molecular docking and sitedirected mutagenesis tests further verified the close interaction of ligustrazine with PPARγ. Additionally, ligustrazine could suppress the production of profibrotic cytokines, inhibit the expression of migration and adhesion-associated molecules,

22
Oxidative Medicine and Cellular Longevity and restrict the expression of cytoskeleton proteins. Herein, ligustrazine could be served as a potential strategy for the prevention of PPA. Although surgical procedures, especially laparoscopic surgery, have made considerable progress, PPA remains a concerning problem. After the peritoneal injury, a series of cascade reactions are triggered, during which hypoxia plays the central role. Mounting evidence suggests the close relations between hypoxia and adhesion. Hypoxia remarkably alters various biomarker generations, irreversibly induces adhesion phenotype and inflammatory phenotype development [40,41], and simultaneously increases the deposition of extracellular matrix [42], following enhancing EMT process [38] in vitro studies. To induce hypoxic conditions, PMCs were cultured under 1% O 2 , 94% N 2 , and 5% CO 2 at different time. The cell viability of PMCs was in a timedependent manner, which indicated that hypoxia was indeed involved in the pathogenesis of adhesion formation. The pharmacology study showed that ligustrazine had special effects on anti-inflammatory activities [43]. Our previous studies demonstrated that ligustrazine has positive preventive effects on PPA both in vivo and in vitro [16,17]. Herein, we sought to clarify the functional roles of ligustrazine in the key pathological link of the occurrence and development of  25 Oxidative Medicine and Cellular Longevity peritoneal adhesion formation. Peritoneal fibrosis is one of the major factors of PPA formation. We found that ligustrazine could downregulate the expression of profibrotic cytokines and abolish the expression of migration and adhesion-associated molecules. The result was consistent with a previous study. It indicated that ligustrazine had positive effects on antifibrosis and prevented PMCs from injury [17]. Cytoskeletal change is a prerequisite for adhesion formation, which is necessary for cell migration [31]. In the current study, we found that ligustrazine could restrict PMC phenotypic transition to acquire myofibroblast-like phenotypes, and inhibit cytoskeletal remodeling reorganization. Hypoxia can induce adhesive phenotype to peritoneal fibroblast or myofibroblast change and promote ECM accumulation and EMT transition during the following progress [31]. Moreover, the altered expression of EMT phenotypic markers contributes to the increased acquisition of myofibroblast-like phenotype and is critical to ECM remodeling. Ligustrazine was found to suppress hypoxia-induced PMC ECM deposition and EMT transition effectively, but the underlying functional mechanisms needed to be further elucidated.
In this study, we highlighted PPARγ as a pivotal factor that is responsible for EMT-like changes and causes abnormal ECM accumulation in the effector mechanisms of ligustrazine. Mechanistically, ligustrazine activated the nuclear distribution of PPARγ. The prior observations were all reversed by the knocking down of PPARγ. And the results of the following bioinformatics analysis further confirmed our hypothesis that the activation of PPARγ is a key step of ligustrazine to suppress the PMC functions. Besides, PPARγ was regarded as a potential target for peritoneal fibrosis in either normoxic or hypoxic conditions, and the key step to abolish the changes of adhesion phenotype in fibroblast was to elevate PPARγ level [44]. To determine the downstream effector of PPARγ activation, we found that the transcription of HIF-1α inhibited by ligustrazine was negatively regulated by PPARγ. In particular, HIF-1α is a pivotal regulator of hypoxia. We inferred that the activated PPARγ could inhibit the transcription of HIF-1α in some dependent manner. It triggered us to explore the exact underlying mechanisms on how PPARγ regulates HIF-1α, thereby influencing the ECM deposition and EMT change.
Fortunately, based on the evidence [45,46], coregulators SMRT or NCoR recruitment by PPARγ was involved in regulating the expression of the downstream genes. HPMCs were transfected with si-SMRT and si-NCoR, respectively. Both the mRNA levels and protein expression of HIF-1α were significantly changed in the SMRT siRNA group rather than in the NCoR siRNA group. It indicated that SMRT was the crucial ligand in the regulatory mechanism of ligustrazine to activate PPARγ on the suppression of HIF-1α transcription. Subsequently, we speculated that ligustrazine and PPARγ might have some physical-binding manner. We predicted that ligustrazine could be docked successfully into several pocket residues to activate the receptor PPARγ, evidenced by the molecular docking analysis. Among these binding sites, the residue Val 339/Ile 341 was critical for ligustrazine to activate PPARγ by site-directed mutagenesis assay. And it is also reported that the conformational changes of PPARγ can help the receptor to bind with dynamic ligand conformations [47]. This further confirmed that different bind sites were involved in the engagement of ligustrazine interaction with PPARγ. All these findings indicated that PPARγ is a direct target of ligustrazine, which provided a new perspective from the molecular aspects. Importantly, we conducted PPA rodent models to validate our observations in vivo. However, pharmacokinetic studies reported that there were several disadvantages of ligustrazine on injured tissues, such as rapid absorption, fast metabolism, short half-time, low bioavailability, and poor tissue distribution [45,46], which limited its biomedical application. To maintain an effective medication concentration, a controlled release preparation; that is, PLA nanoparticles loaded with ligustrazine were used on injury sites. It is well known that PLA as the lipophilic, biodegradable, and biocompatible polymer was widely reported to load with microparticles or nanoparticles [48][49][50]. The distinct physicochemical properties of the LN had good biocompatibility and surface activity. The novel agent could increase the drug particle solubility by nanotechnology, as evidence of high entrapment efficiency and loading capacity results. The results suggested that LN had site-specific drug delivery properties for preclinical application. The adhesion score/grade system showed a significant advantage of antiadhesion efficacy in the LN group. The findings indicated that LN was beneficial in preventing PPA. Masson staining further validated the advantage of LN in suppressing collagen fiber deposition objectively. Likewise, the results of the immunohistochemical analysis, Western blot, qPCR assay, and immunofluorescence analysis were all consistent with prior observations. All evidences suggested that LN may be regarded as an ideal approach for minimization of the adhesion effects.

Conclusion
In summary, our study uncovered that ligustrazine activated PPARγ and interacted with PPARγ in a specific site, following transrepression HIF-1α in an SMRT-mediated manner, thereby inhibiting hypoxia-induced PMC functions, ECM remodeling, and EMT phenotype changes ( Figure 12). And we discovered a novel nanoparticle agent with sustained release properties, drug delivery efficiency, and good tissue penetration in rodent models. Our findings may provide new insights into PPA prevention either in preclinical or experimental research.