Salvianolic Acid B Improves the Effect of Fat Grafts by Inhibiting the NF-kB Signalling Pathway in Macrophages

Jia-Ming Sun Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Chia-Kang Ho Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Ya Gao Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Chio-Hou Chong Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Yang-Dan Liu Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Dan-Ning Zheng Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Yi-Fan Zhang Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery Yu Li (  yuoli@163.com ) Shanghai 9th Peoples Hospital A liated to Shanghai Jiaotong University School of Medicine Department of Plastic Surgery


Methods
In vivo, 0.2 ml of Coleman fat was transplanted into nude mice with salvianolic acid B. The grafts were evaluated by HE and IF at 2, 4 and 12 weeks posttransplantation and by micro-CT at 4 weeks posttransplantation. In vitro, the proliferative and anti-in ammatory activities of salvianolic acid B were analyzed in cultured RAW264.7 cells to detect the mechanism by which salvianolic acid B affects graft survival by inhibiting in ammation.

Results
In vivo, the degree of adipose tissue brosis and in ammatory cell in ltration in the salvianolic acid B treatment group was lower, and the in ltration of M1 macrophages in fat grafts was also less than that in the control group. In vitro, salvianolic acid B inhibited the proliferation and activation of in ammatory pathways in RAW264.7 cells.

Conclusions
This study demonstrates the use of salvianolic acid B as a possible treatment to improve the effect of fat transplantation.

Background
Tissue defect repair is a key component of plastic and reconstructive surgery. In recent years, autologous fat grafting (AFG) has been an appealing approach to repair soft tissue defects because it is abundant, ubiquitous, biocompatible, and relatively simple to access [1]. AFG is used for various cosmetic or reconstructive indications, such as facial contour re nement [2], breast augmentation and reconstruction [3], posttraumatic deformity [4], congenital deformity [5] and other elds. According to the American Society of Plastic Surgeons[6], 24892 AFG procedures for breast reconstruction and augmentation and 43177 AFG procedures for facial lling were performed in 2018. These values represent increases by 37.3% (breast) and 3% (face) compared with the number of AFG procedures performed in 2015.
Despite the advantages of AFG, the high absorption rate and various complications, such as oily cyst formation, brosis, calci cation and fat necrosis, have in uenced the lling effect. A systematic review [7] showed that the volume retention rate of AFG in facial lling varies from 26 to 83%. Another study reported that the overall rate of patient complications is 2.1% for breast cancer, 2.26% for facial lling and 10.5% for buttock lling [8]. Therefore, developing methods to improve the survival rate of fat transplantation and reduce complications is crucial. Some studies showed that lipid droplets and cell debris induce a tissue in ammatory response [9]. Excessive in ammation will further induce fat necrosis. Therefore, reducing the in ammation of fat grafts provides new treatment inspiration.
The anti-in ammatory role of traditional Chinese medicine is increasingly prominent. Salvia miltiorrhiza (SM) is one of the most widely used Chinese medicines in clinical practice. It has the functions of promoting blood circulation, removing blood stasis, anti-in ammatory effects, and antioxidative stress effects. It is widely used to treat artery diseases such as atherosclerosis [10].
Because Salvia miltiorrhiza is a mixed traditional Chinese medicine, different sources of production or different processing methods will affect the proportion of each component in Salvia miltiorrhiza, leading to inaccurate curative effects [11]. Salvianolic acid B (Sal-B, Fig. 1A) is the most abundant and bioactive water-soluble compound in Salvia miltiorrhiza [12]. Studies have shown that salvianolic acid B has antiin ammatory and anti-immune effects [13,14]. In this study, we attempted to investigate whether Sal-B improves the survival rate of fat transplantation through anti-in ammatory effects.

Animals
All the animal experiments were approved by Shanghai Ninth People's Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai, China. Female nude mice (aged 6 to 8 weeks) were housed in individual cages with a 12-hour light/dark cycle and provided with standard food and water ad libitum.

Fat Grafting Model and Treatments
The mice were randomly divided into three groups: saline, 10 μmol/L, and 50 μmol/L. Each mouse was injected subcutaneously on the back with 0.2 ml of Coleman fat using a 1 ml syringe with a blunt in ltration cannula (Supplementary material Fig. 1A). The grafts were injected into a spherical shape. The mice were locally injected with 0.2 ml of saline or salvianolic acid B (10 μmol/L, 50 μmol/L) once every two days. The mice were sacri ced after 2, 4, and 12 weeks (n = 5 per time point per group), the grafts were harvested and carefully separated from surrounding tissue, and their volumes and weights were measured. Each harvested sample was assessed histologically and immunohistochemically.
2.3 RAW264.7 Culture RAW264.7 cells (ATCC, USA) were grown in culture medium, which was changed every 48 h. When the cells reached approximately 90% con uence, they were washed using high-sugar DMEM (Gibco, USA) with 10% foetal bovine serum (FBS; Gibco, USA). The cell suspension was further divided into three culture asks containing growth medium.

Cell viability assays
The effects of salvianolic acid B on the viability of RAW264.7 cells were tested using Cell Counting Kit-8 (Beyotime, China) according to the manufacturer's instructions. Brie y, 5000 cells/well were seeded in a 96-well plate. The cells were cultured in growth medium with various concentrations of salvianolic acid B (0, 10, 25, 50, 75, 100 μmol/L). At 72 h of culture, 10% CCK-8 reagent was mixed with medium and added to each well. The 96-well plate was incubated at 37 °C for 2 h. The relative number of cells was measured at an absorbance of 450 nm using a microplate reader (Thermo, USA).

Flow cytometry
After 3 days in culture with various concentrations of salvianolic acid B pretreatment, RAW264.7 cells were resuspended in PBS buffer according to the number of cells (5,000/ml). For Annexin V and propidium iodide staining, 195 μL of cell suspension was mixed well with 5 μL of Annexin V-FITC, followed by incubation at room temperature for 10 minutes. The cells were washed with PBS and resuspended in 190 μL of deliquated binding buffer, and then 10 μL of 20 µg/ml propidium iodide was added. The samples were analyzed by ow cytometry using CytoFLEX LX (Beckman Coulter, USA). The data were analyzed using CytExpert (Beckman Coulter, USA).

Histological analysis and immuno uorescence staining
Tissues were xed in paraformaldehyde overnight, embedded in para n, cut at a thickness of 5 μm and then stained with haematoxylin and eosin. We used the methods of Shoshani O [15] and Yu P[16] to evaluate histologic parameters, such as cell integrity, tissue in ammation, the presence of cysts/vacuoles, and the extent of brosis. Each parameter was scored as follows: 0 = absence, 1 = minimal presence, 2 = minimal to moderate presence, 3 = moderate presence, 4 = moderate to extensive presence, and 5 = extensive presence. The scoring was performed independently by 3 authors who were unaware of the grouping. For immunocyto uorescence, RAW264.7 cells were incubated with a primary antibody against p-p65 (#3033; 1:500; Cell Signaling Technology, USA) diluted in blocking solution overnight at 4 °C. After incubation with Alexa Fluor 488-conjugated goat anti-rabbit immunoglobulin G (#A-21206; 1:500; Invitrogen, USA), the nuclei were stained with 4′,6-diamidino-2-phenylindole (Southern Biotech, USA).
ImageJ software was used for quantitative analysis. Image analysis was performed according to the website (https://imagej.net/imaging/image-intensity-processing) and the method of Keskin [17].

Micro-CT analysis
The fat grafts were scanned using micro-CT (PerkinElmer, USA) and analyzed by ProPlan CMF 3.0.

RNA Extraction and Real-time RT-PCR
To investigate the polarization level of RAW264.7 cells, RAW264.7 cells were incubated under standard conditions with various concentrations of salvianolic acid B (0, 10, 50, and 100 μmol/L). After 72 hours of culture, the cells were stimulated with ultrapure LPS (10 ng/ml; Sigma, USA), and then the transcriptional levels of iNOS and TNF-α in RAW264.7 cells were assessed by real-time PCR. Initially, the total RNA of RAW264.7 cells was extracted using a total RNA miniprep kit (Axygen, USA), and RT-qPCR was performed using an ABI 7900HT system and SYBR Premix (Takara, Japan) according to the manufacturer's instructions. mRNA quanti cation was performed using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) for normalization. The SYBR green primers for qRT-PCR are listed in Supplementary Table 1.

RNA-Seq analysis
RNA sequencing samples were acquired after the addition of LPS (10 ng/ml)+Sal-B (50 μmol/L), LPS (10 ng/ml) or solvent to RAW264.7 cells for 3 days in growth medium. The RNA quantity and quality were measured using the NanoDrop ND-1000 system. The cDNA library was constructed using the KAPA Stranded RNA-Seq Library Preparation Kit (Illumina) following the manufacturer's protocol. The nal double-stranded cDNA samples were veri ed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). After cluster generation (TruSeq SR Cluster Kit v3-cBot-HS; Illumina), sequencing was performed using an Illumina HiSeq 4000 sequencing platform. Image analysis, base calling, and error estimation were performed using Illumina/Solexa Pipeline (Off-Line Base Caller software, version 1.8).
Quality control was checked on the raw sequence data using FastQC (https://en.wikipedia.org/wiki/FASTQ_format). Raw data were preprocessed using Solexa CHASTITY and Cutadapt to remove adaptor sequences, ribosomal RNA, and other contaminants that may interfere with clustering and assembly. The trimmed reads were mapped to the corresponding reference genome using HISAT2 (version 2.0.4) for RNA sequencing, and StringTie (version 1.2.3) was used to reconstruct the transcriptome. Ballgown software was applied to calculate the fragments per kilobase of exon per million fragments mapped in RNA sequencing data and analyze differentially expressed genes, with the fragments per kilobase of exon per million fragments mapped ≥ 0.5 (Cuffquant) considered for the analysis. The cut-off for de ning which genes were differentially expressed was a fold change greater than 1.5. Gene Ontology functional and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses were performed for differentially expressed genes using the Database for Annotation, Visualization and Integrated Discovery and Kyoto Encyclopedia of Genes and Genomes Orthology-Based Annotation System online tools (http://www.geneontology.org and http://www.genome.jp/kegg).

Molecular docking
The PDB le for the crystal structures of IKKβ was obtained using the protein data bank code 4KIK. The molecular docking procedure was performed under the C-DOCKER protocol of Accelry's Discovery Studio Professional 17.0 software, saved in the SDF le format and minimized using Accelry's Discovery Studio 2019 software. The protein structure was cleaned and inspected for errors, hydrogens were added, and the water molecules were deleted. The IKKβ proteins were de ned as receptors, and the centroid of the binding site was de ned based on the ligand in the cocrystal structure. Next, the original ligand was removed, and the molecule of Sal-B was placed in the sphere position to perform molecular docking. For energy minimization, the CHARMM force eld was used within Accelry's Discovery Studio 2019 software. Finally, the types of interactions between the docked proteins and Sal-B were analyzed.

Statistical analysis
In the present study, all in vitro experiments were conducted 3 times. Single blinding was used for statistical analysis. Two blinded data analysts independently analyzed the data. The nal data were consistent between the two analysts. The data were expressed as means ± SD. The continuous variables between the groups were compared by the independent samples t-test. One-way ANOVA with Tukey's post hoc test was employed for pairwise comparisons among multiple groups. The signi cance of differences between the control and treated groups was set at P<0.05 and assessed using GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA).

Results
3.1 Tissues from the salvianolic acid B-treated group show higher retention rates and lower in ammation after transplantation After 3 months of salvianolic acid B treatment, the fat graft retention in the treatment groups was higher (Fig. 1B). Additionally, by HE staining, the degree of adipose tissue brosis and in ammatory cell in ltration in the salvianolic acid B treatment groups was lower (Fig. 1C, D), and the structural integrity of adipose tissue and formation of oil cysts were lower (Supplementary material Fig. 1B). Furthermore, Perilipin staining further validated the above facts, and the salvianolic acid B treatment groups had more viable adipocytes ( Fig. 2A, B). Interestingly, in the middle stage of fat transplantation, we analyzed the survival of subcutaneous fat transplantation and related complications, and the salvianolic acid B treatment groups exhibited more fat survival and less absorption (Fig. 2C). Thus, salvianolic acid B can improve the retention of transplanted fat, and the effect of a 50 μmol/L concentration is better.

Salvianolic acid B reduces the activation ratio of M1-type macrophages in fat grafts
Macrophages play an essential role in mediating in ammation in fat transplantation. Studies [18] have shown that in the early period after fat transplantation, the immune cells of the graft are mainly M1 macrophages, which release a large amount of in ammatory factors. In the late period after fat transplantation, the immune cells of the graft are mainly M2 macrophages, which release a large amount of anti-in ammatory factors. Therefore, we further used immuno uorescence staining to determine whether salvianolic acid B affects macrophages. We de ned M1 macrophages as F4/80-and CD11c-positive and M2 macrophages as F4/80-and CD206-positive as described by J. Cai [19]. Consistent with other studies, in the control group, M1 macrophages in ltrated in the early stage after fat transplantation, while M2 macrophages in ltrated in the later stage of fat transplantation (Fig. 3A, 3B, 3C, 3D). Interestingly, in the experimental group, early M1 macrophage in ltration was reduced compared with that in the control group (Fig. 3A, 3B), further explaining the results observed in the previous HE staining.
Additionally, the level of M2 macrophages also decreased at 12 weeks, but no signi cant difference was found (Fig. 3C, 3D). However, interestingly the ratio of M2/M1 increased at 4 and 12 weeks after salvianolic acid B treatment (Fig. 3E) compared with the control group. Thus, salvianolic acid B can reduce the in ltration of M1 macrophages in fat grafts.

Salvianolic acid B inhibits the proliferation of macrophages
In in vitro experiments, we used RAW264.7 cells as a macrophage model to clarify the effect of salvianolic acid B on macrophages. First, RAW264.7 cells were treated with different concentrations of salvianolic acid B. We found that when the concentration was greater than 10 μmol/L, the viability of RAW264.7 cells decreased (Fig. 4A). A decrease in proliferation or an increase in apoptosis can lead to a decrease in activity. Therefore, we used EdU staining and ow cytometry to detect the effect of salvianolic acid B on the proliferation and apoptosis of RAW264.7 cells. Interestingly, the proliferation level of RAW264.7 cells was inhibited at concentrations of 50 μmol/L and 100 μmol/L (Fig. 4B, 4D). At a concentration of 100 μmol/L, the apoptotic level of RAW264.7 cells increased (Fig. 4C, 4E). Therefore, we believe that a concentration of 50 μmol/L can inhibit the proliferation of macrophages without causing the apoptosis of macrophages.

RNA-Seq shows that salvianolic acid B reduces LPS-induced in ammation of macrophages and inhibits the activation of macrophages
To further clarify the effect of salvianolic acid B on the function of macrophages, we used the LPSinduced macrophage in ammation model. Through RNA-Seq analysis, compared with the control group, the expression of many genes in the LPS treatment group was changed (Fig. 5A, 5B). Through KEGG analysis, we found that many in ammatory pathways were upregulated, such as the IL-17 signalling pathway, TNF-α signalling pathway and NF-κB signalling pathway (Fig. 5C). In the salvianolic acid B treatment group, the expression of many genes was also changed (Fig. 5A, 5B), and the abovementioned in ammatory pathways were downregulated (Fig. 5C). Additionally, by analyzing the biological process of macrophages in GO analysis, many biological processes of macrophages were changed (Fig. 5D). By merging the upregulated biological process after LPS treatment and the downregulated biological process after salvianolic acid B treatment, salvianolic acid B mainly affected biological processes related to macrophage activation (Fig. 5E). Thus, through RNA-Seq analysis, salvianolic acid B inhibits the activation of in ammatory pathways and macrophages.

Salvianolic acid B inhibits LPS-induced macrophage polarization to the M1 type by inhibiting the LPS signalling pathway
To verify the above sequencing results, the expression of iNOS and TNF-α was detected by PCR. Consistent with the above results, salvianolic acid B inhibited the expression of in ammatory genes (Fig.  6A). Furthermore, the activation of NF-κB (p65) and JNK plays an important role in activating in ammatory macrophages. Therefore, the phosphorylation level of the abovementioned proteins was tested, and the phosphorylation of p-65 and JNK was inhibited by salvianolic acid B (Fig. 6B, 6C, 6D, 6E). Additionally, through computer simulation, salvianolic acid B might have potential binding sites with IKKβ (Supplementary material Fig. 1C).
Taken together, salvianolic acid B functions to inhibit the activation of in ammatory macrophages by inhibiting the NF-kB signalling pathway.

Discussion
Autologous fat grafts have been widely used in reconstructive surgery and cosmetic surgery to solve soft tissue defects because of their convenient use, low cost, and good biocompatibility [20]. However, its unpredictable absorption rate, as well as subsequent complications, such as the formation of oil cysts, calci cations and nodules, all affect the nal clinical effect [3]. Previous studies have attempted to solve the abovementioned problems by improving the blood supply or increasing adipose stem cells. For example, stromal vascular components (SVFs) [21], cell-assisted fat transplantation [22] and cytokines have been used [23]. Although substantial progress has been made, many complications persist, and the cumbersome operation is also a hindrance. Therefore, identifying a simpler and more effective method is required. Salvia miltiorrhiza has been widely used as a traditional Chinese medicine in clinical practice [24]. Our previous studies have found that Salvia miltiorrhiza improved the survival rate of fat transplantation and promoted the proliferation and differentiation of adipose stem cells [25][26][27]. Because Salvia miltiorrhiza is a mixed traditional Chinese medicine, different sources of production or different processing methods will affect the proportion of each component in Salvia miltiorrhiza, leading to inaccurate curative effects [11]. Therefore, identifying the key small molecules in Salvia miltiorrhiza is critical. Salvianolic acid B (Sal-B) is the most abundant and bioactive water-soluble compound in Salvia miltiorrhiza [12]. Many studies have shown that salvianolic acid B has anti-in ammatory effects, particularly inhibiting in ammation induced by macrophages [28][29][30].
Macrophages play an important role in fat transplantation. Studies have shown that most macrophages are M1 polarized with a high level of TNF-α expression in the early stage of transplantation. However, in the middle and late stages of transplantation, M2 macrophages dominate the macrophage population and produce a high level of TGF-β, promoting tissue repair [19]. According to the results of the control group, we further con rmed the existence of this phenomenon. Over time, the proportion of M1 macrophages gradually decreased, while that of M2 macrophages gradually increased. M1 macrophages and M2 macrophages play different roles in the in ammatory response. Studies have reported that M1 macrophages release TNF-α to promote angiogenesis, but too many M1 macrophages will inhibit the proliferation and differentiation of adipose stem cells, leading to more adipocyte necrosis and tissue brosis [9,31]. M2 macrophages improve the blood supply by secreting large amounts of proangiogenic factors and recruiting more stem cells [19,32]. Our ndings showed that salvianolic acid B reduces the ratio of M1 macrophages in fat grafts. Additionally, in vitro salvianolic acid B inhibits the proliferation and in ammation of macrophages induced by LPS. Furthermore, salvianolic acid B does not affect the ratio of M2-type macrophages but can increase the M2/M1 ratio at 4 and 12 weeks.
In fat transplantation, because of the intervention of various surgical instruments, tissue trauma is inevitable, and the damaged tissue will induce an in ammatory response by activating the in ammationrelated signalling pathways of macrophages, such as the TNF-α signalling pathway. Therefore, many proin ammatory factors, such as TNF-α, iNOS, and IL-6, are released [33,34]. In our RNA-Seq results, we observed that salvianolic acid B inhibits the activation of in ammatory signalling pathways, such as TNF-α, IL-17 and NF-κB, after LPS treatment. The phosphorylation of NF-kB (p65) and JNK is related to the activation of proin ammatory macrophages [35,36], and we also detected the phosphorylation level of the abovementioned proteins in macrophages. Thus, salvianolic acid B can inhibit the phosphorylation of NF-kB (p65) and JNK.
Our study revealed the mechanism of salvianolic acid B in regulating in ammation of autologous fat grafts and its potential value in translational applications. In summary, on the one hand, salvianolic acid B inhibits the activation of macrophage in ammatory signalling pathways by inhibiting the phosphorylation of NF-κB and JNK. On the other hand, from in vitro experiments, salvianolic acid B inhibits the proliferation of macrophages (Fig7). The above two reasons lead to a decrease in the level of in ammation in fat grafts. Our study has shortcomings. First, this study adopted the method of local drug injection, the clinical safety and feasibility of which must be further evaluated. Additionally, our study only focused on the effect of salvianolic acid B on macrophages but not on the changes in other in ammatory cells, such as neutrophils, warranting further study.
This study suggests salvianolic acid B as a promising treatment to improve the effect of fat transplantation.

Conclusions
Our study reveals the mechanism of salvianolic acid B in regulating the in ammation of autologous fat grafts and its potential value in translational applications. We found that salvianolic acid B inhibits the activation of macrophage in ammatory signalling pathways by inhibiting the phosphorylation of NF-κB and JNK. Additionally, salvianolic acid B inhibits the proliferation of macrophages. Thus, salvianolic acid B may be a promising agent to reduce the complications of fat transplantation.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics declarations
Ethics approval and consent to participate       Immuno uorescence assays were performed to determine the translocation of p-p65 in cells. The data are represented as means ± SD. * P<0.05, ** P<0.01. SD, standard deviation.