Inhibition of Adipocyte Necroptosis Alleviates Fat Necrosis and Fibrosis After Grafting in a Murine Model

Abstract

Background

Because of the delicate structure of the adipose tissue, fat necrosis accounts for 43.7% of all complications after autologous fat grafting; however, its regulation remains unclear.

Objectives

The purpose of this study was to examine the role of necroptosis in fat graft remodeling after grafting.

Methods

Clinical fat graft necrosis samples were collected, and the expression levels of the necroptosis marker phosphorylated(p)-MLKL were analyzed. Transcriptome analysis was performed on fat grafts before and 1 week after transplantation in C57BL/6 mouse fat grafting models. Additionally, the in vivo effects of RIPK1 inhibitor Nec-1s or RIPK3 inhibitor GSK′872 on the fat grafting complications, including fat necrosis and fibrosis, were investigated.

Results

Necroptosis markers were observed and associated with higher occurrence of fibrosis in clinical fat graft necrosis samples compared to normal fat tissue. Amplification and RNA-Seq were conducted on RNA isolated from fat grafts before and after grafting. MLKL, RIPK1, and RIPK3's expression levels were significantly upregulated in comparison to controls. Higher expression levels of necroptotic RNAs were associated with higher levels of DAMPs, including Cxcl2, HMGB1, S100a8, S100a9, Nlrp3, and IL33, and activated proinflammatory signaling pathways, including the TNF, NF-kappa B, and chemokine signaling pathways. Necroptotic inhibitor Nec-1s and GSK′872 robustly suppressed the p-MLKL expression level and significantly inhibited necroptotic cell death, especially in adipocytes. Moreover, administration of Nec-1s and GSK′872 significantly alleviated fat necrosis and subsequent fibrosis in fat grafts.

Conclusions

Collectively, our study findings highlight the potential therapeutic applications of necroptosis inhibitors in preventing fat necrosis and fibrosis after grafting.

Level of Evidence: 4

Fat grafting is an important surgical procedure in the field of tissue reconstruction, with the aim of filling tissue defects and restoring contours.^1-3 However, an estimated one-third of patients undergoing large-volume fat grafting develop fat necrosis following fat grafting. The prevalence of fat necrosis following fat grafting ranges from 2% to 43.7%.^4-6 Postoperative fat necrosis manifests as cysts, calcifications, or sclerotic nodules in these patients.^7-11 The localized necrotic graft fat eventually becomes surrounded by dense fibrous tissue. Therefore, fat graft necrosis on breast imaging can be mistaken for intraductal carcinoma.¹² Previous studies have focused on the retention of fat grafts, and the normalization of graft tissue structure has been poorly researched. It has been shown that even if the grafts consist of necrotic adipose tissue, a better retention rate is obtained with an increased graft volume.¹³ Therefore understanding cell death and developing therapies to prevent fat graft necrosis are essential.

Historically, necrosis has been characterized as an uncontrollable form of cell death that occurs when severe stimuli trigger it.^14-16 Recently, several research studies have confirmed that genetically encoded molecular mechanisms regulate necrosis. Necroptosis, a form of programmed necrosis, is a controlled cell death process that is closely linked to elevated inflammation through the production of damage-associated molecular patterns (DAMPs), including HMGB1, nuclear DNA, mitochondrial DNA, and ATP from necroptotic cells.^17-19 The activation of mixed lineage kinase domain–like pseudokinase (MLKL) and receptor-interacting protein kinases (RIPK) 1 and 3 causes necroptosis. To create ribosomes, RIPK1 recruits and phosphorylates RIPK3. The ribosomes then phosphorylate MLKL to create necroptotic vesicles, which are complexes made up of RIPK1, RIPK3, and MLKL. MLKL (p-MLKL) is further phosphorylated by RIPK3.²⁰ Following conformational changes and oligomerization, p-MLKL is moved to the plasma membrane's phosphatidylinositol-rich patch, where it forms macropores. This causes cell enlargement and membrane rupture, which results in cell death and the release of chemicals dependent on DAMPs inside the cell.^21-23 It is thought that these substances are responsible for initiating and enflaming inflammatory processes. The DAMPs released from necroptotic cells bind to pattern recognition receptors on innate immune cells, causing them to release proinflammatory cytokines.^24-26

After grafting, adipocytes die quickly under ischemic conditions, causing fat necrosis within the cystic cavity, which eventually becomes surrounded by dense fibrous tissue.²⁷ Recently, as a result of ischemic diseases, necroptosis is reported to play an important role. Acute kidney ischemia-reperfusion injury can be alleviated by inhibiting necroptosis, ventricular remodeling after myocardial infarction can be improved, and ischemic stroke–related neuron loss can be reduced.²⁸ Therefore, we hypothesized that fat necrosis after fat grafting was closely related to necroptosis in adipocytes. Inhibiting necroptosis could reduce adipocyte necrosis, thereby alleviating the development of fat necrosis and fibrosis after grafting. In this study, we first identified and demonstrated that adipocytes underwent necroptosis after fat grafting and that the occurrence of fibrosis could be prevented by inhibiting cell necroptosis.

METHODS

Patients

From May 2022 to December 2022, 3 female patients were selected for this study. All of them developed nodules or calcification in their breasts after breast reconstruction with autologous fat grafting for micrognathia. Tissue samples were obtained from patients with fat necrosis after autologous fat grafting, fixing and processing them for routine diagnosis. An informed consent was obtained, and the Nanfang Hospital of Southern Medical University's Institutional Review Board approved the study according to the Declaration of Helsinki. Before participating, each participant signed an informed consent form.

Animals

Animal ethics approval was obtained from the Nanfang Hospital Animal Ethics Committee and all studies were conducted in compliance with national standards developed by the National Health and Medical Research Council of China. Male C57/BL6 mice were obtained from Southern Medical University.

Fat Grafting Model and Treatments

Eight-week-old C57/BL6 WT male mice weighing 25 to 30 g were provided by Southern Medical University. We randomly assigned mice to 1 of 4 groups, Control, DMSO, Nec-1s, or GSK′872, in a temperature-controlled, 12-hour light-dark cycle environment. Pentobarbital sodium 50 mg/kg intraperitoneally was administered to anesthetize the mice. The experimental grouping was based on the principle of completely randomized design.²⁹ Adipose tissue was obtained from the inguinal fat pads on both sides of the mice, and the obtained adipose tissue was employed to prepare adipose grafts according to the accepted Coleman method.^30,31

With a 1-mL syringe, we injected 0.3 mL of fat graft at the unilateral lumbar dorsum of mice that had been anesthetized and placed on a heating pad. The selection of the 0.3-mL injection dose was supported by findings from previous studies, including the work of Kokai et al, which indicated that this specific dose was more effective in assessing the histological survival rate of the graft during the transplantation process.¹³ The animals were left on a heating pad for 15 minutes for recovery. There was an ad libitum supply of food and water.

Starting the day after transplantation, a total volume of 0.1 mL of phosphate-buffered saline (PBS) + 1% dimethylsulfoxide (DMSO); PBS + 1% DMSO + 5 mg/kg Nec-1s (Selleckchem, Houston, TX); and PBS + 1% DMSO + 5 mg/kg GSK′872 (Selleckchem, Houston, TX) were injected every 2 days around the transplanted area for the DMSO, Nec-1s, and GSK′872 groups for 30 days, respectively. The Control group was not treated. The type of inhibitor, concentration administered, and frequency of administration were based on previous studies; Nec-1s and GSK′872 were ultimately selected.^32-35 The choice of administration route took under consideration that the local area after fat grafting was an ischemic and hypoxic environment, and the effect of water feeding or intraperitoneal injection might be poor; therefore subcutaneous injection around the transplantation was preferred.

At 7, 14, 30, and 90 days postoperatively, fat grafts were obtained from the back of each of the 4 groups of mice after execution by the decapitation method. Photographs of each sample were captured. The tissue volume was measured with the liquid overflow method: a PBS syringe was filled with the graft, and the volume was measured by increasing the level of the liquid. The measured volume of the graft was quickly placed in a 4% formaldehyde solution.

Transcriptome Analysis

For RNA-seq analysis (3 biological replicates per group), RNA was extracted from adipose tissue before and 1 week after grafting to investigate the effects of necroptosis on fat grafting. RNA-seq analysis was accomplished with Novogene (Beijing, China), and total RNA extraction from the tissues was done with TRIzol (ABSIN, Shanghai, China). Transcriptome sequencing was performed with Illumina HiSeq X Ten (Novogene Bioinformatics Technology Co., Ltd., Beijing, China). According to DESeq2, differentially expressed genes (DEGs) were identified by fold change (|FC|) ≥ 1.5 and an adjusted P value >.05. Gene set enrichment analysis was performed with R, with the fgsea (fast gene set enrichment analysis) library to analyze pathway enrichment. The heatmap and volcano plot were then constructed to visualize the DEGs with the “heatmap” and “ggplot2” R packages. KEGG and gene ontology (GO) enrichment analyses was performed with the R cluster Profiler package, and the gene set enrichment analysis was done with GSEA (version 4.1.0) software.

Histological Analyses

For hematoxylin and eosin (H&E) and Masson staining, samples were fixed in paraformaldehyde (4%) for 24 hours, and paraffin-embedded sections were obtained. Histological parameters including the presence of cysts and vacuoles were examined under a light microscope (BX51; Olympus, Tokyo, Japan) and evaluated with the method described by Shoshani et al, including the presence of cysts and vacuoles.³⁶ Fibrosis-positive areas were measured with ImageJ software (National Institutes of Health, Bethesda, MD).

For immunohistochemical analyses, adipose tissue slices were subjected to immunohistochemical examination after treatment with rabbit anti-mouse p-MLKL primary antibody (1:100; Cell Signaling Technology, Beverly, MA). The slices were magnified under a light microscope (BX51). The extent of p-MLKL–positive regions was measured with ImageJ software.

For immunofluorescence analysis, adipose tissue sections were incubated with the primary antibody, rabbit anti-mouse perilipin (1:100, Abcam, Cambridge, UK). The coverslips were then sequentially labeled with species-specific fluorescent secondary antibodies and DAPI (Sigma-Aldrich, St. Louis, MO). The samples were observed with a fluorescence microscope (Olympus, Tokyo, Japan).

Western Blot Analysis

Proteins were extracted with the Protein Extraction Kit (Biocolors, Shanghai, China). Steady-state levels of RIPK1, p-RIPK1, RIPK3, p-RIPK3, MLKL, and p-MLKL were determined by immunoblotting. Briefly, with sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), 50 g of total protein extract and soluble or insoluble protein fractions were separated from the proteins. Incubations at 4°C were performed with primary rabbit polyclonal antibodies sourced from a Mouse Reactive Necroptosis Antibody Sampler Kit (Cell Signaling Technology, Beverly, MA) against RIPK1, p-RIPK1, RIPK3, p-RIPK3, p-MLKL, and MLKL. After washing with PBS 3 times and 1 hour of incubation at 4°C with a horseradish peroxidase–conjugated secondary antibody (Bio-Rad Laboratories, CA), the blots were rinsed. A 5% milk solution was given to block the membranes after electrophoretic transfer to them. The immunocomplexes were detected with a Western Breeze Chemiluminescent Detection Kit (WB7108; Thermo Fisher Scientific, MA) following secondary antibody incubation. Internal controls were performed with β-actin. The film was scanned, and the gray intensity of the target band was analyzed with ImageJ software.

Statistical Analyses

The statistical program GraphPad Prism (GraphPad Software, Inc., La Jolla, CA, USA) was utilized to examine the data. Mean ± standard deviations expressed the data. To assess statistical significance, a t test with independent samples and a 1-way or 2-way analysis of variance (ANOVA) were conducted with Bonferroni's post hoc analysis. Statistical significance was set at P < .05.

RESULTS

Necroptosis Markers Increase in Patient Fat Necrosis Samples After Grafting

All 3 patients selected were female. The mean age of the patients at the time of surgery was 37 years (range 30-46 years) and the mean body mass index (BMI) was 20.10 kg/m² (range 17.82-23.21 kg/m²). No patient smoked or had comorbidities such as diabetes mellitus that could affect fat grafting viability. The follow-up time from the fat grafting procedure to the acquisition of the samples was 6 months in all cases. During patient follow-up 6 months after fat grafting, specimens of fibrotic necrotic fat nodules were collected and analyzed. The macroscopic appearance of the nodular tissue removed by the incision revealed that the nodular tissue consisted of a cystic cavity filled with yellow oily mud-like contents and a thick fibrous layer of the cystic wall. The cystic cavity was affected by fat necrosis (Figure 1A-1C). Histopathological examination revealed that the cystic wall layer consisted of a large amount of fibrous connective tissue with scattered portions of healthy adipose tissue containing round adipocytes. Adipocytes were irregularly shaped closer to the cystic lumen. Infiltration of various inflammatory cells, including lymphocytes and macrophages, was also observed. It was suggested that the site of cell death and inflammatory response occurred at the junction of the cyst lumen and wall (Figure 1D, 1E). A large accumulation of fibrous tissue is an important feature of fibrous nodules in adipose grafts (Figure 1F, 1G). Moreover, the entire tissue was disorganized. The samples were analyzed for the necroptosis marker p-MLKL, and it was found that the expression level of p-MLKL in fat necrosis samples was higher than that in normal adipose tissue (Figure 1H-1J). The areas positive for the marker were mostly located at the junction of the fibrous capsule wall and lumen, and the shape of the positive adipocytes was mostly irregular. Combined with adipocyte morphology, we can assume that the bulk of necroptotic cells were partially localized in adipocytes. This suggested that necroptosis was associated with fat grafting prognosis.

Transcriptomic Analysis Suggested That Necroptosis Was Involved in the Fat Grafting Process

Tissue from the inguinal fat of normal mice and fat grafts of mice a week after grafting were analyzed with RNA-seq to understand the adverse effects of fat graft regeneration. In fat grafts obtained 1 week after grafting, out of the 23137 genes detected, 6007 and 6156 genes were significantly upregulated and downregulated, respectively (adjusted P value < .05, log2 (fold change) > 1) (Figure 2A, and the Appendix, available online at www.aestheticsurgeryjournal.com). Enrichment plots from GSEA showed that necroptosis was activated after fat grafting and that genes related to the necroptosis pathway were upregulated (Figure 2B, 2C). These findings highlighted the importance of necroptosis in fat grafting. GO enrichment analysis was performed on the DEGs (Figure 3A). DEGs are involved in biological processes such as positive regulation of proteolysis and regulation of inflammatory responses and apoptotic signaling pathways. DEGs are involved in cellular components such as the inflammasome, caspase, and pore complexes. In addition, DEGs have molecular functions, such as death receptor and cytokine activity and heat shock protein and cytokine receptor binding. These biological processes, cellular components, and molecular functions are involved in cell rupture, release of intracellular substances, production and release of DAMPs, and the inflammatory response after the onset of necroptosis. The expression levels of the key genes RIPK1, RIPK3, and MLKL, which are closely related to necroptosis, were also significantly upregulated after grafting (Figure 3B) (P < .05). The expression levels of DAMPs, including Cxcl2, HMGB1, S100a8, S100a9, Nlrp3, and IL33 and proinflammatory factor–related genes released after necroptosis, including TNF, IL6, IL17a, and IL10, were activated (Figure 3C, 3D). Taken together, the transcriptomic analyses indicated that necroptosis was involved in the fat grafting process and might play an important role in its outcome.

Necroptosis Was Activated After Fat Grafting in Mice

To validate our findings in human samples, we created animal models for fat grafting. The inguinal fat of normal mice and the fat grafting samples were collected for analysis on days 7 and 14. The expression of p-MLKL was also detected (Figure 4). At days 7 and 14 after grafting, the amount of p-MLKL–positive cells in the adipose tissue was significantly greater than before. Immunohistochemistry (IHC) showed that only the adipose tissue of mice that underwent fat grafting was positive for p-MLKL at days 7 and 14 postgrafting, suggesting that necroptosis occurred only in the postgrafting adipose tissue. The amount of p-MLKL–positive areas observed was significantly higher at Day 14 than that at Day 7 postgrafting (P < .05). This suggested that the onset of necroptosis progressed over the 14 days after grafting. Most cells in the p-MLKL–positive areas had irregular cell morphology and were interspersed with normal adipocytes, indicating that the positive areas were located in non-normal adipocytes. Moreover, IHC at Day 14 postgrafting revealed that most of the irregular cells in the positive area were located near the oil capsule and the fibrous tissue, therefore we hypothesized that the necrotic contents in the oil capsule were an unfavorable factor for adipocyte survival. These data suggested that necroptosis of adipocytes occurred in the grafted adipose tissue.

Blocking Necroptosis Prevented Adipocyte Death in Fat Grafts in Mice

To assess whether blocking necroptosis reduced fat necrosis, we evaluated p-MLKL levels after necroptotic inhibitor Nec-1s and GSK′872 therapy in a mouse fat grafting model (Figure 5B-5R). Immunostaining of fat grafts from the Control and DMSO groups revealed numerous p-MLKL–positive adipocytes. However, the number of p-MLKL–positive adipocytes significantly decreased in Nec-1s and GSK′872 treated groups compared to controls. In both the Control and DMSO groups, the positive areas were gradually increasing on days 7 and 14 postgrafting, peaking on Day 14, and then progressively less on days 30 and 90, indicating that necrosis was aggravated before Day 14 and progressively decreased thereafter. The distribution of necrotic areas was not obvious on Day 7, whereas on Day 14 and thereafter the necrotic areas were mainly distributed near oil sacs and fibrous tissues. The DMSO and Control groups did not differ significantly (P < .05). Contrarily, in the Nec-1s and GSK′872 groups, the positive areas were irregularly dispersed among normal adipocytes and significantly reduced from Day 7 to the end of observation on Day 90. In comparison with GSK′872, Nec-1s showed no statistically significant difference (P < .05). At Day 90, the absence of statistically significant differences in the positive areas between the 4 groups indicated that, in the mouse model, the grafts stabilized, and no necroptosis occurred. Perilipin levels were assessed following treatment with the necroptotic inhibitor Nec-1s (Figure 6A-6F). Immunofluorescence analysis revealed a substantial increase in perilipin-positive adipocytes within the fat grafts of the Nec-1s group, while the number of perilipin-positive cells was significantly reduced in the Control group at Day 14 postgrafting. On Day 30, small neonatal adipocytes were observed to emerge adjacent to the existing adipocytes in the Control group, however the total cell count was notably lower compared to the Nec-1s group. By Day 90, the adipose grafts in both groups demonstrated regional stability, yet the Control group exhibited a significantly lower number of viable adipocytes in comparison to the Nec-1s group. Supplemental Figure 1, available online at www.aestheticsurgeryjournal.com, illustrates the fluorescent staining of the 4 groups at various time points. Overall, both the Control group and DMSO group exhibited lower expression of perilipin compared to the Nec-1s group and GSK′872 group at corresponding time points. On days 7 and 14, a significant decrease in adipocyte count was observed in the Control group and DMSO group in comparison to the Nec-1s group and GSK′872 group, indicating a higher incidence of cell death. By Day 30, small newborn adipocytes were observed surrounding the surviving adipocytes in all 4 groups. By Day 90, the adipose grafts in all groups reached stability, with the Control group and DMSO group showing a significantly lower number of surviving adipocytes compared to the Nec-1s group and GSK′872 group. The specific type of cell death occurring in the cells beyond those that had initially survived following inhibitor application remained unclear. The experimental results suggested that the use of the necroptosis inhibitor led to increased adipocyte survival. Western blotting also revealed a significant decrease in p-RIPK1/RIPK1, p-RIPK3/RIPK3, and p-MLKL/MLKL expression in the Nec-1s and GSK′872 groups compared to the Control group (Figure 6G-6J). No significant differences were observed between the Control and DMSO groups. This confirmed that GSK′872 and Nec-1s significantly inhibited the development of necroptosis after fat grafting (P < .05).

Blocking Necroptosis-Attenuated Fat Graft Necrosis and Fibrosis in Mice

Grafts were obtained on days 7, 14, 30, and 90 after the fat grafting. From the general diagram, it was observed that the volume of grafts was larger in the Nec-1s and GSK′872 groups compared to the Control group (Figure 7). As a result, the volume measurements showed that the Nec-1s and GSK′872 groups had significantly higher graft volumes than the Control group (P < .05). There was no significant difference between the Control and DMSO groups, however. On days 7 and 14, the grafts from the 4 groups showed a similar appearance and texture, with the grafts covered with a thin fibrous capsule membrane. From Day 30 to Day 90, the Control and DMSO groups showed a significant reduction in fat graft volume. The fat graft volume in the Nec-1s and GSK′872 groups were reduced by a lower percentage than that of the Control and DMSO groups at the same time point. Among the 4 groups, the fat graft retention rates at the 90-day time point were higher for the Nec-1s and GSK′872 groups compared to the Control and DMSO groups (68%±8% and 71%±13% vs 49%±8% and 48%±4%, respectively, P < .05).

A higher number of oil sacs in the grafts indicated more necrotic fat. Histological analysis showed that on Day 7, oil sacs were found in the grafts of all 4 groups; however, the number of oil sacs in the Control and DMSO groups was greater than that in the Nec-1s and GSK′872 groups at all 4 time points (Figure 8). The trend in the number of oil sacs in the grafts of the 4 groups was roughly similar, showing a gradual increase on days 7 and 14, a peak on Day 30, and a decrease on Day 90 (P < .05).

To analyze extracellular matrix (ECM) remodeling in the tissues, the 4 groups were stained with Masson's trichrome method. On Day 7, no significant fibrous tissue production was observed in any of the 4 adipose graft groups (Figure 9). Starting on Day 14, the Control and DMSO groups produced more fibrous tissue than the Nec-1s and GSK′872 groups. By Day 30, the difference between the 4 groups was at its maximum, with the Control and DMSO groups producing significantly more fibrous tissue than the Nec-1s and GSK′872 groups, accompanied by an expansion in the population of oil vesicles and disorganization of the tissue structure. By Day 90, some fibrous tissues were gradually replaced with newly formed tissues. The Control and DMSO groups had fewer fibrous tissues compared to Day 30 (P < .05).

Regarding the histological assessment of tissue integrity, oil sac formation, and the presence of fibrosis, notable differences were observed between the Nec-1s and GSK′872 groups as compared to the Control group. However, no significant variations were noted between the Control and DMSO groups (P < .05). These data suggest that fat graft necrosis and fibrosis can be effectively prevented by regulating necroptosis.

DISCUSSION

Due to its wide source and ease of access, fat grafting is widely applied in tissue reconstruction, however complications such as fat necrosis have greatly limited its use.^1,2,37,38 Fat grafting in the breast has a 27.8% complication rate overall, with fat necrosis accounting for 43.7% of all complications.⁶ Chronic inflammation and progressive calcification due to fat necrosis are the worst outcomes of fat grafting.¹⁰ Therefore, preventing graft fat necrosis is particularly important for improving fat grafting outcomes. However, the process of the development and progression of fat necrosis in grafted fat tissue is not clear. As far as we are aware, this is the first study to consider this question. Our study provides evidence that necroptosis occurs in transplanted adipose tissue, leading to the progression of fat necrosis in the adipose grafts. Necroptosis occurs in adipocytes after adipose tissue grafting. RIPK3-mediated phosphorylation of MLKL at the necrosome occurs before MLKL's translocation to the plasma membrane, where it compromises membrane integrity and causes cell death. DAMPs, such as Cxcl2, HMGB1, S100a8, S100a9, Nlrp3, and IL33, are released, attracting immune cells to trigger an inflammatory response and stimulate fibroblasts to initiate fibrogenesis, resulting in fibrosis. Our results showed that inhibitors targeting RIPK1 and RIPK3 were effective in reversing necroptosis, lowering inflammation and fibroblast activation and reducing fibrosis (Figure 10).

According to current theories, it is only after fat grafting that mature adipocytes die due to inadequate oxygen and nutritional support from the recipient tissue, and only a few adipocytes survive and are replaced by stem cells derived from newly differentiated adipose tissue .²⁷ In addition, fat necrosis is related to graft volume; the higher the injection volume, the higher the chance of fat necrosis. Persistent fibrosis and calcification are associated with the histology of retained necrotic fat. Unabsorbed necrotic fat may have multiple manifestations, such as oily cysts, sclerotic induration, and calcified solid tumors. Therefore, we focused on adipocyte viability after fat grafting. Necroptosis is a recently identified proinflammatory programmed mode of cell death that involves ischemia-induced injury during lung or liver transplantation.^39,40 Activation of necrosis was evidenced by immunoassays showing increased RIPK1 expression and the downstream phosphorylation of RIPK3 and MLKL. As shown in this study, a significantly higher expression of p-MLKL in adipocytes was observed in clinically grafted fibrotic-necrotic fat nodules than in normal fat tissue. In addition, p-MLKL–positive areas were mostly located at the junction of the fibrous capsule wall and the lumen, and the shape of the positive adipocytes was mostly irregular.

When fat necrosis remains unresolved, fibrosis occurs due to fibrinogen or collagen secretion from damaged blood vessels or activated fibroblasts. This leads to the formation of dense fibrous tissue surrounding the localized necrotic fat within the cystic cavity. The link between necroptosis and tissue and organ fibrosis is well established in the lungs, liver, and kidneys.^33,41,42 By inhibiting necroptosis, fibrosis in the corresponding organs can be alleviated, the degree of fibrosis can be reduced, and organ function can be maintained. This includes liver, pulmonary, myocardial, and interstitial fibrosis.^{15,33,34,43-47} However, there has been little research on its role in fat grafting. In our study we found that necroptosis could still be detected in the fibrotic-necrotic fat nodules in the grafted area after fat grafting. Although monitoring could not be achieved in the short postoperative period, it can be hypothesized based on the available experimental results that necroptosis is involved in the fat grafting process.

To validate these findings in human samples, we conducted animal experiments to simulate the biological processes of human fat grafting with a fat-grafting animal model. Transcriptomic analysis was first performed for normal mouse adipose tissue and fat grafts 7 days after fat grafting. The reason that the transcriptomics analysis only considered samples from the time point of Day 7 was because we only needed to derive evidence that the necroptotic pathway was activated, and subsequent evidence for longer periods of time was demonstrated by in vivo experiments. Enrichment plots from GSEA showed that necroptosis was activated after fat grafting and that genes related to the necroptosis pathway, including RIPK1, RIPK3, and MLKL, were upregulated. DAMPs, including Cxcl2, HMGB1, S100a8, S100a9, Nlrp3, and IL33, as well as activation of proinflammatory signaling pathways, including TNF, NF-kappa B, and chemokine signaling, were associated with higher levels of necroptotic RNAs. This was likewise corroborated by the high expression of inflammation-related factor genes. HMGB1 is thought to be 1 of the DAMPs, and others have shown that its expression is elevated after necroptosis and inflammatory responses in cells.^48,49 The level of HMGB1 can be reduced by regulating necroptosis, thereby reducing inflammatory response and fibrosis. This process is achieved through the RIPK3/MLKL/HMGB1 and TGF-β1 signaling pathways. This hypothesis was confirmed by the high expression of HMGB1 in the sequencing results of our study.^50-53 The specific role of HMGB1 in the fat grafting process, whether inhibition of HMGB1 can successfully alleviate the fibrotic process of fat grafts, and the signaling pathway responsible for this mechanism remain to be elucidated. Furthermore, IL33 is an “alarmin” with a proven role in immunomodulation and inflammation in many human diseases.^54,55 During asthma development, IL33 produced by bronchial epithelial cells can exacerbate asthma by enhancing macrophage necroptosis and inflammatory responses.⁵⁶ In atherosclerosis, IL33 increases cellular oxidative stress in macrophages, leading to necroptosis and enhanced release of atherogenic factors. Therefore, the IL33/ST2 pathway may play an important role in this process.⁵⁷ This also provides other possibilities for further exploration of fat necrosis and fibrosis formation during grafting, highlighting the importance of necroptosis and its related molecular pathways in the fat grafting process.

Next, we detected necroptosis-related markers in normal mouse adipose tissue and grafts at days 7 and 14 after transplantation. The p-MLKL expression is closely associated with necroptosis and has been widely utilized as a necroptosis marker.^58,59 The results of the current immunohistochemical experiments showed that p-MLKL was abundantly expressed in adipose tissue at days 7 and 14 after transplantation compared to normal mouse adipose tissue. This further validates the results of the transcriptomic analysis.

As mentioned previously, necroptosis can be modulated, and by intervening in this process, the reversal of fibrosis and restoration of organ function can be achieved. Inhibition of necroptosis results in decreased p-MLKL expression and cell death.⁶⁰ Currently, there are a large number of necroptosis inhibitors, some of which have entered clinical trials; from these we selected Nec-1s and GSK′872s, based on previous studies.^34,61-63 In our study, the expression of p-MLKL in adipose tissue was significantly increased after grafting. This result suggested necrosis of fat cells. The expression of p-MLKL was significantly reduced by the inhibition of the key necroptosis factors, RIPK1 or RIPK3. Taken together, the inhibition of necroptosis reversed the death of grafted adipocytes. Western blot analysis revealed that the levels of all 3 phosphorylated proteins associated with necroptosis, p-RIPK1, p-RIPK3, and p-MLKL, decreased after treatment with inhibitors. Likewise, the inhibition of necroptosis elevated the retention of fat grafts. Histological analysis also revealed that the experimental group to which the inhibitor was applied had a more homogeneous tissue structure, fewer oil sacs, and reduced fibrosis compared to the Control and DMSO groups.

The proposal to study the mode of cell death as a key determinant of the outcome of fat grafting opens up a new line of thought in the study of fat grafting, compared to the previous focus on fat graft retention. Recently, there has been a greater clinical focus on protecting adipocyte activity in fat grafts. As reported by Li et al, the results of this study showed that adipocyte viability and structure were critical components of fat grafting, and to some extent live adipocytes have positive effects.⁶⁴ Preventing adipocyte death is also a feasible strategy for reducing the levels of vacuoles, necrotic areas, inflammation, and fibrosis during fat grafting. However, we are also clearly aware that it is not appropriate to attribute all fat necrosis during fat grafting to necroptosis. Programmed death is not only a kind of necroptosis. It has also been shown that the survival of fat grafts is correlated with apoptosis.⁶⁵ Our research could not prove the specific form of fat necrosis during fat grafting. It could be fully demonstrated that necroptosis plays an important role in this process. Regulating necroptosis can indeed improve the outcome of fat grafting.

Previous studies have also described that not only adipocytes are involved in the process of fat grafting, but also endothelial cells, smooth muscle cells, pericytes, white blood cells, fibroblasts, mast cells, preadipocytes, and adipose-derived stem cells (ADSCs). In previous studies, our team also found that macrophages were involved in the formation of fibrosis after fat grafting.⁶⁶ This showed that although fat grafts contained a large number of adipocytes, fat grafting was not a single cell–involved process. It was a complex process involving multiple cells and might be related to the formation of fibrosis after fat grafting. This is also one of the important directions for exploring the mechanism of cell death and fibrosis formation during fat grafting: evaluation of the synergistic or antagonistic effects between multiple cells. ADSC therapy also improves the efficiency of fat grafting, and the exact effect of ADSCs has been fully confirmed.^65,67 However, the problem of rapid death of ADSCs has not been solved, which to some extent affects the efficacy of ADSCs and the final outcome of fat grafting. Inflammatory and immune responses in the recipient area may induce SC death.⁶⁸ We also noticed the important role of ADSCs in previous studies. Therefore, exploring the evolution of ADSCs in fat grafts, whether ADSCs undergo necroptosis, and whether there are effective interventions are important research directions for our team in the future.

Adipocyte survival plays an imperative function in fat grafting, and the inhibition of cell mortality can contribute to graft retention, offering a new solution for the prevention of fat graft–associated necroptosis. There are also some clinical studies that report on applying inhibitors of necrotic apoptosis to treat related diseases. For example, GSK2982772 (Compound 5), a clinical candidate that is a small molecule inhibitor of RIPK1, is undergoing phase 2a clinical studies for the treatment of psoriasis, rheumatoid arthritis, and ulcerative colitis.⁶⁹ It is believed that more inhibitors of necrotic apoptosis will be administered in the clinical and fat grafting fields in the near future.

This study had several limitations. First, because the species differences between animal models and human fat grafting are not well understood, we cannot be certain that our results in animal models will be the same in human patients. In this study, necroptosis was still detected in human 6-month fat grafts (Figure 1C, 1D). In sharp contrast, the outcomes of the animal experimentation analysis of necroptosis-associated markers in the 4 groups of samples collected at the 3-month time point did not have statistically significant differences between them (Figure 4A, 4B). This may be related to the route of administration, that is, whether the inhibitory effect persists after cessation of administration because of localized administration. Although it was mentioned in the previous section that local administration may be preferable to systemic administration given the ischemic and hypoxic environment of the transplanted area, further validation is needed. In future studies, we will further investigate the effect of the route of drug administration on the experiments. Second, we have not established that the manner in which fat is obtained and handled has an effect on cell viability and the degree of necroptosis, so we cannot include this in our interpretation of the results. Last, the length of time during which necroptosis happens in human fat grafts is to be elucidated in future studies. Our findings contribute to a deeper understanding of the critical role of necroptosis in fat grafting and help improve graft retention in clinical practice. Future studies should focus on addressing the limitations described.

CONCLUSIONS

In conclusion, in this study we showed that necroptosis plays a key role in fat necrosis and fibrosis. Fat necrosis and fibrosis can be attenuated by necroptosis regulation. In our study we provide novel ideas for the prevention of clinical complications, particularly fat necrosis and fibrosis, after fat grafting.

Supplemental Material

This article contains supplemental material located online at www.aestheticsurgeryjournal.com.

Acknowledgments

The authors thank the Institute of Research Center of Clinical Medicine, Nanfang Hospital, for providing comprehensive experimental services.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

This work was supported by the National Natural Science Foundation of China (82202476, 82072197) and the Science and Technology Projects in Guangzhou, China (202201011571).

REFERENCES

Author notes