Abstract Decellularized Adipose-Derived Matrix (DAM) has the function of inducing adipogenesis, but the distribution of adipogenesis is uneven. We found for the first time that DAM contains two structural components: The tough fibers DAM (T-DAM) and the fine fibers DAM (F-DAM). T-DAM was a dense vortex structure composed of a large number of coarse fibers, characterized by myoblast-related proteins, which cannot achieve fat regeneration and forms a typical "adipose-free zone". While the F-DAM was a loose structure consisting of uniform fine fibers, has more matrix-related proteins and adipose-related proteins. It can not only better promote the adhesion and proliferation of adipose stem cells in vitro, but also achieve the regeneration of adipose tissue in vivo earlier and better, with a uniform range of adipogenesis. The F-DAM is the main and effective kind of DAM to initiate adipose tissue regeneration, which can be picked out by ultrasound fragmentation. Keywords: Decellularized adipose-derived matrix, Fiber, Structure, Adipose regeneration Graphical abstract [27]Image 1 [28]Open in a new tab Highlights * • The decellularized adipose-derived matrix is not a uniform scaffold, and it is classified for the first time. * • DAM can be divided into tough fibers DAM (T-DAM) and fine fibers DAM (F-DAM). * • F-DAM is an effective DAM to initiate adipose tissue regeneration. 1. Introduction At present, tissue engineering regeneration and repair reconstructions can be divided into the following three types: one is transplantation of other body parts into the target area [[29]1,[30]2], the other is the scaffold or patch alone or with cells, growth factors, or exosomes [[31][3], [32][4], [33][5]], and the third is regeneration in situ of scaffolds or analogs [[34]1,[35]6,[36]7]. The first method is currently the most commonly used, while regeneration in situ has always been the ideal way to repair tissue defects. It does not require donor tissue and avoids additional trauma to the donor area. Moreover, there is no in vitro culture process, avoiding complex processes such as cell variation, contamination, complex cell culture and tissue establishment. It also avoids problems such as transplant trauma, immunogenicity, infection, pathogen transmission, and medical ethics. Since tissue and blood vessels regenerate simultaneously, there are no problems with revascularization. Finally, it requires no secondary transfer operation and is economical and convenient. Decellularized Adipose-derived Matrix (DAM) is one of the few biological materials capable of inducing fat regeneration in situ. It does not require exogenous mesenchymal cells, but relies only on the recruitment of endogenous seed cells and the continuous induction of cell homing to the scaffold. DAM induces tissue regeneration rather than tissue repair, which means that it is eventually replaced by healthy, functional host adipose tissue [[37]8,[38]9]. In previous studies of adipose tissue regeneration induced by DAM, the effect of adipogenesis varied widely after transplantation in vivo. Even at the later stage of implantation, there is still uneven adipogenesis, meaning that adipogenesis does not occur in some areas [[39][10], [40][11], [41][12]]. This could be related to the different tissue sources or preparation methods of DAM. We found that in the DAM, which were prepared using conventional enzymatic methods, some swirling, dense fiber structures could be detected in the HE staining, which were clearly different from those of the loose fibers. These fibers, characterized by a spiral structure, are defined as "Tough DAM" (T-DAM), while the other, more delicate and loose fibers are defined as "Fine- DAM" (F-DAM). In previous studies, DAM was considered to be a unitary acellular scaffold. However, we have found that the structure and composition of DAM is not uniform. We inferred that the reason for the uneven range of adipogenesis in DAM might be the different distribution and proportion of T-DAM and F-DAM. However, what the difference is between the composition and structure of T-DAM and that of F-DAM. And what the difference is in the effect of inducing fat regeneration in vivo. The answers to these questions will not only have a significant impact on the effect and efficiency of DAM inducing adipose tissue regeneration, but may also change the previous experimental conclusions and thus our understanding of DAM. To this end, the morphological structure, composition, and ability of adipogenesis in T-DAM and F-DAM were investigated. 2. Methods 2.1. Preparation of decellularized adipose-derived matrix Adipose tissue was obtained from 10 healthy female patients, aged 37.27 ± 4.86 years with a BMI of 20.93 ± 1.38 kg/m^2. The collection and use of adipose tissue was reviewed and approved by the Ethics Committee of the Plastic Surgery Hospital of the Chinese Academy of Medical Sciences (No. ZX201843). Separation of tough and fine fibers: DAM preparation was performed according to our previous method [[42]13]. Briefly, the coarse fibers were picked out and cleaned as much as possible with sharp forceps. Then, the adipose tissue was repeatedly crushed with ultrasonic cell crusher at 90W power under ice bath stirring. In this process, the tough fibers were aggregated into a group and easily separated. The remaining suspension was centrifuged at 4000 rpm for 3 min, and the fine fibers were obtained after the top layer of oil is discarded. Decellularized treatment: The separated tough and fine fibers were soaked in 1 % Triton-X100 solution in a constant temperature shaker (37 °C, 100 rpm) for 48h, respectively. And then rinsed with sterile distilled water under the shaker at step intervals of 10min, 30min, 1h, 2h and overnight (>12h). After adding the isopropyl alcohol on shaker for 6h to remove remaining lipids. Rinse with sterile distilled water and 75 % alcohol 3 times for 30min each time. Finally, T-DAM and F-DAM were obtained and stored in 1 % penicillin-streptomycin solution at 4 °C. 2.2. Morphological and histological contrast The morphology of adipose tissue, T-DAM and F-DAM were observed. Comparison of structure and composition was performed on DAMs from three batches (n = 3). The DAMs were fixed with 4 % paraformaldehyde and cut into 4 μm thick sections after paraffin embedding. Hematoxylin-eosin (HE) staining was performed according to standard procedures, including dewaxing, soaking, staining, differentiation, blue staining, dehydration, transparency, etc. For Sirius red staining, the paraffin slices after dewaxing were immersed in the Sirius red dye solution for 8 min after deparaffinization, and the images were taken under polarized light after dehydration and transparency. Type I collagen is visualized as an orange or red fiber, while type III collagen presented as a green one. For immunofluorescence staining, after dewaxing, antigen repair and sealing, the paraffin sections were incubated with anti-laminin (Abcam, Cambridge, UK) and the secondary antibody CoraLite594 (Proteintech, Wuhan, China). The results of fluorescence staining were observed after panoramic scanning using the Microdigital scanning system (Motic EasyScan Pro 6). 2.2.1. Scanning electron microscopy The DAMs were fixed with Gluta fixing solution (P1126, Solarbio, Beijing, China) at room temperature for 2h. After fixation, gradient dehydration and drying, gold spraying for 30s, the two DAMs were observed and photographed by scanning electron microscope (SU8100, HITACHI Ltd., Tokyo, Japan). 2.3. Glycosaminoglycan quantification Glycosaminoglycan (GAG) is considered the most important active protein of DAM. Quantification was performed using the Blyscan GAG detection kit (Biocolor, United Kingdom). DAMs were freeze-dried and incubated with papain (25 mg/ml) at 60 °C for 3 h. Papain digested samples were analyzed according to the manufacturer's instructions (n = 3). Absorbance was measured at 650 nm, 3 compound holes were repeated for each sample, and GAG content was quantified using a standard curve. 2.4. Growth factor quantification The concentrations of bFGF and EGF in both DAMs were measured using the Luminex® Multifactor Assay Kit (R&D Systems®, LXSAHM-04) (n = 3). After homogenization, total tissue protein was quantified and a standard curve was generated. According to the manufacturer's instructions, 50 μl of standard and a diluted protein sample were added to each test well, followed by 50 μl of diluted biotin-antibody mixture and 50 μl of diluted PE-labelled streptavidin. Incubation was performed on a microplate oscillator (IKA, Staufen, Germany) at room temperature for 1 h in the absence of light. After purification and suspension with the designated purification buffer, the assay was performed on a flow analyzer (LSRFortessa SORP, BD, NJ, USA) with 3 compound holes in each sample. After detection, the concentration of the growth factor was determined using the corresponding fluorescence value of the standard curve. 2.5. Biomechanical properties test The mechanical properties of the two DAMs were measured with a biomechanical tester (Instron 5967, Norwood, MA). Using a corneal trephine, the F-DAM and T-DAM specimens were cut into cylindrical pieces 6 mm in diameter and approximately 2 mm thick (n = 3) and compressed at 10 mm/min using a 10 N load sensor. The Young's modulus was calculated from the slope of the initial linear portion of the stress-strain curve. The stress-strain curve was plotted using Origin9 after Green's strain transformation was applied. 2.6. Protein label-free quantification and bioinformatics analysis The F-DAM and T-DAM were subjected to label-free quantification (LFQ) proteomic analysis. After quantification of total BCA protein in two groups of samples (n = 3), protein reduction, alkylation and digestion were performed to obtain peptide segments. Then, 1 μg of total peptide was collected from each sample and separated by high-performance liquid chromatography (nanoUPLC EASYnLC1200, Thermo Scientific, USA). Data acquisition was performed in conjunction with a mass spectrometer equipped with a nanoliter ion source (QE-xactive HFX, Thermo Scientific, USA). Proteome Discoverer software (version 2.4.0.305, Thermo Fisher Scientific) and Sequest HT search engine were used for database search analysis, and L-DAM and S-DAM differential proteins were obtained. The screening criteria for differentially expressed proteins were p-value ＜0.05 of Student's t-test or chi-squared test, with fold change≥1.2 [[43]14]. Intensity-based absolute quantification (iBAQ) has been used to calculate the content of important proteins [[44]15,[45]16]. Gene Ontology (GO) annotation enrichment analysis, KEGG annotation enrichment analysis and gene set enrichment analysis (GSEA) were performed. 2.7. Co-culture of DAM with adipose-derived stem cells Adipose-derived stem cells (ADSCs) were extracted from the fat aspirates of the above patients. ADSCs were isolated by 0.1 % type I collagenase at 37 °C for 45 min. They were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10 % fetal bovine serum (FBS). P3 of ADSCs were selected as the research object in this project. 5 mg T-DAM and 5 mg F-DAM was taken, respectively. The DAMs in vitro were all derived from three batches of DAMs (n = 3). After the DAMs were washed with sterile PBS, 10 μl of cell suspension containing 1 × 10^5 ADSCs and DMEM were slowly added to each DAM. Subsequently, 490 μl DMEM was added to a 24-well plate for culture. The next day, the recellularized DAMs were transferred to a new well plate for further culture. 2.8. Evaluation of cell adhesion and cell activity to DAMs LIVE/DEAD® viability/cytotoxicity detection kit (Thermo Fisher) was used to observe cell adhesion of FDAM-ADSC group and TDAM-ADSC group on day 1 and day 7 after co-culture (n = 3). The confocal microscope (Leica, Allendale, N.J.) was used to capture images of the DAM scaffold for observation. The Image J software was used to calculate the number of adipocytes per field. 2.9. The ability to induce adipogenic differentiation in vitro FDAM-ADSC group and TDAM-ADSC group (n = 3) were co-cultured in 24-well plates for 7 days each and then cultured with human adipogenic mesenchymal stem cells adipogenic induction differentiation kit (HUXMD-90031, Oricell, CN) for 2 weeks. At the same time, FDAM without transplanted cells was used as a blank control. The ADSCs-DAM complex was fixed with 4 % paraformaldehyde, stained with oil red O solution, added with 100 % isopropyl alcohol solution, incubated at room temperature for 20 min, and the oil red in the complex was eluted into isopropyl alcohol. After the elution was uniformly blown, 100 μL eluent was absorbed and added to the 96-well plate. The OD value was measured at a wavelength of 520 nm using a microplate reader and zeroed by the blank control group. Lipid formation of the FDAM-ADSC group and the TDAM-ADSC group was observed by Bodipy and Hoechst fluorescence staining. Prepare a working solution of Hoechst 33342+Bodipy (Thermo, United States) according to the kit protocol, shake in the dark for 30 min, and rinse twice with PBS. The confocal microscope (Leica, Allendale, N.J.) was used to capture images for observation. The Image J software was used to calculate the number of adipocytes per field. 2.10. Animal experiment in vivo of adipogenesis on DAMs Balb/c nude male mice aged 6–8 weeks were used in this experiment (Huafukang, Beijing, China). The study was approved by the Ethics Committee of the Plastic Surgery Hospital of the Chinese Academy of Medical Sciences [2023(2)]. Animal experimentation procedures strictly followed the regulation and standards for the protection and use of laboratory animals formulated by the Chinese Academy of Medical Sciences and Peking Union Medical College. T-DAM and F-DAM 50 μl were thoroughly cut, mixed with normal saline, and injected into the left and right subcutaneous back of mice using an 18-gauge needle. Mice were euthanized at 1, 4, and 8 weeks after injection (n = 5 at each time point). 2.11. Animal experiment in vivo of adipogenesis between powdered T-DAM and bulk T-DAM Considering the swirling scaffold structure in T-DAM may be the reason that makes it difficult to achieve adipogenesis. The T-DAM is lyophilized using vacuum freeze-dryer. Part of the T-DAM is cut with scissors into about 1mm3 particles, which is called bulk T-DAM (B-TDAM). The other part was ground into powder using a tissue grinder (Tissuelyser-24L, Jingxin Ltd., Shanghai, China) after freezing with liquid nitrogen. It was found that there was no obvious spiral structure in HE staining, which was named powder L-DAM (P-TDAM). Both T-DAMs were sterilized with cobalt-60 radiation before use. B-TDAM and P-TDAM 5 mg were fully mixed with normal saline and injected into the left and right back of mice, respectively. Implant samples were collected by euthanizing mice 4 weeks after injection (n = 5). 2.12. Histological and immunohistochemical staining After the specimens were fixed in 4 % paraformaldehyde for 24–48 h, the tissues were cut at the center of the longest diameter, embedded in paraffin and sectioned. HE was used to evaluate the morphology of the implants at each time point. The lipogenic marker perilipin-1 (Abcam, Cambridge, UK) was stained by immunohistochemistry. CD31 (Abcam, Cambridge, UK) was used to localize vascular endothelial cells. Panoramic scanning of sections was performed with Microdigital section scanning system (Motic EasyScan Pro 6). Image J software was used to calculate the area percentage of panoramic perilipin-1-positive adipocytes to assess in-implant adipocyte regeneration. Adipocytes were randomly selected to measure cell diameter (the longest diameter was calculated for non-circular adipocytes). The area percentage of CD31-positive vessels at 100x was calculated to assess in-implant angiogenesis. 2.13. Statistical analysis All data are expressed as mean ± standard deviation (SD). Statistical analysis was performed using Prism 9.0 software (Graphpad Software, USA). The independent samples t-test was used to evaluate the physical properties of the two DAMS. And the statistical significance of data collected over multiple time periods in vivo was analyzed at the 95 % confidence level using analysis of variance. p < 0.05 is considered statistically significant (*p < 0.05，**p < 0.01，***p < 0.001). 3. Results 3.1. Characteristics of T-DAM and F-DAM After the adipose tissue was crushed by ultrasound ([46]Fig. 1A), the initial DAM products were divided into two types: T-DAM (fibrous tissue) and F-DAM (suspension). Then, the acellular operation was continued by the enzyme-free method. DAM, which were prepared by different individual and liposuction methods, vary greatly in the ratio of T-DAM and F-DAM, ranging from 9:1–3:2. Both T-DAM and F-DAM are white solids when viewed in general ([47]Fig. 1B). In contrast, T-DAM is white in color and its tissue contains a hard, crimped and thickened part. Most of F-DAM is soft and yellowish flocculent. In HE staining, the corresponding structures of both DAMs can also be found in human white adipose tissue ([48]Fig. 1C). The F-DAM is a uniform fine fibrous structure, while the T-DAM contains a large number of vortex-like dense fibrous structures. In the scanning electron microscope, the structure of DAMs consists of small fibers that are intertwined and wound into large fiber clusters. It can be seen that the F-DAM has smaller fiber clusters and looser fiber structure, while the F-DAM has larger fiber clusters and more dense structure. Fig. 1. [49]Fig. 1 [50]Open in a new tab Macroscopic and microscopic characteristics of T-DAM and F-DAM. (A) The initial of T-DAM and F-DAM immediately broken and separated after ultrasonication. (B)General appearance comparison of T-DAM and F-DAM. Biomechanical curves of T-DAM and F-DAM. (C) HE staining of adipose tissue, red arrow for T-DAM, black arrow for F-DAM (Scale bar = 30 μm), SEM staining (Scale bar = 10 μm), Sirius red staining (Scale bar = 100 μm) and laminin immunofluorescence staining (Scale bar = 100 μm). (D)Elastic modulus of T-DAM and F-DAM.. (n = 3, each group of samples derived from three batches of DAMs.). (For interpretation of the references to color in this figure legend, the