CD54+ rabbit adipose-derived stem cells overexpressing HIF-1α facilitate vascularized fat flap regeneration

Fat flap transplantation is frequently performed in patients suffering from soft tissue defects resulting from disease or trauma. This study explored the feasibility of constructing vascularized fat flaps using rabbit adipose-derived stem cells (rASCs) and collagen scaffolds in a rabbit model. We evaluated rASCs proliferation, paracrine function, adipogenesis, vascularization, and CD54 expression, with or without HIF-1α transfection in vitro and in vivo. We observed that adipogenic differentiation potential was greater in rASCs with high CD54 expression (CD54+rASCs) than in those with low expression (CD54–rASCs), both in vitro and in vivo. HIF-1α overexpression not only augmented this effect, but also enhanced cell proliferation and paracrine function in vitro. We also demonstrated that HIF-1α-transfected CD54+rASCs showed enhanced paracrine function and adipogenic capacity, and that paracrine function increases expression of angiogenesis-related markers. Thus, CD54+rASCs overexpressing HIF-1α enhanced large volume vascularized fat flap regeneration in rabbits, suggesting CD54 may be an ideal candidate marker for ASCs adipogenic differentiation.


INTRODUCTION
Soft tissue regeneration via adipose tissue engineering is a growing field addressing current clinical demands related to various soft tissue defects [1][2][3][4]. Soft tissue substitutes are often required following disease or tissue trauma, such as in adipose or muscle tissues. Soft tissue defects may affect patient physical appearance, cause emotional disturbances, and/or impair function. Currently, autologous tissue flaps, fat, or commercially available fillers are commonly used in reconstructive surgeries, specifically for repairing facial tissue defects and tumor resection sites. Although these strategies achieve clinical success, subsequent volume losses and secondary donor-site deformities challenge long-term efficacies. Emerging tissue engineering strategies represent

Research Paper
innovative approaches for many clinical challenges. Such strategies include incorporating seed cells, biodegradable scaffolds, and microenvironments to provide specific inductive signals for tissue regeneration.
Adipose-derived stem cells (ASCs) represent a readily available supply of mesenchymal stem cells (MSCs) from lipo-aspirates [5][6][7]. ASCs exhibit enhanced proliferation and reduced senescence, and do not trigger immunological rejection [8]. ASCs may differentiate into chondrocytes, osteoblasts, adipocytes, myocytes, or neuronal-phenotype cells in vitro or in vivo [9][10][11][12][13]. Efforts directed toward promoting ASCs adipogenic differentiation have included growth factor or induction agent administration, other angiogenic methods, and the use of appropriate scaffolds [14][15][16][17][18][19]. ASCs exhibit poor adipogenic differentiation capacity under nonadipogenic-inducing conditions. We hypothesized that some cluster of differentiation (CD) markers could be closely related to adipogenic differentiation-related functions in ASCs. Our previous work showed that transfecting human breast adipose-derived stem cells (HBASCs) with CXCR4 effectively improved survival of free fat transplants [20]. We also found that ginsenoside Rg1 and platelet-rich fibrin co-administration improved HBASCs function in soft tissue regeneration engineering [21]. In this study, we purified CD54 + rASCs using immunomagnetic beads, transfected them with HIF-1α, and seeded them on three-dimensional, porous, collagen type I sponge scaffolds in vitro. Paracrine function was assessed prior to implantation. Finally, we assessed vascularized fat flaps constructed using HIF-1αtransfected CD54 + rASCs for optimizing adipose tissue regeneration in vivo.

RESULTS rASCs growth characterization, immunophenotype, and multipotency
Homogeneous rASCs growing in a monolayer with a spindle-shaped morphology were observed by phasecontrast microscopy after primary culture for 7-10 d, reaching 80-90% confluence ( Figure 1). These rASCs proliferated rapidly. During subculture, P3 rASCs reached the same level of confluence within 3-4 d with a 1:3 split ratio ( Figure 1). This demonstrated that rASCs resemble other ASCs in terms of morphology and proliferative capacity. rASCs expressed the mesenchymal surface markers CD29, CD44, CD54, CD73, CD90, CD105, and HLA-ABC, as determined by flow cytometry (Table 1). In contrast, < 7% of rASCs expressed the hematopoietic markers CD31, CD34, and CD45, and the histocompatibility antigen class II HLA-DR-DP-DQ. After DiI labeling, rASCs cytoplasms showed red fluorescence ( Figure 1). Subconfluent P3 rASCs were cultured with adipogenic, osteogenic, and chondrogenic induction media, and lineage-specific cell morphologies were observed following 2, 3, and 2 weeks of culture, respectively. Adipocytes, osteoblasts, and chondrocytes can be identified by positive oil red O, alizarin red, or alcian blue staining, respectively. Lineage-specific histological staining was performed using these dyes, and results confirmed that rASCs differentiated into adipocytes, osteoblasts, and chondrocytes following culture in the relevant induction medium (Figure 1). These results validated rASCs multipotency Effect of HIF-1α transfection on rASCs proliferation in vitro CCK-8 assays were performed on HIF-1αtransfected, blank virus-transfected, and control rASCs. Beginning 3 d after culture, HIF-1α-transfected cell proliferation rate was higher than that of the blank virustransfected and control cells, and HIF-1α-transfected cell doubling time was shorter than that of the blank virus -transfected and control cells (P < 0.001, Figure 2).

Effect of HIF-1α on rASCs adipogenic differentiation in vitro
The four rASCs groups exhibited different adipogenic potentials in vitro. CD54 + rASCs contained more large oil red O-positive lipid droplets in their cytoplasm, and HIF-1α transfection augmented this effect. However, CD54 -rASCs transfected with HIF-1α (group C) and untransfected CD54 -rASCs contained fewer oil red O-positive lipid droplets ( Figure 4). There were differences between groups A and B, groups A and C, groups A and D, groups B and C, groups B and D, and groups C and D in adipocyte density and lipid concentration ( Figure 4).

Neogenetic fat flaps macroscopic findings and histopathological assessment
No animals died during the 12 weeks following implantation. Macroscopic findings indicated that regenerated tissue flaps had formed, and these were excised carefully for testing ( Figure 5). Neogenetic fat flap wet weights were measured using an electronic balance ( Table 2). Differences were found between groups A and B, groups A and C, groups A and D, groups B and C, groups B and D, and groups C and D ( Figure 6). H&E staining results showed that regenerated tissue flaps in all groups were composed of mature adipose tissue. Compared with the other three groups, group A contained the most mature adipose tissue and showed no signs of fibrosis. Group B also consisted predominantly of mature adipose tissue and showed less fibrosis than groups C or D. Groups C and D partially consisted of mature adipose tissue and showed more fibrosis than groups A and B. Moreover, the collagen sponges were observed to have degraded over the course of 12 weeks after implantation in vivo ( Figure 7).

Microvessel density (MVD) of regenerated fat flaps
We histologically evaluated 10 fields/section from the centers of regenerated fat tissues to assess MVD. MVD was much higher in groups A than that of in the other three groups. In contrast, few microvessels were observed in group D. However, HIF-1α-transfected CD54 -rASCs (group C) promoted neovascularization more so than untransfected CD54 -rASCs (group D) or untransfected CD54 + rASCs (group B). Thus, HIF-1α transfection enhanced revascularization. (Figures 7-8).

Oil red O staining and quantitative measurement of adipogenesis
Red-fluorescent cells were observed in regenerated fat flaps by fluorescence microscopy, indicating that these mature adipocytes had differentiated from DiIlabeled rASCs ( Figure 7A4-7D4). Adipogenesis and lipid vacuole formation in neogenetic mature adipose tissue 163.5 ± 38. 9 *P < 0.01, **P < 0.01, # P < 0.01. sections were assessed by oil red O staining. CD54 + rASCs (group B) contained more large oil red O-positive lipid droplets in their cytoplasm, and transfection with HIF-1α (group A) augmented this effect. CD54 -rASCs transfected with HIF-1α (group C) and CD54 -rASCs contained fewer oil red O-positive lipid droplets (Figure 7). There were differences between groups A and B, groups A and C, groups A and D, groups B and C, groups B and D, and groups C and D in adipocyte density and lipid concentration ( Figure 8).

Adipogenesis-related genes and proteins expression in neogenitic fat flaps
Relative levels of PPARγ, C/EBPα, and ADD1 were detected by real-time qPCR and western blot. PPARγ expression was lower in neogenetic fat flaps of groups A (HIF-1α-transfected CD54 + rASCs) and C (HIF-1αtransfected CD54 -rASCs) than in the other two groups (untransfected rASCs). However, C/EBPα and ADD1 levels were increased in groups A and B compared with groups C and D (Figures 11-12).

DISCUSSION
Surgical reconstruction procedures using autologous fat flaps are effective treatment approaches for various soft tissue defects caused by congenital deformities, traumatic injuries, or tumor resection [22][23][24]. However, limitations in clinical practice include poor retention of autologous fat and the occurrence of secondary donor-site deformities.  . Histological evaluation of 10 fields/section taken from the center of the mature adipose tissue showed that the MVD was much higher in group A than in the other three groups, and MVD in group C was much higher than that in group B and A, as determined by the t-test C. *P < 0.01, **P < 0.01, # P < 0.01. HIF is a transcription factor induced under hypoxic conditions. HIF-1α migrates into the nucleus to form a stable heterodimer with HIF-1β or binds to histone acetylation enzymes to avoid degradation [25]. HIF-1α induces VEGF, platelet-derived growth factor (PDGF), and b-FGF expression, thereby enhancing angiogenesis and improving oxygenation of the environment [26][27][28][29][30][31]. However, the effects of HIF-1α in ASCs remained unclear. Some studies showed that HIF-1α is involved in cell proliferation during hypoxia [32][33][34].
This study assessed the adipogenic capacity of rASCs based on the CD54 marker, and found that rASCs with high CD54 expression exhibited stronger adipogenic differentiation potential in vitro. We examined rASCs proliferation and paracrine function with or without HIF-1α transfection in vitro and in vivo, and found that HGF, VEGF, ET-1, b-FGF, IGF-2, and HSP70 were upregulated in transfected rASCs. These results showed for the first time that rASCs modified by HIF-1α exhibit enhanced proliferation and paracrine function.
The transcription factor nuclear receptor PPARγ is an essential and sufficient factor for inducing adipocyte differentiation, and serves as a key marker of the adipogenic switch. However, in our work, PPARγ expression was decreased with rASCs adipogenic differentiation, which was promoted by HIF-1α transfection. Other work showed that hypoxic conditions enhanced the adipogenic potential of MSCs and that HIF-1α overexpression induced lipid droplet accumulation, while PPARγ inhibition upregulated adipocyte-specific gene expression [40]. These results demonstrated that HIF-1α-induced adipogenic differentiation is independent of the PPARγ-regulated pathway. Thus, HIF-1α may induce rASCs adipogenic differentiation through a C/EBPαdependent pathway. Based on our results, the PPARγ pathway may not be uniquely necessary for adipogenic differentiation.
ASCs directly form new vascular networks by differentiating into endothelial cells [41], and paracrine functions are also involved, e.g., in the secretion of angiogenic factors, such as VEGF, ET-1, b-FGF, and PDGF [42][43]. Our results showed that paracrine function and MVD were both increased in fat flaps containing modified rASCs. Angiogenesis is not the only process affected by paracrine signals; previous studies demonstrated that paracrine signals from MSCs promote cell proliferation, migration, and secretion independently of promoting angiogenesis [44]. Moreover, growth factors are necessary for efficient differentiation of ASCs into adipocytes [36]. Since HIF-1α-transfected rASCs not only exhibited enhanced proliferation and paracrine function, but also facilitated vascularized fat flap regeneration, HIF-1α may promote early transplant neovascularization and enhanced rASCs adipogenic differentiation ( Figure 13).

rASCs isolation, expansion, and identification
Twenty subcutaneous adipose tissue samples were obtained from the groin fat pad of adult rabbits under anesthesia. Fibrous tissue was excised from fat pads and minced well with scissors at room temperature (RT). Experimental methods describing rASCs expansion, culture, and identification were described previously [45][46].

HIF-1α-transfected rASCs
HIF-1α was introduced into a pAdEasy-1 adenoviral vector system. The virus was packaged, and titer was assessed. P5 CD54 + and CD54 -rASCs were infected with the HIF-1α-encoding adenovirus at 90% confluence, as determined by fluorescence microscopy. Prior to infection, cells were washed with infection buffer three times and incubated at RT for 10 min. Infection was carried out with a multiplicity of infection of 50 pfu/ cell in 1 ml of infection buffer at RT for 60 min. After infection, rASCs were incubated at 37°C with 5% CO 2 in a humidified atmosphere. Three d later, rASCs were subcultured for use in the following experiments.

Influence of HIF-1α on rASCs adipogenic differentiation in vitro
For adipogenesis assays, HIF-1α-transfected CD54 + rASCs (group A), CD54 + rASCs (group B), HIF-1α-transfected CD54 -rASCs (group C), and CD54-rASCs(group D) were cultured in 6-well plates with adipogenic induction medium containing 200 μM indomethacin, 10 μM insulin, 0.5 mM 3-isobutyl-1methylxanthine, and 1 μM dexamethasone at a seeding density of 500 cells/mm 2 for up to 14 d. Adipogenic differentiation was determined by oil red O staining. Briefly, cultured cells were air dried for 30 min, and then fixed in 10% formalin for 5 min. Cells were rinsed three times with distilled water, and immersed in absolute propylene glycol for 5 min to avoid transferring water into the oil red O. Cells were stained in oil red O solution for 8 min at 60°C in an oven, rinsed with 85% propylene glycol for 5 min,. rinsed with distilled water, and then stained in Gill's hematoxylin solution for 30 sec. Cells were thoroughly washed with running water for 3 min, rinsed twice with distilled water, and mounted with glycerin jelly or aqueous mounting medium. Upon observation of the samples, lipids appeared red and nuclei appeared pale blue. Adipocyte density for each sample was measured in 10 visual fields under the same magnification in a double-blind manner. Cell numbers were normalized to square millimeters. Lipids were extracted from cells with 100% isopropanol and gentle shaking for 5 min. Lipid concentrations were measured based on absorbance at 510 nm in triplicate for each sample.

rASCS-loaded scaffold implantation in vivo
All animal procedures were performed in accordance with the guidelines of the Guangxi Medical University Animal Care and Use Committee. Twenty New Zealand rabbits (aged 3-4 months old, weighing 2.0 ± 0.3 kg) were used to establish an autologous cell implantation model. CD54 + and CD54 -rASCs were cultured with adipogenic induction media for one week before implantation. Collagen scaffolds loaded with cell suspensions were essentially gel masses containing 1.0 ml of 1 × 10 7 /ml HIF-1α-transfected CD54 + rASCs (group A), 1.0 ml of 1 × 10 7 /ml CD54 + rASCs (group B), 1.0 ml of 1 × 10 7 /ml HIF-1α-transfected CD54 -rASCs (group C), or 1.0 ml of 1 × 10 7 /ml CD54 -rASCs (group D). All compounds were implanted into the muscular fasciae on the backs of rabbits. Each implantation site randomly received one of these four scaffolds.

Neogenetic fat flap wet weight measurement and histopathological examination
At 12 weeks post-implantation, all transplants (i.e., neogenetic fat flaps) were excised from the muscular fasciae and weighed using a standard electronic balance. Transplants were then fixed in formalin and embedded in paraffin. Tissue sections from the center of the dissected regenerative tissue biopsy were stained with hematoxylin and eosin (H&E) using standard procedures and examined by light microscopy. MVD was assessed by counting microvessels in 10 fields of each H&E-stained slide (40× magnification; performed by two blinded reviewers).

Oil red O staining and quantitative measurement of adipogenesis in vivo
Neogenetic fat flap tissues samples were embedded in paraffin, cut into 5-µm-thick slices, and observed by fluorescence microscopy. Cyrosections were stained with oil red O and washed twice with PBS. Adipocyte densities were measured in a double-blind manner in 10 visual fields under the same magnification. Cell numbers were normalized to square millimeters. Lipids were extracted from cells with 100% isopropanol and gentle shaking for 5 min. Lipid concentrations were measured based on absorbance at 510 nm in triplicate for each sample.

Real-time quantitative PCR (qPCR)
Total RNA was extracted from each group using TRIzol according to the manufacturer's protocol (Invitrogen, USA). First-strand cDNA was synthesized from 1 μg of RNA with viral reverse transcriptase (TaKaRa, Japan) and used for real-time qPCR. HIF-1α, VEGF, b-FGF, ET-1, IGF-2, HSP70, PPARγ, C/EBPα, and ADD1 levels were quantified using an ABI 7300 real-time PCR system (Applied Biosystems, USA) with SYBR green PCR reaction mix (TaKaRa, Japan). Primers are listed in Table 3. The following program was used: 95°C for 5 min, 40 cycles of 95°C for 15 sec, annealing for 1 min, and 72°C for 30 sec. A melting analysis and agarose gel electrophoresis were performed to confirm PCR product specificities. Relative gene expression was analyzed using the 2 -ΔΔCt method, normalized to GAPDH expression, and data are presented as fold change relative to the control.

Protein extraction and western blotting
Neogenetic fat flap samples from each group were harvested for western blot analysis, and whole cell extracts were obtained. Cell pellets were sonicated in extraction buffer, and extracts were quantified using a Bio-Rad DC protein assay kit (Bio-Rad, Hercules, California, USA). Equal amounts of protein were resolved by 4-12% SDS-PAGE and transferred to PVDF membranes (Millipore, Bedford, Massachusetts, USA). Membranes were blocked with blocking solution (Pierce, Rockford, Illinois, USA). The following primary antibodies (all anti-rabbit) were used: HIF-1α, VEGF, b-FGF, ET-1, IGF-2, HSP70, PPARγ, C/EBPα, and ADD1 (all from Abcam, London, UK). Horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence substrates (Supersignal

Statistical analysis
Data are shown as means ± standard deviations (SD). We used analysis of variance (ANOVA) to compare group means. This approach accounts for intra-and intergroup variation. When ANOVA of the four means revealed statistically significant differences, the means of any two groups were compared using paired t-tests. P < 0.05 (twotailed) was considered statistically significant. All data were analyzed using SPSS for Windows version 17.0 (Chicago, Illinois, USA).