Mechanical micronization of lipoaspirates for treatment of bleomycin-induced scleroderma in nude mice

Scleroderma is a chronic autoimmune disease that causes hardening of the skin. Adipose tissue is used as a regenerative treatment for scleroderma, and mechanical micronization of adipose tissue can concentrate stem cells, improving its therapeutic ecacy. Stromal vascular fraction gel (SVF gel) is produced by subjecting lipoaspirates to a series of mechanical processes. The present study aimed to assess the therapeutic effect of SVF gel on dermal scleroderma. in nude mice by daily subcutaneous injections of for lipoaspirates analyses, and qRT-PCR were conducted 4 weeks post-transplantation. therapies in the scleroderma animal model suggests the potential clinical application of this approach in scleroderma patients.


Introduction
Scleroderma is a chronic autoimmune disease that results in hardening of the skin, vascular pathologies, and generation of autoantibodies [1]. Scleroderma causes various cutaneous manifestations, including morphea and en coup de sabre [2]. Although not life-threatening, its manifestations are di cult to conceal, and lead to disability and decreased quality of life. At present, there is no consensus treatment for scleroderma skin lesions.
Fat grafting has signi cant potential as a regenerative treatment in various diseases [3,4]. The therapeutic effect of fat grafting was also tested in scleroderma patients, identifying robust bene cial effects [5]. Nicolas et al. injected various fat-derived products into the skin lesions of a mouse scleroderma model, demonstrating the regenerative potential of fat grafts, which improved vascular formation and suppressed brosis in scleroderma lesions [6]. However, the mechanisms for the therapeutic effects of transferred fat in scleroderma lesions are incompletely understood.
Several studies have demonstrated that the regenerative effect of transferred fat is associated with the abundance of ASCs in the tissue [7]. Several studies demonstrate that ASCs exert potent immunosuppressive effects via paraendocrine mechanisms [8]. The pathogenesis of scleroderma involves a complex interplay between vascular damage, in ammation, and brosis. ASCs upregulate IL-10 and VEGF, inhibiting the autoreactivity of T-cells, which helps to maintain self-tolerance, and inhibits the Th1 and Th17 responses, thereby decreasing the production of pro-in ammatory cytokines [9].
Macrophages are important cells for the induction of scleroderma. Factors released from innate immune in ammatory cells in the early stages of disease lead to the formation of myo broblasts that produce collagens, resulting in brosis [10]. Thus, increasing the ASC level in fat grafts might contribute to improving its regenerative effect for scleroderma lesions.
Cell-assisted lipotransfer (CAL), which increased stem cell concentration in transferred fat by directly mixing additional stem cells, has been proved to have a better survival rate than traditional fat grafting [11]. Another way to concentrate the ASCs in lipoaspirates is mechanical process, including chopping, shredding, pureeing, and mincing [12,13].We previously reported a re ned method for processing lipoaspirates. Stromal vascular fraction gel (SVF gel), an injectable mixture, was generated from lipoaspirates with a simple mechanical process. This process destroyed most of the adipocytes by intersyringe shifting, and removed oil by centrifugation [13]. The product is rich in ASCs, vascular endothelial cells (ECs), and adipose extracellular matrix (ECM), which improved wound healing and scars. It has been reported that the concentration of ASCs in SVF-gel reached 2.7±0.3x10 5 cells/mL [13].
In the present study, we hypothesized that the immunosuppressive and therapeutic effects of transferred fat were dependent on the relative abundance of ASCs. Two different ways to increase ASC concentration in fat tissue were applied in this study: SVF gel (increased ASC concentration by a mechanical method), and CAL (increased ASC concentration by an enzymatic method). These products were used to compare with Coleman fat (CF) to evaluate their therapeutic, anti-in ammatory, and anti-brotic effects in BLMinduced scleroderma in nude mice.

Animals
All experiments were approved by the Nanfang Hospital Animal Ethics Committee and were conducted in accordance with the guidelines of the National Health and Medical Research Council of China. Eight-week-old pathogen-free male nude mice were obtained from Southern Medical University, Guangzhou and Nanjing University, Nanjing, P. R. China. All mice were maintained on a 12 h day/night cycle under speci c-pathogen-free conditions, and were fed with normal chow and water ad libitum.
Bleomycin and fat graft treatment Bleomycin (BLM; Sano -Aventis, Paris, France) was administered to the nude mice as previously [14]. Brie y, BLM was dissolved in 0.9% phosphate buffered saline (PBS) at a concentration of 300 μg/mL. Thirty mice (8 weeks old) were injected with BLM (100 μL, 30 μg) subcutaneously into both anks using a 26 G cannula daily for 4 consecutive weeks.
Preparation of products from fat grafts Human lipoaspirates were obtained from three healthy women, 33.4 ± 6.3 years of age (mean ± standard deviation (SD)) and body mass index (BMI) 23.2 ± 1.9 g/m2, with no systemic diseases. All volunteers underwent liposuction with a 3 mm multiport cannula, containing sharp side holes 1 mm in diameter at -0.75 atm of suction pressure. The lipoaspirates were centrifuged at 1200 x g for 3 min to remove the oil and liquid fractions. The middle layer after centrifugation was considered CF.
Red blood cell lysis buffer (Leagene Biotechnology, Beijing, China) was used to remove erythrocytes. Supplemental cells (250,000 cells) were added in a small volume (20 μL) to a larger volume of fat (5 mL), which was considered CAL.
SVF gel was generated as described previously [13]. Brie y, CF was mechanically emulsi ed by shifting CF between two regular disposable syringes connected by a female-to-female Luer-Lok connector with an internal diameter of 1.4 mm at a rate of 10 mL/sec for 1 min. The emulsi ed CF was centrifuged at 2000 x g for 3 min. Finally, the substance under the oil layer was de ned as SVF gel.

Fat product injections
Mice were anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg). Fifteen mice treated with BLM for 4 weeks as described above were injected subcutaneously with CF, CAL and SVF gel on both anks of skin lesion (Fig. 1). Six BLM-treated mice received PBS injection in the lesions as a blank control. Animals were sacri ced 4 weeks after injection of fat graft preparations. Lesion skin and fat grafts were harvested.

Page 5/21
The lesion skin were xed in paraformaldehyde for at least 48h and embedded in para n. The sections (5 µm) were dewaxed using graded xylene and ethanol. Masson trichrome (MT) staining was conducted according to the manufacturer's instructions. The dermal thickness and collagen area were calculated with Image J software (National Institutes of Health, Bethesda, Md.) to assess the relative degree of brosis.
Immunohistochemical analysis of lesion samples was performed following standard procedures. Brie y, endogenous peroxidase activity was blocked by immersing sections in 3% H 2 O 2 in methanol for 30 min.
Antigen retrieval was performed by steaming slides in 0.01 M citrate buffer (pH 6.0) for 30 min. Slides were blocked in 1% bovine serum albumin for 30 min at room temperature, and subsequently incubated at 4°C overnight with antibodies against α-smooth muscle actin (α-SMA). Slides were then incubated at 37°C for 1 h with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG secondary antibody.
Between each incubation, sections were washed three times with PBS. Sections were developed with 3,diaminobenzidine tetrahydrochloride and hydrogen peroxide, and subsequently counterstained with hematoxylin. Sections were imaged using a Nikon Eclipse 80i microscope (Nikon, Badhoevedorp, Netherlands).
For immuno uorescence staining, the following primary antibodies were used: rat anti-mouse Mac2

Statistical analyses
Data were expressed as means ± standard error. The results were analyzed with repeated analysis of variance. Independent Student's t-test (two groups at a single time point) and one-way analysis of variance (more than two groups at a single time point) were performed. A value of p < 0.05 was considered to indicate a statistically signi cant difference.

Pathological changes of BLM-induced scleroderma
Daily subcutaneous BLM injections for 4 weeks induced stable skin sclerosis (evaluated at 4 and 8 weeks post-BLM treatment). MT staining revealed that all BLM-treated sites developed stable sclerosis (Fig. 2B) compared with untreated mice ( Fig. 2A). A dense collagen network was observed, and elastic bers were more abundant and denser in the dermis of BLM-treated skin than in that of control skin. The dermis was signi cantly thickened in BLM-treated skin (mean dermal thickness = 391 ± 33.73 μm in BLM-treated mice vs. 152.0 ± 15.23 in untreated mice, P<0.01 (Fig. 2C)). The thicknesses of subcutaneous fat tissue were signi cantly decreased in BLM-treated mice (P<0.01, (Fig. 2D)). These manifestations were indicative of typical scleroderma changes, and were similar to histopathological features reported previously for skin brosis [15]. Body weights of the BLM-treated mice (4th week) were signi cantly decreased compared with untreated mice (P<0.01, (Fig. 2E)). The weight loss continued for 4 weeks after withdrawal of BLM treatment, but did not affect viability.

Histopathological changes after treatment with fat preparations
After different fat preparation treatments for BLM-induced scleroderma, MT staining was used to evaluate the therapeutic effects of fat-derived products. PBS injection did not affect sclerosis, as PBSinjected lesions demonstrated a characteristically dense collagen network and thickened skin. However, all fat-derived cell therapy products signi cantly decreased sclerosis. In the SVF gel and CAL groups, the dermis exhibited decreased condensed collagen and increased follicles and vessels, which was more characteristic of normal skin. (Fig. 3A). Dermal thicknesses of samples from all four groups were measured (Fig. 3B)

Sclerosis lesion in ammation
Macrophages contribute to the onset of scleroderma and were associated with the progess of the disease [16]. M2 macrophages are the speci c type of macrophage closely related to collagen deposition and brosis [17]. In the BLM-induced scleroderma model, macrophages are signi cantly elevated and tend to be M2-polarized. Thus, CD206 was used as an M2 macrophage marker to measure levels of M2 macrophages in skin lesions (Fig. 4A). In PBS-treated scleroderma lesions, high numbers of CD206 + macrophages in ltrated. Injection of SVF gel and CAL signi cantly decreased CD206 + macrophage levels in skin lesions (vs. PBS group, P<0.05, Fig. 4B). SVF gel exhibited the most robust effects in relieving macrophage in ltration (P<0.05, Fig. 4B). However, injection of CF did not signi cantly reduce skin lesion CD206 + macrophage numbers (vs. PBS group, P>0.05, Fig. 4B).
Skin lesion in ammation was further evaluated by qRT-PCR. Expression of Mcp-1, a chemokine related to macrophage recruitment, was signi cantly downregulated in the SVF gel and CAL groups relative to the PBS group, and was lowest in the SVF gel group (P<0.05, Fig. 4D). Mcp-1 expression was decreased in the CF group relative to the PBS group, but was still signi cantly higher than that in the SVF gel and CAL groups (P<0.05, Fig. 4D). Il-6 is a pro-in ammatory cytokine and exhibited a similar expression trend to that of Mcp-1 (Fig. 4C).

In ammation of transferred fat tissue
Mac-2 (a pan-macrophage marker) and CD206 (an M2 macrophage marker) double immuno uorescence was used to identify the macrophage levels and polarization in the fat grafts. Few Mac-2 + /CD206 + M2 macrophages and Mac-2 + /CD206macrophages were present in the SVF gel group (Fig. 5A). CF exhibited the most severe macrophage in ltration (Fig. 5A). In ammation level in transferred fat was further evaluated by qRT-PCR. Expression of Mcp-1 was signi cantly decreased in the SVF gel group relative to other groups (P<0.05, Fig. 5B). CAL exhibited lower expression of Mcp-1 than did CF (Fig. 5B). Il-6 exhibited a similar expression pattern among groups to that of Mcp-1 (Fig. 5C).

Apoptosis in transferred fat
Apoptotic cells in transferred fat were detected by TUNEL assay. Apoptotic nuclei were irregularly distributed in the stromal sites around mature adipocytes in the transferred fat (Fig. 6A). The number of apoptotic cells in the CF group was signi cantly higher than that of the SVF gel and CAL groups (P<0.05, Fig. 6B). Apoptotic cell numbers did not differ between the SVF gel and CAL groups.

Skin lesion myo broblasts
In skin lesions, -SMA + spindle-shaped cells are de ned as myo broblasts, which play a key role in brosis during the scleroderma disease process [18]. In the CF group, signi cantly more myo broblasts (red arrows) appeared in the dermis within the brotic lesions (Fig. 7A). SVF gel exhibited the lowest level of skin lesion myo broblast in ltration among the three groups (Fig. 7A). The relative abundance of myo broblasts differed signi cantly between groups (P<0.05, Fig. 7B). Tgf-β expression [19] was measured in the lesions and transferred fat in the three groups (Fig. 7C). Tgf-β expression was higher in transferred fat than in skin lesions, which was likely associated with active ECM remodeling after fat grafting. In the CF group, transferred fat Tgf-β was higher than in the other groups. Skin lesion Tgf-β levels exhibited a similar trend between groups (Fig. 7C).

Discussion
Scleroderma is a clinically heterogeneous disease with a complex phenotype. Generally, scleroderma is characterized by autoimmunity, vascular dysfunction, tissue brosis, and in some cases, internal organ pathologies [20]. Skin lesions present on most patients and cause dis gurement [21]. Thus, treatment of skin lesions is a priority for clinical management of scleroderma. Topical corticosteroids are the standard topical treatment for localized scleroderma (LS) [2]. However, corticosteroids should be applied only during the active phase of disease, and total treatment time should be restricted to 3 months. This protocol has not been fully evaluated in well-controlled clinical trials. Other topical therapies such as methotrexate [22] and calcipotriol [23] target in ammation and brosis. Conventional treatments are generally well-tolerated but are not su ciently effective [24].
LS is characterized by a markedly thickened and brotic acellular dermis in the form of plaques or macules, which are packed with dense collagen and extracellular matrix proteins, accompanied by in ammatory cell in ltration, loss of the microvasculature, and subcutaneous fat atrophy [25]. Lipotransfer lls in and repairs the subcutaneous tissue defects of LS [26]. Adipose tissue contains abundant ASCs, which have well-characterized angiogenic, anti-in ammatory, and anti-apoptotic properties [27]. Accordingly, we suggest lipotransfer as an optimal treatment for LS.
Lipotransfer has already been explored as a therapeutic approach for reversing brosis in a number of brotic conditions [28]. In 2012, Ould-ali et al. demonstrated that implantation of human adipose tissue signi cantly improved brosis in sclerotic skin lesions of the BLM-induced mouse model, suggesting autologous adipose tissue as a promising novel therapy for scleroderma [29]. Clinical studies have further demonstrated the promising therapeutic potential of adipose implantation for sclerosis [30]. Fat grafting can alleviate skin sclerosis, facial handicap, mouth opening limitations, sicca syndrome, and hand impairment by improving skin pliability and quality [5]. Thus, fat grafting is e cient and welltolerated for the treatment of localized forms of scleroderma skin lesions.
Current exploration of the therapeutic bene ts of fat grafting for scleroderma focused on the effects of ASCs. Fat-derived products vary between conventional CF, CAL, and mechanically processed fat grafts such as SVF gel. Because CF has been extensively applied, many fat-derived products have been developed to condense ASCs for not only improved volume retention, but also for improved therapeutic e cacy in wound healing and skin diseases [27,31]. CAL attaches ASCs to the aspirated fat, converting ASC-poor aspirated fat to ASC-rich fat [32]. As separating ASCs introduces exogenous collagenase and leaves them vulnerable to the attack of immune system in recipient sites, the isolation procedure is legislatively restricted for clinical use. However, SVF gel mechanically destroys and eliminates most mature adipocytes within the adipose tissue, getting rid of employing the collagenase and retaining the ECM scaffold. The whole process can be accomplished in 2 hours and ensures a high stromal vascular fraction cell density (>4.0 × 10 5 cells/mL) and ASCs (2.7±0.3x10 5 cells/mL) synchronously [13].
Although adipose-tissue-based therapies are clinically effective in scleroderma, the relative e cacies of conventional CF and ASC-enriched preparations such as CAL and SVF gel had not yet been evaluated [33]. The present study was designed to compare the therapeutic e cacies of these preparations in the nude mouse BLM-induced scleroderma model.
The present study demonstrated the therapeutic e cacy of three different fat-derived products in induced scleroderma skin lesions. All three products signi cantly decreased dermal thickness and reversed dermal sclerosis. SVF gel and CAL deceased collagen and increased follicles and vessels in BLM-induced scleroderma, and the lesions were closer to the appearance of normal skin. All three experimental groups reduced dermal thickness. CAL signi cantly alleviated dermal sclerosis, although SVF gel had the most robust therapeutic effect. Both SVF gel and CAL successfully decreased in ammation in scleroderma lesions. SVF gel and CAL treatments signi cantly decreased Mcp-1, Il-6, and Tgf-β in transferred fat, with parallel effects in skin lesions. Further CAL and SVF gel treatment signi cantly decreased macrophage in ltration into scleroderma lesions. SVF gel more signi cantly decreased local in ammation than other treatments. By fragmenting large pieces of adipose tissue and removing thick brous tissue, the SVF gel isolation process removes more lipid droplets, which could release in ammatory mediators, and leaves very few at or fragmented adipocytes. Accordingly, the density of ASCs increases markedly after mechanical processing in SVF gel [12].
Though all treatment groups were fat-derived therapies, the prominent effect of SVF gel in the MCFinduced scleroderma model could be related to high level of ASCs in this preparation. ASCs possess antibrotic and regenerative properties in many brotic conditions, including scars, radiation-induced brosis, and scleroderma [34]. ASCs downregulate expression of the pro-brotic markers Tgf-β, α-SMA, CTGF, NF-κB, IL-1, and IL-13, and upregulate expression of anti-brotic broblast growth factor and proangiogenic VEGF and HGF [35]. Topical application of ASCs to animal brosis models softens and thins brotic tissue, and decreases dermal collagen content with a looser and better-organized pattern of residual collagen bers with lower expression of collagens 1-6 [36]. The anti-brotic properties of ASCs could be attributed to remodeling of the brotic matrix, tilting the balance between ECM deposition and ECM degradation in favor of degradation [37]. These results suggest that ASCs are a promising therapeutic agent for localized scleroderma. In a prior study using the BLM-induced scleroderma mouse model, ASC therapy signi cantly attenuated dermal brosis, and reduced skin thickness and hydroxyproline levels, accompanied by decreased Tgf-β [38].
Macrophages are activated in patients with skin disease, including systemic sclerosis, and are potentially important sources of brosis-inducing cytokines [39]. The transferred adipose tissue was under severe ischemic environment and some fragile cells underwent apoptosis. This process recruited numerous macrophages to the injection sites. This study revealed that SVF-gel and CAL group had less cell apoptosis and macrophage in ltration. The local accumulation of macrophages in the transferred fat under the skin lesion may affect the in ammation level in skin lesion. The SVF-gel and CAL group, which had decreased cell death and restrained in ammation level, had lower macrophage in ltration level in the skin lesion.
ASCs were capable to inhibit activated macrophages and mitigate in ammation [40]. Dou et al. demonstrated that ASCs alleviate lung in ammation and skin brosis via immunomodulation in a bleomycin-induced murine model [41]. M2 macrophages are closely related to inhibition of collagen deposition and brosis. Several studies have demonstrated the capacity of ADSCs to promote phenotype switching in macrophages, inhibiting expression of M1 polarization genes. Consistent with previous reports [28], SVF gel decreased macrophage in ltration and pro-in ammatory cytokines, indicative of decreased in ammation. Further, apoptotic adipose cells and skin lesion brosis were decreased in mice treated with SVF gel. These parallel results in transferred fat and skin lesions may indicate a correlation between internal fat graft in ammation and host skin lesion in ammation and brosis in scleroderma patients.

Conclusions
The present study assessed the e cacy of three different adipose-based therapies in scleroderma. SVF gel and CAL signi cantly reversed dermal sclerosis that developed after 4 weeks of BLM treatment. CF modestly decreased dermal sclerosis but did not signi cantly affect the other measured parameters.
Macrophage in ltration and in ammatory cytokine levels in both transferred fat and skin lesions were lowest in mice injected with SVF gel. SVF gel also exhibited the most potent anti-brotic effects. The e cacy of adipose-derived therapies in the scleroderma animal model suggests the potential clinical application of this approach in scleroderma patients.