Exosomes From ADSCs Attenuate Bleomycin-Induced Skin Fibrosis And Oxidative Stress In Scleroderma Via Circ-Zfyve9 Delivery


 Background: Systemic sclerosis (SSc) is autoimmune trait affecting several organs, which is identified by thickening of dermis and connective tissue affected by collagen accumulation, as well as vascular injuries inducing hypoxia. Methods: In this investigation, adipose-derived stem cells (ADSCs) were separated from the ADSC exosomes and a bleomycin-induced SSc mouse model was constructed. We employed high-throughput sequencing to study abnormal expression of circular RNAs (circRNAs) in SSc skin tissues with or without ADSC exosome treatment. The regulatory mechanism and targets were studied using bioinformatics analysis, luciferase reporting analysis, angiogenic differentiation experiments, and RT-qPCR detection. Results: ADSC exosome treatment prevented dermal thickening and fibrosis in bleomycin-induced scleroderma. In addition, circ-Zfyve9 was demonstrated to have an important function in ADSC exosome-mediated skin tissue protection. GPX4 and miR-135 were shown to be circ-Zfyve9 downstream targets. Overexpressing miR-135 or downregulating GPX4 reversed circ-Zfyve9 promotion effects rupon angiopoiesis by promoting lipidosome ROS in EPCs under hypoxic conditions. Overexpressing miR-135 or downregulating GPX4 reversed the circ-Zfyve9 inhibition effect on fibrosis in myofibroblasts under hypoxic conditions. Overexpressing circ-Zfyve9 increased the therapeutic effect of ADSC exosomes. Conclusions: Taken together, the present study results show that the exosomes from ADSCs attenuate bleomycin-induced skin fibrosis and oxidative stress in scleroderma via circ-Zfyve9 delivery.


Introduction
Systemic sclerosis (SSc) is well recognized as scleroderma, which is an autoimmune connective tissue disorder identi ed by adaptive and innate immunity dysregulation, microvascular damage, and generalized skin brosis, as well as multiple organs [1]. Skin brosis is thus an important SSc symptom. Pathological alterations in organs including heart, kidneys, lungs and gastrointestinal tract, determine possibility of clinical improvements. Ten-year survival was 55% in a diffuse cutaneous disorder cohort [2].
Fibrosis usually affects the mouth and face, leading to skin tightening, lip retraction, and irreversible scarring. The oral aperture that narrowed in uences speech, dental structure, psychological status and facial expression together with life quality [3].
Former investigations have illustrated that mesenchymal stem cells (MSCs) serve as the therapy target to halt SSc progression [4]. It is inferred that adipose-derived stem cell (ADSC) transplantation can mitigate SSc-related brosis pathogenesis [5][6][7]. Several studies have con rmed the therapeutic effect of ADSC exosomes on SSc via exosome delivery. Exosomes are small (~ 30-100 nm) membrane vesicles that are released from disparate classes of cells in response to speci c cellular stimulation [8]. They have a crucial messenger function in intercellular communication of delivering pre-wrapped cargoes, like miRNAs, circRNA, and proteins, to recipient cells [9]. Previous studies have found that ADSC-derived exosomes (ADSC-Exos) are internalized in broblasts and help cutaneous wound healing via promoting cell proliferation, collagen synthesis, and migration [10]. The ADSC exosome roles in SSc remain unknown. The present investigation aimed to reveal the regulatory role and mechanism of ADSC exosomes in SSc.

Animals and ethics statement
Animal Care and Utilization Committee at Huashan hospital of Fudan University supervised all animal procedures. Our lab purchased male C57BL/6 mice with four-week age from SLAC Laboratory Animal Co.
Ltd (Shanghai, China). We housed animals individually and independently in ventilated cages at 24-26°C and constant humidity in a 12-h light/dark cycle environment. Ethics Committee in Huashan Hospital of Fudan University oversaw all experiments. All surgical procedures were performed under anesthesia with an intraperitoneal injection of 30 mg/kg sodium pentobarbital.
ADSC culture, isolation, and identi cation Technician washed harvested adipose tissue samples from healthy subjects (consent was obtained from patients via a volunteer) or normal rats with phosphate-buffered saline (PBS), minced in 0.2% collagenase I (Sigma-Aldrich, St. Louis, MO, USA), and digested for one hour at 37°C with intermittent shaking. The samples were then cleaned and further digested in DMEM (Sigma-Aldrich, St. Louis, MO, USA) containing 15% FBS (Gibco BRL, Frederick, MD, USA). The tissues samples were centrifuged at 1000 rpm for ten minutes to erase mature adipocytes. Tissue pellets were subsequently resuspended in DMEM supplied with FBS of 15%, 100 μg/mL streptomycin, and 100 U/mL penicillin, which were cultured at 37°C with 5% CO 2 . We treated ADSCs at ~80-90% con uency with 0.02% ethylenediaminetetraacetic acid/0.25% trypsin (Sigma-Aldrich, St. Louis, MO, USA) for ve minutes under room temperature, which were then replated. Fluorescein isothiocyanate-conjugated CD90, CD29, CD105, vWF, and CD44 antibodies were used for phenotypic analysis. IgG-matched isotype was used as an internal control for every antibody. Our lab cultured normoxic ADSCs in 95% air (O 2 of 20%) and CO 2 of 5% .

Multilineage ADSC differentiation
To validate ADSC multilineage differentiation, third-passage mouse ADSCs were cultured in adipogenic differentiation medium (Sigma-Aldrich, St. Louis, MO, USA). They were then stained leveraging Oil Red O in two weeks or cultured in osteogenic differentiation medium (Sigma-Aldrich, St. Louis, MO, USA), which were stained utilizing Alizarin Red after three weeks.

ADSC-Exo identi cation and isolation
After getting ~80-90% con uency, we rinsed ADSCs with PBS and then cultured them in FBS-free endothelial cell growth medium-2MV supplied using 1 × serum replacement solution (PeproTech, NJ, USA) for two more days. We collected and centrifuged culture medium at 300 g for ten minutes and then at 2000 g for ten minutes to erase cellular debris and apoptotic cells. After centrifugation at 10,000 g for 0.5 h, we ltered supernatant using 0.22-μm lter (Millipore, Billerica, USA). Then, 15 mL of the supernatant were transferred to Amicon Ultra-15 Centrifugal Filter Unit (100 kDa; Millipore), which were centrifuged at 4000 g to achieve a volume of ~1 mL. Our team cleaned ultra ltration unit using PBS twice, which was centrifuged at 4000 g to achieve 1 mL volume. We resuspended exosome pellets in 500 μL PBS at 4ºC. Exosome protein content was detected employing the Pierce bicinchoninic acid (BCA) Protein Assay Kit (Thermo Fisher Scienti c, MA, USA). We maintained exosomes at -80°C until further use. We applied western blotting (WB) and transmission electron microscopy to characterize exosomes.
We determined size using dynamic light scattering with Nanosizer (Malvern Instruments, Malvern, UK). Size distribution with particle radius (nm) were represented on the x-axis and percentage on the y-axis.

Model treatment and establishment
We generated bleomycin-induced murine brosis model following a previous protocol [4]. Then, 100 μL bleomycin (1 mg/mL) were subcutaneously injected into shaved area on mouse's back (1 cm 2 ) with 27gauge needle. Our lab carried out injections daily for three consecutive weeks. Mice in control group received 100 μl of PBS.

Tubule formation assay
In vitro neovascularization was assayed in human brin matrices. Following treatment, serum-starved EPCs in endothelial basal medium were seeded onto six-well plates coated with Matrigel (10 5 cells/well; BD Biosciences, Franklin Lakes, NJ, USA), which were incubated at 37°C for 0.5 d. Tubular structures that formed in the Matrigel were observed and photographed using phase-contrast microscopy. The newly formed tube lengths in ten randomly selected elds per well were measured.

Histological examination
We xed skin tissue samples in each group in 10% formalin solution, which we embedded in para n. We stained thin 5-μm sections using Masson's trichrome, followed by CD31 staining to establish histopathological angiogenesis alternations. Our team tested sections utilizing Axiophot light microscope (Zeiss, Oberkochen, Germany) or ECLIPSE E600 uorescence microscope (Nikon, Tokyo, Japan), which were photographed.

Statistical analysis
Results were expressed as means ± standard deviation (SD). GraphPad Prism (GraphPad, La Jolla, CA, USA) was leveraged to analyze differences among groups. P-values ≤0.05 were regarded as statistical signi cance.

Characterization of ADSC-Exo
Our previous study demonstrated that exosomes have an important function in MSC-mediated wound healing [11]. In order to con rm whether exosomes from ADSCs function in bleomycin-induced skin brosis and oxidative stress. ADSCs isolated from mouse adipose tissue had classic cobblestone-like morphology ( Figure 1A). Immuno uorescence data demonstrated that ADSCs expressed cell surface markers CD44, CD105, CD29 and CD90, while not endothelial marker vWF ( Figure 1B Transmission electron microscopy illustrated that ADSC exosomes had ~100 nm size ( Figure 1J). Nanosizer analysis of exosome particle size also showed that exosomes isolated from ADSCs were ~100 nm in size ( Figure 1K). WB detection showed that ADSC exosomes expressed CD81 and CD63 ( Figure  1L), suggesting that nanoparticles were actually exosomes.

ADSC exosome treatment prevents dermal thickening and brosis in bleomycin-induced scleroderma
Masson's trichrome staining results in bleomycin-induced murine brosis model showed that collagen deposition was increased in bleomycin-induced scleroderma, while ADSC exosome treatment signi cantly decreased collagen deposition (Figure 2A and B). Immuno uorescence detection of ROS staining validated that ADSC exosome treatment decremented ROS accumulation signi cantly in bleomycin-induced scleroderma ( Figure 2C and D). Immunohistochemical CD31 staining showed that angiogenesis was decreased in bleomycin-induced scleroderma. ADSC exosome treatment signi cantly reversed the inhibitory effect of scleroderma on angiogenesis ( Figure 2E and F). WB for broblastassociated proteins TGF-β, collagen I, and α-SMA showed that ADSC exosome treatment signi cantly decreased collagen I, α-SMA, and TGF-β expression in bleomycin-induced scleroderma ( Figure 2G-J).
circ-Zfyve9 has an important function in ADSC exosome-mediated skin tissue protection Several studies have found that circRNA has an important function in organizational microenvironment regulation [12]. In order to determine whether ADSC exosome-mediated skin tissue protection via circRNA delivery. Our team utilized high-throughput sequencing to detect abnormal circRNA expression in scleroderma skin, with or without ADSC exosome treatment. Heatmap analysis demonstrated that ADSC exosome treatment resulted in different circRNA expression levels ( Figure 3A). A total of 574 upregulated and 1707 downregulated circRNAs were found in ADSC exosome-treated scleroderma skin when compared to scleroderma skin without treatment ( Figure 3B). RT-qPCR showed that seven circRNAs from the sequencing analyses were upregulated, including mmu_circ_0000376, mmu_circ_0001240, mmu_circ_0002870, mmu_circ_0009688, mmu_circ_0011768, and mmu_circ_0011922. Data illustrated that only the mmu_circ_0001240 expression was upregulated in ADSC exosome-treated scleroderma skin when compared to scleroderma skin without treatment ( Figure 3C). Bioinformatics analysis found that mmu_circ_0001240 originated from the Zfyve9 gene exon (2024 bp) and was located in chr4:108390393-108392417. The mmu_circ_0001240 was named circ-Zfyve9 ( Figure 3D).
Bioinformatics results demonstrated that GPX4 is miR-135 downstream target. To further show a correlation between GPX4 and miR-135, we incorporated WT or MUT 3'UTR-GPX4 sequences including an miR-135 binding sequence into luciferase reporter vector ( Figure 4D). Our team transfected luciferase reporter vector into HEK293 cells, with or not miR-135 mimic. Luciferase reporter analysis showed that miR-135 inhibited luciferase function in WT, yet not in MUT cell lines ( Figure 4E), verifying that GPX4 is an miR-135 target.
RT-qPCR results illustrated that circ-Zfyve9 expression was enriched in both EPCs and myo broblasts after transfection using circ-Zfyve9 overexpression vector. Treatment with an miR-135 mimic or GPX4 silencing had no effect on circ-Zfyve9 expression in EPCs and myo broblasts ( Figure 4F and G), suggesting that both miR-135 and GPX4 are located downstream of circ-Zfyve9. RT-qPCR detection also found that circ-Zfyve9 overexpression decreased miR-135 expression. GPX4 silencing had no effect on circ-Zfyve9-induced miR-135 inhibition ( Figure 4H and I), suggesting that miR-135 is located between circ-Zfyve9 and GPX4. The results also found that circ-Zfyve9 overexpression increased GPX4 expression, while miR-135 upregulation reversed the promotion effect of circ-Zfyve9 on GPX4 expression. GPX4 expression decremented signi cantly after transfection with a GPX4 silencing vector ( Figure 4J and K), advising that circ-Zfyve9 enhanced GPX4 expression through sponging miR-135. miR-135 overexpression or GPX42 downregulation reversed the promotion of circ-Zfyve9 on angiopoiesis in EPCs under hypoxic conditions EPC tube formation capability evaluations showed that circ-Zfyve9 overexpression promoted angiogenic differentiation of EPCs under hypoxic conditions. Overexpression of miR-135 or GPX4 silencing reversed the promotion of circ-Zfyve9 on the angiogenic differentiation of EPCs under hypoxic conditions ( Figure  5A and B), suggesting that miR-135 overexpression or GPX42 downregulation reversed circ-Zfyve9 promotion on angiopoiesis in EPCs under hypoxic conditions. Immuno uorescence detection showed that circ-Zfyve9 overexpression decreased liposomal oxidative stress by decreasing oxidized liposomes. Overexpression of miR-135 or GPX4 silencing reversed inhibitory circ-Zfyve9 effect upon hypoxia-induced liposome oxidative stress ( Figure 5C and D).
Overexpressing circ-Zfyve9 increased the therapeutic effect of ADSC exosomes Immunohistochemical detection of apoptotic skin tissues showed the ADSC exosomes after treatment with overexpressed circ-Zfyve9. Apoptosis in skin tissues of bleomycin-induced scleroderma was decreased when compared to the ADSC exosome treatment group. Overexpression of miR-135 or GPX4 silencing reversed the protective effect of circ-Zfyve9 exosomes on apoptosis inhibition in bleomycininduced scleroderma skin ( Figure 6A and B), suggesting that circ-Zfyve9 overexpression increased the therapeutic effect of ADSC exosomes.

Discussion
Several studies have found that ADSC transplantation therapy can reverse scleroderma-induced cutaneous brosis [13,14]. Exosome secretion by ADSCs plays an important role in microenvironmental regulation via delivery of bioinformatics molecule [15,16]. The present study found that ADSC exosome treatment avoids dermal thickening and brosis induced by bleomycin administration. In addition, ADSC exosome treatment inhibited scleroderma-induced ROS and reversed the inhibitory effect on angiogenesis in scleroderma, suggesting that ADSC exosomes have a therapeutic effect in scleroderma.
Accumulating studies have reported that circRNA has an important function in the regulation of the stress microenvironment [17]. Whether circRNA has an indispensable regulatory role in scleroderma remains unclear. High-throughput sequencing found that the ADSC exosome treatment group had an abnormal expression of circRNA compared to the scleroderma group. RT-qPCR analysis showed that mmu_circ_0001240 has an important function in ADSC exosome-mediated skin tissue protection.
Luciferase reporter analysis discovered that miR-135 can interact with GPX4. GPX4 catalyzes lipid peroxide reduction in complex cellular membrane environment [20]. Lipid peroxidation has been reported in knockout models, strengthening the GPX4 to protect cells from detrimental lipid peroxide effects [21,22]. In this study, circ-Zfyve overexpression promoted GPX4 expression. Overexpression of miR-135 or GPX4 silencing reversed circ-Zfyve protective effect upon angiopoiesis in EPCs under hypoxic conditions. Overexpression of miR-135 or GPX4 silencing also reversed circ-Zfyve inhibitory effect upon hypoxiainduced brosis in myo broblasts. Immuno uorescence detection showed that circ-Zfyve9 overexpression decreased liposomal oxidative stress by decreasing oxidized liposomes. However, miR-135 overexpression or GPX4 silencing reversed circ-Zfyve9 inhibitory effect upon hypoxia-induced liposome oxidative stress. This suggests that GPX4 and miR-135 are downstream targets of circ-Zfyve9. Overexpressing circ-Zfyve9 increased the therapeutic effect of ADSC exosomes on scleroderma.

Conclusion
In conclusion, current investigation determined that ADSC exosomes attenuate bleomycin-induced skin brosis and oxidative stress in scleroderma via circ-Zfyve9 delivery.  showed that ADSCs have a differentiation potential after induction of osteoblast and adipocyte differentiation (H-I). Transmission electron microscopy illustrated that ADSC exosomes had ~100 nm size (J). Nanosizer analysis of exosome particle size also showed that exosomes isolated from ADSCs were ~100 nm in size (K). WB detection showed that ADSC exosomes expressed CD81 and CD63 (L), suggesting that nanoparticles were actually exosomes.

Figure 2
Masson's trichrome staining results in bleomycin-induced murine brosis model showed that collagen deposition was increased in bleomycin-induced scleroderma, while ADSC exosome treatment signi cantly decreased collagen deposition (A and B). Immuno uorescence detection of ROS staining validated that ADSC exosome treatment decremented ROS accumulation signi cantly in bleomycininduced scleroderma (C and D). Immunohistochemical CD31 staining showed that angiogenesis was decreased in bleomycin-induced scleroderma. ADSC exosome treatment signi cantly reversed the inhibitory effect of scleroderma on angiogenesis (E and F). WB for broblast-associated proteins TGF-β, collagen I, and α-SMA showed that ADSC exosome treatment signi cantly decreased collagen I, α-SMA, and TGF-β expression in bleomycin-induced scleroderma (G-J).

Figure 3
Heatmap analysis demonstrated that ADSC exosome treatment resulted in different circRNA expression levels (A). A total of 574 upregulated and 1707 downregulated circRNAs were found in ADSC exosometreated scleroderma skin when compared to scleroderma skin without treatment (B). RT-qPCR showed that seven circRNAs from the sequencing analyses were upregulated, including mmu_circ_0000376, mmu_circ_0001240, mmu_circ_0002870, mmu_circ_0009688, mmu_circ_0011768, and mmu_circ_0011922. Data illustrated that only the mmu_circ_0001240 expression was upregulated in ADSC exosome-treated scleroderma skin when compared to scleroderma skin without treatment (C).

Figure 4
The results showed that miR-135 can decreament signi cantly uorescein intensity, suggesting that miR-135 is circ-Zfyve9 downstream target (A). Luciferase reporter analysis veri ed that miR-135 inhibited luciferase activity in wild-type (WT), but not in mutated (MUT) cell lines (B and C), saying that miR-135 is a circ-Zfyve9 target. Bioinformatics results demonstrated that GPX4 is miR-135 downstream target. To further show a correlation between GPX4 and miR-135, we incorporated WT or MUT 3'UTR-GPX4 sequences including an miR-135 binding sequence into luciferase reporter vector (D). Our team transfected luciferase reporter vector into HEK293 cells, with or not miR-135 mimic. Luciferase reporter analysis showed that miR-135 inhibited luciferase function in WT, yet not in MUT cell lines (E), verifying that GPX4 is an miR-135 target. RT-qPCR results illustrated that circ-Zfyve9 expression was enriched in both EPCs and myo broblasts after transfection using circ-Zfyve9 overexpression vector. Treatment with an miR-135 mimic or GPX4 silencing had no effect on circ-Zfyve9 expression in EPCs and myo broblasts (F and G), suggesting that both miR-135 and GPX4 are located downstream of circ-Zfyve9. RT-qPCR detection also found that circ-Zfyve9 overexpression decreased miR-135 expression. GPX4 silencing had no effect on circ-Zfyve9-induced miR-135 inhibition (H and I), suggesting that miR-135 is located between circ-Zfyve9 and GPX4. The results also found that circ-Zfyve9 overexpression increased GPX4 expression, while miR-135 upregulation reversed the promotion effect of circ-Zfyve9 on GPX4 expression. GPX4 expression decremented signi cantly after transfection with a GPX4 silencing vector (J and K), advising that circ-Zfyve9 enhanced GPX4 expression through sponging miR-135.   Immunohistochemical detection of apoptotic skin tissues showed the ADSC exosomes after treatment with overexpressed circ-Zfyve9. Apoptosis in skin tissues of bleomycin-induced scleroderma was decreased when compared to the ADSC exosome treatment group. Overexpression of miR-135 or GPX4 silencing reversed the protective effect of circ-Zfyve9 exosomes on apoptosis inhibition in bleomycininduced scleroderma skin A and B), suggesting that circ-Zfyve9 overexpression increased the therapeutic effect of ADSC exosomes.