Exosomes from Adipose-Derived Mesenchymal Stem Cells Overexpressing Stanniocalcin-1 Promote Reendothelialization after Carotid Endarterium Mechanical Injury

Background and objective: Endothelial cell inammation caused by mechanical injury of endovascular treatment remains the major obstacle to reendothelialization, which leads to arterial restenosis. We investigated the reendothelialization effect of exosomes from adipose-derived mesenchymal stem cells (ADSC) overexpressing Stanniocalcin-1 (STC-1). Methods: Primary ADSCs were extracted from the adipose tissue of the inguinal area of C57/BL mice. ADSCs were transfected with lentivirus vectors containing STC-1. Exosomes were puried from culture medium using the Exo-Quick kit and characterized by transmission electron microscopy, nanoparticle tracking analysis and western blot. PHK-26 as molecular probe was used to track the exosomes engulfed by mice arterial endothelial cells (MAEC). The role of STC-1-ADSC-Exosome (S-ADSC-Exo) in MAECs was veried through scratch test and tube forming experiment. Carotid endarterium mechanical injury was induced by insertion with a guidewire into the common carotid artery lumen. Exosomes were administered by tail vein injection. Content of Reactive oxygen species (ROS) was measured using commercial kits. Carotid arteries were harvested for histological examination, immunouorescence staining, and Evan’s blue staining. Results: Transfection of STC-1 signicantly enhanced STC-1 levels in ADSCs, their exosomes, and MAECs. Compared with the control group and the ADSC-Exo group, STC-1 enriched exosomes markedly enhanced STC-1 level, inhibited the expression of NLRP3, Caspase-1, and IL-1β in MAECs, exhibited good lateral migration capacity, and promoted angiogenesis. Exosome-pretreating groups exhibited lower levels of ROS than those of controls. In vivo administration of S-ADSC-Exo had reendothelialization effect on post-injury carotid endarterium as evidenced by thinner arterial wall, low-expressed NLRP3, and more living endothelial cells. Conclusions: The reendothelialization effect of exosomes from ADSCs on post-injury carotid endarterium can be enhanced by genetic modication to contain elevated STC-1. of IL-1β expression. e Quantication of thickness of arterial wall. IF, immunouorescence. HE, hematoxylin-eosin. *P< .05, vs control. Scale bar = 20 mm.

For patients at high risk of surgical complications, combination carotid angioplasty with stent implantation has become an accepted alternative to endarterectomy [2]. Despite re nements of interventional technique, the mechanical injury to carotid endarterium is still unavoidable completely and subsequent in ammatory restenosis is not conducive to ensure the long-term patency [3]. Post-injury restenosis just like wound-healing comprises complex pathophysiological mechanisms consisting of in ammation, proliferation, and migration followed by remodeling of arterial wall [4]. Persistent and exaggerated in ammation triggers the release of numbers of cytokines and growth factors, resulting in hyperplasia of neointima and proliferation of smooth muscle cells [5]. Therefore, inhibition of in ammation, restoration of endothelial cell function, and promotion of reendothelialization are effective approaches to prevent and treat restenosis after endovascular treatment.
Exosomes carry a wide range of bioactive molecules, such as DNAs, mRNAs, microRNAs, cytoskeletal elements, proteases, signaling molecules and play important roles in intracellular information communication [6,7]. Bioactive molecules can be transported to target cells under physiologic and pathologic conditions, which induce functional and expressional changes [7]. Recently, with regard to the role in the process of in ammation, exosomes derived from different cell types have drawn much interest.
Via exosomes, IL-4 modi ed macrophages foster M2 polarization and inhibit in ammation to retard atherosclerosis through transferring their regulatory microRNAs to target cells [8]. Exosomes generated from cardiomyocytes and endothelial cells followed acute myocardial infarction are taken up by macrophages and regulate local in ammation to attenuate ischemic injury [9]. Mesenchymal stem cells (MSC) are multipotential stem cells that can promote immunomodulation, angiogenesis, matrix remodeling in the injured vessels by secreting factors to activate the signaling pathways involved in vascular repair. The powerful paracrine capacity of MSCs, not their differentiation potential, is the principal mechanisms of their repair action [10]. Local transplantation of adipose-derived mesenchymal stem cells around arteriovenous stula alleviates restenosis and restores patency after PTA in mice [11].
MSCs have been reported to exert their powerful immunomodulatory and anti-in ammatory effects in the pathological process of atherosclerosis [12]. However, the microenvironment of target tissue may affect ADSCs survival and migration, which leads to poor long-term prognosis [13]. Compared with other cell types, MSCs are more likely to secrete a great quantity of exosomes, and exosomes can partially overcome the above de ciencies due to their low immunogenicity [14]. Umbilical cord mesenchymal stem cell-derived exosomes have been indicated to attenuate TNF-α induced in ammation in endothelial cells [15]. Mesenchymal stromal cell-derived exosomes are shown to ameliorate diabetic peripheral neuropathy by suppression of proin ammatory genes [16]. With the development of exosome research, it has been found that the function of exosomes can be enhanced by genetic modi cation of stem cells to load target proteins [17].
Interleukin (IL) -1β has been reported to participate in the chronic in ammation and restenosis after endovascular treatment [11,18]. As an important component of innate immunity, NLRP3 in ammasome consisting of NLRP3 receptor protein, apoptosis-associated speck-link protein containing a CARD (ASC), and Caspase-1 has recently emerged as a protein complex for inducing the release of IL-1β and IL-18 [19].
ROS are key molecule signal for activating NLRP3 [20], and proved to be triggered after stent implantation and PTA [21]. STC-1 is a kind of glycoprotein directly act on mitochondria and then inhibit ROS through uncoupling [22]. Thus, it is plausible that the axis of STC-1-NLRP3-IL-1β may be the new target for protecting endothelial cells against in ammation induced by mechanical injury and promoting reendothelialization. In the current study, we delivered STC-1 to endothelial cells by exosomes from ADSCs that were genetically modi ed to overexpress STC-1. We highlight the reendothelialization effect of STC-1-modi ed exosomes on post-injury carotid endarterium in the setting of in ammation caused by mechanical injury. Genechem (Shanghai, China). 1´10 6 ADSCs were seeded in 10 mL of low glucose DMEM supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin overnight and were subsequently transfected with 200nM STC-1 vector at a multiplicity of infection of 10 for 24 hours. After that, the envelope suspension was transferred to normal culture medium, and the transfected ADSCs were cultured for 48 hours. The green uorescent protein (GFP) encapsulated in lentivirus vectors was observed using uorescence microscope.

Exosomes extraction and identi cation
At 48 hours after transfection, the exosomes were isolated at 4°C according to the method previously described [23]. Firstly, the culture medium of ADSCs was ltered to remove large debris and oating cells. Secondly, small debris was removed by centrifugation at 10,000´g for 30 min. Lastly, the remaining supernatants were further centrifugated at 10,000´g for 3 h. The precipitate was resuspended in phosphate-buffered saline (PBS) and the diameters of exosomes were measured using nanoparticle tracking analysis. The morphologic characteristics of exosome were observed using transmission electron microscopy (TEM). Expression of CD9, CD63 and CD81, as markers of exosomes, were con rmed by Western blot.

Mice arterial endothelial cells (MAECs) culture and tracking of exosomes engulfed by MAECs
MAECs were purchased from Daixuan Biosciences Inc (Shanghai, China) and cultured in high glucose (4500 mg mL -1 ) DMEM supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. MAECs were cocultured with PHK26 labelled exosomes for 24 h. The exosomes engulfed by MAECs were identi ed by uorescence microscopy. DAPI was used to stain nucleus, and Phalloidin was used to stain cytoskeleton.
NLRP3 in ammasome activation NLRP3 in ammasome was activated according to the method previously described [22]. MAECs (1´10 5 ) were seed in 6-well plates. The culture medium was added with LPS (Ultra-pure InvivoGen, San Diego, CA) at the concentration of 2 mg mL -1 in high glucose DMEM for 6 hours. After washing with PBS three times, the MAECs were incubated with 5 mM ATP (InvivoGen, San Diego, CA) for 45 min. After that, the MAECs were washed with PBS for three times and cultured in high glucose DMEM. Scrapping with a pipette tip was conducted in the upper 3 wells of 6-well plate, with the lower 3 wells receiving no procedure as control. 24 hours later, the cells were collected to determine the expression of NLRP3 in ammasome.

Scratch test and Tube forming experiment
The lateral migration capacity of MAECs was assessed by scratch test. 1´10 5 MAECs were seed in 6-well plates and scraped by a pipette tip to generate uniform wounds when reached at approximately 90% con uent. Each well was washed three times with PBS to remove oating cells, and then the MAECs were cultured in medium with no serum. The blank area at the time intervals (0 h, 6 h, 12 h, and 24 h) were observed under inverted microscope and calculated using Image-Pro Plus 6.0 software.
10 mL of Matrigel per well was evenly placed on the angiogenic slide (Ibidi, Germen). After that, the slide was placed in an incubator for 2 hours to solidify the Matrigel. About 1´10 4 MAECs were seeded on per well and observed under inverted microscope after 4-6 hours.

Animal model of carotid endarterium wire-injury and tail vein injection
After administering anesthesia (pentobarbital sodium, 0.5 mg g -1 ), the mice were xed on a heating plate maintained at 37 °C. A middle neck incision was performed, the left common carotid artery and its branches were skeletonized. After dissection, mechanical injury to carotid endarterium was induced by insertion with a guidewire into the left common carotid artery. At the end of procedure, the guidewire was removed, the proximal and distal sutures were tied off gently. The procedure is shown in Fig. S1. The administration of exosomes was performed through tail vein injection.

Experimental protocol
Experimental groups and protocol are described as followed.
In vitro. To evaluate angiogenic effect of STC-1overexpressing exosomes on MAECs after scrapping, 3 groups were enrolled. The NLRP3 in ammasome was activated by addition of LPS + ATP according to the method as described above. The scratch test: Control group: PBS (2 mL) was added to the 6-well plate cultured with MAECs (1´10 5 per well) after scraped by a pipette tip. ADSC-Exo group: ADSC-Exo (2 mL, 2mg mL -1 ) was added to the 6-well plate cultured with MAECs (1´10 5 per well) after scraped by a pipette tip. S-ADSC-Exo group: S-ADSC-Exo (2 mL, 2mg mL -1 ) was added to the 6-well plate cultured with MAECs (1´10 5 per well) after scraped by a pipette tip.
In a parallel series of experiments, MAECs were collected from the 3 groups 24 hours after scratching to determine the expressions of NLRP3 in ammasome.
In a parallel series of experiments, the left common carotid arteries were additionally harvested from the 3 groups (n=5 per group) to evaluate the expressions of NLRP3 in ammasome.

Enzyme-Linked Immunosorbent Assay
Contents of ROS were measured using enzyme-linked immunosorbent assay kits according to the manufacturer's instructions (Thermo Fisher, Shanghai, China).

Histological examination
On the 14 th day after procedure, the post-injury carotid arteries were harvested for histological analysis.
Para n-embedded sections (4 mm) of left common carotid arteries were stained with hematoxylin-eosin (HE) dye to determine the thickness of arterial wall. Immuno uorescence staining of the left common carotid arteries was performed to determine the expression of NLRP3 in ammasome. To determine the reendothelialization effect on post-injury carotid endarterium, 1% Even's blue dye was infused into the carotid arteries to delineate the injured area from reendothelialized area.

Characterization of ADSCs
ADSCs extracted from C57/BL mice at passage 3 ( Fig. 1a) were used in the following experiments. Flow cytometry analysis showed that speci c surface antigens of stem cell (CD29 and CD44) were strongly positive, while the speci c surface antigens of hematopoietic cell (CD31and CD34) were negative (Fig.  1b). After transfection with STC-1 lentivirus vectors, GFP encapsulated within lentivirus vectors made ADSCs present green uorescence (Fig. 2a). Exosomes derived from ADSCs were membrane vesicles (Fig. 2b).

Reendothelialization effect in vitro
After activation of NLRP3 in ammasome, the scratch test was performed to assess the lateral migration capacity of MAECs. Compared with the control group and the ADSC-Exo group, excellent lateral migration capacity was observed in the S-ADSC-Exo group (Fig. 5a). Blank area calculation showed that the S-ADSC-Exo group had the fastest migration speed at time intervals (6 h, 12 h, 24 h) (Fig. 5c). Meanwhile, compared with the control group, recovery after scratching was faster in the ADSC-Exo group.
To assess the angiogenesis effect of S-ADSC-Exo on MAECs, tube forming experiment was performed.
The results revealed that tube-forming ability of MAECs was signi cantly improved in the group incubated with S-ADSC-Exo. As shown in Fig. 5b, d, tube density in the S-ADSC-Exo group was highest compared with the control group and the ADSC-Exo group. Representative western blot pictures of NLRP3, Caspase-1, and IL-1β expressions were shown in Fig. 6b. Densitometric analysis revealed that transfection of STC-1 markedly decreased the expression of NLRP3 in the S-ADSC-Exo group (P< .05, vs control, Fig. 6c). There was no signi cant difference in the expression of NLRP3 between the control group and the ADSC-Exo group. Densitometric analysis revealed that transfection of STC-1 markedly decreased the expression of Caspase-1 in the S-ADSC-Exo group (P< .05, vs control, Fig. 6c). There was no signi cant difference in the expression of caspase-1 between the control group and the ADSC-Exo group. Densitometric analysis revealed that transfection of STC-1 markedly decreased the expression of IL-1β in the S-ADSC-Exo group (P< .05, vs control, Fig. 6c). There was no signi cant difference in the expression of IL-1β between the control group and the ADSC-Exo group.

Ros levels in carotid arteries
Compared with the control group, levels of ROS in the ADSC-Exo and S-ADSC-Exo group were lower (Fig.   7). Compared with the ADSC-Exo group, level of ROS in the S-ADSC-Exo group was lower. Fig. 7 The level of ROS in carotid arteries after mechanical injury. *P< .05, vs control. #P< .05, S-ADSC-Exo vs ADSC-Exo.

Mechanism of S-ADSC-Exo promoting reendothelialization in carotid artery tissue
Representative western blot pictures of NLRP3, Caspase-1, and IL-1β were shown in Fig. 8a. Densitometric analysis revealed that transfection of STC-1 markedly decreased the expression of NLRP3 in the S-ADSC-Exo group (P< .05, vs control, Fig. 8b). There was no signi cant difference in the expression of NLRP3 between the control group and the ADSC-Exo group. Densitometric analysis revealed that transfection of STC-1 markedly decreased the expression of Caspase-1 in the S-ADSC-Exo group (P< .05, vs control, Fig. 8b). There was no signi cant difference in the expression of Caspase-1 between the control group and the ADSC-Exo group. Densitometric analysis revealed that transfection of STC-1 markedly decreased the expression of IL-1β in the S-ADSC-Exo group (P< .05, vs control, Fig. 8b). There was no signi cant difference in the expression of IL-1β between the control group and the ADSC-Exo group. Representative sections of carotid endarterium stained with CD31 (red), NLRP3 in ammasome (green), and merge (orange) are shown in Fig. 9a. As summarized in Figure 9b, d, the expressions of NLRP3, Caspase-1 and IL-1β in the S-ADSC-Exo group were much lesser than that in the control group and that in the ADSC-Exo group. HE staining (Fig. 9a, d) results showed that the thickness of arterial wall was much thinner in the S-ADSC-Exo group, compared with the control group and the ADSC-Exo group. Moreover, Even's blue staining demonstrated that the number of living endothelial cells was signi cantly greater than that in the control group and that in the ADSC-Exo group (Fig. 9e).

Discussion
The salient ndings revealed by this current study are that the reendothelialization effect of S-ADSC-Exo on post-injury carotid artery is associated with enhanced lateral migration capacity of MAECs, promoting angiogenesis, decreased expression of NLRP3 in ammasome in vitro and in vivo, suppressed negative remodeling of arterial wall.
ADSCs are abundant in source, easy to harvest and isolate, can robustly release exosomes, and play a crucial role in tissue repair [24]. Collective data have indicated that exosomes released from ADSCs conduct vascular repair by promoting vascular plasticity [25], enhancing angiogenesis [25], improving post-injury vascular regeneration [26], and regulating autophagy [27]. We revealed that exosomes from genetically modi ed ADSCs induced reendothelialization effects on post-injury carotid endarterium as evidenced by improvement of carotid artery remodeling and survival of endothelial cells. In ammatory response is consequence of mechanical injury using endovascular techniques and may be a principal contributor to complex in ammatory reactions of endothelial cells [28]. NLRP3 in ammasome are a group of intracellular protein complexes produced during in ammation activation, and act as innate immune signal receptor to initiate in ammatory response. IL -1β, as an important member and a major executor of NLRP3 in ammasome, is strongly expressed in the chronic in ammation and restenosis after endovascular treatment [11,18]. In the current study, the elevation of NLRP3 in ammasome (NLRP3, Caspase-1, and IL-1β) after scratching was signi cantly inhibited by exosomes from ADSCs overexpressing STC-1, suggesting that exosomes may account for reendothelialization via antiin ammation effects. It is worth noting that compared with the control group, the lateral migration capacity and the tube density is higher in the ADSC-Exo group. Although STC-1 was highlighted, some cargoes aside from STC-1 in ADSC-Exo may contribute to the reendothelialization and angiogenesis, which needs to be further investigated.
RNAs are important regulatory factors delivered by exosomes to target cells. Increasing reports indicates that ADSC-derived exosomes regulate target cell protein expression and cell morphological change. ADSC-derived exosomes promote wound-healing in diabetic mice through microRNA-128-3p/SIRT1 mechanism [27]. Via exosomes, ADSCs communicate with cardiomyocytes and macrophages to ameliorate ischemic injury by activating S1P/SK1/S1PR1 signaling pathway and foster M2 polarization [29]. Microglia-induced neural in ammatory injury can be suppressed by ADSC-exosomes through the NF-kB/MAPK pathway [30]. All these data show that the function ADSC-exosome can be enhanced by changing their RNA content. STC-1 is a conserved glycoprotein that can directly act on mitochondria, regulate oxidative phosphorylation and inhibit in ammatory reactions [22,31]. By inhibiting oxidative stress, STC-1restrain renal ischemia-reperfusion injury through ROS-mediated multiple signaling pathways [32]. Through being encapsulated within the exosomes, the STC-1 or other RNAs is protected from the degradation of protease or RNase [33]. Compared with untransfected groups, transfection of STC-1 enhanced the levels of STC-1 in ADSCs, released exosomes, and MAECs engul ng exosomes. The content of ROS was decreased in carotid arteries after in vivo administration of S-ADSC-Exo. However, the same result occurred in the ADSC-Exo group, there may be some other active biomolecules within the exosomes played the alike role of STC-1, which needs to be further studied.
To elucidate the potential reendothelialization effect of STC-1 on post-injury carotid endarterium, possibilities for the related proteins were explored. STC-1 is indicated to be an endogenous regulator of ROS [34], and ROS is the key signal to activate NLRP3 in ammasome [35]. ROS is a major mediator of the in ammation follows mechanical injury of endovascular treatment. Mechanical injury triggers an overproduction of ROS, which leads to complex change of intracellular protein expressions and results in cell apoptosis and pyroptosis [36]. ROS can activate NLRP3 in ammasome by promoting the link of Thioredoxin-interacting protein (TXNIP) and NLRP3 [37]. The expressions of NLRP3, Caspase-1, and IL-1β were detected to be inhibited in MAECs and in carotid arteries, indicating that the expressions of NLRP3 in ammasome were in consistent with the content of ROS and ROS was involved in the activation of NLRP3 in ammasome. Another important nding in this study was that STC-1 was transferred into carotid arteries by STC-1enriched ADSC-exosomes, and the expressions of NLRP3 in ammasome were decreased as evidenced by immuno uorescence staining. In the meanwhile, the negative remodeling of arterial wall and the survival of endothelial cells were promoted. Therefore, it is plausible that STC-1 is mediator of NLRP3 in ammasome by which STC-1 enriched ADSC-exosomes attenuated the oxidative stress after mechanical injury of carotid artery and mediated reendothelialization effects on post-injury carotid endarterium compared with ADSC-derived exosomes.