Preparation of ROS‐Responsive Exosome Coating of Nitinol Material for Making the Neurointerventional Stent

Nitinol (NiTi) alloy is an ideal material for preparing neurointerventional stents due to its excellent mechanical properties and biocompatibility. However, NiTi stents without surface‐specific functionalization cause a high rate of stent thrombosis and in‐stent restenosis after implantation. In ischemic stroke, exosomes have a role in regulating nervous system development, regeneration, vascular remodeling, and neuroinflammation. In this work, the effect of exosome coating on the biocompatibility of NiTi alloy is evaluated. NiTi alloy is successively immersed in relevant solution (sodium alginate, 3‐aminophenylboronic acid, distearoylphosphatidylethanolamine, and exosomes) to form ROS‐responsive exosome coated surfaces. In vitro experiments (platelets, endothelial cells, smooth muscle cells, and macrophages) and in vivo subcutaneous implantation experiments are performed to test coatings for biocompatibility. The results show that the modified NiTi alloy surface has high hydrophilicity, which has the functions of inhibiting platelet aggregation, promoting endothelialization, inhibiting smooth muscle cell migration, anti‐inflammatory, and good tissue biocompatibility. This study develops exosome neurointerventional stent coating as a bioactive drug via reactive oxygen species accumulation in lesions, thus targeting drug release, inhibiting intimal hyperplasia, and reducing inflammation and thrombosis.


Introduction
Intracranial atherosclerotic stenosis (ICAS) is one of the common causes of ischemic stroke. [1] As a novel treatment modality, stent implantation has become the preferred treatment strategy for ICAS with the continuous development of to an imbalance in the body's oxidative and antioxidant systems, causing the accumulation of excessive ROS. [7] At this time, excess ROS can induce vascular endothelial damage, inflammatory cell recruitment, lipid peroxidation, metalloproteinase activation, and extracellular matrix deposition, ultimately leading to atherosclerotic plaque formation. [8][9][10] Therefore, reducing ROS destruction of the vessel wall by reducing ROS accumulation in target lesions using ROSsensitive agents may help inhibit the development and progression of AS.
Exosomes are multifunctional nanoparticles (50-100 nm) secreted in the extracellular area (blood, urine, and cerebrospinal fluid). They contain complex RNAs (encoding messenger RNA, micro RNA, and transfer RNA) and proteins, and hence are regarded as specifically secreted membrane vesicles. They are involved in intercellular communication and have a strong potential for regenerative repair applications. [11,12] Studies have shown that exosomes promote angiogenesis, inhibit apoptosis, and regulate immune responses; [13][14][15] coating exosomes on the surface of cardiovascular scaffold materials has the same efficacy. [16,17] Only a few studies have been conducted on using exosomes in treating stroke, which have displayed excellent performance in remodeling blood vessels, inhibiting neuroinflammation, and improving cerebral ischemia. [11,[18][19][20] However, studies on exosome-coated neurovascular stents have not been performed to date. Therefore, combining bioactive plasma exosomes with neurovascular stents may be a promising therapeutic approach.
This study aimed to develop an intracranial vascular stent coating using exosomes as bioactive drugs by exploiting the characteristics of ROS accumulation in lesions, which can not only target release but also inhibit intimal hyperplasia and reduce inflammation and thrombosis. This study used in vitro and in vivo experiments to verify the properties of the ROS-responsive exosome coating to promote endothelialization, inhibit VSMCs migration, and improve anti-inflammatory abilities (Figure 1). The results of this study provided support for the clinical translation of exosomal neurovascular coated stents and revealed their potential application value in treating ICAS.

Exosome Extraction and Identification
The venous blood of patients (All participants gave an informed written consent before participating, ethics approval number 2018-53) was collected using ethylenediaminetetraacetic acid (EDTA) tubes (BD367863; Nobled www.advmatinterfaces.de Technology Co., Ltd., Beijing, China), and the plasma was obtained by centrifugation at 1000 rpm for 20 min; at 2000 rpm for 20 min, and at 5000 rpm for 10 min at 4 °C successively. Then, exosomes were extracted using the exosome kit (4 484 450; Invitrogen, USA) following the manufacturer's protocols. The exosome surface markers, CD81 (ab109201, Abcam, RRID: AB 10866464) and TSG101 (ab275018; Abcam), were identified using Western blot. The exosome morphology was examined via transmission electron microscopy (HT7800; Hitachi, Tokyo, Japan) and nanoparticle tracking analysis (ZetaView PMX 110, Meerbusch, Germany). The particle size was analyzed using Origin 2021 to generate the size profilerelated images.

Surface Morphology Characterization
BN, SN, SMN, SMPN, and SMEN samples were dried at 37 °C for 12 h. X-ray photoelectron spectroscopy (XPS; K-Alpha Spectrometer; Thermo Fisher Scientific, UK) was employed to determine the abundance of component elements on the aforementioned alloy sample surface. The data were analyzed using Avantage software v5. At the same time, atomic force microscopic (AFM; Dimension Icon; Brucker, Germany) was used to examine the roughness and surface morphological characteristics of all the samples. Scanning electron microscopy (SEM; JSM-7800F; JEOL, Japan) and contact-angle meter (SDC-200S; Shengding Precision Instrument Co., Ltd., Guangdong, China) were used to characterize and assess the hydrophilic properties of the BN or SMEN sample surfaces. Upright fluorescence microscope (BX53; Olympus, Tokyo, Japan) was used to characterize the distribution of exosomes (10 µg mL −1 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine,4-chlorobenzenesulfo nate salt (DiD, C1039; Beyotime) incubated at 37 °C for 30 min) on the surface of SMEN.

Coating Release Test
The sample SMEN was immersed in 0.5 mL of phosphatebuffered saline (PBS) or 100 µm H 2 O 2 solution. Then, 50 µL of release solution was obtained at fixed time points (0, 1, 1.5, 3, 5, 10, 13, and 16 h). At the same time, the same volume of PBS or 100 µm H 2 O 2 solution was added to keep the total volume constant. The protein concentration was measured using the Bradford protein assay kit (PC0010; Solarbio), and the release curve of the coated exosomes was obtained based on the standard curve. The schematic diagram of the responsiveness principle and release procedure of coating are shown in Figure 2B,C.

Acute Hemolysis
The arterial blood samples from patients with ICAS were collected using EDTA tubes (All participants gave an informed written consent before participating, ethics approval number 2018-53). The samples were centrifuged (1500 rpm, 15 min) to obtain red blood cells. The BN or SMEN samples were immersed in 3 mL of 2% red blood cell solution (diluted with 0.9% saline). At the same time, a negative control group with normal saline and a positive control group with distilled water were prepared by incubating the samples at 37 °C for 2 h and then centrifuging at 3000 rpm for 5 min. The supernatant (150 µL) was aspirated, absorbance was measured at 545 nm using a microplate reader, and hemolysis rate was calculated using the following formula where HR is the hemolysis rate, OD s is the absorbance value of the BN or SMEN samples, OD n is the negative control value, and OD p is the positive control value.

Platelet Adhesion
The EDTA tubes were used to collect the arterial bloods of patients with ICAS. All the samples were centrifuged at 1500 rpm for 15 min, and 3000 rpm for 10 min to obtain platelet-rich plasma (PRP). The diluted PRP solution was incubated with BN or SMEN alloys at 37 °C for 1 h and washed with PBS solution thrice. These alloy-fixed (2.5% glutaraldehyde, 15 min) samples were gradually dehydrated using 30%, 50%, 70%, 95%, and 100% ethanol. PRP-treated samples were dried and examined using SEM to characterize the adhesion status of PRP onto the alloy surface. www.advmatinterfaces.de

HUVEC Adhesion
HUVECs (5 × 10 4 cells per well, 1 mL of ECM) were seeded in 24-well plates containing BN or SMEN samples to evaluate the adhesion of HUVECs on the SMEN surface. These samples were cultured for 1, 3, and 5 days. After washing with PBS solution thrice, the cells were fixed in 2.5% glutaraldehyde (P1126; Solarbio, Beijing, China) and dehydrated sequentially using 30%, 50%, 70%, 95%, and 100% graded ethanol for 5 min. The alloy samples were washed and flown-dried, a gold layer was sprayed onto the alloy surface for SEM examination, and the cell adhesion on the surface of each group of samples was observed using SEM.

HUVEC Migration
The cell migration could not be observed using a microscope in bright field due to the number of cells growing on the surface of the NiTi alloy sheets. Moreover, when scratching, NiTi alloy sheet and well-plate bottom produced relative movement under force, and it was difficult to ensure scratch uniformity. BN and SMEN leaching solution (1 mL per sample, 24 h) were used for scratch experiments. The HUVECs were seeded at an initial density of 4 × 10 5 in six-well plates, and 2.5 mL of medium was added to each well. When the density reached about 85%, scratching was carried out using a 200 µL pipette tip and the solution was washed six times with 1 mL of PBS solution. The 2.5 mL of the aforementioned extracting solution was added. The cell migration was recorded using a light microscope (Olympus) after 0 and 6 h, and the migration area was quantified using ImageJ 1.53k.

Tube Formation Assay
NiTi alloy wire and HUVECs were used in this study to better observe the tube formation abilities of HUVECs around the BN or SMEN samples. The 48-well plate was coated with a mixture of 100 µL of Matrigel (082 7265, ABW, China) and the medium (1:1). After coating, the wells were incubated at 37 °C for 1 h. The HUVECs (3.5 × 10 4 per well, 100 µL of ECM) were added. Then BN or SMEN wires were gently placed on the chamber surface. After 3 h of incubation, the images were obtained using the light microscope (Olympus) and Nb branches were quantified using ImageJ 1.53k.

Cellular Immunofluorescence
The BN or SMEN samples were placed in a 24-well plate, and the surface was seeded with HUVECs (5.5 × 10 4 cells per well).

NO Release
The HUVECs (6 × 10 4 cells per well) were seeded into 24-well plates containing BN or SMEN samples. Then, the cell culture of the supernatant was collected after 48 h. The examination was conducted for NO release using the NO detection kit (S0021S, Beyotime) following the manufacturer's protocols. The absorbance at 540 nm was recorded using a microplate reader to quantitatively measure the NO concentration. [16]  , and the samples were incubated with secondary antibody CoraLite488 for 1 h. The nuclei were counterstained using DAPI, observed under an upright fluorescence microscope, and quantified using ImageJ 1.53k.

Inflammatory Factor Expression
The IL-6/IL-10 concentrations in the M1/M2 macrophage supernatant were determined using the ELISA kit (m1028583-C/ m1028605-C; Mlbio, Shanghai, China). Briefly, 50 µL of the cell supernatant (1000 g, 20 min) and 100 µL per well horseradish peroxidase (in each group were added to a 96-well plate at 37 °C for 1 h. The liquid was aspirated, and the plate was washed five times using 350 µL per well of the detergent. Then, 50 µL of substrate A and 50 µL of substrate B were added to each well, and the mixture was incubated at 37 °C for 15 min. Finally, 50 µL of the stop solution was added to block the reaction, and the absorbance at 450 nm wavelength was recorded using a microplate reader. All the data were analyzed using the methods used in previous studies. [22] 2.6. In Vivo Experiments to Evaluate the Anti-Inflammatory and Anti-Fibrotic Properties of SMEN Samples

Alloy Sheet Implantation
Healthy female mice (n = 6) aged 6-8 weeks were obtained from the Viton Lever Laboratory Animal Center, and the experimental process complied with the ethical review of experimental animal welfare (ethics approval number HUSOM2022-006). The samples were cut and ground into 0.5 × 0.5 × 0.06 cm 3 alloy sheets, and the coating preparation was conducted as discussed earlier.
Before implantation, all the mice were anesthetized using 6% chloral hydrate (6 mL kg −1 by intraperitoneal injection). After anesthesia, the BN samples (n = 3) were implanted into the left back of the mice subcutaneously and the SMEN samples (n = 3) were implanted into the right back of the same mice subcutaneously. The coated modified surface adhered closely to the abdominal musculature. After the incision was treated with the skin, an appropriate amount of 3M tissue repair glue was added dropwise. The specimens were obtained on the first and seventh days, and embedded in paraffin.

Immunohistochemistry, Hematoxylin-Eosin Staining, and Masson Staining
Immunohistochemistry (IHC) was employed to examine the expression levels of pro-inflammatory cytokine IL-6 (ab290735, Abcam) and identify the inflammation response of implanted alloy samples. All the experimental procedures were performed using the streptavidin-peroxidase (SP) kit (SP-9000; Zhongshan Golden Bridge Biotechnology, Beijing, China) following the manufacturer's protocols. At the same time, hematoxylineosin (HE) and Masson staining were conducted to examine the inflammation and fibrosis of tissues in proximity to the implanted alloy samples. All the experimental procedures were performed using the Prussian Blue staining kit (Cat. 1424; Solarbio) and Masson triple staining kit (Cat. G1340; Solarbio).

Statistical Analysis
Data analysis was performed using GraphPad Prism 7.0 statistical software. The measurement data conforming to a normal distribution were described using mean ± standard deviation, and the Student's t-test was used to compare the two groups. P ≤ 0.05, P ≤ 0.01, and P ≤ 0.001 indicated statistically significant differences.

Ethics Approval and Consent to Participate
The mice used in this study were approved by the Henan

Coating Characterization
We first employed XPS to test the elemental profile onto the alloy surface to validate whether the alloy was successfully www.advmatinterfaces.de modified by related materials. As shown in Figure 3A, the disappearance of the Ti 2p peak and the appearance of the Na 1s peak implied that SA successfully adhered to the BN surface via electrostatic interaction. With the subsequent stepwise surface modification, we could observe that C1s peak gradually increased, and O 1s and Na 1s peaks gradually decreased and disappeared after the appearance of N1s peak, implying that M-APBA, DSPE-PEG, and exosomes were gradually successfully grafted on the BN surface. The AFM measurements were performed on each layer to observe the surface morphology and roughness of modified BN. As shown in Figure 3B, the surface morphology of modified BN changed layer by layer and the roughness changed from BN (7.03 nm), SN (5.91 nm), SMN (14.8 nm), and SMPN (5.81 nm) to SMEN (27.9 nm), indicating that the alloys were successfully modified by the related materials. Figure 3C shows the surface morphologies of NiTi alloy sheets before and after coating. The fluorescence map (upper panel) shows that no red fluorescent spots existed on the BN surface, and the SMEN surface had a red fluorescent spot distribution, which was due to the fact that the exosomes on

Blood Compatibility
Platelet adhesion and acute hemolysis assays were always conducted to examine the blood biocompatibility of the implanted alloy. [23,24] Platelet aggregation induced by implantation in the blood circulation system could cause serious blood diseases. Besides promoting thrombosis, platelet adhesion also acts as an "inciter" and a "participant" of the inflammatory response and plays a crucial role in the initiation of inflammation and immune regulation. [25] As shown in Figure 4A, SEM images revealed that the density of cells adhering to the SMEN surface was significantly lower than that of the bare alloy, and the morphology of adhered platelets was pseudopodic, dendritic, and aggregated with high activity. These results implied that SMENcoated alloys prevented the cell aggregation. The platelets adhering to the surface of the SMEN sample were round, with no accumulation and low activity. [26,27] Furthermore, the acute hemolysis of red blood cells was also examined by exposing the BN or SMEN samples at 37 °C for 2 h. As shown in Figure 4B, no significant difference was observed between BN and SMEN samples. However, the median value of hemolysis rates (0.001073 ± 0.0002184%) for SMEN samples was higher than that for BN samples (0.0008075 ± 0.0002675%). These results suggested that SMEN-coated alloys displayed the excellent blood biocompatibility compared with BN samples.

SMEN Promoted Endothelial Cell Proliferation and Migration
When the stent was implanted into the human blood vessel as a "foreign body," the immune system of the body was rapidly activated, resulting in complications such as inflammatory response, delayed endothelialization of the stent, and intimal hyperplasia. [5] First, we examined the proliferation of HUVECs on the surface of a BN or SMEN sample. As shown in Figure 5A, bright-field images clearly demonstrated that the cell density of HUVECs adhering to SMEN samples was significantly higher than that of HUVECs adhering to BN samples, which was corroborated by optical density (OD) values ( Figure 5B). Next, we examined the cell migration of HUVECs adhering to SMEN samples compared with BN samples. Bright-field images ( Figure 5C) and relatively quantitative measurement ( Figure 5D) showed that the migration rates of HUVECs adhering to SMEN samples were higher than that of HUVECs adhering to BN samples. The angiogenesis properties ( Figure 5E,F) of HUVECs adhering to SMEN samples were consistently higher than those of HUVECs adhering to BN samples. These features of HUVECs were highly associated with the concentration of released NO. [28,29] Consequently, the concentration of released NO was examined in HUVECs adhering to SMEN samples. The concentration of released NO was higher for SMEN samples compared with BN samples ( Figure 5G), confirming the roles of SMEN in promoting endothelial function. Finally, we examined the expression levels of vWF and VE-cadherin, which were associated with angiogenesis and barrier permeability maintenance. [30,31] As shown in Figure 5H,I, the expression levels of vWF and VE-cadherin in HUVECs adhering to BN samples were higher than those in SMEN samples. The expression level of VE-cadherin in SMEN samples was relatively high, and the expression level in BN samples was relatively low, which was similar to the NO-release results.

SMEN Inhibited VSMCs Proliferation and Migration
α-SMA and SM22α are two marker proteins of the contractile VSMCs. As shown in Figure 6A,B, the immunofluorescence results showed that both α-SMA and SM22α were significantly expressed on the surface of SMEN samples, while the expression of α-SMA and SM22α was significantly reduced  www.advmatinterfaces.de

Macrophage Phenotype Switch and Inflammatory Factor Expression
This study induced M0 macrophages into M1 and M2 macrophages with IFN-γ, LPS, and IL-4 to study the phenotypic switch of macrophages. The results showed (Figure 7A,B) that the M1 marker protein iNOS was highly expressed on the surface of BN samples and had a low expression on the surface of SMEN samples. On the contrary, the M2 marker protein Arg I was expressed at a low level on the surface of BN samples and highly expressed on the surface of SMEN samples ( Figure 7D,E). The ELISA results showed ( Figure 7C,F) that the secretion of pro-inflammatory cytokine IL-6 was higher on the surface of BN samples than on SMEN samples, and the secretion of anti-inflammatory cytokine IL-10 was lower on BN samples.

Anti-Inflammatory and Anti-Fibrotic Effects of SMEN In Vivo
IHC, HE staining, and Masson's trichrome staining were performed on the tissue samples after implantation of the mice samples to observe the histocompatibility of samples in vivo.
On the first day, a low level of expression of pro-inflammatory cytokine IL-6 was observed around the two groups of samples, and the expression level was comparatively higher in BN samples than in SMEN samples (Figure 8 top). And on the seventh day, the level of expression of IL-6 decreased, but still higher in BN samples than in SMEN samples (Figure 8 top).
A few inflammatory cells were seen around the tissues of the two groups of samples, mainly centrocytes, as shown by HE (Figure 8

Discussion
Surface modification is very helpful for improvement of medical devices. An outstanding representative of surface modification to regulate cell-material interactions is the nanocoating of occluder of left atrial appendage to enhance cell migration pioneered by the Chinese Ding team, [32] which has been translated from bench to bedside in many countries. The principle of material surface, cell adhesion, cell migration, and other cell behaviors has been revealed in their publications based on a unique nanopatterning technique for specific cell adhesion [33] and on a biomimetic surface for nonspecific cell adhesion. [34] Nevertheless, our present paper distinguishes itself from a surface modification of a NiTi material with a ROS-responsive exosome coating to make a neurointerventional stent. In this study, SA-based ROS-responsive exosome coatings were successfully prepared through physical and chemical reactions. First, the negatively charged highly hydrophilic polymer SA [35,36] was electrostatically adsorbed to form a uniform highly hydrophilic coating on the positively charged NiTi [37,38] alloy surface. The

www.advmatinterfaces.de
coating improved hydrophilicity and reduced the roughness of the NiTi alloy surface. The highly hydrophilic coating surface preferentially adsorbed albumin without platelet receptors, reducing platelet accumulation and thrombosis. [39][40][41] Moreover, SA cross-linked with Ca 2+ in the blood to form a hydrogel during the delivery of the coating to lesions compared with other polymer coatings, [5,[42][43][44]

www.advmatinterfaces.de
and DSPE were bridged in the outermost layer of the coating. When electron-rich ROS attacked the electron-deficient boronic ester, the aryl boronate group was oxidized and the self-cleaving methylphenoxy group was decomposed and then hydrolyzed to phenol. [45] This helped achieve the responsiveness release, which not only reduced the loss of exosome layers during the implantation of stents but also consumed ROS in the lesion site. Although Wang et al. [46] also modified the surface of the stent material using the ROS-responsive principle, the mixed encapsulation method of the coating resulted in insufficient coating adhesion and stability. Previous studies demonstrated that plasma exosomes helped improve ischemic stroke injury. [47,48] In this study, the use of DSPE to bridge exosomes could improve the biocompatibility and stability of exosomes compared with the direct encapsulation of exosomes with polymers, contributing to vascular remodeling. [4,17] This study represents a significant departure from previous work on ROSresponsive exosome coatings, as exemplified by Hu et al. [17] While Hu and colleagues utilized DSPE bridging exosomes to coat chromium-based cardiovascular interventional stents; this study employed a NiTi alloy with super compliance and weak paramagnetism, which is the primary raw material for neurointerventional stents. Moreover, the response release point in this study was triaminophenylboronic acid and dihydroxyl-modified DSPE-PEG, whereas Hu et al. used tetracarboxyphenylboronic acid and dihydroxyl-modified DSPE-PEG. Although the experimental results did not exceed those of Hu et al., we attribute this discrepancy to the smaller amount of exosomes coated. Notably, this study utilized plasma exosomes, which are easier to obtain than stem cell exosomes, and has important implications for the clinical translation of exosome-eluting neurointerventional stents. Platelet adhesion and acute hemolysis experiments were performed in this study using arterial blood from patients with ICAS. Compared with the studies of inhibition of platelet adhesion by Tim et al. [49] (hydrophilic stent coating) and Obiweluozor et al. [50] (thrombus-resistant semi-IPN hydrogel coating), the hemocompatibility results of this study were closer to the real situation of the lesion. Both bare alloy and SMEN samples exhibited lower hemolysis rate, well below the International Organization for Standardization standard of 5%, [51] confirming that the coating had good hemocompatibility. Compared with BN samples, the amount of platelet adhesion on the surface of SMEN samples was significantly reduced. The platelets were round, non-aggregated, and less active, which could reduce the activation of the coagulation system, thereby reducing stent thrombosis, [41,52] with similar efficacy to the CD31 mimetic layer of the coronary stent developed by Diaz-Rodriguez et al. [53] The WCA observed in this study was much lower than that in the study by Hou et al., [16] possibly because DSPE-PEG was a highly hydrophilic polymer.
The stent surface is critical for endothelial cell adhesion and rapid endothelialization. [54] In this study, HUVECs proliferation, adhesion, and migration experiments were carried out on the coating. After coating modification, the NiTi alloy showed better cell proliferation than BN samples, with 1.5 times higher cell number on day 5, which might stem from the involvement of exosomes in information exchange between HUVECs, thereby promoting the proliferation of HUVECs, which was consistent with previous findings that exosomes promoted cell proliferation. [55,56] Several studies showed [5,16,57] that the speed of endothelial cell proliferation and migration was critical for stent endothelialization. A scratch assay was employed to assess endothelial cell migration. The result showed that the cell migration area of SMEN samples was 1.3 times than that of BN samples within 6 h, indicating that SMEN samples had better rapid endothelialization. In this study, endothelial cell immunofluorescence showed that vWF expression in the SMEN group was downregulated compared with that in the BN group, and VE-cadherin was relatively upregulated, which was beneficial to maintain the integrity of the vascular endothelial cell barrier and reduce platelet adhesion. [30,31,58,59] Hence, these results suggested that the ROS-responsive exosome coating could promote the proliferation and migration of endothelial cells and positively affect the endothelialization of NiTi alloy surfaces.
Excessive proliferation and migration of VSMCs are the main causes of ISR. [60,61] In ICAS, VSMCs are mostly dedifferentiated with proliferative and migratory abilities due to the excessive accumulation of ROS at the lesion site. [62] In this study, the ROS-responsive exosome coatings increased the expression of contractile VSMCs marker genes α-SMA and SM22α, as determined via IF analysis, indicating that the coating was beneficial for inhibiting the growth of atherosclerotic plaques to a certain extent. The Transwell experiment in this study showed that the co-culture of SMEN samples and HUVECs inhibited the migration of VSMCs, which might be because the exosome coating promoted the release of related factors from HUVECs [63] and also the transformation of VSMCs from synthetic to contractile. Furthermore, the NO detection showed that HUVECs in the SMEN group generated more NO, which was vital in dilating blood vessels and inhibiting platelet adhesion, activation, and proliferation of VSMCs. [6,[64][65][66] It helped inhibit the progression of ICAS to some extent. This was similar to the rapamycin NOrelease stent developed by Zhang et al. [67] Studies on immobilizing exosomes that mediate inflammation and induce immune responses on the surface of vascular grafts or stents have recently attracted much attention. [16,17,68] The ELISA results in this study indicated that the SMEN enhanced the secretion of anti-inflammatory cytokine IL-10 and inhibited the secretion of pro-inflammatory cytokine IL-6. This might be related to the fact that the ROS-responsive exosome coating inhibited macrophage transformation toward the M1-like subtype (pro-inflammatory type) and, in turn, promoted macrophage transformation toward the M2-like subtype (anti-inflammatory type), which was consistent with Zhu et al.'s finding that adipose stem cell-derived exosomes promoted the polarization of macrophages to the M2-like subtype, [69] as shown by IF, which was conducive to promoting inflammation resolution, tissue repair, and plaque stabilization. [70] Clinical data suggested that more than 80.0% of thromboembolic events occurred within 2 days after the procedure. [5] Because of the bidirectional association between inflammation and thrombosis, [52,71] we hypothesized that thromboembolic events might be related to acute inflammation after stent implantation. Therefore, we considered 1 and 7 days as the time windows to observe the acute inflammatory response www.advmatinterfaces.de and antifibrotic effect after the implantation of samples. The immunohistochemical experiment showed that the number of inflammatory cells around the SMEN samples was less on the first day, predominantly centrocytes; only a few fibroblasts secreted the pro-inflammatory cytokine IL-6, and no tissue proliferation was seen. The collagen deposition around SMEN samples was comparatively lesser and with more neovascularization than that for BN samples, which was consistent with the results of Wang et al. [72] implanting fibrinogen/dopaminemodified titanium alloy into mice subcutaneously. The aforementioned results indicated that SMEN had good histocompatibility and potential application value. Before ending our discussion of a ROS-responsive exosome coating of nitinol for interventional stent, it seems worthy of noting the effect of serum proteins on ROS, pointed out very recently by Ding team in their assessment of the potential cytotoxicity during their research and development of the next-generation biodegradable cardiovascular device for interventional treatment along with the first clinical case, [73] based on their new concept of metalpolymer composite stent. [74] Both of their in vitro and in vivo examinations support the strong effects of physiological conditions, in particular serum proteins, on biocompatibility of an implant. Therefore, any ROS efficacy in vivo or in a biomimetic condition should take the effect of biomacromolecules in the local microenvironment into consideration in understanding of the corresponding experimental results.

Conclusions
In this study, based on the principle that phenylboronic ester bonds are responsive to ROS, M-APBA was used to connect hydrophilic polymers and exosomes on the surface coated with SA NiTi alloy, and the highly hydrophilic ROS-responsive exosome coating was successfully prepared. In vitro cell experiments showed that the coating could effectively promote the proliferation and migration of HUVECs on the surface of NiTi alloy, the raw material of the ICAS stent; inhibit the phenotypic transition, proliferation, and migration of VSMCs; and reduce the inflammatory response. The collagen deposition was thinner and with more neovascularization around SMEN samples compared with BN samples in animal experiment. The in vitro and in vivo experiments fully demonstrated that the highly hydrophilic ROS-responsive exosome coating had potential application value in neurointerventional stent, might have great potential in preventing ISR.