Corroded iron stent increases fibrin deposition and promotes endothelialization after stenting

Abstract Poststent restenosis is caused by insufficient endothelialization and is one of the most serious clinical complications of stenting. We observed a rapid endothelialization rate and increased fibrin deposition on the surfaces of the corroded iron stents. Thus, we hypothesized that corroded iron stents would promote endothelialization by increasing fibrin deposition on rough surfaces. To verify this hypothesis, we conducted an arteriovenous shunt experiment to analyze fibrin deposition in the corroded iron stents. We implanted a corroded iron stent in both the carotid and iliac artery bifurcations to elucidate the effects of fibrin deposition on endothelialization. Co‐culture experiments were conducted under dynamic flow conditions to explore the relationship between fibrin deposition and rapid endothelialization. Our findings indicate that, from the generation of corrosion pits, the surface of the corroded iron stent was rough, and numerous fibrils were deposited in the corroded iron stent. Fibrin deposition in corroded iron stents facilitates endothelial cell adhesion and proliferation, which, in turn, promotes endothelialization after stenting. Our study is the first to elucidate the role of iron stent corrosion in endothelialization, pointing to a new direction for preventing clinical complications caused by insufficient endothelialization.


| INTRODUCTION
A stent implantation is widely used to treat coronary stenosis. 1,2 However, intimal injury during implantation surgery is inevitable because of the mechanical damage caused by the surgery. 3,4 In addition, intravascular stent implantation causes a persistent prothrombotic milieu until the endothelial cells fully cover the stent struts. 5 Endothelial monolayers not only serve as a dynamic physiological border to isolate vascular surrounding tissue from circulating blood flow but also provide a nonadhesive surface for leukocytes and platelets. 6 Moreover, endothelial monolayers have been shown to regulate platelet activation and aggregation, which are closely correlated with occlusive thrombosis. 7 Clinicians usually recommend dual antiplatelet therapy to prevent poststent restenosis caused by insufficient endothelialization. However, patients are required to undergo dual antiplatelet therapy for up to 12 months until the endothelial monolayer is fully formed. Unfortunately, antiplatelet agents, such as aspirin and clopidogrel, pose a high risk of life-threatening hemorrhage. 8,9 Insufficient endothelial monolayer recovery is the primary cause of long-term complications after stenting. [10][11][12] Thus, promoting endothelialization not only reduced the demonstrated that a catalytic surface improved endothelial cell compatibility but inhibited VSMC proliferation by depositing Ti-Cu coatings on stainless steel substrates. 13 Ming et al. provided evidence that fabricated shellac coatings decrease neointimal hyperplasia by regulating VSMC phenotypic transformation. This coating design also exhibited excellent blood compatibility and antibacterial activity. 14 In addition, vascular endothelial growth factor (VEGF) and anti-CD34 antibodies were immobilized onto the surface of the Ni-Ti alloy sheet, which improved endothelial cell proliferation. 15 Similarly, the endothelialization process was promoted by conjugating human blood exosomes onto the surface of a polydopamine-coated 316L stainless steel stent. 16 Carbon coatings deposited using a mixed-mode high power impulse magnetron sputtering (HiPIMS) process provide biocompatible interfaces suitable for blood-contacting devices. 17 Moreover, optimized Ti-xCu coatings with Ti and Cu/CuTix crystals improved endothelial cell compatibility and inhibited VSMC proliferation. 18 In contrast to modifying surface coatings or biological coating surfaces, special attention has also been paid to drug-loaded coatings. By encapsulating paclitaxelloaded mesoporous silica nanoparticles within electrospun polylactic acid fibers, bio-functional stents covered with dual drug-loaded electrospun fibers achieved programmed VEGF and paclitaxel release, which increased endothelialization and reduced long-term stenosis. 19 Additionally, the poly-dopamine/hexanediamine-chondroitin sulfate C coating exhibited rapid endothelialization and better blood compatibility in cardiovascular implants. 20 Our previous study observed rapid endothelialization after the implantation of a nitrided iron stent. Thus, we speculated that the F I G U R E 1 Corrosion pits generated in the surface of the iron stent after degradation. (a) Schematic diagram of the dynamic circulation device: a peristaltic pump was introduced to mimic vessel pulsation, the silastic tube was connected to the peristaltic pump, and the pump continuously transferred plasma from one end to the other. The iron stent was deployed into a transparent conduit. Additionally, a S316L stent was placed near the iron stent as a control. (b) As the circulation time increased, the appearance of the iron stent turns dark, and black iron corrosion granules were generated around the struts. The red and green arrows indicate the S316L stent and iron stent, respectively. (c, d) Pitting corrosion was extensively generated in the corroded iron stent, and the surface became rough and uneven, whereas no corrosion pits were observed in the S316L stent, and the surface remained polished. rough and uneven surface of the corroded iron stent influences the adhesion of blood components and promotes endothelialization. To verify this hypothesis, we performed corrosion experiments to confirm the degradation characteristics of the iron stent after implantation and detailed investigations of the morphology of the corrosion pits generated by iron stent degradation. An experimental arteriovenous shunt model was introduced to determine blood component adhesion. We found that owing to the generation of corrosion pits, the surface of the iron stent became rough and uneven, which increased fibrin deposition on the corroded iron stent surface. To further evaluate the effects of fibrin deposition on endothelialization, we implanted a corroded iron stent in both the carotid and iliac artery bifurcations. The results showed that corroded iron stents increased fibrin deposition and promoted nonendothelial formation. To explore the effects of fibrin deposition on endothelialization, we conducted co-culture experiments under dynamic flow conditions. These results indicated that fibrin deposition in the corroded stent enhanced endothelialization by stimulating endothelial cell adhesion and proliferation. Our study as a whole is the first to delineate the role of iron stent corrosion in endothelialization, pointing to a new direction for enhancing endothelialization after stenting.
2 | RESULTS 2.1 | Corrosion pits were extensively generated in the corroded iron stent A circulation device was introduced to delineate the corrosion of the iron stent, as shown in Figure 1a. The stent was deployed in the F I G U R E 2 Numerous fibrins deposited on the corroded iron stent surface. (a, b) Schematic diagram of the arteriovenous shunt experiment. A transparent conduit was connected between the femoral and vein arteries in rabbits. The corroded iron stent was deployed in the middle of the conduit. A new iron stent was inserted as the control. (c, d) Numerous fibrins are deposited uniformly over the entire surface of the corroded iron stent. These fibrin networks become increasingly dense with prolonged circulation time. However, considerably less fibrin is present in the control stent. The red arrow indicates the fibrin in the corroded iron stent. (e) No statistically significant differences were identified regarding the platelet, red blood cell, or leukocyte levels between the two stents. Bars represent the mean ± standard deviation (SD). *p < 0.05; "ns" represents no statistical difference. middle of a transparent conduit, and plasma was circulated through it. Additionally, a 316L stainless steel stent (S316L stent) was placed near the iron stent as a control. As shown in Figure 1b, yellow iron corrosion granules were observed around the stent strut. As the experimental time increased, the iron stent became surrounded by corrosion granules. Moreover, corrosion pits were generated in the iron stent, leading to a rough and uneven surface (Figure 1c,d). In contrast, few corrosion granules were observed in the S316L stent, and its surface remained polished. These results demonstrate the surface changes in the corroded iron stent, confirming the degradation characteristics of the iron stent following implantation.

| Numerous fibrins were deposited on the corroded iron stent surface
As the surface of the corroded iron stent became rough and uneven, we speculated that it may influence the adhesion of blood components, such as erythrocytes, platelets, leukocytes, and fibrin. An arteriovenous shunt experiment was performed to address this issue, where a transparent conduit was connected to the femoral artery and vein in the rabbit model. A corroded iron stent was then deployed in the middle of the conduit. An iron stent without corrosion was used as a control. A schematic of this process is presented in Figure 2a,b.
With prolonged experimental time, we found numerous fibrin strands deposited uniformly over the entire surface of the corroded iron stent, forming thick fibrous networks in some areas. However, the surface of the new iron stent was relatively clean, with little fibrin deposition on its surface (Figure 2c,d). In contrast to successful fibrin deposition, no statistical differences were found in the adhesion of red blood cells, platelets, and leukocytes between the two groups (Figure 2e-g). Clinical observations have indicated that the rate of endothelialization is delayed when the stent is implanted at the artery bifurcation. This is due to the dramatic change in the direction of blood flow in the artery bifurcations, leading to a much higher shear stress on the vessel wall and insufficient endothelialization. 21,22 To further address the role of fibrin deposition in artery bifurcation endothelialization, we implanted a corroded iron stent in the iliac artery bifurcation (Figure 4a,b). The fibrin-coated iron stent and the S316L stent were used as controls. All rabbits (n = 14; weight, 5.2 kg ± 0.5 kg; 7 males and 7 females) were successfully stented, and no periproce- Thus, we confirmed that fibrin deposition in a corroded iron stent promoted endothelialization in both carotid artery and iliac bifurcation implantation models.

| Fibrin on corroded iron stent promotes endothelialization
F I G U R E 4 A corroded iron stent promotes endothelialization at artery bifurcation sites. (a) Artery angiography during stent implantation in the iliac artery bifurcation. The red arrow indicates the iliac artery bifurcation. (b) Schematic diagram of iliac artery bifurcation implantation experiment. (c) Complete endothelial cell coverage in the corroded iron stent. However, the struts were exposed to the vascular lumen, and the endothelialization process was delayed in the S316L stent. (d) Endothelial monolayer thickness is higher on the corroded iron stent. (e) Rapid endothelialization process on the corroded stent surface. However, much less endothelial cell coverage is found in the S316L stent. (f) Fibrin deposition is increased in the corroded iron stent. Bars represent the mean ± standard deviation (SD). *p < 0.05; **p < 0.01

| Fibrin deposition facilitated endothelial cell adhesion and proliferation
Endothelial cell adhesion to the stent surface is the initial step in endothelialization. Endothelialization increased following fibrin deposition in the corroded iron stent. We speculated that fibrin deposition promotes endothelialization by increasing endothelial cell adhesion. Therefore, the relationship between fibrin deposition and endothelial cell adhesion was investigated in vitro. Notably, endothelial cell adhesion responds to fluid flow. Therefore, the biological response of endothelial cell adhesion must be examined under dynamic flow conditions, and a parallel-plate flow chamber system was introduced to In addition to endothelial cell adhesion, endothelial cell proliferation plays an essential role in endothelialization. The co-culture medium was used to quantitatively assess endothelial cell proliferation after fibrin deposition. A fibrin-coated plate and a nontreated plate Consistently, the number of endothelial cells is also elevated in the fibrin-coated plate group. The white arrow indicates the positive signal of VEGF-A staining. Bars represent the mean ± standard deviation (SD). *p < 0.05; **p < 0.01; "ns" represents no statistical difference.
that fibrin deposition promotes endothelialization by facilitating endothelial cell adhesion and proliferation.

| Fibrin deposition promotes endothelialization via VEGF-A upregulation
VEGF-A is a strong endothelial cell activator. The endothelialization process increased in the corroded iron stent, which led us to further explore the role of VEGF-A in endothelialization. We first measured VEGF-A expression in artery tissues implanted with corroded iron stents. Coinciding with rapid endothelialization, VEGF-A displayed high-intensity values in corroded iron stent artery tissue, indicating elevated expression of VEGF-A induced by fiber adhesion (Figure 6a,b). In addition to VEGF-A, other molecules that could potentially influence endothelial cell growth, including VCAM-1 and FGF-2, were examined. However, these molecules were not induced after the implantation of the corroded iron stent (Figure 6c,d). To further study the effects of VEGF-A on endothelialization, we performed an ex vivo experiment using a co-culture medium. A fibrin-coated plate and an untreated plate (negative control) were used. As shown in Figure 6e,f, endothelial cell exposure in the fibrin-coated plate increased VEGF-A expression. In accordance with VEGF-A, the number of endothelial cells in the fibrin-coated plate was higher than that

| Hemocompatibility evaluation following corroded iron stent implantation
Hemocompatibility is a basic requirement of biomedical materials.
Therefore, it is essential to evaluate the hemocompatibility of corroded iron stents following implantation. During corroded iron stent degradation, iron ions are released from the stent and high concentrations of iron ions can cause heavy metal intoxication. Thus, iron ions must be dynamically monitored after implantation. As shown in APTT were similar to those of the S316L stent. Corrosion granules were produced from the iron stent during degradation. Corrosion granules may be transported by blood flow and block downstream distal branches. Therefore, the risk of vessel embolism after corroded iron stent implantation must be evaluated, and follow-up angiography was conducted. As presented in Figure 7d, the peripheral blood vessels in the intracranial, iliac, and subclavian arteries were clearly visualized, indicating that no embolism had formed in the distal branch.

| DISCUSSION
According to a research report from the World Health Organization, the morbidity and mortality rates of cardiovascular diseases (CVDs) are much higher than those of other disease types. 23 Currently, interventional surgery is the most effective treatment for acute CVDs, 24 wherein percutaneous transvascular balloon angioplasty and stent implantation are commonly used. 2 However, more than 30% of patients suffer from obstruction of the treated artery within 3 months, owing to insufficient endothelial cell recovery. 3,4 The endothelial monolayer not only functions as a dynamic physiological border isolating the surrounding tissue from circulating blood flow, but it also provides a nonadhesive surface for leukocytes and platelets in the vascular system. 25,26 Moreover, endothelial monolayers regulate platelet activation and aggregation, which are closely associated with occlusive thrombosis after interventional procedures. 6,7 Therefore, promoting neointimal coverage and restoring endothelial function have emerged as crucial concerns after stent implantation surgery.
Bioresorbable stents are heralded as the fourth revolution in interventional technology and are designed to avoid the long-term health risks posed by drug-eluting stents. 27,28 Iron and iron alloys are candidate materials for bioresorbable stents, owing to their excellent mechanical performance and biocompatibility. [29][30][31] Furthermore, we recently observed rapid endothelialization after iron stent corrosion.
With the degradation of the iron stent, numerous corrosion pits are generated on its surface. 32 Therefore, we speculated that with the production of corrosion pits, the surface of the iron stent became rough and uneven, thereby promoting the adhesion of blood components, which ultimately influenced endothelial cell coverage. To verify this, we first described the morphology of the corrosion pit by deploying an iron stent in a plasma-circulation device. Because of the excellent corrosion resistance of the 316L stainless steel stent (S316L sent), we introduced the S316L stent and set it as the control to elucidate the corrosion properties of the iron stent. As expected, yellow corrosion granules were extensively generated around the stents.
Moreover, numerous corrosion pits were generated around the stent, with corrosion granules carried away by the flowing fluid, and the surface became rough and uneven. These findings describe the development of corrosion pits and confirm the degradation characteristics of iron stents after implantation.
Biomaterial properties such as surface roughness and composition have been shown to influence their interaction with surrounding blood proteins. 33 Although the surface changes of corroded iron stents are described above, the effects of corrosion pits on the adhesion of blood components need to be further elucidated. We performed an arteriovenous shunt experiment using a rabbit model and deployed the corroded iron stent in a transparent conduit simultaneously connected to the femoral artery and vein. A new iron stent was used as a control. Notably, we found that short, thin fibrin strands developed uniformly over the entire surface of the corroded stent, forming thick fibrous networks in some areas. Moreover, these fibrin networks grew and became increasingly dense with prolonged circulation time. In contrast, the surface of the new iron stent was relatively clean, with little fibrin being deposited on the surface. In addition to changes in fibrin attachment, a rough surface may also influence the adhesion of other blood components, such as platelets, red blood cells, and leukocytes. Therefore, we measured the adhesion between the components. However, no differences in blood components were identified between the two stents. Hence, we concluded that iron stent corrosion increased fibrin deposition on its surface.
Fibrin is a large, complex, fibrous glycoprotein with three pairs of polypeptide chains linked by 29 disulfide bonds. 34,35 Fibrin deposition is essential for hemostasis and angiogenesis. In particular, fibrin enhances the cellular production of extracellular matrix proteins to serve as a basement membrane, which is essential for achieving stable microvascular networks. 35,36 Therefore, temporary fibrin deposition promotes cell attachment, spreading, migration, and alignment. 37,38 Our previous findings suggest increased endothelialization after iron stent corrosion. Thus, we speculated that rapid endothelialization was involved in fibrin deposition on the corroded iron stent surface. A convincing and novel arterial implantation model was used to test this hypothesis. Clinical observations indicated that the endothelialization process was dramatically delayed if the stent was deployed at the artery bifurcation. The dramatic change in the direction of blood flow at the artery bifurcation and the accompanying shear stress were much higher on the vessel wall in this position, impeding endothelialization. To address the effects of fibrin deposition on endothelialization, we chose the iliac artery bifurcation and implanted a corroded stent in this position. Additionally, the fibrin-coated iron stent and S316L stent were used as controls. Remarkably, rapid endothelialization was observed in fibrin-coated stents during the early experimental period. With a prolonged implantation time, more fibrin deposits were detected on the surface of the corroded iron stent. Subsequently, neointimal coverage of the corroded stent is initiated. However, considerably less fibrin deposition was observed in the S316L stent, and its surface was smooth with little endothelial cell coverage.
In short, by introducing the bifurcation artery implantation model, we demonstrated that iron stent corrosion promoted fibrin deposition, thereby increasing the endothelialization rate.
Fibrin serves as an extracellular matrix protein that provides the basis for endothelial cell adhesion and growth. 37,38 Fibrin coating of polytetrafluoroethylene prostheses increased endothelialization. 39 Moreover, fibrin coating supports endothelial cell maturation and stabilization under blood-flow conditions. 40,41 Indeed, fibrin can bind to a variety of blood proteins, including fibronectin, von Willebrand factor, albumin, and VEGF. Fibrin participates in cell growth, thrombosis, and inflammation. 42,43 Endothelialization was promoted after fibrin deposition on the corroded stent surface. Therefore, we propose that fibrin attachment enhances endothelialization by increasing endothelial cell adhesion. To test this hypothesis, we performed novel ex vivo experiments by introducing fibrin-coated stents. In addition, to mimic the real circumstances of blood flow, the co-culture medium was subjected to dynamic flow conditions, and a parallel-plate flow chamber system was used. As expected, we observed faster endothelial cell adhesion on the fibrin-coated stent surfaces. Moreover, noticeable changes in cell adhesion-related mRNA levels were observed with the stent. In addition to endothelial cell adhesion, cell proliferation was measured quantitatively. However, no significant differences were observed during the early experimental period. In contrast, the number of cells began to increase in the fibrin-coated plate with prolonged incubation time, and the number of endothelial cells was approximately 200% higher in the fibrin-coated plate than in the control plate. Based on these results, we demonstrated that fibrin deposition enhanced endothelialization by promoting endothelial cell adhesion and proliferation.
Endothelialization is a complex process that involves endothelial cell adhesion and proliferation. VEGF-A promotes endothelial cell growth via both direct and indirect mechanisms. VEGF-A stimulates endothelial cell migration and proliferation near microvasculature. [44][45][46] Therefore, VEGF-A is recognized as an important endothelialization factor. As a rapid endothelialization process was observed in the fibrin-coated stent, we further determined the role of VEGF-A in endothelialization. We detected VEGF-A expression in a fibrin-coated stent in vivo using immunofluorescence (IF) staining. Consistent with rapid endothelial cell coverage, VEGF-A IF results displayed highintensity values in the fibrin-coated stent group. Furthermore, an ex vivo experiment was conducted, and an anti-VEGF-A antibody was used to confirm this finding by inhibiting VEGF-A function. As Our study as a whole is the first to describe the role of iron stent corrosion in rapid endothelialization, pointing to a new direction for enhancing endothelialization after stenting ( Figure S2).

| Device parameters and performance
The detailed microstructure and composition of nitrided iron stents (Fe alloyed with 0.074 wt% N) used in this study have been previously described. 47 Nitrided iron stents and 316L type stainless steel stent

| Iron stent corrosion
As shown in Figure 1, the experimental device consisted of three different parts: the pumping system, monitoring system, and conduitconnected setup. First, one end of the silastic conduit was connected to the recycling pump, and the other end was placed into a reservoir filled with plasma. The pump was then immersed in the reservoir, and the plasma was transferred from one end to the other. The monitoring system was added to the conduit system and used to measure experimental parameters, such as temperature, pH, and flow dynamic velocity. The temperature was set between 36 and 37 C, and the pH ranged from 7.35 to 7.45. A new iron stent was deployed in the middle of the conduit, and another S316L stent was used as a control.
Stent appearances and corrosion pits were observed during prolonged corrosion. The experiment was performed in triplicate.

| Rabbit arteriovenous shunt
Ex vivo arteriovenous shunt models in rabbits were used to evaluate the adhesion of blood components to the corroded iron stents. The initial mass of the stent was recorded before implantation. On the day before the arteriovenous shunt experiment, the animals received a loading dose of 25 mg of clopidogrel and aspirin to prevent platelet accumulation. All rabbits received diazepam (1 mg/kg) and ketamine (25 mg/kg) via subcutaneous injection for anesthesia with mechanical ventilation. The right femoral artery was surgically exposed, and a 4F guide catheter was introduced over a 0.356 mm guidewire. The right femoral vein was then exposed, and a transparent silicone conduit was immediately connected to the guide catheter and vein. Two cor-

| Stent implantation
The corroded iron stent (Φ3.0 Â 18 mm, thickness 90 μm) and the S316L stent (Φ3.0 Â 18 mm, thickness 90 μm) were placed into the right carotid artery. In addition, the corroded stent covered with fibrin and the S316L stent (Φ3.0 Â 18 mm, thickness 90 μm) were separately placed in the iliac artery bifurcation; 14 adult New Zealand white rabbits were included in the study (weight, 5.0 kg ± 0.4 kg; 7 males and 7 females). The initial mass of the stent was recorded before implantation. All rabbits received 25 mg of clopidogrel and 25 mg of aspirin to avoid platelet accumulation the day before implantation. All rabbits received diazepam (1 mg/kg) and ketamine (25 mg/kg) via subcutaneous injection for anesthesia with mechanical ventilation. The right femoral artery was surgically exposed, and a 4F guide catheter was introduced over a 0.356 mm guidewire. The stent was then introduced using a guidewire and deployed into the right carotid artery or the iliac artery bifurcation. Heparin (200 IU/kg) was administered to all rabbits via a catheter to maintain an activated coagulation time of >300 s during surgery.

| Scanning transmission electron microscopy
All samples obtained from the vasculature surrounding the implanted prosthesis were processed by sectioning (50 nm) polished cross sections, and the images were captured with a Leica EM UC6 ultramicrotome. Briefly, images of stented arteries were obtained. The embedded tissue samples were then cross-sectioned into 1.5 mm-thick slices and successively ground with 1200, 2000, and 7000 grit silicon carbide grinding papers. Fibrin deposition on the stent surface was observed using STEM (JSM-6510; JEOL, Japan). Three images were acquired for each sample.

| Artery angiography
Carotid and iliac artery angiography images were obtained using the CGO-2100 Cath-Lab system (Wandong, China). An angiographic catheter is inserted proximally to the target artery. The contrast agent was injected into the target artery using an angiographic catheter.
X-ray images of the same location before and after contrast agent injection were collected using a CGO-2100 Cath-Lab system. The subtracted images were immediately processed using a CGO-2100 Cath-Lab system.

| Optical coherence tomography
The C7 XR Fourier-Domain System (LightLab Imaging, Westford, MA, USA) was used to collect OCT images. First, the OCT catheter was placed in the targeted stenting segments to capture the images. A contrast reagent (iodixanol 370; GE Healthcare, Viviparus) was injected via the guiding catheter. Subsequently, the detected fiber was drawn back into the arterial lumen at a speed of 20 mm/s. The generation rate of OCT images was 100 fps.

| Cell cycle analysis
Flow cytometry was performed to measure the cell cycle progression.
Endothelial cells used for flow cytometry were obtained from rabbit vascular tissue that was serum-starved overnight. The cells were then stimulated with platelet-derived growth factor-BB (PDGF-BB) (20 ng/ml) for 20 h. Ethanol (70%) was used to fix the trypsinharvested cells, and the fixed cells were stored at 4 C. The cells were resuspended in phosphate-buffered saline three times. Subsequently, all cells were incubated with a propidium iodide solution (50 μg/ml, 400 μl, and 100 μg/ml RNase A). Fluorescence was detected and measured using a FACSCalibur flow cytometer (Becton Dickinson).
The percentage of endothelial cells in the G0/G1, S, and G2/M phases was calculated. The experiments were performed in triplicate.

| Statistical analysis
SPSS version 18.0 (SPSS Inc., Chicago, USA) was used for the statistical analysis. A one-way analysis of variance was used to assess the statistical differences between groups. Statistical significance was set at a p-value <0.05.

ACKNOWLEDGMENTS
This study was supported by the National Science Foundation of China (No. 81271325).