Abstract

Using poly-L-lactic acid for implantable biodegradable scaffold has potential biocompatibility issue due to its acidic degradation byproducts. We have previously reported that the addition of amorphous calcium phosphate improved poly-L-lactic acid coating biocompatibility. In the present study, poly-L-lactic acid and poly-L-lactic acid/amorphous calcium phosphate scaffolds were implanted in pig coronary arteries for 28 days. At the follow-up angiographic evaluation, no case of stent thrombosis was observed, and the arteries that were stented with the copolymer scaffold had significantly less inflammation and nuclear factor-κB expression and a greater degree of reendothelialization. The serum levels of vascular endothelial growth factor and nitric oxide, as well the expression of endothelial nitric oxide synthase and platelet-endothelial cell adhesion molecule-1, were also significantly higher. In conclusion, the addition of amorphous calcium phosphate to biodegradable poly-L-lactic acid scaffold minimizes the inflammatory response, promotes the growth of endothelial cells, and accelerates the reendothelialization of the stented coronary arteries.

1. Introduction

Although metallic drug-eluting stents have dramatically improved the outcome of patients with cardiovascular disease, in-stent restenosis (ISR) and stent thrombosis (ST) remain major challenges [14]. Stents are only needed during the vessel healing and reendothelialization periods and, if left in place for a longer period of time, can actually prove harmful by sustaining inflammation, increasing the incidence of ISR and ST, and promoting negative remodeling [57]. In this context, stents made of biodegradable polymers such as poly-L-lactic acid (PLLA) appear to be an attractive alternative to durable polymer, as they can be completely replaced by healed tissue and allow positive remodeling [813]. However, the poor biocompatibility of PLLA is a major obstacle to its use as a stent component. Indeed, stents made of PLLA have a higher inflammatory potential than metallic stents and are more thrombogenic because of the acidic byproducts released during their degradation. These effects likely promote the development of ISR and the occurrence of ST [12, 14], interfere with the proliferation of endothelial cell, and delay vessel healing [15, 16]. Since the rapid and complete reendothelialization of the stented segment can prevent the occurrence of inflammatory and thrombotic events, a polymer composite with endothelial cell-positive properties could prove to be of great value for the development of novel stent technologies.

Amorphous calcium phosphate (ACP) is used in the form of coatings, ceramics, and composites in numerous biomaterials and has excellent biocompatibility due to its chemical similarity to human bone tissue [17]. It has been reported that ACP can reduce the inflammatory reaction associated with the hydrolytic degradation of PLLA [9, 10, 18]. When in the human body, ACP releases ions into aqueous media, which are hydrolyzed to form hydroxide ions (OH). In turn, OH then neutralizes the carboxylic acid end group available from the hydrolytic degradation of PLLA. In our previous studies in rats and rabbits, we have reported that the addition of ACP to the PLLA polymer coating significantly improved the biocompatibility of the stent [9, 10]. In this study, we have further investigated the biofunctions of ACP in a biodegradable scaffold consisting of a PLLA/ACP composite when implanted in porcine coronary arteries.

2. Materials and Methods

2.1. Scaffold Preparation

All scaffolds (6 PLLA and 6 PLLA/ACP) were developed and produced by VasoTech, Inc. (Lowell, MA, USA). ACP (size < 150 nm; Ca/P~1 : 1) and paclitaxel were homogeneously mixed with PLLA powder (MW = 250,000 g/mol) with a ratio of PLLA/ACP/PAX at 96/2/2 (w/w/w) using a speed mixer (SpeedMixer DAC 600). The mixture was dried at around 60°C overnight prior to extrusion in a single screw extruder (Genca Engineering Inc., Saint Petersburg, FL). For comparison, 2% paclitaxel was also homogeneously mixed with PLLA powder, and the mixture was then extruded under the same condition with PLLA/ACP/PAX extrusion. The extruded tubes have a uniform wall thickness of 150 μm and an outer diameter at 1.8 mm. The extruded tubes were laser-automated according to design specifications (3.0 mm diameter × 13 mm length × 150 μm width). One radiopaque metal marker was incorporated on each end. All scaffolds were crimped on 3.0 mm × 15 mm balloon catheters and sterilized with gamma radiation prior to implantation (Figure 1).

2.2. Animal Preparation and Scaffold Implantation

Twelve Tibetan miniature pigs of either sex and weighing between 20 Kg to 25 Kg were purchased from Pearl River Laboratory (Dongguan, China). The study protocol was approved by the Institutional Animal Care Committee at the Affiliated Hospital of Jinan University, School of Medicine. All procedures involving animals conformed with the “Guide for the Care and Use of Laboratory Animals” published by the US National Institutes of Health (NIH Publication number 85-23, revised 1996).

The implantation procedures were performed as described previously [9]. Briefly, all animals received dual antiplatelet therapy (325 mg aspirin and 75 mg clopidogrel daily) for 3 days prior to the procedure. After a 12-hour fasting period, the animals were anesthetized with 0.3 mg/Kg subcutaneous ketamine and 30–40 mg/Kg intravenous pentobarbital for continuous sedation. A 7F guiding catheter was inserted percutaneously into the coronary artery through the left femoral artery and, after the infusion of 200 μg of nitroglycerin into the coronary artery, quantitative coronary angiography (QCA) was performed from three different views. Tortuous arteries and arteries with diameters <2.5 mm or >3.0 mm were excluded. Each scaffold was implanted randomly in the left anterior descending coronary artery or the right coronary artery (~diameter of 2.8 mm and no obvious tapering, stent to artery ratio 1.1 : 1), slowly expanded to 8 atmosphere (nominal balloon pressure) for 20 seconds before a second QCA was performed. The animals were then given 7,000 U of heparin through the arterial sheath and 200 μg of intracoronary nitroglycerin to prevent vasospasms.

2.3. Follow-Up Quantitative Coronary Angiography and Histopathological Evaluation

At 28 days after implantation, all animals were anesthetized and underwent a follow-up QCA using the same procedures as described above. After the QCA, the animals were euthanized with an anesthetic overdose (4% pentobarbital sodium) and the hearts were harvested, rinsed with 0.9% heparinized saline, and perfusion-fixed with 10% buffered formalin for an hour for histopathological evaluation.

2.4. Quantitative Coronary Angiography Analysis

The QCA analysis was performed offline with a computer-assisted system using an automated edge detection algorithm (MEDIS, Cardiovascular Angiography Analysis System II, Pie Medical Data, Maastricht, NL) by investors blinded to the treatment groups. The segments with the implanted stents were analyzed using two orthogonal views. The proximal luminal diameter (LDp), the distal luminal diameter (LDd), and the middle luminal diameter (LDm) were measured in each stented artery segment. The mean luminal diameter (MLD) was defined as (LDp + LDd + LDm)/3.

2.5. Histology and Immunohistochemistry

All stented arteries were paraffin-embedded and sectioned into 5 μm thick proximal, central, and distal segments. The sections were then stained with hematoxylin-eosin, as described previously [9, 10].

Two stented arteries from two animals were further analyzed with scanning electron microscopy (SEM) to evaluate the degree of reendothelization. For this experiment, the samples were cross-sectioned with a razor blade, and each cross section was sputter-coated with gold-palladium using a desktop gold sputter coater. Photomicrographs were taken using the SEM (VEGA 3, Czech Republic) set at 20 kV.

The inflammation score of each stented segment was graded as described previously: 0, none; 1, scattered inflammatory cells; 2, inflammatory cells encompassing 50% of a strut in at least 25% to 50% of the circumference of the artery; and 3, inflammatory cells surrounding a strut in at least 50% of the circumference of the artery [9, 10]. The endothelialization score was defined as the extent of the circumference of the arterial lumen covered by endothelial cells and graded from 0 to 3 (0 =< 25%; 1 = 25–50%; 2 = 50–75%; and 3 => 75%).

The expression of nuclear factor-κB (NF-κB), platelet-endothelial cell adhesion molecule-1 (PECAM-1), and endothelial nitric oxide synthase (eNOS) was assessed in all stented arteries by immunohistochemistry staining according to the manufacturer instructions (Immunohistochemistry kits, Biosynthesis Biotechnology Co., Beijing, CN). Three individual sections per segment (proximal, central, and distal) were analyzed, and their average was used as a measure of quantification. The percentage of positive cells and the average optical density were analyzed by using Image-Pro Plus 6.0 analysis software (Media Cybernetics, Inc.). The positive expression of the indicator index was defined as the percentage of positive cell × average optical density × 100. The mean value of the proximal, middle, and distal sections was calculated and used as the final measurement for each specimen.

2.6. Blood Sample Assessment

For each animal, 10 mL of coronary arterial blood was withdrawn before the procedure and at 28 days after implantation. The serum concentration of nitric oxide (NO, umol/L) and vascular endothelial growth factor (VEGF, pg/mL) were determined using ELISA kits (Nanjing Jiancheng Biological Engineering Institute, China, and RB Company, USA, resp.) according to manufacturer instructions.

2.7. Statistical Analysis

Statistical analyses were performed with SPSS software version 17.0 (Statistical Product and Service Solutions Ltd.). All data were expressed as mean ± standard deviation (SD), and categorical variables were expressed as counts (%). Independent -tests were performed to detect between-group differences. All statistical tests were 2-tailed, and a value of was considered statistically significant.

3. Results

3.1. Quantitative Coronary Angiography Analysis

The QCA analysis showed that all 12 scaffolds were successfully deployed at the predetermined diameter, and the stented vessels and the distal branches were open without any sign of peripheral embolization or thrombosis. At 28 days’ follow-up, the MLD and percent diameter stenosis were not significantly different between the two groups (Table 1).

3.2. Histology Analysis

At necropsy, gross examination of the hearts showed no coronary abnormalities, epicardial hemorrhage, myocardial infarction, or aneurysms in any of the six PLLA/ACP stented arteries (Figure 2(a)). However, in the PLLA group, one animal had notable tissue inflammation around the stented segment (Figure 2(b)).

Histology evaluation revealed that all scaffolds had neointimal growth present on their surface (Figure 3(a)), and one animal in the PLLA group had notable arterial wall swelling and inflammation (Figure 3(b)). Specifically, SEM showed that the inner layer of the neointima had a rough and textured surface (Figure 4(a)), indicating that the injured intima was not completely healed. In the PLLA/ACP group, the neointima surface was smooth with some endothelial cells immerging (Figure 4(b)), indicating that the reendothelialization process was in progress.

Animals in the PLLA/ACP group had a significantly lower inflammation score ( versus for PLLA, ) and higher endothelization score ( versus for PLLA, ) (Figure 5(a)).

Immunohistochemistry analysis revealed the positive staining of NF-κB, eNOS, and PECAM-1, among which NF-κB was mainly distributed in the nuclei of the inflammatory cells and eNOS and PECAM-1 were mainly distributed in the nuclei of the endothelial cells (Figure 5(b)). The expression of NF-κB in PLLA/ACP stented arteries was significantly less than that in PLLA stented arteries ( versus , ) (Figures 6(a) and 6(b)), while the expression of both eNOS ( versus , ) (Figures 6(c) and 6(f)) and PECAM-1 ( versus , ) (Figures 6(b) and 6(e)) was significantly higher in the PLLA/ACP group than that in the PLLA group.

Compared with the preimplantation values, animals in both the PLLA/ACP and PLLA groups had significantly lower serum concentrations of NO (PLLA/ACP: versus , ; PLLA: versus , ) and higher serum concentrations of VEGF (PLLA/ACP: versus , ; PLLA: versus , ) at 28 days after implantation. In the PLLA/ACP group, both the serum concentrations of NO ( versus , ) and VEGF ( versus , ) were significantly higher than that observed in the PLLA group (Table 2).

4. Discussion

The polyester family of polymers, including PLLA, poly-D,L-lactic acid (PDLA), polyglycolic acid (PGA), and poly-lactic-co-glycolic acid (PLGA), have been intensely investigated for biodegradable scaffold platform applications [8, 1923]. Among these, PLLA was considered a promising candidate with excellent chemical and mechanical properties, but suboptimal biocompatibility has become a major challenge when used as a coating on coronary artery stents [24].

In the present porcine model of coronary artery stenting, both the histopathology and QCA analyses performed at 28 days after implantation showed, with respect to restenosis formation, that the use of a PLLA/ACP composite to generate a fully biodegradable scaffold performed better than when PLLA was used alone. These results are in agreement with our previous studies in rats and rabbits and clearly demonstrate that ACP plays a significant role in improving PLLA biocompatibility and in vivo performance.

4.1. Inhibition of Inflammation

Our histological analyses showed that arteries stented with the PLLA/ACP scaffold had significantly less inflammatory cell infiltration and, consequently, a lower inflammation score. Immunohistochemistry also revealed that the expression of NF-κB was significantly lower in PLLA/ACP stented arteries than in PLLA stented arteries. NF-κB is a protein complex that controls the transcription of DNA and plays a key role in the regulation of cellular responses to stimuli such as inflammation, infection, cancer, and autoimmune diseases [25]. In human atherosclerotic plaques, it has been clearly demonstrated that NF-κB is activated and plays a major role in the upregulation of the proinflammatory and prothrombotic responses within the plaque [2628]. Our results therefore suggest that ACP can significantly alleviate the tissue inflammatory response caused by the scaffold implantation.

The hydrolytic degradation byproducts of PLLA are lactic acid and its oligomer with –COOH groups at the end, which can induce an inflammatory response. As discussed in our previous reports [9, 10], there are two possible reasons as to the inhibitory potential of ACP on the inflammatory process. First, ACP can produce supersaturated levels of Ca2+ and ions by releasing ions into the aqueous media. The released ions can be further hydrolyzed, generating OH ions. When ACP is blended into PLLA and placed in aqueous media, the OH ions can neutralize the acidic –COOH groups, thus reducing inflammation. Second, the released Ca2+ may also be trapped by –COOH groups to form insolubilized salts, providing another possible route to remove the acid.

4.2. Acceleration of Reendothelialization

Studies have demonstrated that delayed arterial healing, characterized by incomplete reendothelialization, is the most powerful histological predictor of ST [5, 29]. Therefore, the presence of a continuous endothelial layer over the stent scaffold will prevent or, at least, diminish the occurrence of thrombotic events. Furthermore, it has been reported that the degree of ISR could be considerably reduced if the endothelium regenerates rapidly and completely after stenting [30].

Platelet-endothelial cell adhesion molecule-1 (PECAM-1) is a structurally important marker for endothelial cells. It mediates cell-cell communication, upregulates the function of integrins, and plays a pivotal role in the transendothelial migration of leukocytes, the regulation of platelet function, the inhibition of apoptosis, and the mediation of signal transduction [3133]. In the present study, both morphological and morphometric analyses showed that the inner surface of the PLLA/ACP stented arteries had significantly greater endothelial cell coverage than that observed in the PLLA stented arteries. Additionally, our immunohistochemistry analysis revealed that PECAM-1 expression in the PLLA/ACP stented arteries was significantly greater than that of PLLA stented arteries. These results suggest that ACP promotes the growth of vascular endothelial cells and accelerates the reendothelialization process after stent implantation.

Our study also showed that, 28 days after implantation, the expression of eNOS within the vascular wall and the serum concentration of NO were significantly higher in the PLLA/ACP group than in the PLLA group. Endothelial nitric oxide synthase is mainly distributed in the vascular endothelium and is a key enzyme necessary for the production of NO by endothelial cells [3436]. Under physiological conditions, eNOS is continuously activated to sustain the constant production of NO by vascular endothelial cells, which is an indicator of endothelial cell integrity. The secreted NO acts as a barrier between endothelial cells and the proinflammatory mediators present in the circulating blood. Because the nanomaterial ACP upregulates the function of eNOS, it may prevent endothelial cell dysfunction following vascular injury as is observed during stenting.

Vascular endothelial growth factor is a highly specific vascular endothelial mitogen, which selectively enhances the mitosis of vascular endothelial cells and degrades the extracellular matrix to allow the migration of endothelial and perithelial cells [37, 38]. By specifically targeting the vascular endothelial cells through the upregulation of A-1 and Bcl-2 expression, it also inhibits endothelial cell apoptosis. At the same time, it upregulates the decay accelerating factor (DAF) to protect endothelial cells from complement-mediated damage and promotes the proliferation of vascular endothelial cells and neovascularization. In the present study, the significantly higher expression of VEGF in the PLLA/ACP group suggests that the addition of ACP to the PLLA scaffold can better facilitate the repair of vascular endothelium.

In addition, because of the addition of nanoscale ACP, the PLLA/ACP scaffolds generate multiple nanopores during the degradation process. This microporous structure provides an excellent surface for cell attachment and possibly a delivery vehicle for cell-based therapies. In this case, blending ACP with PLLA creates nanometer pores that enlarge gradually to a micrometer scale as degradation process could greatly promote the reendothelialization of the injured vessel thereby leading to partial endothelial coverage within 1 month. Therefore, the risk of stent thrombosis is dramatically reduced and, possibly, completely avoided. Compared with the long-term foreign body inflammatory reaction and delayed endothelial repair observed with currently available stents, we believe that the PLLA/ACP scaffold has definitive clinical advantages.

4.3. Limitations and Future Studies

The study was performed in normal porcine coronary arteries, in which the different vascular endothelial repair processes may differ from those observed in atherosclerotic plaques. Also, 28 days of follow-up is short, and because the injured arterial wall is not fully remolded, a study performed using an atherosclerotic model with a longer follow-up period is necessary to fully understand the biological activities of ACP when being part of a PLLA/ACP composite.

5. Conclusion

ACP alleviates the inflammatory response following the implantation of a PLLA scaffold, promotes the growth of endothelial cells, accelerates reendothelialization, and restores endothelial cell structure and function. The addition of ACP to biodegradable scaffolds appears to have a promising future in cardiovascular applications.

Competing Interests

The authors declare that they have no competing interests.

Authors’ Contributions

Dongsheng Gu, Gaoke Feng, and Guanyang Kang equally contributed to this work.

Acknowledgments

The study was supported by grants from the Key Program for International S&T Cooperation of China (no. 2011DFA33290 to Weiwang Gu and Tim Wu), the Industrial-Academic Research Collaboration Program of Guangdong, China (no. 2011A091000022 to Weiwang Gu and Tim Wu), the Guangdong Innovative and Entrepreneurial Research Team Program (no. 2014ZT05S008 to Tim Wu and Xuejun Jiang), and the International S&T Cooperation Project of Dongguan, China (no. 2013508150019 to Jianmin Xiao, Tim Wu, Xuejun Jiang, Gaoke Feng, and Guanyang Kang). The authors gratefully acknowledge the help of Chaoshi Qin in analyzing all quantitative coronary angiograms. They thank Weiguo Wan and Lin Xu for the surgical and interventional procedures. Thanks are due to China Scholarship Council.