|Home | About | Journals | Submit | Contact Us | Français|
Endothelial-cell function is important in the healing of damaged endothelium after percutaneous coronary artery damage. In 3 different animal models, we sought to determine whether rapamycin (sirolimus) affects the proliferation and migration of endothelial cells and endothelial progenitor cells.
First, after we implanted stents in dogs, we found that re-endothelialization was impeded more by drug-eluting stents than by bare-metal stents, 30 days after percutaneous coronary intervention. Second, in vitro in rats, we found that 1–100 ng/mL of rapamycin time- and dose-dependently inhibited proliferation over 72 hr (with effects evident as early as 24 hr) and also dose-dependently induced endothelial progenitor-cell apoptosis. Finally, in vivo in rats, we observed that vascular endothelial growth factor expression was decreased after 5 days of rapamycin treatment. We conclude that rapamycin impedes re-endothelialization after drug-eluting stent implantation by inhibiting the proliferation and migration of coronary endothelial cells, inducing endothelial progenitor-cell apoptosis, and decreasing vascular endothelial growth factor expression in the circulation.
Percutaneous coronary intervention (PCI) with lesion-stenting is the preferred treatment for patients who have acute coronary syndromes.1–3 The introduction of balloon-expandable bare-metal stents (BMSs) reduced coronary remodeling and restenosis rates in comparison with angioplasty alone.4,5 However, because the risk of restenosis with the use of BMSs remained greater than 10%, drug-eluting stents (DESs) were designed so that pharmacologic agents could be released after deployment. Rapamycin (sirolimus) or paclitaxel was added to 1st-generation DESs in order to reduce restenosis by inhibiting vascular smooth-muscle cell migration and proliferation, in turn reducing restenosis and target-vessel revascularization to less than 10% after DES implantation.6–8 Clinically, DESs are superior to BMSs and decrease major adverse cardiac events as well as restenosis rates.6,7,9 Although their restenosis rates are lower than those of BMSs, DESs have similar rates of in-stent thrombosis.9–16 Several hundred cases of stent thrombosis have been reported in rapamycin-coated stents,17 and reports imply that thrombosis rates of DESs may be even higher in “real-world” patients than rates reported in clinical trials.18,19 The reason for the discrepancy between lower restenosis rates and unaltered or higher thrombosis rates with DESs in comparison with BMSs is unclear.9,11
Several factors are involved in the pathogenesis of in-stent thrombosis. These include procedure-related factors, such as mechanical vessel injury or incomplete stent apposition; patient-related factors, such as vessel size or coagulation activity; and the thrombogenicity of the stent itself.20 The drugs used for coating the stent could also be involved in the development of in-stent thrombosis.20
Rapamycin, a macrocyclic lactone, is an immunosuppressive agent that arrests cells in the G1 phase of the cell cycle and induces apoptosis, inhibiting vascular smooth-muscle cell proliferation and migration.21 It becomes pharmacologically active by binding to ubiquitous, predominantly cytosolic immunophilin receptors. The binding of rapamycin to the mammalian target of rapamycin (known as RAFT 1, RAPT 1, and FRAP22–24) inhibits kinase activity and subsequently decreases the phosphorylation and activation of p70 S6 kinase, the translation of mRNA-encoding ribosomal proteins and elongation factors (eukaryotic initiation factor 4E binding protein [pH acid-stable protein I] and eEF-2), and the enzymatic activity of the cyclin-dependent kinase cdk2-cyclin E complex, resulting in a mid-to-late G1 cell-cycle arrest.25–29
However, the effect of rapamycin on re-endothelialization is still unclear. In this study, we measured the effect of rapamycin on the function of endothelial cells (ECs) and endothelial progenitor cells (EPCs), both of which contribute to endothelial healing.
Our supply of rapamycin (suspended in 100% methanol) was a gift from Microport Medical Co., Ltd. (Shanghai, China). Our experiments were performed on dogs and rats. Dogs were chosen in order to evaluate the effects of rapamycin after DES and BMS implantation. Two groups of rats were used: 1 group for the in vitro evaluation of rapamycin's effects on the function of ECs and EPCs, and another group for the in vivo evaluation of rapamycin's impact on vascular endothelial growth factor (VEGF) expression. All experiments conformed to our Institutional Guidelines for the Care and Use of Laboratory Animals.
Twelve male mongrel dogs (weight, 18–22 kg) were randomly assigned: 6 to the DES group and 6 to the BMS group. All 12 were given oral clopidogrel (25 mg/d) and aspirin (100 mg/d) 3 days before PCI.
The PCI procedure was performed as previously reported.30,31 On the day of the PCI, all 12 dogs were anticoagulated with 100 IU/kg of heparin after insertion of the catheters. In the DES group, a Cypher® DES (Cordis Corporation, a Johnson & Johnson company; Miami Lakes, Fla) was implanted in the left distal internal thoracic artery. In the BMS group, a Bx SONIC™ BMS (Cordis) was implanted in the left proximal internal thoracic artery. The stents were 2.25 mm in diameter and 18 to 28 mm long. The release pressure was 10 atm. Rapamycin-eluting stents were designed so that 80% of the rapamycin was eluted by 30 days.6,7 After the procedure, the animals were allowed to recover, were returned to their pens, and were given oral clopidogrel (25 mg/d) and aspirin (100 mg/d) for 1 month.
After 30 days, all of the stent-implanted dogs were deeply anesthetized, fully heparinized (100 U/kg), and euthanized. The injured vessels were perfusion-fixed at 100 mmHg in 1 L cold (4°C) 4% paraformaldehyde in 0.1 mol/L phosphate-buffered saline (PBS), pH 7.4, as previously reported.30,31 After perfusion fixation, the arteries of interest were excised and immersed in fresh fixative. The arteries were dissected vertically, and the stents were scanned by electron microscopy in order to evaluate the endothelial coverage and the sticking and activation of platelets on the stents.
Adult male Sprague-Dawley rats (weight, 85–100 g) were used for cell isolation, without grouping before interventions were performed on cells. The rats were anesthetized with 3% pentobarbital sodium. Their hearts were rapidly excised and placed in a dish that contained oxygenated Krebs-Henseleit bicarbonate buffer. After connective tissue was removed, the left ventricular tissue was excised and minced finely in 0.2% collagenase in calcium-free buffer, then incubated for 10 min at 37°C in a shaking-water bath. Trypsin (0.25%) was added, and the result was incubated for 3 min. Dissociated cells were filtered through a 100-μm mesh filter and washed with nominally calcium-free Krebs-Henseleit bicarbonate buffer, followed by filtering and washing in the same buffer containing calcium, then centrifugated at 1,000 × g for 10 min. Cells were resuspended in Dulbecco's modified Eagle's medium (DMEM) that was supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 IU/mL penicillin and 100 pg/mL streptomycin) at 15,000 cells/mL, then plated at 2.5 × 103 cells/cm2 on culture dishes that had been pretreated for 24 hr with rat-tail collagen. After 4 hr, the attached cells were washed with DMEM and cultured in DMEM with 10% FBS in a concentration of 5% CO2–95% O2 at 37°C.
The EPCs were cultured according to techniques described previously.32 Briefly, hollow femurs and tibias from Sprague-Dawley rats were prepared using standard surgical procedures, and whole bone marrow was harvested by flushing the marrow with Medium 199 (M199) using a sterilized syringe. Total mononuclear cells were isolated from a cell suspension containing marrow through Ficoll-density gradient centrifugation, and then were washed twice. Isolated cells were subsequently plated onto dishes coated with human fibronectin at 1 × 106 mononuclear cells/cm2 in M199 that was supplemented with 10% FBS, 10 ng/mL of recombinant rat vascular endothelial growth factor (VEGF), 10 ng/mL of recombinant human basic fibroblast growth factor, 100 U/mL of penicillin, and 100 μg/mL of streptomycin. On day 4, the adherent cells were washed extensively to remove unattached cells, fresh growth medium was added, and the culture was maintained for 7 days. All experiments were performed using 1st-passage cells. Day-7 EPCs were lifted with trypsinization, resuspended in M199 supplemented with 10% FBS, and then replated onto a culture plate for 24 hr before treatment with reagents (0.1, 1, 10, and 100 ng/mL of rapamycin).
Endothelial cells or EPCs were seeded on glass coverslips coated with collagen in 24-well plates and then were fixed with 4% paraformaldehyde. After incubation at 37°C for 30 min, the cells were treated with 0.1% Triton™ X-100 (Dow Chemical Co.; Midland, Mich) in PBS-15% for 5 min, and nonspecific binding sites were saturated with 10% FBS for 30 min. The cells were next incubated with different primary antibodies for 40 min in 1% FBS, and then with goat secondary antibodies. The coverslips were observed with a Leica TCS SP2 confocal microscope (Leica Microsystems; Wetzlar, Germany).
Cell proliferation was quantified by cell counts and methyl thiazolyl tetrazolium (MTT) assay. Briefly, cells were plated into 96-well microplates at 5 × 103 cells/well. When the cells reached 70% to 75% confluence, they were treated with rapamycin (0.1, 1, 10, and 100 ng/mL) or carrier (0.1% methanol) in reduced-serum media (DMEM containing 5% FBS). After incubation for 24 hr, 20 μL of the MTT reagent was added to each well, and the multiwell plates were incubated in a humidified atmosphere for 4 hr. The supernatant was then removed from each well, and 150 μL/well of dimethyl sulfoxide was added in order to solubilize the formed formazan salt crystals. The formazan product was spectrophotometrically quantified at 490 nm using an enzyme-linked immunosorbent assay (ELISA) reader. Data were expressed as a percentage of control.
Endothelial cells were seeded onto 6-cm gelatin-coated dishes and were allowed to form confluent monolayers. The cell monolayer was scraped with a 1-mL pipette tip in order to create a cell-free zone. Thereafter, cells were washed with PBS and were cultured with or without rapamycin (0.1, 1, 10, and 100 ng/mL). Each scratch was randomly photographed at 4 separate sites along the length of the scratch, starting proximally and ending distally. Photographs were taken on an inverted microscope immediately after the scratch was made, and then again at 24 hr. Cells that migrated across the wound edges were counted manually.
An ApoAlert® Annexin V apoptosis detection kit (BD Biosciences Clontech; Palo Alto, Calif) was used. Approximately 5 × 105 cells were resuspended in 200-μL 1 × binding buffer. Annexin V (5 μL) and propidium iodide (PI; 10 μL) was added. After 20 min of incubation at room temperature in the dark, the cells were analyzed by flow cytometry using a FACScan™ flow cytometer (BD Biosciences; San Jose, Calif). Treated cells were then evaluated by fluorescence-activated cell-sorter analysis for identifying cells at different stages of the cell cycle. Annexin V+/PI-indicated apoptotic cells.
Thirty Sprague-Dawley rats (weight, 100–150 g) were randomly and equally divided into 4 experimental groups and a control group. Each experimental group was intraperitoneally administered 0.05, 0.1, 0.15, or 0.2 mg/kg/d of rapamycin, for 5 days. The control group was intraperitoneally administered 10% methanol (0.1 mL/kg/d) over the same time course.33,34 Blood samples were collected immediately before the animals were killed. The VEGF levels in plasma were measured in triplicate or quadruplicate with an ELISA test according to the manufacturer's instructions. The optical density was measured at 490 nm.
Data were presented as mean ± SEM from at least 3 independent experiments. All values were processed by GraphPad Prism® 5.0 (GraphPad Software, Inc.; La Jolla, Calif), and analysis of variance was performed for serial analysis. A P value less than 0.05 was considered statistically significant.
All of the experimental dogs survived stent implantation. After 30 days, we saw smooth blood flow, no luminal loss, and no thrombosis in the in situ, proximal, or distal vessels. Electron microscopy showed endothelium covering the BMS, ECs arranged in rows, no stent exposure, and no significant platelet activation (Figs. 1A and 1B). However, we observed poor EC junction formation on the DES (Fig. 1C) and stent exposure to the vascular lumen (Figs. 1D and 1E). Platelets adhered to the endothelium, and some could be seen with lamellipodia, which indicated activation (Figs. 1C and 1F).
Fluorescent staining with derivative of acetylated low-density lipoprotein (DiI-Ac-LDL) and 4′6-diamidino-2-phenylindole-2HCl confirmed that the cells isolated from the rats were ECs (Fig. 2A).
Rapamycin inhibited cell migration into the scratch area at 24 hr (Fig. 3A). To quantitate EC migration into the scratch area, each scratch was divided into two 25% border regions and one 50% middle region. Treatment with 10 ng/mL or 100 ng/mL of rapamycin reduced EC proliferation to 71.5% ± 6.4% (P <0.01) and 57% ± 9% (P <0.001) of control values, respectively, after 24 hr. Longer incubations with low rapamycin levels (1 ng/mL) also decreased EC proliferation after 48 hr (55.3% ± 13.4% of control, P <0.01). After 72 hr, rapamycin effected clear reductions in cell proliferation (Fig. 2B).
Rapamycin dose-dependently inhibited EC migration into the 50% middle region, with no effect at 0.1 ng/mL, inhibition to 75.5% ± 7% at 1 ng/mL (P <0.05), 65.2% ± 4.3% at 10 ng/mL (P <0.001), and 39.5% ± 8% at 100 ng/mL (P <0.001) (Fig. 3B).
Endothelial progenitor cells are immature, bone marrow-derived cells that can transform into mature ECs and promote postnatal angiogenesis and vasculogenesis.35 We tested the ability of rapamycin to influence EPCs by double-staining with DiI-Ac-LDL and FITC-UEA-I (Fig. 4A). We performed MTT in order to measure proliferation, and we used Annexin V+/PI– staining and flow cytometry to measure apoptosis. Rapamycin dose-dependently inhibited EPC proliferation, with 0.1 ng/mL producing 98.7% ± 8.9% (P >0.05) of control-cell growth, 1 ng/mL producing 84.9% ± 6.7% (P <0.001), 10 ng/mL producing 77.2% ± 6.1% (P <0.001), and 100 ng/mL producing 69% ± 4.1% (P <0.001) (Fig. 4B). Rapamycin also induced EPC apoptosis. The normal apoptotic index in EPCs was 4.2% ± 0.5%, and all higher rapamycin levels increased this index, with 1, 10, and 100 ng/mL producing apoptotic levels of 10.8% ± 0.8%, 14.4% ± 1%, and 18.6% ± 1.2%, respectively (all P <0.001; Fig. 4C).
A 5-day treatment with rapamycin decreased VEGF expression at higher doses (0.15 and 0.2 mg/kg/d; P <0.05 and P <0.01, respectively) but not at lower doses (0.05 and 0.1 mg/kg/d) (Fig. 5).
Our study shows that rapamycin inhibits re-endothelialization by reducing EC proliferation and migration, inhibiting EPC proliferation and inducing apoptosis, and reducing VEGF expression. The rapamycin levels used in our in vivo experiments were comparable to or even higher than those that have been used in patients in clinical settings. Maximal systemic concentrations of rapamycin after deployment of rapamycin-eluting stents are 1 ng/mL (≈1.15 × 10–9 mol/L).36 Local concentrations, although difficult to evaluate, are likely to be significantly higher, partly because of rapamycin's lipophilic properties that promote accumulation in the vascular wall.10,37–39 Therefore, the concentrations used in our study may be relevant for patients who are treated with DESs.
Preclinical studies of rapamycin-eluting stents show a range of biological effects on arterial wall healing, inflammation, and neointimal growth. Studies of 28 days38,39 or longer40 showed significant suppression of neointimal growth with rapamycin-eluting stents compared with polymer-coated stents or BMSs. Other preclinical studies that used systemic delivery of paclitaxel or everolimus showed delayed healing, with significantly decreased endothelialization at 28 days.41,42 However, we found incomplete re-endothelialization in DES sites in comparison with BMS sites, with increases in fibrin and microthrombosis.
Endothelial cells are mature cells that have detached from the intimal monolayer of blood vessels in response to endothelial injury.43 Disruption of the endothelial layer contributes to postangioplasty restenosis and graft failure.44 Restoration of an intact endothelium is therefore crucial in reestablishing an antithrombotic vascular surface. Endothelial cells proliferate and migrate from intact neighboring coronary segments, eventually leading to the re-endothelialization of the injured segment. In vitro, rapamycin and paclitaxel inhibit proliferation and migration of vascular smooth-muscle cells and suppress ECs,36,45–50 thereby potentially impeding re-endothelialization. We observed poor EC junction formation and microthrombi of focal platelet aggregation at 1 month after rapamycin-stent implantation.
The growth of new microvessels (or angiogenesis) and the re-endothelialization of impaired endothelium involve EPCs.31,51,52 Endothelial progenitor cells are immature, bone marrow-derived cells with the capacity to transform into mature ECs and promote postnatal angiogenesis and vasculogenesis.35 Of note, rapamycin inhibits proliferation and migration of human EPCs in vitro53,54 via interaction with the mammalian target of rapamycin and, at least in part, by induction of apoptosis. Therefore, in contrast with BMSs, implantation of a rapamycin-eluting stent leads to a decrease in circulating CD34+ cells.55
The angiogenic growth factor VEGF may induce migration, proliferation, and tube formation in EPCs.56 Rapamycin reduces VEGF secretion by tumor cells,48 perhaps through down-regulation of the hypoxia-inducible factor-1α protein, which is upstream of VEGF expression.57 However, because exogenous VEGF was present in the M119 during the proliferation and differentiation of EPCs, this mechanism cannot be responsible for our findings. Second, VEGF-receptor signaling may be disrupted by rapamycin, because it can inhibit VEGF-driven proliferation and tube formation in human umbilical vein ECs.58 This mechanism might prevent proliferation and differentiation of circulating EPCs and block neo-angiogenesis from bone marrow-derived cells. Further studies into the mechanism of action of rapamycin alone or in combination with different antiangiogenic or cytotoxic drugs are warranted.
Drs. F. Li and H.-T. Liu contributed equally to this paper.
This work was supported by the National Natural Science Foundation of China (No. 30770784).
Address for reprints: Hai-Chang Wang, MD, PhD, Department of Cardiology, Xijing Hospital, Fourth Military Medical University, 15 Changlexi St., Xi'an, Shaanxi 710032, PRC