F-127 pluronic gel was purchased from Sigma-Aldrich; Merck Millipore (Darmstadt, Germany). Honokiol was purchased from Selleck Chemicals (Houston, TX, USA). 3F Fogarty embolectomy catheters were obtained from Edwards Lifesciences (Irvine, CA, USA). The primer sequences used in mRNA analysis are presented in Table I. The primary antibodies used in western blotting analysis were: Smad2/3 (Santa Cruz Biotechnology, Inc., Dallas, TX, USA; catalog no. sc-6032), phosphorylated-Smad2/3 (p-Smad2/3; Santa Cruz Biotechnology, Inc.; catalog no. sc-11769), transforming growth factor-β1 (TGF-β1; Novus Biologicals, LLC, Littleton, CO, USA; catalog no. MAB240), TGF-β receptor type I (TGF-βRI; Santa Cruz Biotechnology, Inc.; catalog no. sc-398-G), TGF-βRII (Santa Cruz Biotechnology, Inc.; catalog no. sc-17792) and GAPDH (Abcam, Cambridge, UK; catalog no. ab8245). Mouse monoclonal α-smooth muscle actin (α-SMA; Abcam; catalog no. ab7817) and rabbit monoclonal anti-proliferating cell nuclear antigen (PCNA; Abcam; catalog no. ab19166) were used for immunohistochemistry analysis. The secondary antibodies were: Mouse anti-goat IgG horse radish peroxidase (HRP) (Santa Cruz Biotechnology, Inc.; catalog no. sc-2354) (for Smad2/3, p-Smad2/3, TGF-βRI), goat anti-mouse IgG HRP (Abcam; catalog no. ab205719) (for TGF-β1, TGF-βRII, GAPDH and a-SMA), goat anti-rabbit IgG HRP (Abcam; catalog no. ab205718) (for PCNA). Remaining solvents and reagents were analytical grade.

F-127 pluronic gel (30% w/v) in saline was prepared and stored at −20°C for future use. Honokiol (10 mg) was dissolved in 1 ml dimethylsulphoxide (DMSO) and honokiol-containing pluronic gel was prepared by adding 1 ml of this solution to 9 ml of the previously prepared 30% pluronic gel at 4°C. Therefore, 1 ml of this solution contained 1 mg of honokiol. The resulting honokiol-containing pluronic gel was stored at −20°C for future use. It is of note that DMSO was used for the dilution because honokiol is insoluble in water and has a relatively poor solubility in ethanol.

Male New Zealand White rabbits (3.0–3.5 kg, n=24, aged 4–5 months) were obtained from the Laboratory Animal Center of The Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine (Shanghai, China), animals were housed at a temperature of 23±1°C and a humidity of 50±20% for 1 week with free access to food and water. General anesthesia was induced by intravenous injection of 3% pentobarbital sodium (1 ml/kg). A midline neck incision was made and the left carotid artery (left common carotid artery, left external carotid artery, left internal carotid artery) was exposed. Subsequently, a permanent ligature was made on the left external carotid artery ~5 mm away from the bifurcation, the left internal carotid artery was ligated by a bulldog clamp, another bulldog clamp was also placed on the proximal end of left common carotid artery. A 3F Fogarty embolectomy catheter was then introduced into the left common carotid artery through left external carotid artery to establish the animal model of vascular balloon injury. The balloon catheter was inflated with ~0.1 ml saline solution three times to denude the aorta endothelium, the length was about 3 cm. Immediately following balloon injury, 1 ml Pluronic gel containing honokiol or DMSO was applied around the injured carotid artery. Rabbits were randomly divided into three experimental groups: i) Honokiol treated group (1 mg, n=8); ii) vehicle (equal quantity of Pluronic gel containing DMSO) treated group (vehicle alone, n=8); and iii) untreated group (balloon injury without any treatment, n=8). The left external carotid artery was ligated immediately after treatment, the bulldog clamps were removed and the incision was closed. All animals recovered after 40–60 min surgery and showed no symptoms of stroke. Animals were euthanized and sacrificed in order to collect the carotid arteries 14 days post-surgery, three segments were cut from each artery, one segment was fixed in 10% buffered formalin and the remaining two segments were stored in liquid nitrogen. The protocols were approved by the Animal Care and Use Committee of Shanghai Jiao Tong University School of Medicine.

Carotid arteries were fixed in formalin fixative solution for 24 h, then dehydrated in graded ethanol and subsequently embedded in paraffin. Transverse 5 µm slices were cut and then stained with hematoxylin and eosin (H&E) (3 min for hematoxylin and 30 sec for eosin at room temperature) and Masson (5–10 min at room temperature) following the manufacturer's protocol. The cross-sectional intima area, media area and intima-to-media (I/M) area ratio were calculated from the H&E stain slices, collagen fibers were calculated on Masson staining slices (photographed by a Nikon light microscope). All the data was analyzed and calculated using the Image-Pro Plus version 6.0 (Media Cybernetics, Inc., Rockville, MD, USA).

Paraffin-embedded tissues were cut into 5-µm thick slices, the sections were incubated with primary antibodies (α-SMA, 1:2,000 dilution; and PCNA, 1:2,000 dilution) at 4°C overnight and then stained with a labeled DAB Detection kit (Fuzhou Maixin Biotech Co., Ltd., Fuzhou, China), followed by counterstaining with hematoxylin (30 sec at room temperature). Negative control was treated with PBS. A total of 5 sections of each carotid artery were selected for quantification, and 4 random fields were imaged at magnification of ×200 in each section. The number of positively stained cells was quantified using Image-Pro PLUS version 6.0 (Media Cybernetics, Inc.).

TRIzol reagent (Molecular Research Center, Cincinnati, OH, USA) was used to isolate total RNA from the tissues according to the manufacturer's protocol. cDNA was synthesized from 1 µg of total RNA using iScript cDNA Synthesis kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at 42°C for 60 min, 70°C for 10 min and 4°C for storage). qPCR was performed using SYBR Premix Ex Taq kit (Takara, Bio, Inc., Otsu, Japan) on a Stratagene Mx3005 RT-qPCR thermocycler (30 sec at 95°C, followed by 50 cycles of denaturation at 95°C for 5 sec, annealing step at 60°C for 35 sec and extension at 72°C for 15 sec). The mRNA expression levels of samples were measured using the 2^−ΔΔCq method (14). The primer sequences used in the current study are presented in Table I.

Carotid arteries were homogenized in radioimmunoprecipitation assay (RIPA) lysis buffer, the protein concentration was determined using Bicinchoninic acid Protein Assay reagent (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Equal quantity protein lysates (30 µg) was separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to polyvinylidene difluoride membrane (EMD Millipore, Billerica, MA, USA). Membranes were blocked with 5% non-fat milk for 1 h and subsequently incubated with primary antibodies (Smad2/3 1:1,000 dilution; p-Smad2/3 1:1,000 dilution; TGF-β1 1:1,000 dilution; TGF-βRI 1:200 dilution; TGF-βRII 1:200 dilution; GAPDH 1:1,000 dilution) at 4°C overnight, followed by incubation with a secondary antibody for 1 h (at room temperature, 1:200 dilution for all). Protein bands were detected with enhanced chemiluminescence. Protein bands were determined by Quantity One version 4.62 software (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and each band was determined three times.

Data are presented as the mean ± standard deviation and were analyzed by one-way analysis of variance with the SPSS version 22.0.0.0 (IBM Corporation, Armonk, NY, USA). P<0.05 was considered to indicate statistically significant difference.

H&E staining was used to assess the effect of honokiol on intimal thickening following vascular injury, as presented in Fig. 1B. Morphometric analysis of the three treatment groups revealed that the intima area was significantly reduced in the honokiol treatment group when compared with the vehicle and untreated groups (Fig. 1C), the I/M area ratios followed a similar pattern (Fig. 1C). No significant difference was identified between intimal area and I/M area ratios in the vehicle and the untreated groups. These findings suggested that honokiol was effective in inhibiting intimal thickening and the vehicle treatment had no effect on intimal hyperplasia.

Immunohistochemistry staining for α-SMA and PCNA was performed to investigate whether honokiol reduced neointima formation by inhibiting proliferation of VSMCs. It was determined that the majority of neointimal cells were positive for α-SMA, indicating that the cells that proliferated in the intima were VSMCs (Fig. 2A). Additionally, intimal proliferation was observed with immunohistochemistry staining of PCNA (a marker for cell proliferation) and presented in Fig. 2B, the ratio of PCNA-positive cells was significantly reduced in the honokiol-treated group in comparison with vehicle and untreated groups (Fig. 2C). These findings indicated that local honokiol treatment significantly inhibited intimal VSMCs proliferation.

Collagen contents in the arteries 14 days after vascular injury were assessed with Masson staining (Fig. 3A), as presented in Fig. 3B, honokiol treatment group had a significantly reduced the collagen content compared with the vehicle and untreated groups, the vehicle treatment group exhibited similar collagen deposition compared with the untreated group.

MMPs and TIMP-1 have essential roles in intimal thickening following vascular injury; therefore, the expression of MMPs and TIMP-1 was investigated using RT-qPCR analysis. As presented in Fig. 4, there was a significant increase in the MMP-1, MMP-2 and MMP-9 mRNA expression levels when the honokiol-treated group was compared with the vehicle and untreated groups. TIMP-1 mRNA expression was significantly reduced in the honokiol treated group compared with the vehicle and untreated groups. No significant difference was identified between the vehicle-treated group and untreated group in terms of the expression of MMPs and TIMP-1.

TGF-β1/Smad2/3 signaling is a critical regulator in the pathological process of intimal hyperplasia and inhibition of the TGF-β1/Smad2/3 signaling pathway may reduce VSMC proliferation and collagen deposition, thus attenuating intimal hyperplasia (15–17). Therefore, the present study investigated whether the reduced VSMC proliferation and collagen deposition were mediated via TGF-β1/Smad2/3 pathway. It is of note that western blot analysis revealed that honokiol treatment significantly reduced the expression of p-Smad2/3, whereas TGF-β1 and Smad2/3 expression showed no obvious change in the honokiol treatment group compared with vehicle or untreated groups (Fig. 5). TGF-βRII phosphorylates TGF-βRI and then TGF-βRI in turn phosphorylates Smad2/3 (18); therefore, the expression of TGF-βRI and TGF-βRII was subsequently analyzed. However, no significant difference was identified between the honokiol, vehicle or untreated groups (Fig. 5).