Baicalin, BLT-1, GW9662, rosiglitazone, geranylgeranyl pyrophosphate (GGPP) and GW3965 were purchased from Sigma-Aldrich (Merck Millipore, Darmstadt, Germany). Rabbit polyclonal antibodies to GAPDH (cat. no. ab9485), SR-BI (cat. no. ab106572), PPAR-γ (cat. no. ab45036) and liver X receptor-α (LXRα; cat. no. ab3585) were purchased from Abcam (Cambridge, UK). SR-BI small interfering RNA (siRNA) was from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA).

THP-1 cells were purchased from American Type Culture Collection (Manassas, VA, USA). THP-1 cells were maintained in RPMI-1640 (Hyclone; GE Healthcare Life Sciences, Logan, UT, USA) supplemented with 10% fetal bovine serum (Hyclone), streptomycin (100 mg/ml) and penicillin (100 U/ml) at 37°C in a humidified atmosphere with 5% CO₂. In order to stimulate the differentiation of cells into macrophages, THP-1 cells in 6-well culture dishes were treated with 160 nmol/ml phorbol 12-myristate 13-acetate for 72 h.

Following differentiation of the cells into macrophages, 50 mg/l oxidized low-density lipoprotein (oxLDL; Biomedical Technologies, Inc., Stoughton, MA, USA) and (³H)-cholesterol (1.0 µCi/ml; New England Nuclear Corp., Boston, MA, USA) were added to the THP-1 cell medium for another 24 h. The cholesterol-loaded macrophages were then exposed to baicalin of different concentrations (0, 2, 10 and 50 µM) for 48 h, or at a concentration of 50 µM for different times (0, 6, 12, 24 and 48 h) in the presence of ApoA-1 (10 µg/ml; Sigma-Aldrich), HDL subfraction 2 (HDL₂; 50 µg/ml) or HDL subfraction 3 (HDL₃; 50 µg/ml) (both purchased from USBiological, Swampscott, MA, USA. To confirm the effect of the PPAR-γ signaling pathway, the cells were pre-treated with PPAR-γ antagonist GW9662 (20 µM) for 2 h before stimulation with baicalin, or treated with PPAR-γ agonist rosiglitazone (1 µM) for 12 h. To confirm the effect of the LXRα signaling pathway, the cells were pre-treated with LXRα antagonist GGPP (5 µM) for 2 h before stimulation by baicalin, or treated with LXRα agonist GW3965 (5 µM) for 12 h. For analysis of cholesterol efflux, the medium and the THP-1 macrophages were collected, respectively, and the radioactive content was determined by liquid scintillation (Beckman LS6500; Beckman Coulter, Inc., Fullerton, CA, USA). The percentage of cholesterol efflux was calculated by dividing the radioactive content in the medium by the sum of the radioactive content in the medium and in the cells (17).

Radioimmunoprecipitation assay buffer and phenylmethanesulfonyl fluoride (Beyotime Institute of Biotechnology, Haimen, China) were used to extract the protein from the THP-1 macrophages following the various treatments. The bicinchoninic acid method was used to measure the protein concentration of the lysates. Moderate quantities (5–10 µl) of protein lysates were separated using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis at 90 V. The proteins were transferred to nitrocellulose membranes (EMD Millipore, Billerica, MA, USA), which were then blocked with 5% non-fat milk for 1 h at room temperature and incubated with primary antibodies against GAPDH (1:1,000), SR-BI (1:500), PPAR-γ (1:1,000) and LRXα (1:500) overnight at 4°C. Finally, following incubation with goat anti-rabbit secondary antibody (cat. no. ZDE-5209; for GAPDH, 1:10,000; for SR-BI, 1:2,000; for PPAR-γ, 1:5,000; for LXR, 1:1,000; ZSGB-Bio, Beijing, China) for an additional 2 h at room temperature, the antigen-antibody complex was detected using an electrochemiluminescence detection system (Immobilon Western Chemiluminescent HRP Substrate; EMD Millipore). The blots were quantified using Adobe Photoshop CS5 (Adobe Systems, Inc., San Jose, CA, USA)

Following treatment, total RNA was extracted from the THP-1 macrophages using TranZol Up (Transgen Biotech, Beijing, China). Then, RNA samples were treated with RNase-free DNase (Transgen Biotech). Following extraction, 1 µg mRNA was reversely transcribed using a First Strand cDNA Synthesis kit (Fermentas; Thermo Fisher Scientific, Inc., Waltham, MA, USA). After that, total cDNA was amplified using the FastStart Universal SYBR Green Master (ROX) mix (Roche Diagnostics, Basel, Switzerland) in the Light Cycler real-time PCR detection system (Roche Diagnostics) for 40 cycles at 95°C for 10 sec, 60°C for 20 sec and 72°C for 30 sec. 18S was chosen as the reference gene and the primer sequences for qPCR analyses were as follows: 18S, forward primer: 5′-CTTAGTTGGTGGAGCGATTTG-3′, reverse primer: 5′-GCTGAACGCCACTTGTCC-3′ (17). SR-BI, forward primer 5′-TCCTCACTTCCTCAACGCTG-3′ and reverse primer 5′-TCCCAGTTTGTCCAATGCC-3′ (18). Melting curves were assessed to confirm the specificity of the products generated for each set of primers. The 2^−ΔΔCq comparative method was then used to normalize the relative levels of gene expression (19).

SPSS statistical analytical software, version 18.0 (SPSS Inc., Chicago, IL, USA) was used to analyze the data. The normally distributed data were analyzed by one-way analysis of variance and the nonparametric variables were analyzed by Mann-Whitney U test. P<0.05 was considered to indicate a statistically significant difference.

To determine the effect that baicalin had on cholesterol efflux in THP-1 macrophages, THP-1 cells were first exposed to oxLDL and (³H) cholesterol for 24 h, and then the cholesterol-loaded THP-1 macrophages were cultured with different concentrations of baicalin (0, 2, 10 and 50 µM) for 48 h in the presence of ApoA-1, HDL₂ or HDL₃ in the medium (Fig. 1). As shown in Fig. 1A-C, baicalin at concentrations of 2 and 10 µM failed to promote cholesterol efflux to ApoA-1, HDL₂ and HDL_3. However, 50 µM baicalin significantly promoted cholesterol efflux to HDL₂ and HDL₃ by ~3.4- and 2.2-fold respectively compared with the control group (P<0.05), while this concentration of baicalin failed to promote cholesterol efflux to ApoA-1 (P>0.05). The results demonstrated that baicalin accelerated cholesterol efflux to HDL₂ and HDL₃, but not ApoA-1, when its concentration reached 50 µM.

Subsequently, 50 µM baicalin was used to stimulate the cholesterol-loaded THP-1 macrophages for various times (0, 6, 12, 24 and 48 h). As shown in Fig. 1E and F, incubation with baicalin for 6 h did not enhance cholesterol efflux to HDL₂ and HDL₃, while ≥12 h incubation significantly increased cholesterol efflux (P<0.05), which peaked at 48 h (P<0.01). However, as shown in Fig. 1D, incubation with baicalin did not increase cholesterol efflux to ApoA-1. These results indicate that baicalin is able to accelerate cholesterol efflux to HDL₂ and HDL₃, but not ApoA-1, in THP-1 macrophages in a concentration- and time-dependent manner.

SR-BI has been reported to be the major mediator of cellular cholesterol efflux (20). In the present study, in order to determine whether the effect of baicalin on cholesterol efflux in THP-1 macrophages was mediated by SR-BI, the expression of SR-BI was estimated. The cells were first treated with different concentrations (0, 2, 10 and 50 µM) of baicalin for 48 h. SR-BI was detected by western blotting and RT-qPCR at the protein and mRNA levels, respectively. As shown in Fig. 2A, following stimulation with 50 µM baicalin for 48 h, the expression of SR-BI was significantly increased compared with that in the control group (P<0.01). However, although 2 and 10 µM baicalin also slightly increased the expression of SR-BI, the increases were not statistically significant. Subsequently, baicalin (50 µM) was cultured with the cells for different time periods (0, 6, 12, 24 and 48 h). As shown in Fig. 2B, the relative expression of SR-BI was significantly increased when the stimulation time reached 12 h (P<0.05), and peaked at 24–48 h (P<0.01). RT-qPCR was used to examine the mRNA level of SR-BI after the treatment, and similar results were obtained; the expression of SR-BI at the mRNA level exhibited changes similar to those of the protein with different concentrations of baicalin (Fig. 2C) and different treatment durations (Fig. 2D). These data showed that baicalin induced SR-BI production at the protein and mRNA levels in a concentration- and time-dependent manner.

To further confirm the pivotal role of SR-BI in baicalin-induced cholesterol efflux, BLT-1 (a specific inhibitor of SR-BI) and SR-BI siRNA were each used to pre-treat the cells for 2 h before baicalin was added into the cells. As shown in Fig. 2E and F, BLT-1 and SR-BI siRNA significantly inhibited baicalin-induced cholesterol efflux compared with that in the baicalin group (P<0.05). Thus, all the results above suggest that baicalin enhanced cholesterol efflux in THP-1 macrophages via SR-BI.

PPAR-γ has been reported to promote cholesterol efflux and regulate the expression of SR-BI (19,21). In order to determine whether PPAR-γ is involved in the regulation of cholesterol efflux by baicalin, the expression of PPAR-γ in THP-1 macrophages following treatment with baicalin was examined. As shown in Fig. 3A, following stimulation with baicalin (50 µM) for 3 h, the expression of PPAR-γ significantly increased (P<0.05), and a peak was reached at the 12-h time point (P<0.01). An agonist and antagonist of PPAR-γ were then respectively applied to further clarify the role of PPAR-γ. As shown in Fig. 3B, pre-treatment with PPAR-γ antagonist GW9662 (20 µM) significantly inhibited the upregulated expression of SR-BI (P<0.01) stimulated by baicalin, while treatment with the PPAR-γ agonist rosiglitazone (1 µM) significantly increased the expression of SR-BI (P<0.01). Cholesterol efflux was then estimated following the treatment with GW9662 and rosiglitazone. As shown in Fig. 3C and D, cholesterol efflux was significantly inhibited following pre-incubation with PPAR-γ antagonist GW9662, but increased after incubation with PPAR-γ agonist rosiglitazone. Thus, these results suggest that baicalin-induced cholesterol efflux was mediated by PPAR-γ.

LXRα has been identified as a target gene of PPAR-γ. The binding of PPAR-γ with the peroxisome proliferator hormone response element (PPRE), which is located in the promoter region of LXRα, is reported to directly upregulate the expression of LXRα (11). PPAR-γ and LXRα are involved in a metabolic cascade that regulates the expression of downstream genes and increases cholesterol efflux (9,22). Thus, in the present study, the expression of LXRα in THP-1 macrophages was detected following the treatment with baicalin.

As shown in Fig. 4A, although the expression of LXRα only increased slightly after 3 h stimulation with baicalin (P>0.05), it significantly increased after stimulation with baicalin (50 µM) for 6 h (P<0.05), and reached a peak at the 12-h time point (P<0.01). Subsequently, an agonist and antagonist of LXRα were also respectively applied to further confirm the role of LXRα. As shown in Fig. 4B, pre-treatment with the LXRα antagonist GGPP (5 µM) significantly inhibited the upregulating effect of baicalin on the expression of SR-BI (P<0.01), while LXRα agonist GW3965 (5 µM) significantly increased the expression of SR-BI (P<0.01). Finally, the effects of GGPP pre-treatment and GW3965 treatment on cholesterol efflux were also determined. As shown in Fig. 4C and D, cholesterol efflux was also significantly inhibited following pre-incubation with GGPP, but elevated following incubation with GW3965. The results above indicate that the PPAR-γ/LXRα pathway plays a pivotal role in cholesterol efflux, and baicalin-accelerated cholesterol efflux is mediated by the PPAR-γ/LXRα pathway in THP-1 macrophages.